3ULFH5HJXODWLRQRI

$LUSRUW6HUYLFHV ,QTXLU\5HSRUW
5HSRUW1R
-DQXDU\
 Commonwealth of Australia 2002

ISBN 1 74037 075 9

This work is subject to copyright. Apart from any use as permitted under the
Copyright Act 1968, the work may be reproduced in whole or in part for study or
training purposes, subject to the inclusion of an acknowledgment of the source.
Reproduction for commercial use or sale requires prior written permission from Info
Products. Requests and inquiries concerning reproduction and rights should be
addressed to the Manager, Legislative Services, Info Products, Department of
Finance and Administration, GPO Box 1920, Canberra, ACT, 2601.

Publications Inquiries:
Media and Publications
Productivity Commission
Locked Bag 2 Collins Street East
Melbourne VIC 8003
Tel: (03) 9653 2244
Fax: (03) 9653 2303
Email: maps@pc.gov.au

General Inquiries:
Tel: (03) 9653 2100 or (02) 6240 3200

An appropriate citation for this paper is:

Productivity Commission 2002, Price Regulation of Airport Services, Report
no. 19, AusInfo, Canberra.

The Productivity Commission

The Productivity Commission, an independent Commonwealth agency, is the
Government’s principal review and advisory body on microeconomic policy and
regulation. It conducts public inquiries and research into a broad range of economic
and social issues affecting the welfare of Australians.
The Commission’s independence is underpinned by an Act of Parliament. Its
processes and outputs are open to public scrutiny and are driven by concern for the
wellbeing of the community as a whole.
Information on the Productivity Commission, its publications and its current work
program can be found on the World Wide Web at www.pc.gov.au or by contacting
Media and Publications on (03) 9653 2244.
 0HOERXUQH2IILFH
/HYHO&ROOLQV6WUHHW
0HOERXUQH9,&
/RFNHG%DJ&ROOLQV6WUHHW(DVW
0HOERXUQH9,&
7HOHSKRQH 
)DFVLPLOH 
&DQEHUUD2IILFH
7HOHSKRQH 

23 January 2002

The Honourable Peter Costello MP

Treasurer
Parliament House
CANBERRA ACT 2600

Dear Treasurer

In accordance with Section 11 of the Productivity Commission Act 1998, we have pleasure
in submitting to you the Commission’s final report on Price Regulation of Airport
Services.

Yours sincerely

Gary Banks Richard Snape Neil Byron

Chairman Deputy Chairman Commissioner
Terms of reference
I, ROD KEMP, Assistant Treasurer, pursuant to Parts 2 and 3 of the Productivity
Commission Act 1998, hereby refer Prices Regulation of Airports to the
Commission for inquiry and report within 12 months of receipt of this reference.
The Commission is to hold hearings for the purpose of the inquiry.

Background

2. During 1997 and 1998, long-term leases (50 years with an option to renew for a
further 49 years) were sold over seventeen airports previously operated by the
Federal Airports Corporation, to private sector operators. In July 1998 two wholly
Australian Government-owned companies were formed to acquire leases over the
four Sydney basin airports (Sydney, Bankstown, Camden, Hoxton Park) and
Essendon airport.

3. All 22 of the leased airports are regulated under the Airports Act 1996 and
twelve of the airports (Adelaide, Alice Springs, Brisbane, Canberra, Coolangatta,
Darwin, Hobart, Launceston, Melbourne, Perth, Sydney and Townsville) are
currently subject to prices regulation under the Prices Surveillance Act 1983.

4. Except for Sydney Airport, the prices regulation of the airports comprises a
CPI-X price cap on declared aeronautical services, prices monitoring of aeronautical
related services, and special provisions for necessary new investment at airports. At
Sydney Airport aeronautical services are subject to prices surveillance and
aeronautical related services are subject to price monitoring. The arrangements are
administered by the Australian Competition and Consumer Commission.

5. In establishing the current prices regulations in 1997, the Government

announced that the arrangements would only apply for the first five years of
operation of the lease. Subsequent regulation would be determined after a review of
the arrangements before the end of the first five year period, with the review to be
based on the premise that the price caps applied to aeronautical services will no
longer operate.

6. The purpose of this inquiry is to examine whether new regulatory arrangements,

targeted at those charges for airport services or products where the airport operator
has been identified as having most potential to abuse market power, are needed to
ensure that the exercise of any such power may be appropriately counteracted.

IV TERMS OF
REFERENCE
Scope of Inquiry

7. The Commission is to report on whether there is a need for prices regulation of

airports, and the appropriate form of any prices regulation, taking into account the
following principles:
(a) the CPI-X price cap applied to aeronautical charges during the first five year
period of private ownership will no longer operate;
(b) future prices regulation should be applied to those aeronautical services and
those airports where airport operators have most potential to abuse market
power;
(c) airport operators may propose to the inquiry alternative approaches to prices
regulation which would provide equal or better protection to users;
(d) prices regulation should minimise compliance costs on airport operators and
the Government;
(e) prices regulation should promote the efficient operation of airports;
(f) prices regulation should facilitate benchmarking comparisons between
airports, competition in the provision of services within airports (especially
protecting against discrimination in relation to small users and new entrants),
and commercially negotiated outcomes in airport operations; and
(g) the Commission may recommend more effective forms of prices regulation
than are currently in place, where this may be necessary.

8. In making its recommendations, the Commission is to:

(a) review the operation of the existing prices regulation of airports;
(b) identify the rationale for any future prices regulation at airports;
(c) identify relevant alternatives to the current arrangements, including the prices
oversight arrangements at relevant airports in other countries, and the extent
to which these alternatives would achieve the rationale in (b);
(d) analyse and, as far as practical, quantify the benefits, costs and economic and
distributional impacts of the current arrangements and the alternatives
identified in (c);
(e) identify the different groups, including the travelling public, likely to be
affected by the current prices oversight arrangements and the alternatives
identified in (c); and
(f) list the individuals and groups consulted during the review and outline their
views.

TERMS OF V
REFERENCE
9. In undertaking the review, the Commission is to advertise nationally, consult
with key interest groups and affected parties, and produce a report.

10. The Government will consider the Commission’s recommendations, and the
Government’s response will be announced as soon as possible after the receipt of
the Commission’s report.

ROD KEMP
21 December 2000

VI TERMS OF
REFERENCE
Contents

Terms of reference IV

Abbreviations and explanations XI

Glossary XIII

Key messages XVI

Overview XVII

Recommendations and findings XLVI

Recommendations XLVI
Findings XLVIII

1 Introduction 1
1.1 Terms of reference 1
1.2 Background to the current inquiry 2
1.3 The Commission’s approach 4
1.4 Conduct of the inquiry 6

2 Australian airports and their markets 9

2.1 Airport facilities and services 9
2.2 Australian airports 11
2.3 The airport business at core-regulated airports 15
2.4 The airport business at other Australian airports 42

3 The regulatory environment 45

3.1 Price regulation until October 2001 45
3.2 Price regulation since October 2001 54
3.3 Quality of service monitoring 56
3.4 Access regulation 58
3.5 Other relevant regulation for core-regulated airports 64
3.6 Regulation of airports not subject to price regulation 76

CONTENTS VII
4 Assessing the need for price regulation of airport services: some
preliminary issues 79
4.1 Why regulate prices of airport services? 79
4.2 Potential efficiency and distributional effects of market power 82
4.3 Principles of efficient pricing of airport services 87
4.4 Principles of good regulation 93

5 Market power of airports 95

5.1 Introduction 95
5.2 Airports and barriers to entry 97
5.3 Price elasticity of demand for an airport’s services 106
5.4 Summing up 143

6 Market power in particular airport services 145

6.1 Introduction 145
6.2 Aircraft movement facilities 146
6.3 Passenger processing facilities 147
6.4 Lounge space (VIP and business) 152
6.5 Landside vehicle facilities 154
6.6 Aircraft refuelling 166
6.7 Maintenance facilities 170
6.8 Flight catering facilities 172
6.9 Freight handling and ground service equipment storage sites 173
6.10 Freight facility sites and buildings 174
6.11 Waste disposal facilities 175
6.12 Administrative office space 175
6.13 Commercial and retail activities 177
6.14 Summary 179

7 Conduct of unregulated airports 181

7.1 Influences on unregulated airport pricing 181
7.2 Conduct of unregulated airports: efficiency and distributional
effects 204

8 Review of airport price regulation: price-cap and prices-

notification arrangements 219
8.1 Introduction 219

VIII CONTENTS
 8.2 The price cap: CPI-X and prices notification of aeronautical
services 221
8.3 The price cap: allowable price increases above the cap 231
8.4 Recent changes to price regulation 246
8.5 Price regulation of Sydney Airport services 248

9 Review of airport regulation: price monitoring, quality monitoring

and access regulation 261
9.1 Price monitoring of airport services 261
9.2 Monitoring of quality of service provision 265
9.3 Access regulation 271

10 Regulatory options: cost-based and incentive regulation 281

10.1 Cost-based and incentive regulation 281
10.2 Future regulation: price caps and Australian airports 293
10.3 Coverage of price caps and cost-based regulation 309

11 Regulatory options: price monitoring, access provisions and

general competition law 315
11.1 Price monitoring 315
11.2 Access provisions and general competition law 329

12 Appropriate future regulation: the Commission’s assessment 343

12.1 The Commission’s task 343
12.2 The Commission’s assessment of the need for regulation 344
12.3 Option A: dual-till price caps 345
12.4 Option B: price monitoring 350
12.5 Weighing the options 354

A Terms of reference: correspondence 363

B Public consultation 365

B.1 List of submissions 365
B.2 Visits 366
B.3 Public hearing participants 367

C Aeronautical and non-aeronautical services 369

C.1 The effect of non-aeronautical profits on aeronautical charges 369
C.2 Non-aeronautical profits: locational rents and/or market power? 372

CONTENTS IX
 C.3 Arguments for and against a single till 373

D Characteristics of demand and competition at core-regulated

airports 383

E Fuel throughput levies and taxi charges 391

E.1 Fuel throughput levy 391
E.2 Taxi charges 395
E.3 Issues arising from price regulation 398

F Valuation of aeronautical land and efficient land use 401

F.1 Efficient land use 404
F.2 Sydney Airport and the ACCC decision 407
F.3 Distributional issues 410

G Economic regulation of privatised airports: international

experiences 411
G.1 United Kingdom 411
G.2 New Zealand 422
G.3 Germany 427
G.4 Denmark 428
G.5 Summary of international experiences 429

H Issues in allocating scarce airport slots 431

H.1 Airports and congestion 431
H.2 Price or quantity rationing? 433
H.3 Assessing the options 440

References 445

X CONTENTS
Abbreviations and explanations

Abbreviations
AAA Australian Airports Association
AAL Adelaide Airport Limited
ACA Airport Coordination Australia
ACCC Australian Competition and Consumer Commission
ACTO Australian Cargo Terminal Operators
Airports Act Airports Act 1996
APAC Australia Pacific Airports Corporation
APAM Australia Pacific Airports (Melbourne)
APS Australian Protective Services
ASA Airservices Australia
BAC Brisbane Airport Corporation
BARA Board of Airline Representatives of Australia
BTE Bureau of Transport Economics
BTR Bureau of Tourism Research
CAA Civil Aviation Authority (UK)
CASA Civil Aviation Safety Authority
CBD Central business district
CC Commerce Commission (NZ)
CPA Competition Principles Agreement
CPI Consumer Price Index
DoTRD Department of Transport and Regional Development
DoTRS Department of Transport and Regional Services
FAC Federal Airports Corporation

ABBREVIATIONS XI
AND
EXPLANATIONS
GAA General aviation aircraft
GATS General Agreement on Trade in Services
IATA International Air Transport Association
IC Industry Commission
ICAO International Civil Aviation Organization
JUHI Joint user hydrant installation
MTAA Super Fund Motor Trades Association of Australia
Superannuation Fund
MTOW Maximum take-off weight
MUDT Multi-user domestic terminal
NCC National Competition Council
NEC National Electricity Code
NECG Network Economics Consulting Group
NNI Necessary new investment
NTTC Northern Territory Tourist Commission
OECD Organisation for Economic Co-operation and
Development
PC Productivity Commission
Pers. comm. Personal communication
PS Act Prices Surveillance Act 1983
PSA Prices Surveillance Authority
RPT Regular public transport
RRCB Revised Regulatory Cost Base
SACL Sydney Airports Corporation Limited
s. Section
ss Sections
Sub. Submission
Sydney Airport Sydney Kingsford Smith Airport
TP Act Trade Practices Act 1974
Trans. Transcripts

XII ABBREVIATIONS AND

EXPLANATIONS
TRL Transport Research Laboratory
VFR Visiting friends and relatives
VHCA Victorian Hire Car Association
WAC Westralia Airports Corporation
WACC Weighted average cost of capital

Explanations
Recommendations RECOMMENDATION

Recommendations in the body of the report are

highlighted using bold italics with a heading, as
this is.
Findings FINDING

Findings in the body of the report are paragraphs

highlighted using italics with a heading, as this is.

Glossary

Access The terms and conditions under which an airport owner will
undertaking provide access, as agreed with the Australian Competition
and Consumer Commission under Part IIIA of the Trade
Practices Act 1974.

Aeronautical Services provided by infrastructure that facilitates aircraft

services movements (eg runways), and passenger processing
facilities.

Common-user Terminals and associated infrastructure managed by the

terminals airport operator and used (potentially) by a number of
different airlines. All international terminals at core-
regulated airports are common-user terminals as are some
domestic terminals.

GLOSSARY XIII
Core-regulated Leased airports designated as such under the Airports Act
airports 1996. They comprise Adelaide, Alice Springs, Brisbane,
Canberra, Coolangatta, Darwin, Hobart, Launceston,
Melbourne, Perth, Sydney and Townsville airports.

Dual till An arrangement for setting airport charges whereby only the
costs and revenues of providing aeronautical services are
included in the assessment of allowable aeronautical prices.
In other words, aeronautical services are priced on a ‘stand-
alone’ basis, without regard to any net revenues from non-
aeronautical services.

Ground-handling Handling services directly associated with the aircraft

services including cleaning and catering, provision of power and fuel,
and loading and unloading of baggage and freight.

Load factor The number of passengers carried expressed as a percentage

of the number of seats available.

Locational rent Payments to land above opportunity cost that derive from its
locational advantages for a particular use. Locational rents
are consistent with efficient use of a scarce resource and do
not reflect an abuse of market power (which occurs where
scarcity is created by the supplier).

Non-aeronautical Services provided by or at airports that are not aeronautical

services services (eg freight facilities, car parking, and retail shops
and food outlets).

Passenger A passenger arriving or departing on a scheduled regular

movement public transport service.

Phase 1 airports Airports leased to private operators in 1997 — Brisbane,

Melbourne and Perth.

Phase 2 airports Airports leased to private operators in 1998 — Adelaide,

Alice Springs, Canberra, Coolangatta, Darwin, Hobart,
Launceston, Townsville, Mount Isa, Tennant Creek,
Archerfield, Jandakot, Moorabbin and Parafield.

Regional ring A feature of the Slot Management Scheme 1998 for Sydney
fence Airport which effectively creates a separate pool for regional
slots.
XIV GLOSSARY
Revenue Passengers paying any fare on a scheduled regular public
passengers transport service, including passengers travelling on tickets
acquired through frequent flyer programs.

Single till An arrangement for setting airport charges whereby all

airport revenues and costs are taken into account in setting
aeronautical prices. Allowable aeronautical prices are set on
a ‘residual basis’, after subtracting from total airport costs
the revenue derived from non-aeronautical activities.

Slot A permission for an aircraft movement.

Weight-based Airport charges for the use of airport services based on the
charges weight of the aircraft (usually maximum take-off weight).

Yield management Discriminatory pricing by airlines designed to maximise

revenue from each flight.

GLOSSARY XV
 Key messages
• Detailed analysis of the demand and supply characteristics of airport markets
reveals that only four airports — Sydney, Melbourne, Brisbane and Perth — have
substantial market power.
– Adelaide and, to a lesser extent, Canberra and Darwin, have moderate market
power.
– Any market power of other airports does not warrant special regulation.
• The airport services where market power is strongest include facilities for aircraft
movements (runways, taxiways, aprons) and ‘front-door’ vehicle access.
• The scope for airports with market power to use (or abuse) that power is
constrained by commercial pressures and opportunities, particularly the substantial
‘non-aeronautical’ income to be had from promoting airline passenger traffic.
• In these circumstances, because of the risks and potential costs of strict price
controls relative to more light-handed price regulation, such controls are judged not
to be required even at the four airports with substantial market power.
• The uncertain outlook for the aviation market since September 2001 also calls for a
more flexible approach, as the Government has recognised in removing price caps
from other key airports, including Adelaide, Canberra and Darwin.
• The Commission’s preferred approach is to put in place a light-handed regulatory
regime (additional to general competition law) in which all seven airports assessed
as having some market power would have their pricing and other behaviour
monitored for a ‘probationary’ five-year period.
– A review towards the end of the five years would assess the need for continued
regulation (if any) of these airports.
– Monitoring would promote more productive commercial relationships between
airports and airlines, while providing the discipline of the possible reintroduction
of stronger regulation after five years.
• If the Government nevertheless were to opt for a stricter form of price regulation,
CPI-X price caps are preferred because they can offer incentives for efficient airport
operation.
– Unlike the existing price caps, however, any new price caps should be explicitly
based on a ‘dual till’ and factor in anticipated investment.
– Conventional price caps would be confined to Melbourne, Brisbane and Perth
airports. For a capacity-constrained Sydney Airport, arrangements should not be
such as to force prices down.
• Under both options, the general access and anti-competitive conduct provisions of
the Trade Practices Act would apply to all airports.
• Whatever the regulatory framework decided for Sydney Airport, that policy, in
particular the pricing and investment provisions, should be clearly and publicly
articulated to bidders so that the sale price can adequately reflect it.

XVI KEY MESSAGES

Overview

Is price regulation of airport services required at privatised airports? If so, in what

form? The main purpose of this inquiry has been to address these questions,
including for the soon-to-be-privatised Sydney Airport.

Though the terms of reference encompass all Australian airports, the Commission
has mainly focused on the ‘core-regulated’ airports: that is, the eleven so-called
Phase 1 and Phase 2 privatised airports plus Sydney Airport (box 1). These
twelve airports account for 96 per cent of international, 93 per cent of domestic and
63 per cent of regional passenger movements.

Box 1 Price-regulated airports

Phase 1 airports are the three major capital city airports that were privatised in 1997:
Melbourne, Brisbane and Perth. Prices at each of these airports have been regulated.
Phase 2 airports comprise 14 smaller airports that were privatised in 1998. Only eight
of these have been subject to price regulation: Adelaide, Alice Springs, Canberra,
Coolangatta, Darwin, Hobart, Launceston and Townsville.
Sydney Airport, which was leased in 1997, but not privatised, also has been subject to
price regulation.
The twelve airports that have been subject to price regulation are designated as core-
regulated airports in the Airports Act 1996.

The Commission’s task

In making its recommendations, the Commission is required to take into account a
number of principles. Most of these were foreshadowed when the airport leases
were sold in 1997 and 1998. The thrust of the principles is that any future price
regulation should be targeted at those aeronautical services and airports where
operators are assessed as having most potential to abuse market power.
Complementary objectives include:
• fostering efficient operation of airports and commercially-negotiated outcomes;
• minimising compliance costs on all parties; and

OVERVIEW XVII
• promoting transparency and competition.

Box 2 Some key terms

Aeronautical services — services provided by infrastructure that facilitates aircraft
movements (eg runways), and passenger processing facilities.
Non-aeronautical services — services provided by or at airports that are not
aeronautical services (eg freight facilities, car parking, and retail shops and food
outlets).
Single till — an arrangement for setting airport charges whereby all airport revenues
and costs are taken into account in setting aeronautical prices. Allowable aeronautical
prices are set on a ‘residual basis’, after subtracting from total airport costs the revenue
derived from non-aeronautical activities.
Dual till — an arrangement for setting airport charges whereby only the costs and
revenues of providing aeronautical services are included in the assessment of
allowable aeronautical prices. In other words, aeronautical services are priced on a
‘stand-alone’ basis, without regard to any net revenues from non-aeronautical services.
Locational rent — payments to land above opportunity cost that derive from its
locational advantages for a particular use. Locational rents are consistent with efficient
use of a scarce resource and do not reflect an abuse of market power (which occurs
where scarcity is created by the supplier).

Thus the key tasks for the inquiry have been to:
• analyse the functions of airports and the markets in which they operate;
• identify the objectives of price regulation;
• identify which airports have significant market power and in which services;
• assess the potential for exploiting market power by those airports, including any
commercial constraints;
• identify and assess options for future price regulation, taking into consideration
the performance of existing price regulation; and, finally,
• determine the extent and type of future price regulation required, if any.

Background to the inquiry

Prior to 1997, most major airports were operated by the Federal Airports
Corporation (FAC). The FAC set its landing and terminal charges on a network-
wide, ‘single-till’ basis (box 2). More profitable airports subsidised unprofitable or

XVIII OVERVIEW
less profitable ones, and revenues from complementary (non-aeronautical)
activities, such as car parking and retailing, subsidised aeronautical services.

In 1997 and 1998, 50-year leases for 17 major Australian airports (excluding
Sydney) were sold to private operators. The stated rationale was to ‘improve the
efficiency of airport investment and operations in the interests of users and the
general community, and to facilitate innovative management’ (Harris 1997, p. 2).
Most leases were sold for individual airports.

Competition was to be encouraged by restricting cross-ownership of certain airport

pairs and limiting airline ownership to 5 per cent. In addition, the Government noted
that the single-till pricing approach that had been followed by the FAC would not
be mandated. Individual airport operators could retain profits, at their discretion,
from their non-aeronautical activities.

At the same time, however, there was a concern that privatised airports might abuse
their market power with respect to aeronautical services. Thus, privatisation was
accompanied by what were described as transitional price-regulation measures,
designed to allow all parties to adjust to the new operating environment for airports.

Price regulation comprised a five-year, CPI-X annual cap on prices for aeronautical
services at 11 of the largest privatised airports — Melbourne, Brisbane, Perth,
Adelaide, Alice Springs, Canberra, Coolangatta, Darwin, Hobart, Launceston and
Townsville. Essentially, these arrangements extended, for the first five years of
privatisation, the FAC aeronautical prices that were in place prior to privatisation.
The cap was complemented by cost pass-through provisions for ‘necessary new
investment’ and government-mandated security services as well as by quality
monitoring and by special access arrangements designed to facilitate new airline
entrants.

Though Sydney Airport was not privatised, its lease was transferred to the
government-owned Sydney Airports Corporation Limited (SACL). It has been
subject to prices notification under the Prices Surveillance Act 1983 (where any
price increases have to be vetted by the ACCC) but not to a price cap or any
requirement to reduce aeronautical charges each year.

Developments since the draft report

The Commission released its draft report on 24 August 2001. In the following
month, the Australian aviation market was rocked by two events: the terrorist
attacks in the United States and the cessation of services by Australia’s second-
largest carrier, Ansett. Immediately following these events, airports reported

OVERVIEW XIX
reductions in traffic of 30–40 per cent, though the level of services has recovered
somewhat since then. The effect of reduced domestic and international services on
several airports (especially the smaller Phase 2 airports) prompted the
Commonwealth Government on 5 October 2001 to announce major changes to
airport price regulation. It described these changes as ‘consistent with the findings
in the Productivity Commission’s draft report’.
• Price caps on aeronautical charges at the eight Phase 2 airports were removed.
They were replaced with price monitoring of aeronautical charges at Adelaide,
Darwin and Canberra airports while remaining Phase 2 airports were no longer
subject to any price regulation.
• Phase 1 airports (Melbourne, Brisbane and Perth) remained subject to price caps
but were allowed to implement one-off average price increases for price-capped
services of up to 6.2, 6.7 and 7.2 per cent, respectively.

When announcing the changes, the Minister for Financial Services and Regulation
said that ‘the Government will give further consideration to the prices oversight
arrangements at airports in the light of developments in airport prices and the
Productivity Commission’s final report’ (Hockey 2001).

Australian airports
Australian airports not only provide a range of services to airlines, they also provide
or facilitate ‘non-aeronautical’ services to passengers and others (box 3).

Box 3 Facilities and services provided by airports

Aircraft movement facilities include runways, taxiways and aprons.
Passenger processing facilities include aerobridges, baggage systems, check-in
counters, public areas in terminals, flight information displays and landside roads.
(Australian airports do not provide air traffic control services — these are the
responsibility of Airservices Australia.)
Most domestic terminals at the main Australian airports have been operated by the two
major domestic airlines under long-term lease arrangements (up to 30 years),
established well before airport privatisation. Common-user domestic terminals
constructed more recently are operated by the airports, as are international terminals.
Non-aeronautical services include restaurants, hotels, shopping and car parking.

Since privatisation, the proportion of airport revenue earned from non-aeronautical

activities has increased from about 62 per cent to around 69 per cent of total

XX OVERVIEW
revenue of the 11 privatised core-regulated airports. Sydney Airport earned about
60 per cent of its revenue from non-aeronautical activities in 2000-01.

Aeronautical charges

There are three main types of aeronautical charge levied by Australian airports.
• A landing charge that covers use of runways, taxiways and aprons. Landing
charges at airports still subject to price caps averaged around $6.30 per aircraft
tonne (GST inclusive) in January 2002.
• A terminal charge at airport-operated terminals (averaging about $4 per
international passenger at the larger international privatised airports).
• Additional charges sometimes are levied for aircraft parking ($50 per day at
Melbourne Airport, but around $10 per day is more common).

Aeronautical charges at Australia’s price-capped airports appear low by

international standards. The recent ACCC-vetted price increases for Sydney Airport
(of around 100 per cent) still leave that airport’s charges for international flights
below the average for major world airports, even though Sydney’s new charges are
now considerably higher than regulated charges at Melbourne, Brisbane and Perth
airports.

Since the removal of price caps for Phase 2 airports on 5 October 2001, list prices
have risen at most of those airports (Adelaide, Alice Springs, Canberra,
Coolangatta, Darwin and Townsville). The airport operators claim that this has been
necessitated by significantly lower traffic volumes, and consequently higher unit
costs, and increased security costs. At Hobart and Launceston airports, charges have
remained at levels that applied under price caps.

Charges for regular public transport (RPT) flights at other Australian airports often
are not publicly available because rates are negotiated with airlines. For those that
are available, comparisons are very difficult because the airports concerned, which
are mostly owned by local municipalities, usually supply terminal and other
services not supplied by core-regulated airports, and their traffic volumes typically
are much lower.

Why regulate airport prices?

The prima facie rationale for price regulation of airports (additional to the general
provisions of the Trade Practices Act 1974 (TP Act)) is their perceived market

OVERVIEW XXI
power — that is, an ability to raise prices above the efficient costs of supplying
aeronautical services — and their scope to use it.

Such pricing could increase airfares and reduce the consumption of air travel below
efficient levels. Airports that exercise market power could earn excessive profits or
operate inefficiently; they also could allow quality to deteriorate and not invest
appropriately. It is also possible that market power could be used selectively to deny
access to the airport, either by way of higher prices or by imposing other
unacceptable conditions of access.

Any abuse of market power would also have distributional effects, involving
transfers between airports, airlines and consumers. However, while such transfers
may be substantial in total, there do not appear to be major differences between
income profiles of the various groups. Moreover, attempting to use airport charges
to redistribute income would be a very blunt and inefficient approach. Hence, in
assessing the need for appropriate regulation, it has been the possible efficiency,
rather than distributional, outcomes that have guided the Commission’s assessment.

The existence of market power is not, of itself, sufficient justification for price
regulation of airports. Price regulation should promote more efficient outcomes than
those otherwise likely to be achieved by the market.

Which airports have market power?

The terms of reference indicate that regulation should be targeted at only those
airports and those aeronautical services where there is most potential to abuse
market power. This will depend on barriers to entry to the industry and the
sensitivity of users to price rises.

Barriers to entry are significant

In most Australian cities (and cities of similar size elsewhere) there is only one
airport catering for RPT services. This is because:
• basic airport infrastructure, such as runways and taxiways, must be used as a
package and thus requires a single provider of a large, lumpy investment;
• there is some evidence of economies of scale, at least in relation to runways and
taxiways, though economies of scale for the airport as a whole may be exhausted
at quite low levels of output;

XXII OVERVIEW
• most major aeronautical assets, other than land, typically cannot be used for
other purposes and thus, once invested, are ‘sunk’. This may deter new entrants
by raising the cost of failure; and
• airports require very large land allocations and buffer zones for environmental
and safety reasons and this may make duplication prohibitively costly (or, for
planning and regulatory reasons, impossible) in convenient locations.

Network or coordination benefits also accrue to airlines and their passengers from
using one airport. Airlines can coordinate aircraft and passenger transfers at lower
cost and passengers do not have the inconvenience of transferring between airports.
Even given the choice of two similar (uncongested) airports serving the same
destination, airlines would be unlikely to spread similar services across both.

Though these natural monopoly characteristics of airports arise from efficiency

benefits (one airport in a particular city being able to provide services at lower cost
than two), they inevitably also limit scope for direct competition in the provision of
airport services for that city.

The strength of demand for particular airports varies

Lack of effective competition between airports in a particular location is a necessary

but not sufficient condition for market power. The degree of market power also will
depend on the sensitivity of users to an increase in an airport’s charges. This, in
turn, depends on:
• the extent to which airfares rise if airport charges increase, which in turn
depends on the share of airport charges in the airfare and the supply responses of
other input providers to air services; and
• the sensitivity of passengers to changes in airfares to a particular destination and
airport (which depends on their willingness to substitute other destinations or
alternative transport and communication modes and the scope for using another
airport).

On average, charges for airport services account for a small share of airfares. In the
absence of any substitution possibilities, this suggests low price sensitivity and
strong market power. But, substitution possibilities for some airports may be quite
strong, so the sensitivity of some travellers to price changes at particular airports
may still be quite high.

For example, tourists are more likely to compare package prices for different
destinations and to substitute different travel modes to a destination (self-drive
versus flying, for example). International tourists may be flexible as to their

OVERVIEW XXIII
entry/exit point. On the other hand, destination-specific, time-sensitive business
travellers are likely to be less responsive to price changes than tourists. But even
business travellers are likely to have some scope to adjust their frequency of travel
or, increasingly, to use forms of electronic communication such as teleconferencing.

Assessment of substitution possibilities and market power

Table 1 summarises the Commission’s assessment of substitution possibilities for

domestic passengers at the 12 core-regulated airports. This is based on an analysis
of the types of traveller served by particular airports.

Table 1 Summary of demand and competition characteristics of core-

regulated airports for domestic passenger traffic
Main market
segment to Potential for Potential for Potential for Overall
destination for destination modal airport degree of
Airport interstate travel substitution substitution substitution market power

Adelaide Business/VFRa Low Moderate Low Moderate

Alice Springs Holiday High Moderate High Low
Brisbane Business/VFR Low Moderate Low High
Canberra Business/VFR Low High Low Low/Moderate
Coolangatta Holiday High Moderate High Low
Darwin Holiday High Moderate Low Low/Moderate
Hobart Holiday High Moderate High Low
Launceston Holiday High Moderate High Low
Melbourne Business/VFR Low Moderate Low High
Perth Business/VFR Low Low Low High
Sydney Business/VFR Low Moderate Low High
Townsville Holiday High Moderate Low Low
a Visiting Friends and Relatives.

The Commission agrees with the ACCC that Alice Springs, Coolangatta, Hobart
and Launceston airports appear to have little market power because they rely
heavily on the tourism market and there is scope for competition from ‘nearby’
airports. The Commission also considers that Townsville Airport has limited market
power because it relies on the holiday market (and there is scope for competition
from other holiday destinations and other transport modes).

On the other hand, it appears that Brisbane, Melbourne, Perth and Sydney airports
possess significant market power in domestic markets due to high proportions of
business travellers and those ‘Visiting Friends and Relatives’, and their status as the
main international ports of arrival and departure in the country. Competition among
those airports for international traffic is likely to moderate, but not eliminate, this
latter effect.
XXIV OVERVIEW
The degree of market power held by Adelaide, Canberra and Darwin airports is less
clear. On balance, the Commission’s assessment is that these airports appear to have
at most a moderate degree of market power.

More than 200 non-core-regulated airports cater for some RPT services (rather than
just general aviation). No charges are imposed at some regional airports, while at
others charges can be very high. Of themselves, high charges do not necessarily
indicate an abuse of market power. While the Commission has not been able to
assess the likely market power of individual non-core-regulated airports, the small
traffic volumes of most regional airports suggest that unit costs would often be high.
High charges at many regional airports may also reflect revenue-raising objectives
of the local municipalities that own them. Privately-owned airports servicing resorts
generally operate in the highly competitive tourist market.

In which services do the key airports have market power?

For those airports assessed to have a moderate to high degree of market power, that
power will extend to the bundle of services for which there are few alternative
suppliers and which an airline and its passengers must consume. The Commission’s
assessment of the degree of market power across the range of services provided by
airports is summarised in table 2.

Table 2 Market power in particular airport services

Service Degree of market power

Aircraft movement facilities High

Passenger movement facilities Moderate/High
Lounge space (VIP and business) Low
Vehicle access facilities High
Car parking Low/Moderate
Taxi facilities Low/Moderate
Aircraft refuelling Moderate/High
Aircraft light/emergency maintenance sites Moderate
Aircraft heavy maintenance facilities Low
Flight catering facilities Low
Freight and ground equipment storage sites Low
Freight facility sites and buildings Low
Waste disposal facilities Low
Administrative office space Low/Moderate
Commercial and retail activities Low

OVERVIEW XXV
Market power appears to be strongest for:
• facilities for aircraft movements, including access to runways, taxiways and
aprons; and
• vehicle access, including front-door access to the airport for passengers,
transport providers and off-airport car-parking providers.

A range of other airside services are necessary for passenger and cargo processing,
but there is some, albeit limited, discretion as to quality and quantity of service —
for example, aerobridges, check-in facilities and baggage handling. This suggests
that an airport’s market power in these services may be somewhat less than for
services provided by aircraft movement facilities but, in most cases, it would still be
significant.

The degree of airport market power in relation to aircraft refuelling has been a
particularly contentious issue. For some flights and some destinations, there is a
degree of choice as to where an aircraft refuels, but this discretion is limited and, at
more remote airports, probably non-existent.

Areas where market power is likely to be more limited include international airline
lounges, catering and freight storage facilities, and services that can easily be
located off-airport. Airports (and State Governments) also compete keenly for
aircraft heavy maintenance facilities.

While passengers value the convenience of shops and food outlets at airports, the
purchase of these services in most cases is likely to be highly price sensitive. The
services do not have to be consumed as part of the airport ‘bundle’ and there are
many alternatives to buying at the airport. Passengers also have choices in relation
to duty-free purchases — destination, transfer, or origin airports; on-board duty-free
and downtown outlets; and whether to buy duty free or not.

Passengers require transport to and from an airport but typically there is a choice of
several modes (taxi, self-drive, public transport) that are not provided by the airport.
For those who require parking, major airports (eg Sydney and Melbourne) have
substantial off-airport competition, at least for long-term (one day or more) parking.
Higher parking charges at airports relative to off-airport locations appear largely to
reflect greater convenience. Nevertheless, by controlling access to their ‘front door’,
airport operators could attempt to limit competition from off-airport providers of car
parking or other transport providers (eg taxis).

XXVI OVERVIEW
The potential for abuse of market power
Assessing the likely conduct of airports with market power is prerequisite to
assessing whether regulation is required and, if so, in what form. There are two
aspects that need to be considered: first, the existence of any constraints on the
abuse of market power; and second, the economic consequences of exercising
market power.

Commercial constraints on the abuse of market power

While some major Australian airports appear to have substantial market power in
the core aeronautical services they provide, they may not use this power to its full
extent, because of commercial incentives and constraints.

Non-aeronautical revenues can constrain aeronautical pricing

There is fairly strong evidence that the potential earnings from passenger spending
on non-aeronautical services (including retailing, car parking and restaurants)
provide airports with an incentive to encourage extra passengers to the airport.
• In 1999-00, for every dollar of operating profit earned from the provision of
aeronautical services at Australia’s major international airports, more than four
dollars of profit were earned from non-aeronautical activities. Though there are
some issues regarding the classification of earnings (eg lease payments which
for domestic terminals are classified as non-aeronautical), and the data do not
necessarily indicate the incremental earnings from an extra passenger, non-
aeronautical activities clearly are of great importance to airports.
• There are numerous examples of financial incentives being offered by airports to
encourage additional airline services, suggesting that most airports consider that
airlines are responsive to such incentives. Airports also regard service quality as
an important factor contributing to passenger spending (box 4).

Box 4 Attracting customers to airports

‘High levels of customer satisfaction drive greater turnover of discretionary expenditure in
airports and this is highly profitable.’ And ‘[Airline] market development is expensive and is
often recognised as a barrier to entry to new carriers. Melbourne Airport has been in active
partnership with successive Victorian Governments in developing markets both to and from
Melbourne.’ (Melbourne Airport, sub. 7, pp. 9 and 12)
‘… airport operators do price to attract new entrants and small operators, as evidenced by
the discounts offered in Sydney Airport’s … Conditions of Use.’ (SACL, sub. 27, p. 55)

OVERVIEW XXVII
Long-term growth strategies

To a considerable extent, airports’ market power relies on the relatively small effect
of an increase in airport charges on air travel costs. But this also provides airports
with some incentive to cooperate with other input providers (eg in the tourism and
aviation industries). In other words, airports, by themselves, will have little
influence over the attraction of their location, but they may be able to increase
demand and/or reduce demand uncertainty by working with other providers who are
in a better position to influence demand.

State Government departments commented on the close working relationships that

have developed with airports since they were privatised. That privatised airports
have sought and maintained such relationships does not sit well with the notion of
them exploiting whatever market power they may have.

Airline countervailing power?

Any attempts by airports to exercise their market power might be prevented by

countervailing action by airlines. Airlines and airports hold strongly opposing views
on whether airlines have countervailing power (box 5). Available evidence suggests
that scope for competition in the aviation market and the importance of major
airports to airline networks will limit (though not necessarily rule out) airline
countervailing power in their dealings with the major capital city airports. Airline
market power is likely to be far stronger in their dealings with smaller airports,
which have less commercial clout and a greater reliance on holiday markets.

Box 5 Opposing views on airline countervailing power

‘Adelaide Airport Limited [AAL] is not able to abuse any market power that it may have in the
domestic market as AAL’s main customers, Qantas and Ansett, hold significant
countervailing market power. Put simply, Qantas and Ansett are much more important for
Adelaide Airport than Adelaide Airport is for Qantas and Ansett.’ (AAL, sub. 20, p. 1)
‘Countervailing power can only arise if the buyer has choices available to it … However, if a
buyer has no choices other than to purchase goods or services from a particular seller, no
countervailing power will arise, no matter how large the buyer is relative to the market.
Airlines have no choice but to use the services of airport operators to which customers want
to fly.’ (Qantas, sub. 48, p. 12)

The reduction in demand for airport services since September 2001 is likely to have
enhanced airline countervailing power, at least in the short term. In the current
market environment, threats by international airlines to withdraw services, or threats
by domestic carriers not to increase services to an airport, would carry more weight
than previously.
XXVIII OVERVIEW
Likely effects of airport market power

Pricing efficiency

Regardless of market power considerations, it is likely, if aeronautical prices at the

three airports still subject to price caps were unregulated, that average aeronautical
charges would tend to rise above current levels. This is because, under the
transitional price-cap arrangements, prices have continued to reflect the single-till
prices established by the FAC. The UK airports regulator, the Civil Aviation
Authority, recently observed for the UK regulated airports that ‘the incremental
costs of additional capacity are generally significantly above single till levels’.

How far beyond their efficient levels or, indeed, whether aeronautical prices at
airports with market power would increase in the absence of any airport-specific
price regulation, no-one can say with certainty. However, for reasons outlined
above, the Commission considers that any price increases would be constrained by a
range of market forces, including commercial interests of the airports themselves.

Moreover, to the extent that airports can discriminate in pricing, such that charges
are allocated across airlines, aircraft and (through fare structures) across passengers
in a way that roughly reflects different passengers’ willingness to pay, the net effect
on consumption of air travel of a given increase in airport charges will be reduced.
Airports, like any business, have a strong incentive to discriminate among users in
their pricing to increase their profits. Though such price discrimination will not be
‘perfect’, there are numerous examples of airport price structures designed to
promote or retain marginal users of the airport, including direct incentives designed
to encourage additional flights and new entrant airlines.

Distributional effects

If airports increased their prices, then broadly speaking the losers would be
passengers (who may pay higher fares) and airline shareholders (who may have to
accept lower returns). While the Commission is not in a position to make a
judgement about preferred income distributions, it makes the following
observations:
• there does not appear to be any great difference between the income profiles of
passengers and airline shareholders on the one hand and the shareholders of
airports (including large Australian superannuation funds) on the other;
• owners of airports and airlines comprise both Australian residents and non-
residents (foreign interests can own up to 49 per cent of privatised airports); and

OVERVIEW XXIX
• many passengers are not Australian. In 1999, almost 60 per cent of passengers
on international flights were non-residents. This represents about 15 per cent of
all passenger movements at the major international airports. Non-residents also
fly domestic and regional routes but the data on the proportions are not available.

Other consequences of market power

Several airline participants considered that airports with market power also would
allow their costs to rise to inefficient levels (at the expense of potentially higher
airport profits). The Commission has not been persuaded by this argument. The
ability of owners to monitor managers in this respect is unlikely to be significantly
less for privatised airports than for firms operating in other markets.

Airlines also have suggested that an airport with market power would allow quality
to deteriorate to increase profits. This is a possibility, particularly if prices are
constrained by regulation. However, an airport is unlikely to allow quality to
deteriorate if higher prices can be charged for higher quality and if passenger
spending on non-aeronautical services is sensitive to quality levels. Both of these
conditions seem to apply.

Airports with market power may also have an incentive to under-invest in order to
limit supply and maximise monopoly profits over time. New investment will occur
only when the expected additional profits from that investment exceed any profits
forgone from lower prices at the existing facility. But to the extent an airport can
discriminate in pricing, its incentives to under-invest would be reduced.

Moreover, if an airport allowed capacity constraints to develop and to remain for

lengthy periods (eg by not building an additional taxiway or runway), then the door
may be opened to potential competitors. (There is evidence that international flights
to and from Melbourne and Brisbane airports have increased due to the curfew and
movement limits imposed by regulation at Sydney Airport.) Under-investment in
capacity and quality enhancement could also reduce non-aeronautical revenues.

Market power could be used to deny access to the airport, either by way of higher
prices or by imposing unacceptable access conditions. An airport is unlikely to have
an incentive to deny access to its major customers, the airlines. (On the whole, the
response of airports to new airline entrants seems to confirm this point.) But, it may
have an incentive to deny or frustrate access to potential competitors, such as off-
airport car-parking operators or providers of other forms of transport to the airport.

XXX OVERVIEW
 Box 6 Contrasting views on airport behaviour without price control

The airlines
‘In BARA’s view, failing to effectively regulate airport charges will weaken the incentives for
airport operators to provide airport services efficiently … In the absence of competition there
is scope for productive inefficiencies. Effective regulation, through acting as a surrogate for
some of the cost pressures on a firm facing effective competition, can reduce these costs.’
(Board of Airline Representatives of Australia, sub. 41, pp. 29–30)
‘Left unregulated, airport operators can be expected to use their market power through
monopoly pricing, diminishing service quality and the imposition of unreasonable terms and
conditions of access to the airport.’ (Qantas, sub. 48, p. 29)

The airports
‘Airports in an unregulated environment will have strong incentives to consult with their
customers on investments and quality outcomes to avoid wasteful investment … while
facilitating an optimal level of demand.’ (SACL, sub. 27, p. 34)
‘It is Melbourne Airport’s view that the commercial incentives for airports in Australia are
such that … they will pursue pricing and investment policies that are likely to maximise
economic welfare in the long run. In contrast, regulatory intervention through price controls is
likely to reduce welfare in the long run.’ (Melbourne Airport, sub. 7, p. i)

Feasible regulatory options

The case for any form of regulation rests on the net benefits it generates compared
with other forms of regulation and, indeed, no regulation. There are numerous
regulatory options for airports, several of which already apply in Australia and
overseas. These range from heavy-handed cost-based regulation, to incentive
regulation (price caps) and to more light-handed approaches such as price
monitoring.

Cost-based and incentive regulation

Cost-based regulation aims to ensure that a regulated firm earns no more than its
production costs, including a ‘normal’ rate of return on capital. It covers rate-of-
return regulation and cost-justified price increases (as currently apply under the
prices-notification regime for Sydney Airport). Regulated price increases based on
costs typically provide insufficient incentives for firms to operate efficiently — the
incentive to reduce costs is weakened because increased profits cannot be retained.
Cost-based regulation also requires detailed assessment by the regulator of the
firm’s capital and operating costs, asset base, and expected sales.

OVERVIEW XXXI
Incentive regulation is designed to reduce some of the problems of cost-based
regulation. It has usually taken the form of a CPI-X price cap. Over the period of
the price cap, regulated firms have an incentive to reduce costs (particularly beyond
the cost savings implied by the cap) to increase profits. But if prices are not to
persist at levels that deliver excessive profits or losses, eventually price caps must
be adjusted to allow prices to converge to levels that (just) cover costs. While the
use of benchmarks can reduce the need to consider costs incurred by the regulated
firm, it is unlikely that benchmarking can remove the need for cost assessment
altogether. In other words, the incentive advantages of the price caps tend to be
eroded over time, as the caps converge to cost-based regulation. For this reason,
price caps are regarded generally as providing a transitional form of price
regulation.

A range of issues has arisen in relation to price-cap arrangements applying to

Australian airports. The inherited FAC starting prices (set to provide a modest
return on capital, based on income from all sources) were not adequate to provide
for efficient replacement aeronautical investment on a dual-till basis. These prices
then combined with X values that incorporated expected productivity and passenger
growth and some undefined investments, and (initially ill-defined) investment cost
pass-through provisions.

At best, a lack of clarity has promoted strategic behaviour by all parties, increased
compliance costs and discouraged commercial negotiation. At worst, the
arrangements, which combine elements of incentive and cost-based regulation, have
discouraged efficient investment by sending poor price signals both to airport
operators and users about the costs of providing aeronautical services and by
requiring very detailed regulatory assessment of every investment proposal.

Some of these problems could have been avoided by a more transparent process and
clearer guidelines from the outset; others reflect the design of the price caps.
Though price caps could be redesigned to incorporate planned investment, and
benchmarks could be used to provide independent estimates of achievable
productivity growth, inevitably there will have to be an eventual assessment by the
regulator of an airport’s operating and capital costs, land values, risk, and demand
growth. Given the substantial conceptual and information problems in asset
valuation and cost assessment (assessing the opportunity cost of airport land being
especially problematic), the risks of regulatory error and the potentially damaging
consequences of this for investment correspondingly increase.

Accordingly, the Commission considers that price caps, while preferable to explicit
rate-of-return regulation, should be reserved for situations where there is a strong
likelihood of excessive pricing and where such pricing is likely to impose major
costs on the community.
XXXII OVERVIEW
Single or dual till?

Leases for privatised core-regulated airports were sold with an explicit undertaking
that a single till would not be mandated. It appears that Sydney Airport will be sold
on the same basis. Nonetheless, various airline participants have argued that price
regulation should ensure that some or all non-aeronautical profits are applied to
reduce aeronautical charges.

Though airports are likely to offer incentives to encourage aeronautical traffic in

order to increase profits from related commercial activities, mandating the transfer
of non-aeronautical rents is likely to discourage development by the airport of both
aeronautical and non-aeronautical services, generating large efficiency losses in the
long run. Indeed, reversion to a regulated single till, even on a partial basis, could
stifle the risk-taking, innovation and development of the airport site that are
regarded as major benefits of privatisation (as well as raising issues of sovereign
risk).

Price monitoring

Price monitoring is a light-handed alternative to price caps or rate-of-return

regulation. Under price monitoring, firms generally are required to provide
information on prices, costs and profits, but there is no direct regulatory control
over prices charged or revenues or profits earned. The impact of monitoring on
firms’ pricing decisions is more indirect, through moral suasion, providing
customers with better information, publicity, and the threat of stricter forms of price
regulation being re-introduced. Monitoring can be a less explicit or intrusive
method for regulating prices than price caps or cost-based regulation. It can achieve
the same objectives as stricter forms of regulation but at lower cost and with less
distortion of incentives. Perhaps most importantly, as compared with more intrusive
regulation, price monitoring can facilitate commercial negotiations between airport
operators and users (provided there is no automatic recourse to regulatory
determination of prices).

On the other hand, the potential to exercise market power may be greater under a
price-monitoring regime than under price caps or cost-based regulation. However,
such potential can be constrained by a credible threat that the Government would in
time re-introduce stronger regulation at airports where market power clearly was
being abused.

OVERVIEW XXXIII
Access provisions and general competition law

Airport operators currently are subject to s. 192 of the Airports Act 1996 and
Part IIIA (the National Access Regime) of the TP Act, while anti-competitive
practices are subject to Part IV of that Act. The five-year, automatic declarations of
privatised core-regulated airports cannot be renewed under s. 192 as it stands.
However, new airport-specific access arrangements or Part IIIA would continue to
complement any future price regulation of airport services or, indeed, could provide
an alternative to an industry-specific price-regulation regime. Prices of airport
services could be regulated indirectly through regulation of the terms and conditions
of access to a declared airport service.

As already noted, airports do not have strong incentives to deny access to their
major customers, the airlines. They may have some incentives to frustrate access to
those with whom they directly compete (eg car-parking operators) or where they
can control competition in a market through their control of access to the airport
facility in question (eg ground-handling operations). However, there is little
evidence that airports have sought to do this. The Commission has not been
persuaded that there is a case for the continuation of special access provisions that
impose more easily satisfied declaration criteria for airports than other industries.

Commercial agreements

Several participants suggested various ‘midway’ approaches where airports and

airlines would enter commercial agreements on prices, quality and conditions of
access, within a broad regulatory framework.

The notion of promoting commercial agreements has immediate appeal because

they could circumvent the need for a high level of regulatory involvement.
However, the Commission considers that any such agreements, to be successful,
necessarily must be negotiated voluntarily (by both sides), without automatic
recourse to the regulator and without prescriptive requirements. Guidelines for such
agreements (including consultation and model dispute-resolution mechanisms) may
promote such arrangements, as would an indication that the agreements would be
taken into account in future regulatory reviews of the industry.

That such guidance from government may be helpful for commercial entities in a
commercial environment may seem surprising. The reason for it is the historical
legacy of decades of government ownership of this industry and the pervasive
effects of regulation.

XXXIV OVERVIEW
The Commission’s assessment
Major Australian airports — Sydney, Melbourne, Brisbane and Perth — have been
assessed as having a high degree of market power in core aeronautical services.
This suggests that some form of airport-specific price regulation is required for
these airports, additional to general third-party access and competition regulation.

The Commission is not convinced that Adelaide, Canberra and Darwin airports
have substantial market power and they may be comparable in this respect to Cairns
Airport (the sixth-largest in Australia, albeit State owned), which is not subject to
any special price regulation. However, some form of monitoring may be appropriate
for these smaller airports at least as a transitional measure.

Remaining core-regulated airports, because they appear to have much less market
power, or because, given their size, the costs of regulation would far outweigh any
potential benefits, should not be subject to any airport-specific economic regulation,
including price or quality monitoring. (They would continue to be subject to the
TP Act and relevant provisions of the Airports Act.)

For those airports warranting continued regulation, the Commission considers that
the choice essentially comes down to forms of regulation that can provide
incentives for efficient outcomes. Of the stricter forms of price regulation, CPI-X
price caps provide the best incentives but they also bring inevitable risks of
regulatory error. Given that airports face significant commercial constraints and
incentives that will moderate abuse of any market power, the Commission sees
significant advantages in a more light-handed approach involving price monitoring.
Each of the options is outlined and their relative merits then assessed in more detail.

Option A: dual-till price caps

The stricter option essentially would continue the regulatory regime introduced on
5 October 2001, but with some refinements in the setting of price caps (designed to
promote efficient investment at airports and reduce regulatory intervention in
investment decisions) and also some changes to price-monitoring arrangements.
For Melbourne, Brisbane and Perth airports, there would be a CPI-X price cap for
five years, complemented by price and quality monitoring.
• The cap would apply only to the aeronautical services in which those three
airports have been assessed as having a high degree of market power: aircraft
movement facilities (including refuelling services), vehicle access facilities and
most passenger movement facilities. In broad terms, this coverage is very similar
to current price caps, except for refuelling services.

OVERVIEW XXXV
• Average prices allowed under the cap should broadly reflect the anticipated cost
of efficiently providing regulated aeronautical services on a dual-till basis
(box 7). In other words, unlike current price caps where inherited single-till
prices are adjusted incrementally as new investment is put in place, the cap
should correspond to anticipated efficient dual-till prices.

Box 7 Setting of price caps under Option A

Any price caps should be developed on a current-cost basis. They should incorporate
the cost of required (optimised) aeronautical assets (including anticipated efficient
capital expenditure) and anticipated efficient operating costs (with reference to
productivity benchmarks where feasible) over the five-year regulatory period. In
addition, they should be calculated on a dual-till basis. That is, assets used and profits
earned in non-aeronautical activities should not be taken into account in determining
allowable aeronautical prices.
Implementation of this current cost, dual-till approach would require both an
assessment and revaluation of the aeronautical asset base. While historical costs may
make for a simpler assessment, they are less likely to encourage efficient production
and investment in this industry, especially where the current opportunity cost of
significant assets is substantially above their historic cost (eg aeronautical land at
Sydney Airport).
Capped prices at these airports would then be calculated on a broadly similar basis to
current aeronautical prices at Sydney Airport (but with land appropriately valued).
Moreover, whereas at Sydney, prices were increased by around 100 per cent (from
previous FAC prices) in one step, any significant price changes could be phased in
over the regulatory period via the X factors (appropriately adjusted for anticipated
productivity and demand growth), to give airports and airport users time to adjust to the
new basis for price setting.
With price caps calculated in this way, prices would not require adjustment for
aeronautical investment foreseen and agreed to at the commencement of the
regulatory period. However, there still may need to be provision for appropriate cost
pass-through of large, unanticipated investments (and possible downward adjustment
if planned investments do not proceed).
Price caps should continue to be calculated on a tariff-basket basis and hence allow
price restructuring within the cap.
The Commission does not consider that setting price caps on a dual-till basis
necessarily would deliver windfall gains (or losses) to airport operators. The outcome
would depend on the expectations about future prices and regulation that were
factored into their bids for the purchase of the leases.

• Price caps should continue to be complemented by monitoring of the quality of

regulated services. While quality monitoring should identify any failure to invest
appropriately on the part of airports (because relevant quality indicators would

XXXVI OVERVIEW
 deteriorate), with most investment incorporated in allowable prices, there may
also need to be monitoring to ensure that investments are not being delayed
intentionally by the airport to increase profits. Airlines would be expected to
play a role in developing the forward investment program and their input should
be sought in monitoring its implementation.
• Price monitoring would continue for some services currently designated as
‘aeronautical-related’ in which market power has been assessed as moderate (eg
check-in counters, staff car parking). Some other services currently designated
aeronautical related (eg public car parks), would not be subject to monitoring.
Reporting of aggregate financial data for both aeronautical and non-aeronautical
services would continue.

For Sydney Airport, for reasons discussed in box 8, in the absence of slot sale or
trading mechanisms, regulation preferably should allow aeronautical prices to
reflect more closely the opportunity costs incurred by airlines and their passengers
in using the facility (that is, the value of slots to those who miss out at peak times),
rather than simply reflecting the calculated costs of production. At the very least,
provided capacity remains constrained, aeronautical prices should not be required to
decline in real terms and should be adequate to encourage efficient, feasible
expansion of aeronautical capacity at that facility.
• This could be implemented either by requiring notification of aeronautical price
increases above the CPI, or by imposing a CPI-X cap with the X set at zero.
• In either case, price increases should be allowed to reflect peak-period demand
(as provided for in the current regulatory framework for Phase 1 airports), and to
accommodate necessary investment.

For Adelaide, Canberra and Darwin airports, there would be a continuation of

(modified) price-monitoring arrangements (as outlined below in Option B) rather
than price caps. Quality monitoring would continue at these airports.

For all other airports, there would be no airport-specific price regulation or quality
monitoring. This would require amendment of the Airports Act so that Alice
Springs, Coolangatta, Hobart, Launceston and Townsville airports no longer were
designated as ‘core-regulated’ airports.
All airports should be subject to the generic provisions of the Part IIIA National
Access Regime. An airport-specific access regime should be considered only if
procedural improvements already identified by the Commission, such as scope
for multilateral arbitrations, are not made to the National Access Regime.

OVERVIEW XXXVII
 Box 8 The benefits of price rationing at Sydney Airport
Prior to September 2001, Sydney Airport experienced excess demand for aircraft
movement slots for several hours a day. This situation is likely to resume in the not-too-
distant future. The limited supply of slots is currently due to a movement cap and
curfew, and the regional ‘ring fence’. It is likely that the market-clearing price at peak
times exceeds airport production costs.
Appropriate peak/off-peak charges at Sydney Airport would promote more efficient use
of the airport and, given the massive investment required to establish a second airport,
also would provide important signals to the Government about the need for a new
facility and the costs of the various constraints imposed on the use of Sydney Airport.
Though prices approaching market-clearing levels at peak times would be likely to
promote use of the airport by those who value it most — even if ring-fenced regional
flights were quarantined from such price increases, in line with announced Government
policy — the airport could earn large profits, reflecting the scarcity value of slots.
(Efficient peak/off-peak charging would balance lower off-peak charges against higher
peak charges — this would tend to moderate the extent of profit increases.)
It is important to note that any rise in peak prices is unlikely to lead to higher passenger
fares. There is strong evidence that the scarcity value of the slots is factored into fares
already; it is very difficult to obtain discounted fares to Sydney at peak times. In other
words, it is the airlines that are currently benefiting from the excess demand for landing
slots at peak times.
If bidders were informed prior to the sale of Sydney Airport that efficient peak/off-peak
charges would be permitted, then any anticipated scarcity rents would accrue to the
Commonwealth Government (and taxpayers generally) in the bid price, rather than to
the new airport operator.
Source: appendix H.

Option B: price monitoring

This option would extend price monitoring to Phase 1 airports and Sydney Airport
for a probationary period, and maintain (modified) price monitoring of Adelaide,
Canberra and Darwin airports. As in Option A, there would be no airport-specific
price regulation of any other airports.

For Sydney, Melbourne, Brisbane, Perth, Adelaide, Canberra and Darwin airports,
there would be mandatory price monitoring by the ACCC. The monitoring regime
would continue for five years.
• During this probationary period, the regulator would not have the power to alter
unilaterally the monitoring regime or impose stricter price regulation.

XXXVIII OVERVIEW
• Information requirements would be specified at the commencement of the period
and could not be amended without agreement of the parties. The Commission
envisages somewhat more detailed monitoring of Sydney, Melbourne, Brisbane
and Perth airports, including the weighted average cost of capital (WACC) for
aeronautical assets. Otherwise, proposed information requirements for the seven
airports broadly would be consistent with existing disclosure and reporting
requirements.
• As under current monitoring and reporting arrangements, information would be
formally audited. There also should be provision to treat some data as
commercial-in-confidence where disclosure could undermine commercially-
negotiated agreements. An annual report presenting monitored information
would be made publicly available, with commentary (but not a pre-emptive
assessment of the monitoring regime) by the ACCC (and auditors) where
considered warranted.
• Quality-of-service monitoring would continue at all price-regulated airports.
• Voluntary commercial agreements between airports and users (including non-
airline users) would be encouraged by providing guidelines regarding coverage,
consultation and dispute-settlement mechanisms. (The Commission sees no need
to exempt from access regulation airports that enter into such agreements.)
• An independent public review would be conducted towards the end of the five-
year monitoring period to ascertain whether there should be any future price
regulation of those airports. Other airports could be included in the review only
where there is prima facie evidence of persistent misuse of market power.
• Factors to be taken into account by the review in assessing whether airports had
abused market power should be specified at the start of the regulatory period
(box 9). Efficient pricing principles for airports are discussed in this report and
also have been developed more generally by the Commission in its reviews of
the National Access Regime and Telecommunications Competition Regulation.
They are predicated on the need to avoid excessive prices, without the regulator
attempting to set prices too precisely, with the resultant risk that prices may be
pushed too low and efficient investment discouraged.

For all other airports, there would be no airport-specific price regulation or quality-
of-service monitoring. This would require amendment of the Airports Act so that
Hobart, Launceston, Alice Springs, Coolangatta and Townsville Airports no longer
were designated as ‘core-regulated’ airports.

All airports should be subject to the generic provisions of the Part IIIA National
Access Regime. An airport-specific access regime should be considered only if

OVERVIEW XXXIX
procedural improvements already identified by the Commission, such as scope for
multilateral arbitrations, were not made to the National Access Regime.

Box 9 Proposed review criteria

The proposed review of price regulation of airports to ascertain the need for continued
regulation (if any) in five years time, should be guided by the following criteria.

Pricing principles
• At airports without significant capacity constraints, efficient prices broadly should
generate expected revenue that is not significantly above the long-run costs of
efficiently providing aeronautical services (on a dual-till basis). Prices should allow a
return on (appropriately defined and valued) assets (including land) commensurate
with the regulatory and commercial risks involved.
• At airports with significant capacity constraints, efficient peak/off-peak prices may
generate revenues that exceed the average production costs incurred by the airport.
• Price discrimination and multi-part pricing that promote efficient use of the airport
should be encouraged. This may mean that some users pay a price above the long-
run average costs of providing aeronautical services, whereas more price-sensitive
users pay a price close to marginal cost.

Other criteria
• Whether quality-of-service outcomes have deteriorated and/or failed to meet the
requirements of users.
• The extent to which commercial agreements on prices and quality of service have
been negotiated.
• The degree of consultation with airport users and the extent of the exchange of
information.
• The number and outcome of applications for third-party access and the extent and
validity of user complaints.

Weighing the options

Option A is superior to the existing regime

Option A would continue to apply price-cap regulation to those airports considered

to have a high degree of market power. In addition, Option A would set price caps
that broadly reflect costs of supplying aeronautical services on a stand-alone, rather
than single-till, basis. For these reasons, Option A would, in the Commission’s

XL OVERVIEW
view, perform better than the regulatory arrangements in place before 5 October
2001 and those that have applied since then. In particular:
• by making allowance for anticipated investment in the price-cap parameters, it is
likely that regulatory involvement in and, therefore, scope for distortion of,
investment decisions will be reduced. In addition, airports and their users would
receive much better signals from efficient dual-till prices regarding the real costs
and benefits of supplying aeronautical services;
• airports and their customers would have greater incentives to negotiate price and
quality agreements within the overall limits of the designated caps (at least
where there is unused airport capacity), reflecting incentives to increase
aeronautical and non-aeronautical profits by expanding use of the airport at the
margin;
• use of benchmarking, where feasible, could help to sustain the positive incentive
effects of price caps and avoid them converging to rate-of-return regulation; and
• maximum average aeronautical prices at Melbourne, Brisbane and Perth airports
would be set on a dual-till basis, consistent with prices approved by the ACCC
for Sydney Airport.

But price caps are not warranted

The Commission has not been persuaded, however, that there is a strong case for
the continuation of price caps at any privatised core-regulated airports. This is for
two principal reasons.

The first is the ever-present risk of regulatory failure, given the severe information
problems confronting any regulator. Setting price caps inevitably entails detailed
regulatory assessment of, and involvement in, airport operations and investment
decisions. It should therefore be used only where the potential efficiency costs of
abuse of market power are significant. Even then there is a risk that regulation will
cause its own distortions to production and investment decisions. While the
Commission agrees that some transitional problems with current price-cap
arrangements may have been settled, and that the price caps proposed under
Option A should reduce regulatory involvement in investment decisions, the risk of
regulatory failure remains high. This risk has been amplified by the uncertainty that
currently pervades global aviation markets.

The second and related reason is that the ‘problem’ to be addressed does not
warrant such a heavy-handed regulatory regime. Though the four largest airports
have considerable market power, the prospect of them using that power in a way
that would generate significant costs to the economy or community is supported

OVERVIEW XLI
neither by the evidence nor the analysis. There are strong commercial incentives
pulling in the other direction, including scope for increased profits in non-
aeronautical activities from increasing passenger volumes and incentives to
discriminate and differentiate in pricing.

Option B provides more appropriate regulation

On balance, therefore, the Commission considers that while the market power of the
four major airports warrants some form of regulatory oversight, the continuation of
price caps is not the best approach. Option B offers a much better regulatory
mechanism for promoting the principles for regulation identified in the terms of
reference.

It would promote commercial relationships

In particular, Option B would encourage airports and airlines to negotiate

commercial agreements. While any guidelines for such agreements should not be
overly prescriptive — the parties would be far more skilled in working through the
detail of agreements than regulators — the Commission sees a role for formal
guidance to assist all parties.

While some participants have questioned whether commercial negotiations are

feasible in this industry, the Commission notes that airlines deal directly with
numerous unregulated small airports, both privately- and publicly-owned, and some
unregulated larger ones (eg Cairns). This is not to say that relations would always
be smooth, just that there seems to be considerable scope for reasonable commercial
interaction between airports and users, provided there is some ultimate constraint on
abuse of any market power (which Option B provides, in addition to the TP Act).

In this regard, the proposed length of the monitoring period is crucial. If it were too
short, some parties might be encouraged not to deal in good faith, in order to
increase the likelihood of re-regulation. If the period were too long, airports with
market power might have an incentive to make use of that power. The
Commission’s view is that five years would achieve the right balance.

It would also constrain inefficient outcomes

To give airports guidance as to what behaviour could provide grounds for re-
introduction of stricter price controls (or removal of airport-specific prices
oversight), the Commission has elaborated some broad principles for a review body
to apply in five years’ time in assessing how the industry and regulatory framework

XLII OVERVIEW
were performing (box 9). Some participants expressed concern about the need for
additional public scrutiny of any such principles. However, the broad pricing
principles proposed have already been discussed extensively during this inquiry and
parallel inquiries into infrastructure regulation (the National Access Regime, the
Prices Surveillance Act 1983 and Telecommunications Competition Regulation).

The Commission also draws attention to the continued relevance of third-party

access provisions (Part IIIA) and Part IV of the TP Act to airports. The Commission
does not consider that airports have strong incentives to deny or frustrate access in
order to impede competition in a way that would be likely to lead to declaration of
aeronautical facilities (at least under the modified declaration criteria recommended
in the Review of the National Access Regime (PC 2001a)). Nonetheless, the
potential activation of this mechanism does provide users with regulation of last
resort, providing additional encouragement for airports to enter into reasonable
agreements regarding prices and conditions of airport use.

Concluding remarks

In the Commission’s view, the regulatory approach of Option B provides the best
balance of incentives and disciplines for major airports.

This recommendation for lighter regulation inevitably rests on a judgement about

the likely behaviour of major airports. A number of participants saw this recognition
of the limitations of knowledge about future behaviour as weakening the
Commission’s conclusions in its draft report. Yet those arguing in favour of price
caps have based their arguments on their own judgements about the likely
behaviour of airports. The Commission has independently considered the evidence
and arguments in making its assessment.

Some participants also have argued against discontinuation of price caps at a time of
upheaval and uncertainty in the aviation industry. The Commission is firmly of the
view that the uncertain outlook calls for more, not less, flexibility. Indeed, this has
been recognised by the Government in discontinuing price caps for Phase 2 airports.
If airport operators themselves cannot predict what will happen over the next few
months or years, regulators are unlikely to be able to fix price caps that can deal
efficiently with future market conditions.

Some have cited their interpretation of the recent New Zealand experience with
light-handed regulation in an attempt to rebut the Commission’s price-monitoring
option. But closer examination suggests that the New Zealand system,
notwithstanding some deficiencies, has not been a failure (box 10). For example,

OVERVIEW XLIII
Auckland and Christchurch airports have agreed on new aeronautical prices with
their major airline customers and those price outcomes do not appear excessive.

Box 10 New Zealand’s experience with light-handed regulation:

success or failure?
Since 1988, a light-handed system of price monitoring has applied to New Zealand’s
three largest airports — Auckland, Christchurch and Wellington. The system is
currently being reviewed by the NZ Commerce Commission (CC).
The NZ approach has been criticised by some participants in this inquiry for, in their
view, failing to restrain the abuse of market power by these airports — allegedly
resulting in high prices, high returns and efficiency losses — and for encouraging
extensive and expensive litigation between airlines and airports.
While there may have been shortcomings in the NZ system of price regulation,
evidence of abuse of market power is weak.
• Airports and airlines have agreed on price increases at Auckland and Christchurch
airports. Auckland’s prices have increased by 12.5 per cent. This was the first
increase in landing charges at Auckland Airport since privatisation in 1998, and its
first net increase since 1992.
• Assumptions and modelling underpinning the CC’s analysis of the need for price
controls currently are the subject of intense debate. The CC’s conclusions depend
heavily on its decision to remove from the allowed asset base land purchased by
the airport to build a new runway (an investment even airlines agree is necessary).
If this land had not been excised from the asset base, Auckland Airport’s estimated
rate of return would have fallen below the benchmark allowed.
• Even on the CC’s own reckonings, Auckland Airport’s landing charges in 2000 were
less than 3 per cent above what was estimated to be a competitive benchmark
price. Wellington and Christchurch airports’ charges were estimated to be below the
estimated competitive benchmarks. The difficulty of valuing land makes 3 per cent
an insignificant deviation from zero.
The high incidence of litigation between airports and airlines appears to have been
driven by uncertainty regarding the interpretation of an airport’s legislative requirement
to ‘consult’.
The CC has delayed the release of its final recommendations by six months to
consider further the substantive issues raised in response to its draft report.

While light-handed regulation currently is not the dominant model worldwide, there
are signs that the direction is changing. For example, in the United Kingdom, the
Civil Aviation Authority recently has proposed modifications to regulatory
arrangements that are in part designed to foster commercial negotiation between
airports and airlines at major airports.

XLIV OVERVIEW
The Commission recognises that its preferred option would involve a considerable
shift from current arrangements at the four major airports — though one largely
envisaged by the architects of the regulatory arrangements put in place at the time
of airport privatisations. The Commission also recognises that some parties may
find such a transition difficult, particularly given the long history of government
provision of airport services at those airports, and pricing structures that have
effectively subsidised aeronautical charges. Nonetheless, even if price caps were to
be maintained (in any form), aeronautical charges would still need to rise to
encourage efficient long-run service provision.

The Commission considers that the full benefits of privatisation of airports are
unlikely to be realised if commercial relationships between airports and airlines
continue to be heavily conditioned by intrusive price regulation. The ongoing need
for substantial investments at major airports requires a more commercial and
cooperative approach. The potential for heavy-handed regulation unduly to
constrain prices and commercial relationships poses a real risk and one that could
impose large costs on consumers in the future.

The Commission emphasises that it is not advocating deregulation of major airports.

It is proposing a probationary regulatory package designed to facilitate the transition
to a more commercial environment, while providing credible constraints on the use
of market power by these airports.

Whether or not the Commission’s preferred option is adopted, it is essential that,

prior to implementation of the chosen regulatory approach, parties are consulted on
the practicalities of proposed regulation and made aware of the various
requirements, in order to reduce uncertainty and the potential for disputation. In
particular, bidders for Sydney Airport should have a clear picture of the regulatory
framework for that facility, so that expected future airport charges can be factored
adequately into the sale price.

OVERVIEW XLV
Recommendations and findings

Recommendations
Recommendations 1–8 should be read in conjunction with chapter 12, which sets
out the details of Options A and B.

RECOMMENDATION 1

For Sydney, Melbourne, Brisbane and Perth airports, price caps and prices notification
arrangements should be replaced by mandatory price monitoring arrangements for a
probationary five-year period, as outlined in Option B.
• Airport-specific price-monitoring arrangements could be incorporated either in the
Airports Act 1996 (Airports Act) or the Trade Practices Act 1974 (TP Act), but
should be consistent with any generic price-monitoring provisions that may be
introduced into the TP Act following the Commission’s separate review of the Prices
Surveillance Act 1983 (PS Act).

In the event that the Government opted for a stricter form of price regulation at these
four airports, Option A should apply such that:
• annual price caps of the form CPI-X continue for five years at Melbourne, Brisbane
and Perth airports. Price caps should be set to reflect the efficient costs of providing
aeronautical services in the long run, on a dual-till basis. Price caps should be
complemented by price monitoring of some ‘aeronautical-related’ services; and
• for a capacity-constrained Sydney Airport, prices should not be required to fall in
real terms. Regulation should comprise either prices notification or a price cap of the
form CPI-X, with X set at zero. Price increases should be allowed to reflect peak-
period demand and to accommodate necessary investment.

RECOMMENDATION 2

For Adelaide, Canberra and Darwin airports, mandatory price monitoring of

aeronautical services and some ‘aeronautical-related’ services (as outlined under Option
B) should continue for five years. (Airport-specific price-monitoring arrangements could
be incorporated either in the Airports Act or the TP Act, but should be consistent with
any generic price-monitoring provisions that may be introduced into the TP Act
following the Commission’s separate review of the PS Act.)

XLVI PRICE REGULATION

OF AIRPORT
SERVICES
 RECOMMENDATION 3

Quality monitoring of regulated services (as outlined under Option B) should continue
at all airports subject to price regulation; that is, at Sydney, Melbourne, Brisbane, Perth,
Adelaide, Canberra and Darwin airports.

RECOMMENDATION 4

Neither price monitoring nor price caps should be reintroduced for Alice Springs,
Coolangatta, Hobart, Launceston and Townsville airports. The Airports Act should be
amended so that these airports are no longer designated as ‘core-regulated’ airports.

RECOMMENDATION 5

Commercial agreements should be encouraged and assisted (for example, by providing

guidelines regarding coverage) under price-monitoring arrangements, or price caps, if
they were retained at some airports.

RECOMMENDATION 6

Price regulation of airports should be reviewed towards the end of the five-year
regulatory period. The review should be independent and public. Its objective should be
to ascertain the need for any future price regulation of airports (including price
monitoring or more stringent price regulation). In making its assessment, the review
should be guided by principles of efficient pricing plus several other criteria set out
under Option B. Agreed review criteria should be spelt out at the beginning of the
regulatory period.

Other airports should be included in the review only where there was prima facie
evidence of persistent misuse of market power (namely, evidence of inefficient prices,
poor quality etc).

RECOMMENDATION 7

All airports should be subject to the generic provisions of the National Access Regime in
Part IIIA of the TP Act. An airport-specific access regime should be considered only if
procedural improvements, such as scope for multilateral arbitrations, are not made to
the National Access Regime.

RECOMMENDATION 8

Prior to implementation of the chosen regulatory approach, airports and airlines should
be consulted on the practicalities of the proposed regulation and made aware of its
various requirements, in order to reduce uncertainty and the potential for disputation. In
particular, bidders for Sydney Airport should have a clear picture of the regulatory
framework for that facility so that expected future airport charges can be factored
adequately into the sale price.
RECOMMENDATIONS XLVII
AND FINDINGS
Findings
This section draws together all findings contained in this report. Findings are listed
under the relevant chapter.

Chapter 5 Market power of airports

FINDING 5.1

There are significant barriers to entry in the provision of airports. These arise from
natural monopoly characteristics that are reinforced by regulatory constraints.

FINDING 5.2

The price elasticity of demand for the services of a particular airport is influenced
by a number of factors. Although the typically low proportion of airport charges in
airfares and airline costs suggests low price sensitivity, this will be mitigated by any
potential for destination, modal and airport substitution, and the supply responses
of other input providers to changes in airport charges.
Airports that face more significant substitution possibilities will face more
price-sensitive demand (and hence have lower market power).

FINDING 5.3

Of the core-regulated airports, Sydney, Melbourne, Brisbane and Perth have most
market power. Adelaide Airport is likely to have a moderate degree of market
power. Canberra and Darwin airports also are likely to have a moderate degree of
market power, though they appear to face stronger substitution possibilities than
Adelaide Airport.
Core-regulated airports that do not appear to have significant market power (due
mainly — except for Townsville — to the scope for effective inter-airport
competition) are Alice Springs, Coolangatta, Hobart, Launceston and Townsville.

Chapter 6 Market power in particular airport services

FINDING 6.1

For those airports with a moderate to high degree of market power, the extent of
market power varies across the services provided. It appears to be most significant
in aircraft movement facilities, vehicle access, some forms of passenger processing
facility and aircraft refuelling. With respect to aircraft refuelling, market power
appears to be most significant at Brisbane, Perth and Sydney airports.
XLVIII PRICE REGULATION
OF AIRPORT
SERVICES
Where other providers potentially compete with the airport in the provision of
services, access may be an issue if these providers require access to the airport site.

Chapter 7 Conduct of unregulated airports

FINDING 7.1

Profits from non-aeronautical activities at most core-regulated airports appear

significant, especially when compared with current earnings from regulated
aeronautical charges. Though this earnings disparity may be expected to be
somewhat less if price regulation of aeronautical services were removed, there is an
incentive for airports to temper prices for aeronautical services (particularly for
additional services and new entrants), improve quality and/or increase aeronautical
capacity to encourage passenger growth and non-aeronautical revenue. The
magnitude of non-aeronautical earnings suggests that the effects on aeronautical
prices would be significant.

FINDING 7.2

The countervailing power of airlines in their dealings with major capital city
airports appears limited. However, airlines may have a degree of countervailing
power even at those airports where there is scope for airport substitution (for
example, entry ports for international flights), where airlines form alliances and
bargain as a group, or where selective threats can be made to reduce services that
are highly profitable to airports.
For smaller core-regulated airports, airline countervailing power is likely to be
stronger, due to the commercial strength of major airlines relative to smaller
airports, the market segments served by those airports and greater scope for airport
competition.

FINDING 7.3

Non-airline users of airports do not appear to have significant countervailing

power, though representative bodies acting on behalf of these users may exert some
influence.

FINDING 7.4

Airport operators have a strong incentive to discriminate in pricing in order to

increase their profits. Evidence suggests that they have scope to do this by applying
different pricing structures, including low entry prices and other incentives for
marginal flights and new airlines. Coupled with the ability of airlines to
discriminate in pricing among passengers, such pricing is likely to reduce any

RECOMMENDATIONS XLIX
AND FINDINGS
efficiency losses arising either from the need to cover the fixed costs of providing
aeronautical services or from the exercise of airport market power.

FINDING 7.5

At capacity-constrained facilities (where the shortage of capacity is not created or

perpetuated by the airport), efficiency requires that capacity is rationed such that
consumers with higher valuations of the service obtain access. Where capacity is
restricted, airports not subject to price regulation would generally have an
incentive to set aeronautical prices that promote efficient use of the airport.

FINDING 7.6

Aeronautical charges at some airports could be expected to rise if unconstrained by

regulation, in part because current charges may be below the level required to
cover the costs of investment. Whether and to what extent prices rise above levels
required for efficient provision of aeronautical services, as the result of the exercise
of market power, cannot be forecast with certainty. However, there is reasonably
strong evidence that a range of market influences will moderate any increase in
prices and any negative efficiency effects of higher prices. Chief among these are
the effects of non-aeronautical earnings and the scope for discrimination in pricing
so that airport users are not discouraged from using airport facilities.
Higher airport charges could redistribute income from airlines and their
passengers to airports. The burden of such higher charges would be likely to fall on
both Australian residents and non-residents.

FINDING 7.7

Managers of privatised airports with market power are unlikely to have much scope
to allow inefficiencies in production.

FINDING 7.8

In principle, airports with market power may have some incentive to delay
investment or to allow quality to deteriorate, in order to maximise their profits.
However, in practice, to the extent that airports with market power can discriminate
in pricing or differentiate products for different users, those incentives will be
weakened. The scope to earn additional non-aeronautical profits from higher
quality or expanded aeronautical capacity and passenger throughput, will also
encourage the provision of appropriate quality and investment levels.

L PRICE REGULATION
OF AIRPORT
SERVICES
 FINDING 7.9

An airport with market power has little incentive to deny or frustrate access to its
major customers, the airlines.
An airport with market power may have an incentive to restrict ‘front-door’ access
to off-airport providers of competing services such as car parking, or providers of
competing transport modes, though there is little evidence of this occurring.

Chapter 8 Review of airport regulation: price-cap and prices-

notification arrangements

FINDING 8.1

The single-till basis of the starting prices (also incorporating some cross subsidies
between airports), and the real declines in aeronautical prices at most airports
under the price cap suggest that, for many airports, aeronautical prices by the end
of 2000-01 may have been below the level necessary to justify future aeronautical
investment.

FINDING 8.2

The necessary new investment provisions have not promoted the commercially-
negotiated outcomes that were envisaged by the architects of the regime. This has
been partly due to the need to develop criteria and procedures for necessary new
investment after purchase and for participants to adapt to the very different
business environment following airport privatisation.
However, the observed difficulties also point to some fundamental problems. In
particular:
• the lack of transparency regarding what investment was considered to be
included in the base aeronautical prices and what was to be covered by
necessary new investment, with resultant effects on incentives to invest;
• the incentives for some participants to approach the regulator rather than
achieve commercially-negotiated solutions;
• the high costs of complying with the regime; and
• the regulatory risk due to the uncertainty and delays introduced by the need to
have every investment-related price increase vetted by the regulator.
FINDING 8.3

If regulation of Sydney Airport aeronautical charges continues to involve prices set

by the regulator on a production-cost basis, aeronautical land should be valued at
its opportunity cost rather than at indexed historical cost.
RECOMMENDATIONS LI
AND FINDINGS
FINDING 8.4

The aeronautical price increases implemented at Sydney Airport place its pricing
on a fundamentally different and more appropriate (dual-till) basis than that at
other core-regulated airports. The significant range of possible outcomes for a
number of cost parameters (for example, land values, cost of capital etc) indicates
the imprecision of regulatory price setting and the potential for inefficient prices
being established by regulation.
If significant excess demand returns at Sydney Airport at peak times,
production cost-related or rate-of-return based regulation would be likely to
constrain the setting of efficient aeronautical charges for those times.

Chapter 9 Review of airport regulation: price monitoring, quality

monitoring and access regulation

FINDING 9.1

The overall quality of service provision by airport operators at monitored airports

does not appear to have deteriorated since monitoring commenced. This outcome
reflects to a large extent the commercial incentives for airport operators not to
reduce service quality.
Although there are some problems with the quality monitoring process, it provides
transparency of reporting and facilitates some comparison between airport
operators at relatively low cost.

FINDING 9.2

Though privatised core-regulated airports have facilitated access for new entrant
airlines, airport-specific access provisions (s. 192 of the Airports Act) do not
appear to have been instrumental in achieving this outcome.

Chapter 10 Regulatory options: cost-based and incentive regulation

FINDING 10.1

Incentive regulation has some advantages over cost-based regulation. However,

given the tendency of price caps to converge towards cost-based regulation (with
associated high levels of regulatory involvement and risks of regulatory error, and
the consequences of these for investment and long-term provision of services),
price-cap regulation should be implemented only when there is clear evidence that,
without such regulation, economic efficiency would be seriously impaired.

LII PRICE REGULATION

OF AIRPORT
SERVICES
Where price caps are implemented, the approach adopted for investment should be
spelled out clearly and transparently to all relevant parties from the outset, in order
to reduce the risk of inefficient outcomes and excessive gaming.

FINDING 10.2

The Commission considers that if price-cap regulation were to be continued for any
airports, then the cap should apply only to those aeronautical services in which the
airport has substantial market power. Profits earned in non-aeronautical activities
should not be taken into account in setting the cap.
If an airport exercises significant market power (as opposed to earning locational
rent) in any non-aeronautical activity, separate price monitoring or other
regulation may be appropriate.

Chapter 11 Regulatory options: price monitoring, access provisions

and general competition law

FINDING 11.1

Where airport market power is not substantial, or where there are commercial
constraints on the misuse of market power, price monitoring has significant
advantages over stricter forms of price regulation. Provided there is no easy
recourse to regulatory intervention, a price-monitoring regime can promote
efficient outcomes while reducing the risk of regulatory failure. Price monitoring
also has the potential to reduce compliance costs and promote commercial
negotiations.

FINDING 11.2

There do not appear to be any grounds for airport-specific access arrangements

that set lower declaration thresholds than Part IIIA of the TP Act. An airport-
specific regime that allowed for multilateral arbitrations might be warranted if such
arbitrations were not provided for in the National Access Regime, as recommended
separately by the Commission in its review of that regime.

FINDING 11.3

There are insufficient grounds for an airport-specific access undertaking regime.

FINDING 11.4

Facilitation and encouragement of agreements between airport operators and users

has the potential to promote commercial relationships and efficient outcomes.

RECOMMENDATIONS LIII
AND FINDINGS
1 Introduction

This chapter sets out and clarifies the terms of reference, provides background to
this inquiry, and outlines the Productivity Commission’s approach to the inquiry.

1.1 Terms of reference

The Commonwealth Government has asked the Productivity Commission to report
on whether there is a need for price regulation of airports, and the appropriate form
of any price regulation. The terms of reference for the inquiry are reproduced at the
beginning of this report.

The terms of reference state that the purpose of the inquiry is to:
… examine whether new regulatory arrangements, targeted at those charges for airport
services or products where the airport operator has been identified as having most
potential to abuse market power, are needed to ensure that the exercise of any such
power may be appropriately counteracted. (terms of reference, para. 6)

In so doing, the Commission is to take into account several principles, the first of
which states:
The CPI-X price cap applied to aeronautical charges during the first five year period of
private ownership will no longer operate. (terms of reference, para. 7(a))

Shortly after commencement of the inquiry, the Commission wrote to the Assistant
Treasurer, seeking clarification of paragraph 7(a) of the terms of reference. In its
letter, the Commission stated:
The Commission understands that Paragraph 7(a) essentially conveys the
Government’s intention that the current price cap arrangements for Phase I and II
airports will cease after five years of operation. It is also our view that paragraph 7(a) is
not intended to preclude from consideration price cap arrangements of the general form
of CPI-X as an option for future prices regulation of airport services, should some form
of prices regulation be regarded as appropriate.

The Commission sought confirmation from the Assistant Treasurer that this general
approach to price regulation could be considered as one of the range of options
where there is most potential for abuse of market power. The Commission’s

INTRODUCTION 1
understanding of paragraph 7(a) was confirmed by the Assistant Treasurer in
correspondence with the Commission (appendix A).

Other principles that the Commission is to take into account include that price
regulation:
• be applied to those aeronautical services and to those airports where airport
operators have most potential to abuse market power;
• minimise compliance costs on airport operators and the Government;
• promote the efficient operation of airports; and
• facilitate benchmarking comparisons between airports, competition in the
provision of services within airports, and commercially negotiated outcomes in
airport operations.

1.2 Background to the current inquiry

In 1997, the Commonwealth Government commenced the sale to private operators
of long-term leases (50 years with an option to renew for a further 49 years) for 17
of the 22 airports operated at the time by the Federal Airports Corporation (FAC)
(figure 1.1).

Leases were granted to private sector operators at Brisbane, Melbourne and Perth
airports in 1997 (Phase 1), and for a further 14 airports (Adelaide, Alice Springs,
Canberra, Coolangatta, Darwin, Hobart, Launceston, Townsville, Mount Isa,
Tennant Creek, Archerfield, Jandakot, Moorabbin and Parafield airports) in 1998
(Phase 2). The remaining five federal airports — the four Sydney basin airports
(Sydney (Kingsford Smith), Bankstown, Camden and Hoxton Park) and Essendon
Airport — were leased but not privatised, with their ownership transferred to two
wholly Government-owned companies in 1998. The FAC subsequently ceased
operation.

The sale of Essendon Airport was completed in September 2001. The sale of
Sydney Airport has been deferred from late 2001 to early 2002. The three other
Sydney basin airports are to be sold in the second half of 2002.

At the time of the sale of leases, 12 of the 22 leased airports were designated as
‘core-regulated’ airports under the Airports Act 1996 and were subject to price
regulation under the Prices Surveillance Act 1983 (PS Act). They comprised
Sydney Airport and 11 privatised airports (all Phase 1 airports, and eight of the
Phase 2 airports — Adelaide, Alice Springs, Canberra, Coolangatta, Darwin,

2 PRICE REGULATION
OF AIRPORT
SERVICES
Hobart, Launceston and Townsville). These airports are characterised by significant
interstate and, in some cases, international regular public transport services.

Figure 1.1 Airports leased from the Commonwealth Governmenta

a FAC airports at the end of 1996.

Price regulation of core-regulated airports included prices notification, price

monitoring, price-cap arrangements and special provisions for necessary new
investment at airports. These airports and the other leased airports were also subject
to regulation under the Airports Act and the Trade Practices Act 1974 (TP Act).

In 1996, the then Department of Transport and Regional Development

(DoTRD 1996) described the prices oversight arrangements as transitional
measures, designed to allow airport operators, users and the Government regulator
time to adjust to the new operating environment for airports. It was envisaged that
pricing arrangements would change after experience was gained in the new
operating environment. As such, a review of the pricing arrangements was a key
feature of the privatisation program developed by the Commonwealth Government.
The review will be based on the premise that the price cap applied to aeronautical
charges during the first five years will no longer operate. In its place, the review will
INTRODUCTION 3
 aim to develop arrangements targeted at those charges where the airport operator has
most potential to abuse market power. The existing set of aeronautical charges will be
examined, on an airport-by-airport basis, with the review assessing whether services
should be added or removed from surveillance. (DoTRD 1996, p. 7)

The review was to be completed before the end of the first five-year period of the
leases.

The Commonwealth Government subsequently decided that the review would be

undertaken by the Productivity Commission, rather than by the Australian
Competition and Consumer Commission (ACCC) as foreshadowed in the
Government’s Pricing Policy Paper issued in 1996 (DoTRD 1996), because:
There are strong synergies between this review and the Productivity Commission’s
current review of the Prices Surveillance Act 1983, which provides the framework
within which airport prices are regulated. There are also synergies with its current
reviews of telecommunications specific competition regulation and access
arrangements under Part IIIA of the Trade Practices Act 1974. (Kemp 2000)

The Productivity Commission commenced its review in December 2000. In October

2001, the Commonwealth Government announced changes to prices oversight
arrangements at airports to assist airports to cope with difficulties arising from
domestic and international events. Changes included the removal of price caps on
aeronautical charges at eight Phase 2 airports, and allowance of one-off price
increases at Melbourne, Brisbane and Perth airports of up to 6.2, 6.7 and
7.2 per cent, respectively (chapter 3). The Government noted that the changes were
consistent with the findings of the Commission’s draft report, released in August
2001. The arrangements are to be reviewed in light of developments in airport
prices and the Commission’s final report (Hockey 2001).

1.3 The Commission’s approach

The Commission’s approach takes into account the terms of reference for this
inquiry and the general policy guidelines in the Productivity Commission Act 1998.
Although this inquiry is not a legislative review under the requirements of the
Competition Principles Agreement (CPA), the principles embodied in this
agreement apply. The terms of reference for this inquiry broadly reflect the CPA.1

In considering the issues and formulating its draft recommendations, the

Commission has also taken into account previous Productivity Commission reports

1 In 1995, the CPA was signed by the Commonwealth, States and Territories as part of the
National Competition Policy reform package. The CPA, in essence, sets out the principles to be
followed by governments in relation to the agreed competition policy reforms.
4 PRICE REGULATION
OF AIRPORT
SERVICES
and current inquiries, particularly those noted above by the then Assistant Treasurer,
Senator Kemp.

The Commission’s approach, in broad terms, has been to:

• identify the objectives of price regulation;
• assess whether there is a need for price regulation of airport services by:
- identifying which (if any) airports possess market power and the potential to
abuse it;
- identifying those services in which airports may possess market power; and
- assessing the likely extent and potential impacts of abuse of market power by
airports, including efficiency and distributional effects;
• assess the costs and benefits, including efficiency and distributional effects, of
price regulation;
• identify options for future price regulation, taking into consideration pricing
regulatory arrangements in other countries, and analyse the costs and benefits,
including efficiency and distributional effects, of those options; and
• assess what form of price regulation is required, if any.

Scope of the inquiry

As noted above, the terms of reference ask the Commission to report on whether
‘there is a need for price regulation of airports, and the appropriate form of any
price regulation’. In other words, the terms of reference do not constrain the
Commission to considering only those airports subject to price regulation following
the granting of leases, that is, core-regulated airports. Other airports include those
leased in Phase 2 but not subject to price regulation, regional airports with regular
public transport services, and smaller general aviation airports. Nonetheless, the
inquiry has focused on whether there is a need for continued price regulation at
core-regulated airports and, if so, the appropriate form of such regulation.

Sydney Airport, which is leased but not yet privatised, has not been subject to the
price-cap arrangements of other core-regulated airports. As a core-regulated airport,
it is, however, subject to other forms of price regulation and therefore included in
this inquiry.

This report assesses price regulation as it applied to all core-regulated airports for
the period following the granting of leases until October 2001, when regulation
changed. It also examines price regulation since October insofar as the effects of the
regulatory change are evident between October and the finalisation of this inquiry.

INTRODUCTION 5
Although the regulatory focus is on price regulation under the PS Act, other
regulations are considered insofar as they affect, directly or indirectly, the prices of
services provided by airports. Examples of such regulation are the Airports Act, the
TP Act, environmental regulations, and international aviation agreements.

Since the release of the Commission’s draft report domestic and international events
have, at least in the short term, altered the environment in which airports operate.
The Commission has endeavoured to address the impact of these events in this
report.

Report structure

The airport business in Australia, particularly at core-regulated airports, is described

in chapter 2. Chapter 3 describes the regulatory environment affecting the operation
of Australian airports, focusing on price regulation and access arrangements. The
possible rationales for future price regulation of airport services, together with
principles of good regulation and efficient pricing of airport services, are discussed
in chapter 4. Chapters 5 and 6 examine potential sources of market power of
airports, and the extent to which market power exists for particular airports
(chapter 5) and services (chapter 6). Chapter 7 considers the possible pricing and
other behaviour of airports, if they operated without price regulation. The potential
efficiency and distributional effects of such behaviour are also discussed.
Chapters 8 and 9 evaluate the performance of the post-privatisation framework for
regulating prices: price-cap and prices-notification arrangements are examined in
chapter 8; and price monitoring, quality monitoring and access regulation are
addressed in chapter 9. Options for future regulation of airport services are
discussed in chapters 10 (cost-based and incentive price regulation) and 11 (price
monitoring, access and anti-competitive conduct). Chapter 12 outlines the
Commission’s preferred approach for future price regulation of airport services.

1.4 Conduct of the inquiry

The Commission received terms of reference for this inquiry on 21 December 2000.
As required by the terms of reference, and in line with normal Commission inquiry
procedures, the Commission has encouraged maximum public participation. Soon
after receipt of the terms of reference, it placed advertisements in the national press
and sent a circular to a range of individuals and organisations thought likely to have
an interest in the inquiry. An issues paper was released in early January 2001 to
assist participants in preparing their initial submissions.

6 PRICE REGULATION
OF AIRPORT
SERVICES
The Commission held informal discussions with organisations, companies, and
individuals to seek information and canvass a wide range of views. A list of those
with whom discussions were held is set out in appendix B. Fifty submissions were
received in response to the issues paper (appendix B). Interested parties were given
the opportunity to respond to matters raised in the draft report by way of written
submissions and at public hearings held in October 2001. Twenty-nine
supplementary submissions were received in response to the draft report
(appendix B). Transcripts of hearings and all non-confidential submissions (or
non-confidential parts of submissions) are available on the Internet, at Commission
and State libraries, and from Photobition Digital Imaging Centre.

The Commission thanks participants for their participation in meetings with

Commissioners and in public hearings, and for their submissions.

Professor Richard Snape (Presiding) and Dr Neil Byron were the Commissioners
for the majority of this inquiry. Due to the illness of Professor Snape after the
hearings on the draft report, the Productivity Commission’s Chairman, Gary Banks,
formally joined this Division of the Commission in order to provide additional
assistance to the inquiry. Due to these events, the final report was signed on
23 January 2002, somewhat later than the original reporting date.

INTRODUCTION 7
2 Australian airports and their markets

This chapter provides an overview of the airport business, particularly at core-

regulated airports. The services and facilities provided at airports by airport
operators, and others, are examined. Revenues, charges, costs and profitability of
core-regulated airports are assessed in detail. Demand characteristics of the services
provided by Australian airports are also discussed. Where relevant, the situation at
Australian airports is compared with major airports overseas.

The Australian commercial aviation sector experienced significant change in the

second half of 2001 with a substantial reduction in the number of services provided
by Australia’s then second-largest domestic airline, Ansett Airlines — as well as the
repercussions of the 11 September terrorist attacks in the United States. With the
most recent official data available to 1999-00 and 2000-01, the picture of the
industry described in this chapter is likely to change at least in the short to medium
term. At the time of writing it is not possible to assess confidently whether these
events will constitute a short-term aberration or the start of a fundamentally
different longer-term trend in the demand for aviation and airport services.

2.1 Airport facilities and services

Airports supply a range of products, but their primary function is to provide for
planes to land and depart, and to facilitate the interchange of passengers and freight
between air and surface transport. Airports range in size, the scope of services
provided and the markets served. They serve international, domestic, regional and
general aviation markets.

Airport facilities

Airports comprise a combination of landside and airside facilities. Landside

facilities cover terminals and the infrastructure within them, including flight
information display systems, check-in counters, public amenities, and lounges for
passengers. Terminals also provide space for offices and for commercial operations
such as car hire, retail, and food and beverage activities.

AUSTRALIAN 9
AIRPORTS AND THEIR
MARKETS
In addition, landside facilities encompass facilities outside the terminals, such as
perimeter roads, car parks, and taxi, bus and rail points linked to terminals by
walkways.

Airside facilities include runways, taxiways and aprons, as well as airfield lighting,
aircraft parking bays, visual navigation aids, hangars, freight terminals and facilities
for aircraft maintenance and refuelling, and in-flight catering.

Airport services

Of the range of services provided at airports in most OECD countries, only a small
number is directly supplied by the airport operator (section 2.3). The services and
activities provided at airports can be divided into three groups: essential operational
services, handling services and commercial activities (Doganis 1992).

Essential operational services

Essential operational services are concerned primarily with ensuring the safety of
aircraft and airport users. These include:
• air traffic control and meteorological services;
• runway, building and aircraft maintenance services;
• communications, security, and fire and medical services; and
• aircraft movement services for the runways, taxiways and aprons, such as nose-
in guidance and marshalling.

Handling services

Handling activities cover services directly associated with the aircraft itself
(sometimes referred to as ground handling) and include:
• cleaning and catering;
• the provision of power and fuel; and
• loading/unloading of baggage and freight.

Also commonly classified as handling activities are services associated with the
various stages of processing passengers, baggage and freight through the respective
terminals and onto the aircraft. These services include:
• baggage make-up, handling and reclaim, ticketing and check-in; and
• immigration, customs and quarantine services (for international flights).
10 PRICE REGULATION
OF AIRPORT
SERVICES
Commercial activities

Commercial activities cover a range of services and activities not directly related to
the interchange of passengers and freight between surface and air transport. They
include retail outlets, banks, restaurants and bars, car-hire desks, car parking, and at
some large airports, such as Frankfurt, can extend to activities such as conference
centres and hotels, cinemas, night clubs and supermarkets.

Airport capacity

The (physical) capacity of an airport refers to its ability to accommodate aircraft

movements. The overall capacity of an airport is influenced by the interaction of
several factors, some of which are listed below.
• Runway size and design — the runway is the largest single capital asset of an
airport and tends to be viewed as the controlling element of airport capacity. The
length, width and thickness of concrete of runways determine the type (size) of
planes that can land. Runway design factors include the number, spacing and
orientation of runways, and the nature of taxiways.
• Airport layout — this can affect both aircraft taxi time and the time taken to
unload and process passengers and/or cargo.
• Mix of traffic — capacity is higher at airports that have a larger proportion of
their traffic serviced by larger planes.
• Air traffic control — particularly minimum separation requirements that depend,
among other things, on plane size.
• Environmental constraints (both natural and regulatory), and general regulatory
constraints — for instance, weather conditions, visibility, noise restrictions,
movement caps, and curfews.
• Land availability — at the airport and in surrounding areas, as well as regulatory
constraints on land use, influences airport design and the potential for capacity
expansion.

2.2 Australian airports

While there are around 2000 operational airports and airfields in Australia, most are
small privately-owned facilities catering for general aviation aircraft (GAA). Only a
relatively small number (around 200) receive regular public transport (RPT)
services. Of the airports that received RPT services in 2000-01, the 10 largest
airports accounted for approximately 90 per cent of all scheduled passenger

AUSTRALIAN 11
AIRPORTS AND THEIR
MARKETS
movements.1 RPT services in Australia are divided into three categories:
international, domestic and regional.2 In 2000-01, domestic travel accounted for
68 per cent of approximately 76 million passenger movements, followed by
international passengers with 22 per cent and regional passengers with 10 per cent.

Core-regulated airports

Except for Cairns Airport, the largest airports in Australia are all core-regulated
(chapter 1). In 2000-01, core-regulated airports accounted for 89 per cent of
scheduled passenger movements. Sydney, Melbourne and Brisbane airports are by
far the largest airports (figure 2.1), together accounting for 67 per cent of total
passenger movements at Australian airports in 2000-01.

Figure 2.1 Passenger movements, Australia’s 13 largest airports, 2000-01a

Domestic International

30000

25000
Passenger movements (’000s)

20000

15000

10000

5000

0
in

lle
ey

gs
t
ne

ne

rth

de

rra

ta

on
ar
rn

w
at

vi
dn

rin
ba

ai
ur

st
Pe

ob
be
ai

ar
ng

ns
el

ce
bo
Sy

Sp
is

H
an

D
Ad

w
la
Br

un
el

To
oo

e
M

La
ic
C

Al

a Revenue passengers on RPT services. Domestic totals include (provisional) regional data.
Data source: DoTRS (unpublished).

1 Includes revenue passengers arriving at, and departing from, Australian airports on scheduled
international, domestic and regional RPT services. Revenue passengers are passengers paying
any fare and include passengers travelling on tickets acquired through frequent flyer schemes.
2 ‘Domestic’ and ‘regional’ are data classifications of revenue traffic carried on scheduled RPT
services performed within Australia and its territories. Domestic airline data refer to revenue
traffic carried by high-capacity aircraft (aircraft with more than 38 seats or a payload greater
than 4200 kg). Regional airline data refer to revenue traffic carried by low-capacity aircraft
(aircraft with 38 seats or less, or with a payload of up to 4200 kg) (DoTRS 2000b).
12 PRICE REGULATION
OF AIRPORT
SERVICES
In 2000-01, there were approximately 17 million passenger movements on
international RPT services at Australian airports and nearly 96 per cent of these
movements were at core-regulated airports, with Sydney accounting for around half.

The 10 largest airports in Australia in terms of RPT aircraft movements in 2000-01

are shown in figure 2.2 — nine of which are core-regulated. Of the remaining core-
regulated airports, Hobart Airport was ranked twelfth, Launceston Airport sixteenth,
and Alice Springs Airport eighteenth.

Figure 2.2 Aircraft movements at Australia’s 10 largest airports, 2000-01a

Domestic Regional International

300

250
Aircraft movements (’000s)

200

150

100

50

0
in

lle
ey

s
ne

ne

de

rth

rra

ta
rn

at

vi
dn

ba

ai
ur

Pe

be

ai

ar

ng

ns
el
bo
Sy

C
is

an

D
Ad

w
la
Br
el

To
oo
M

a RPT services. Includes (provisional) regional data.

Data source: DoTRS (unpublished).

The percentage breakdown of passenger movements at core-regulated airports by

airline sector is shown in figure 2.3. The share of international passengers is greatest
at Sydney and Perth airports, where international passengers comprise around
30 per cent of total passenger movements.

AUSTRALIAN 13
AIRPORTS AND THEIR
MARKETS
Figure 2.3 Passenger shares at core-regulated airports, 2000-01a

International Domestic Regional

100

90

80

70

60
Per cent

50

40

30

20

10

0
rth

ne

in

lle
ey

gs
t
ne

de

ta

rra

on
ar
w

at

vi
dn

rin
ba

st
ai
ur
Pe

ob
be
ar

ng

ns

ce
el
bo
Sy

Sp
is

H
an
D

Ad

w
la
Br

un
el

To
oo

e
M

La
ic
C

Al

a Based on revenue passengers carried on scheduled international, domestic and regional RPT services.
Includes (provisional) regional data.
Data source: DoTRS (unpublished).

Other airports

The 20 largest non-core-regulated airports (excluding Cairns) are shown in

figure 2.4. Of these airports, Mount Isa is a privatised ex-Federal Airports
Corporation (FAC) airport,3 and only Broome received (limited) international
traffic. Domestic airlines account for around 60 per cent of passenger movements at
these airports with the remainder made up of passengers on regional airlines.

In addition, a number of capital cities have large airports (in terms of aircraft
movements) that primarily serve the general aviation sector. These include
privatised ex-FAC airports such as Archerfield (Brisbane), Jandakot (Perth),
Moorabbin (Melbourne) and Parafield (Adelaide).

3 Australian Airports Limited owns and operates both Mount Isa and Townsville airports.
14 PRICE REGULATION
OF AIRPORT
SERVICES
Figure 2.4 Passenger movements at Australia’s 20 largest non-core-
regulated airports, 2000-01a

Domestic Regional

500

450

400
Passenger movements (’000s)

350

300

250

200

150

100

50

rt
Br e

lie

ga ove

a
ry
k)

am y
nd

on

Ka me

bo

ie

ne
Ka ur
ch n

ou a
ka

or

ur
th

llin
Is
w

po
g

rn
bu
oc

bo
or
oc sla

ub

to
pt

ag
to

oo

rra
yd

ild
ac

nt

Bu
on

Ba
R

go

Al
am

ds
ar

D
W
I

M
M

ev
rs

H
on

la
kh

illi

oo

M
ye

G
D
fs
ilt

ag
ar

of
(A

am

C
M

W
ra

H
la
Yu

a Excludes Cairns Airport. Revenue passengers carried on scheduled international, domestic and regional
RPT services. Included with the domestic passenger movements at Broome Airport are approximately
4200 international passenger movements. Includes (provisional) regional data.
Data source: DoTRS (unpublished).

2.3 The airport business at core-regulated airports

Australia has progressed further down the path of privatisation of major airports
than most other OECD nations. Most major airports around the world remain
publicly-owned facilities, although there is considerable variation in public
ownership structures across countries.4 Of the OECD countries, the United
Kingdom, Austria, Denmark, New Zealand, Germany and Japan are among those
moving (or which have moved) to privatisation (Button and Stough 2000). In 1997,

4 Ownership can take the form of a central government having control of all, or the majority, of a
country’s airports (such as Greece, Sweden and Norway); alternatively, ownership of airports
can reside with regional or local governments. Another variation on the public ownership
model is for the government to establish an authority to manage one or more airports on its
behalf. In the late 1970s and 1980s, several national airport authorities came into being,
modelled on the British Airports Authority, as it was prior to privatisation. Examples include
the Airports Authority of Thailand, the Israeli Airports Authority, the Mexican Airports
Authority and Aer Rianta (the Irish airports authority).
AUSTRALIAN 15
AIRPORTS AND THEIR
MARKETS
the Federal Aviation Administration in the United States introduced an airport
privatisation pilot scheme with the opportunity for five airports to be privatised —
by 2000, only two sales had been completed.5

Ownership

Nine unlisted companies manage Australia’s 11 privatised core-regulated airports.6

The Airports Act 1996 includes some limits on the amount of foreign ownership of
airport companies and on the degree of common ownership of certain pairs of
airports (chapter 3). Two overseas airport operators have interests in Australian
airport companies: BAA plc7 in Australia Pacific Airports Corporation (APAC),
which operates Melbourne and Launceston airports; and Amsterdam Airport
Schiphol in Brisbane Airport Corporation (BAC). A number of investment and
funds management companies hold substantial interests in airport companies. At
some airports, there is a degree of public ownership with local and State
governments maintaining an interest in the operating company.8

Sydney Airports Corporation Limited (SACL), a corporatised Commonwealth

entity, operates Sydney Airport. SACL also operates the three other Sydney basin
airports (Bankstown, Camden and Hoxton Park). As noted in chapter 1, Sydney
Airport is expected to be sold in the first half of 2002.

Structures and responsibilities

While the range of services and activities provided at airports varies across the core-
regulated airports, the division of responsibility for the provision of these services
(operational, handling and commercial activities) follows a similar pattern. As with
most major airports in developed countries (and as was the case with the core-
regulated airports’ previous operator, the FAC) the airport operator provides
relatively few services directly to airport users. Rather, Australian airport operators
tend to provide and maintain facilities that allow other organisations to carry out
their activities. These organisations include government agencies such as

5 A 99-year lease on Stewart International in New York State has been sold to National Express,
and Niagara Falls International has been sold to a Spanish company, CINTRA (TRL 2000b).
6 Australia Pacific Airports Corporation operates Melbourne and Launceston airports, and
Northern Territory Airports Group operates Darwin and Alice Springs airports.
7 BAA plc operates a number of airports in the United Kingdom, including Heathrow, Gatwick,
Glasgow and Edinburgh airports.
8 For example, the Port of Brisbane Corporation, a wholly State-owned enterprise, has interests
in BAC, and Launceston City Council has an equity share in Australia Pacific Airports
(Launceston) Pty Ltd.
16 PRICE REGULATION
OF AIRPORT
SERVICES
Airservices Australia (ASA), airlines (and other specialist providers of operational
and handling services) and concession holders.

ASA, a government-owned commercial authority, is responsible for air traffic

control and airspace management services at all core-regulated airports and 14 other
major Australian airports where traffic levels are sufficiently large to require air
traffic control services. ASA also delivers aeronautical information,
communications and, at 16 major airports (including all core-regulated airports),
rescue and firefighting services.

While ASA is responsible for air traffic control and airspace management services,
it is not responsible for allocating landing and take-off slots. The Airports Act
(described in chapter 3) provides for the Minister of Transport and Regional
Services to authorise a body to be the slot coordinator at Australian airports.
Currently, Airport Coordination Australia (ACA), a private company, is the
authorised slot coordinator. ACA provides timeslot management services for all
international airline services and new entrant airlines servicing domestic routes at
Australian airports, and all operations at Sydney Airport.9

Other Commonwealth government agencies that provide services at core-regulated

airports include the Civil Aviation Safety Authority (which conducts safety
regulation of civil air operations), the Australian Customs Service, the Australian
Quarantine and Inspection Service, the Department of Immigration and
Multicultural Affairs, and the Australian Federal Police.

Airport operators are required to provide a number of government-mandated

security services. The seven core-regulated airports with international RPT services
are required to engage Australian Protective Services, a Commonwealth
government security agency, to provide a counter-terrorist first response capability.
Other required services are security screening of passengers and hand-baggage, and
checked-baggage. Most airport operators engage specialist companies to provide
these services (there is no requirement to engage any particular provider), although
there is some variation across airports as to whether the airport operators supply the
equipment required for passenger and baggage screening, or lease it from a
specialist provider.

Australian airport operators also provide and maintain runways, taxiways, visual
navigation aids, airfield lighting, roads and aircraft parking areas. Usually, ground-
handling services such as cleaning, the provision of power, baggage handling and

9 Slot management at Sydney Airport is regulated separately from other airports under the
Sydney Airport Demand Management Act 1997 (chapter 3).
AUSTRALIAN 17
AIRPORTS AND THEIR
MARKETS
other operational services (for example, aircraft push-back and towing, and ground
engineering) are supplied by the airlines and/or independent suppliers.

Airport operator involvement in other airside activities, such as refuelling and

freight storage, is limited to the provision of basic infrastructure, or to the provision
of land to allow others to provide the infrastructure and services. At larger airports,
oil companies often provide both refuelling facilities and services at airports, the
airport operator provides land on which oil companies can build storage and
reticulation equipment. Oil companies often operate the storage and reticulation
systems as a joint venture (referred to as a joint user hydrant installation (JUHI)).
Airports historically have charged for the use of land to build refuelling facilities by
way of rent payable under leases and licence fees (appendix E). However, at small
airports where fuel is trucked to aircraft, the situation may differ: the airport
operator provides facilities for access to the aircraft and it may also provide for
storage of fuel and equipment (ACCC 1998b).

Landside, airport operators manage international and domestic common-user

terminals and the infrastructure associated with these facilities, such as check-in
counters, baggage reclaim and handling systems, escalators, flight information
systems and passenger lounges. Airlines tend to operate the check-in and baggage
and freight facilities. Airport operators also manage the provision of retailing and
advertising services at these terminals.

The arrangements for the management of the domestic terminals used by Australia’s
major domestic carriers, Ansett and Qantas Airways, have been quite different from
those at the common-user terminals (both domestic and international). Since the late
1980s, the two major domestic airlines have operated their own domestic terminals
under long-term leases negotiated prior to the establishment of the FAC. Under the
leases, which at most core-regulated airports run to around 2018, airlines are
responsible for all operational aspects at the terminal, including managing the
provision of retailing and advertising activities. In addition, at some airports,
including Melbourne and Sydney, airline responsibility extends to providing and
maintaining terminal infrastructure, with the airport operator providing only the
land for the domestic terminals under the leases. At other airports, such as Brisbane,
the airport operator provides and maintains the terminal infrastructure. While
Qantas continues to operate its domestic terminals under these arrangements, the
future of Ansett’s domestic terminals is uncertain. Currently, Ansett’s domestic
terminals are operated by Ansett’s administrator and are being used by Ansett
Mark II.

The airport operators (or, in the case of domestic terminals, the airlines) engage
specialist operators to provide commercial activities. Retail, car-hire and catering
activities are delivered under this type of arrangement. Car-parking management
18 PRICE REGULATION
OF AIRPORT
SERVICES
appears to be the only commercial activity that some airport operators perform
directly. Airport operators at Brisbane, Perth and Sydney airports, for example,
manage some or all of their car-parking services directly, whereas operators at
Melbourne and Canberra airports have engaged the services of a car-parking
management company.

International comparisons

The arrangements for the provision and management of airport facilities and
services at core-regulated airports generally are similar to those in place at major
airports in other developed countries. Although the ownership structures differ
significantly, airport operators do not tend to provide many services directly.
Instead, as is the case in Australia, they manage the facilities and infrastructure that
allow airlines and other organisations to carry out their business.

Common to nearly all airports is that local or central government authorities provide
air traffic control and navigation services. Even when the airport operator provides
these services, government policies or international arrangements (chapter 3)
heavily influence the nature of the service.

However, at some airports there are significant differences in the organisational

structure of service provision. Some major European airports are directly involved
in the provision of ground-handling services and, until recently, a number of
airports were the sole providers of these services (such as Frankfurt and Milan
airports). This situation altered as a result of a European Commission directive
aimed at introducing competition in ground-handling services at all European
Community airports.10

Some airport operators also are directly involved in the provision of commercial
activities. For example, Aer Rianta, the Irish airport authority, operates the duty and
tax free shops at its airports (including Dublin) and, in the case of Shannon Airport,
it operates catering and bars (Aer Rianta no date). At Fiumicino Airport in Rome,
the airport authority also operates a number of retail outlets (ADR no date).

Perhaps the most significant difference between the organisational structure of core-
regulated airports and those of airports overseas is in the management of domestic
terminals. Airline management of their own terminal infrastructure is unusual in an
international context. The only other developed country where this type of
arrangement is common is the United States, where airlines own or lease most

10 Council Directive 96/67/EC of 15 October 1996 on access to the ground-handling market at

Community airports.
AUSTRALIAN 19
AIRPORTS AND THEIR
MARKETS
terminals in major airports (although, as in Australia, major airports often have a
common-user terminal that is managed by the airport operator).

Demand for airport services

The demand for airport services is, in general, a derived demand: the demand for
airport services is derived from the demand for airline services; in turn, the demand
for airline services is derived from the demand for business meetings, for visiting
friends and relatives, for migration, for tourism, for freight handling and so on. The
demand for airport services is also filtered through the cost structure of other input
providers, including airlines. This section describes in broad terms the different air
travel segments served by Australian airports, and changes in demand over time.
Demand elasticities of airport services are discussed in chapter 5.

Air travel and airports

Most international visitors come to Australia for tourism purposes (figure 2.5) —
around 56 per cent in 1999. These visitors to Australia must choose an international
gateway airport, the five largest being Sydney, Brisbane, Melbourne, Perth and
Cairns.

Figure 2.5 International visitor arrivals, by purpose of visit, 1991–99a

Business Visiting friends and relatives Other Holiday Total

5000

4500

4000

3500
Passengers (’000s)

3000

2500

2000

1500

1000

500

0
1991 1992 1993 1994 1995 1996 1997 1998 1999

a Refers to foreign visitors only (does not include Australian travellers).

Data source: ABS (Overseas Arrivals and Departures, Australia, Cat. no. 3401.0).

20 PRICE REGULATION
OF AIRPORT
SERVICES
The airlines in the domestic market link the principal cities throughout Australia by
operating high-capacity jet aircraft. In 2000, 10 routes (city pairs) accounted for
over two-thirds of all passenger movements in Australia (table 2.1). For each
principal city, the domestic market is concentrated at one airport. The regional
market generally links the smaller centres with the larger principal cities, with each
regional city served by one airport.

Table 2.1 Major routes (city pairs) by passenger movements, 2000a

Rank City pair Passengers

1 Melbourne – Sydney 5 877 971

2 Brisbane – Sydney 3 734 524
3 Brisbane – Melbourne 1 875 504
4 Adelaide – Melbourne 1 373 578
5 Coolangatta – Sydney 1 250 460
6 Adelaide – Sydney 1 180 389
7 Perth – Sydney 1 043 461
8 Brisbane – Cairns 994 818
9 Melbourne – Perth 978 874
10 Canberra – Sydneyb 907 688
Other 8 854 557
Total 28 071 824
a Includes passengers carried on international flights operated by Australian airlines over these city pairs.
b Includes passengers carried by subsidiary regional airlines.

Source: DoTRS (2001).

During the 1990s, passenger movements increased significantly at core-regulated

airports (figure 2.6). Over the period 1991-92 to 2000-01, passenger movements
increased by more than one-third at all core-regulated airports, except Alice
Springs, Coolangatta and Launceston. Passenger movements increased at
Coolangatta and Launceston airports by 26 and 14 per cent respectively over the
same period and remained virtually unchanged at Alice Springs Airport. While
growth in passenger movements has generally been strong, growth in international
passenger movements at some of the major gateway airports (particularly Brisbane
and Cairns) slowed in the late 1990s, partly as a result of the economic downturn in
Asia. Passenger movements increased substantially at a number of core-regulated
airports in 2000-01 with the entry of two discount airlines to the domestic aviation
sector.

AUSTRALIAN 21
AIRPORTS AND THEIR
MARKETS
Figure 2.6 Passenger movements at core-regulated airports, 1991-92 to
2000-01a
Passenger movements (’000s)

Sydney Melbourne Brisbane Perth

30000

25000

20000

15000

10000

5000

0
1991-92 1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-00 2000-01

Adelaide Canberra Coolangatta Darwin

5000

4500

4000

3500

3000

2500

2000

1500

1000

500

0
1991-92 1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-00 2000-01

(Continued on next page)

22 PRICE REGULATION
OF AIRPORT
SERVICES
Figure 2.6 (continued)

Hobart Alice Springs Townsville Launceston

1200

1000

800

600

400

200

0
1991-92 1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-00 2000-01

a Based on revenue passengers carried on scheduled international, domestic and regional RPT services.
Includes (provisional) regional data.
Data source: DoTRS (unpublished).

Demand for airport services in 2001

The first half of 2001 was characterised by high growth in passenger movements on
major domestic routes. The entry of two airlines providing discount services in the
domestic market in 2000 — Impulse Airlines in June and Virgin Blue in August —
contributed to this growth. The total number of domestic passenger movements over
the first six months of 2001 was around 10 per cent higher than the corresponding
figure in the previous year.11

However, in the latter half of 2001 the aviation industry altered substantially.
Impulse Airlines ceased independent operations in May, and in September
Australia’s second-largest domestic carrier, Ansett, entered into voluntary
administration and briefly suspended all operations. By October 2001, Ansett had
resumed operations as Ansett Mark II, albeit with significantly reduced capacity. As
at January 2002, questions over the longer-term structure and organisation of the
airline remained unresolved. This, combined with a general reduction in

11 Includes regional passenger movements (DoTRS, unpublished data).

AUSTRALIAN 23
AIRPORTS AND THEIR
MARKETS
international air travel worldwide in response to the terrorist attacks in the United
States on 11 September and their aftermath, resulted in a significant reduction in
flights and consequently in the demand for airport services. The initial impact of the
reduction in Ansett services was felt heavily at all core-regulated airports, in
particular those dependent on domestic airlines. At Coolangatta Airport, for
example:
In the week following the Ansett collapse domestic seat capacity … was reduced to
64 per cent of that for the week before the collapse. For the week ending 4 October this
had fallen to 62 per cent before increasing to 67 per cent for the week ending
11 October. (sub. DR58, p. 2)

In October 2001, Adelaide Airport noted that the reduction in Ansett services
resulted in:
… a loss of around 182 weekly flight arrivals equivalent to 43 per cent of the Airport’s
domestic landed tonnes/week and a similar short-term reduction in available seat
capacity to and from the city. (sub. DR57, p. 3)

Since September 2001, Ansett Mark II has resumed some services, and Qantas and
Virgin Blue have moved to increase capacity on a number of routes or introduced
services on new routes. However, domestic airline capacity is still below the level
that existed when Ansett was fully operational.

These changes in the structure of the airline industry are expected to result in
Qantas having a significantly larger share of the domestic aviation market in the
short to medium term. In the medium term industry analysts have predicted that
Qantas’ share of the domestic commercial aviation market could remain at
65 per cent at least and potentially much higher (Bartholomeusz 2001; Ferguson
and Meyrick 2001).12

Sources of airport revenue

In 1999-00, total revenue earned by the operators of core-regulated airports was

approximately $783 million. Sydney, Melbourne and Brisbane airports together
accounted for approximately 80 per cent of this revenue.

Airport revenue can be divided into aeronautical and non-aeronautical revenue.

Generally, aeronautical revenue refers to the revenue earned by the airport operator
from the provision of a number of key essential operating and handling services and
facilities.

12 Prior to September 2001, Ansett had 37 per cent of the domestic market, Qantas had
56 per cent and Virgin Blue had 7 per cent (Ferguson and Meyrick 2001).
24 PRICE REGULATION
OF AIRPORT
SERVICES
In Australia, for core-regulated airports over the period 1997-98 to 2000-01,
aeronautical revenue was defined to include revenue earned by airports from
services subject to prices notification under the Prices Surveillance Act 1983
(chapter 3). Included was revenue derived from the provision of a number of
essential operating and handling services and facilities, with most aeronautical
revenue earned from runway charges and (international and domestic common-user)
terminal charges.13 From 1998-99, revenue from vehicle access charges levied on
taxis at Brisbane, Perth and Canberra airports was classified by the ACCC as
aeronautical revenue (appendix E). At all three airports in 1999-00, revenue from
these vehicle access charges amounted to less than 1 per cent of total aeronautical
revenue.

In 1999-00, across all core-regulated airports, only 34 per cent of total revenue was
earned from aeronautical services. Coolangatta Airport reported the highest
proportion of revenue from aeronautical services — 55 per cent. In 1995-96, prior
to privatisation, when all current core-regulated airports were operated by the FAC,
the average share of revenue from the same bundle of aeronautical services at the
same airports was 38 per cent. Since 1995-96, the share of revenue earned from
non-aeronautical services has increased at all privatised core-regulated airports apart
from Coolangatta and Townsville airports (figure 2.7).

A number of airport operators — such as BAC (sub. 8) and Westralia Airports

Corporation (WAC, sub. 21) — stated that the potential growth opportunities from
non-aeronautical services were significant factors contributing to their decision to
become involved in the airport business in Australia, and the final price paid. As
WAC noted:
The bidders for [Perth Airport] were attracted to statements in the information
memorandum that highlighted … unexploited commercial potential in retail, trading,
car parking, ground transport and property development. (sub. 21, p. 6)

Retail operations, in particular, have become an important source of revenue for

airports, and an area that airport operators have sought to develop since
privatisation. APAC (2000a) noted that revenue from retail operations has increased
by 45 per cent at Melbourne Airport since privatisation.14

13 Also included in aeronautical revenue were airport charges levied on airlines to cover the costs
of services provided by Australian Protective Services.
14 Retail revenue is a component of non-aeronautical revenue. In this instance, retail revenue
refers to four revenue streams: (1) duty free, food and beverage, currency exchange and
speciality stores within the international terminal; (2) short- and long-term car parks; (3) car-
rental concessions; and (4) advertising sites.
AUSTRALIAN 25
AIRPORTS AND THEIR
MARKETS
Figure 2.7 Comparison of revenue shares: FAC and core-regulated
airportsa

Aero FAC (1995-96) Aero (1999-00) Non-aero FAC (1995-96) Non-aero (1999-00)

90

80

70

60
Per cent

50

40

30

20

10

0
in

lle
gs

ey
t
de

ne

rra

ta

ne

rth
on
ar
w
at

vi
rin

dn
ba
ai

ur
st
ob

Pe
be

ar
ng

ns
ce
el

bo

Sy
Sp

is

H
an

D
Ad

w
la
Br

un

el
C

To
oo
e

M
La
ic

C
Al

a The high share of non-aeronautical revenue at Adelaide Airport may be explained partly by the importance
of property revenue to the airport. Property revenue accounts for around half of Adelaide Airport’s total
revenue, with major sources including the domestic terminals leased to Qantas and Ansett, and a light-
industrial business precinct.
Data sources: ACCC (2001b–f); FAC (1996).

As noted above, airport operators generally do not provide retail services directly;
rather, they provide space for retailers to carry out their business. The payment
agreements between airport operators and retailers are based primarily on
concession fees (whereby airport operators receive payments based on the volume
of revenue generated by the retailer),15 often complemented by a small fixed annual
rental component. Airports can be an attractive location for retailers because they
provide access to potentially large volumes of relatively affluent customers
(Melbourne Airport, sub. 7). Doganis (1992) noted that concession income is
strongly related to passenger numbers generally, and to the number of international
travellers in particular. The two airports at which the percentage of non-aeronautical
revenue has fallen since privatisation — Coolangatta and Townsville — do not have
significant international RPT services.

15 Concession fees can also be based on passenger numbers, although this form of payment is less
common.
26 PRICE REGULATION
OF AIRPORT
SERVICES
Several privatised core-regulated airports — particularly Brisbane and Adelaide —
see property revenue as a means of increasing non-aeronautical revenue. Brisbane
Airport, for example, has established a light-industrial precinct on airport property
to take advantage of its large land holdings. Current and committed tenants at the
precinct include a mobile phone and accessories wholesaler and Australia Post, as
well as operations more traditionally associated with airports, such as freight
operations (BAC 2001).

Rentals paid to airport operators by the holders of the domestic terminal leases are
classified as property revenue, in both the regulatory accounts and company annual
reports. Property revenue also includes payments to airport operators from the
providers of operating and handling services for the use of land and facilities.
Consequently, property revenue is often a significant component of an airport
operator’s total revenue. It is approximately one-third of total revenue at Hobart
Airport (sub. 11) and nearly half at Adelaide Airport (Adelaide Airport Limited
(AAL), sub. 20).

Several participants noted that a significant proportion of an airport operator’s

revenue, and aeronautical revenue in particular, is generated from a very small
number of airlines. Prior to the reduction in Ansett services, Melbourne Airport
noted that approximately 80 per cent of its aeronautical revenue came from four
customers, with Qantas accounting for around 40 per cent. These same four
customers accounted for 45 per cent of the airport operator’s total revenue (sub. 7;
trans., p. 104 in PC 2001a). At Perth Airport, Ansett and Qantas accounted for
approximately 46 per cent of aeronautical revenue and 64 per cent of property
revenue; in addition, ‘when airline alliances are taken into account, 84 per cent of
aeronautical revenues and 68 per cent of property revenues are earned from the two
Alliances [Star and One World]’ (sub. 21, p. 33).

At smaller airports, the reliance on the major domestic carriers tended to be even
greater. At Launceston Airport, for example, Qantas and Ansett accounted for
98 per cent of aeronautical revenue (Melbourne Airport, sub. 7). In March 2001,
AAL noted:
Qantas and Ansett alone, and their regional airlines … contribute about 75 per cent of
aeronautical revenues at Adelaide Airport (or approximately 23 per cent of total current
revenue) … In total, more than half of AAL’s total revenue is sourced from Qantas and
Ansett. (sub. 20, p. 8)

With Ansett services reduced, the reliance of core-regulated airports on Qantas as

the major source of aeronautical revenue will increase, at least in the short to
medium term. In the longer term, it appears likely that operators of core-regulated
airports will continue to rely on a small number of customers for the majority of
their aeronautical revenue.
AUSTRALIAN 27
AIRPORTS AND THEIR
MARKETS
Charges for aeronautical services

Charges subject to the price cap at core-regulated airports at June 2000 are shown in
table 2.2. Airport operators typically levy a landing charge for the use of runways,
taxiways and aprons (airport operators do not supply air traffic control services)
and, where they operate the terminal, a terminal charge. At some airports additional
charges are levied for aircraft parking. In addition to charges covered by the price
cap, airport operators levy charges to recover the costs of government-mandated
security services, and to recover the costs of necessary new investment as approved
by the ACCC.

Table 2.2 Airservices Australia (ASA) charges, and charges subject to the
price cap at core-regulated airports, 2000 a,b
International
Runway charges terminal charges Aircraft parking ASAc

A$/t MTOW A$/t MTOW A$/aircraft A$/t MTOW

Adelaide 4.72 1.05 11 per day 10.39
Alice Springs 5.55 na 0 12.14
Brisbane 5.30 2.43 11 per day 5.67
Canberra (2.27)d na 0 10.83
Coolangatta 5.32 na 0 11.99
Darwin 5.55 1.02 0 8.89
Hobart 5.55 1.05 11 per day 13.79
Launceston 5.61 na 0 14.53
Melbourne 5.34 3.80 50 per day 4.06
Perth 5.06 2.48 10 per day 8.19
Sydney — 2.92 7.92 11 per daye 4.65
July 2000 charges
Sydney — 6.88f (35.10)g 35 per 15 minh 4.65
May 2001 charges
Townsville 5.75 1.05 0 4.33i
a GST exclusive. Charges are for RPT services greater than 10 tonnes (MTOW). b As at 30 June 2000. c As
at 1 July 2000. ASA charges include terminal navigation charges, and aviation rescue and firefighting charges.
d Per passenger charge. e For each aircraft parked in a designated general aviation parking area (an aircraft
parked for more than two hours is charged the daily rate). A different charge is levied on aircraft parking in a
designated aviation area. Prior to the ACCC May 2001 decision, the charge was $350 per day. f Based on a
runway charge of $3.44 per tonne (MTOW), which is levied at landing and take-off. Other core-regulated
airports currently charge for landing only. g Based on a per passenger charge of $17.55, which is levied on
each arriving and departing passenger (excluding transit passengers), and covers runway, international
terminal facilities, and passenger and checked-baggage security screening services. The charge was
introduced in November 2001 and is a restructuring of the May 2001 ACCC-approved prices. h For aircraft
parking in a designated aviation area. i Terminal navigation charges only. na Not applicable. MTOW Maximum
take-off weight.
Sources: ACCC (2001b–I, n); ASA (2000).

The starting point for runway charges at privatised airports was the FAC charge at
the time of privatisation, which was the same at all these airports apart from

28 PRICE REGULATION
OF AIRPORT
SERVICES
Adelaide. The differences in the runway charges shown across privatised airports in
table 2.2 reflect differences in their respective price-cap arrangements (chapter 3).
The starting point for terminal charges was also the FAC charge — shortly before
privatisation (in January 1997) the FAC increased charges for the use of terminals at
Sydney, Melbourne, Brisbane, Adelaide and Perth airports by varying amounts for
each airport.

Most user charges at core-regulated airports are levied on the same basis as the
system used by the FAC. User charges still tend to be calculated on the basis of the
maximum take-off weight (MTOW) of the aircraft. However, some airports have
introduced passenger-based charges, particularly for the international and common-
user domestic terminals. Sydney Airport now levies a per passenger charge on
international passengers to cover runway, international terminal, and passenger and
checked-baggage security screening services. Melbourne Airport charges for use of
its multi-user domestic terminal on a per passenger basis. Canberra and Coolangatta
airports (see below) are currently the only core-regulated airports to levy runway
charges on a per passenger basis.

Also shown in table 2.2 are the ASA charges for navigation and rescue services.
ASA moved from network charges to location-specific charges (based on full cost
recovery) for firefighting and rescue services in 1997 and for terminal navigation
charges in 1998. The variation in ASA charges is much greater than that in airport
charges and, as costs are largely fixed, unit charges are higher in airports with lower
traffic volumes.

Airport operator charges outside the price cap

Under the price-cap regime, airport operators are permitted to ‘pass through’ to
users 100 per cent of the costs related to government-mandated security
requirements. Such charges do not affect compliance with the price cap. Cost pass-
through is allowed for charges levied on the airport operator by Australian
Protective Services and for the provision of passenger and checked-baggage
screening services.

Pass-through provisions also exist for charges levied to recover costs for necessary
new investment (NNI) (chapter 3). Examples of ACCC-approved NNI charges at
core-regulated airports since privatisation are shown in box 2.1. Airport operators
are permitted to increase approved charges in line with the CPI.

AUSTRALIAN 29
AIRPORTS AND THEIR
MARKETS
 Box 2.1 Examples of ACCC-approved necessary new investment
charges
• In August 2000, the ACCC approved the introduction of a charge of $1.65 per
passenger (GST inclusive) to recover the costs associated with the construction of
the multi-user domestic terminal at Melbourne Airport. The decision was based on a
commercially-negotiated agreement between Melbourne Airport and Impulse and
Virgin Blue airlines. (ACCC 2000g)
• In April 2000, the ACCC approved increases in landing and international terminal
charges at Brisbane Airport to recover the costs of a range of projects (including
apron expansion and improvements in taxiway lighting). Landing charges were
permitted to increase by 3 cents a tonne in 2000-01 and by a further 9 cents a tonne
in 2001-02 (MTOW). International terminal charges were permitted to increase by
38 cents a tonne in 2000-01 and by a further 61 cents per tonne in 2001-02
(MTOW). (ACCC 2000f)
• In April 2000, the ACCC approved an increase in landing charges at Perth Airport of
2 cents per tonne (MTOW), and the introduction of a charge of 10 cents per
international passenger. The charges were allowed in order to recover the costs of a
range of projects over the period 1997-98 to 1999-00. (ACCC 2000c)

Airport charges since October 2001

In recognition of the need to give airports greater pricing flexibility to manage the
sudden and significant decline in the volume of airport traffic, the Commonwealth
Government changed prices oversight arrangements at core-regulated airports in
October 2001 (chapter 3). Among other things, the changes allowed the Phase 1
airports (Brisbane, Melbourne and Perth) a once-only price increase (by varying
amounts for each airport) as a pass-through in the price cap. Price-cap regulation
was removed from all other privatised core-regulated airports (with price-cap
regulation replaced by price monitoring of aeronautical services at some airports).16

Since implementation of the new prices oversight arrangements, there have been
changes in the level and, in some cases, structure of charges for airport services at
most core-regulated airports (all charges GST inclusive). In many cases, airports
offer various rebates. However, details of these often are confidential (chapter 7).
• The Phase 1 airports — Melbourne, Brisbane and Perth — have all increased
charges by the full amount determined by the ACCC to be in accordance with
the Government’s October announcement. However, the structure of the price
changes differs by airport.

16 Prices of aeronautical-related services continue to be monitored at some airports.

30 PRICE REGULATION
OF AIRPORT
SERVICES
 - Melbourne Airport increased average aeronautical charges by 6.2 per cent in
January 2002. Landing charges were increased by 59 cents to $6.40 per tonne
(MTOW), an increase of approximately 11 per cent. At the same time,
international terminal charges were reduced by 59 cents per tonne, leaving
the total charge (per tonne) paid by international airlines unchanged.
- In November 2001, Brisbane Airport increased all aeronautical charges by
6.7 per cent.
- In November 2001, Perth Airport increased domestic landing charges by
19 per cent to $6.30 per tonne (MTOW). Other charges remained unchanged.
This represented a weighted average increase in aeronautical charges of
7.2 per cent.
• Adelaide Airport increased landing charges for domestic RPT services by around
22 per cent to $6.00 per tonne (MTOW) in December 2001. Charges for
international and regional services were not altered. This represented a weighted
average increase in charges of approximately 14 per cent.
• Coolangatta Airport moved from weight-based charges to passenger-based
charges in November 2001. A passenger charge of $5.90 is now levied on
arriving and departing passengers. This represents an increase of approximately
170 per cent on the estimated equivalent per passenger charge of the previous
weight-based charges. (Gold Coast Airport, sub. DR58)
• In October 2001, Canberra Airport increased its per passenger landing charge by
113 per cent to $5.40 per passenger.
• Darwin and Alice Springs airports (both operated by Northern Territory
Airports) moved from weight-based landing charges to passenger-based charges
in December 2001. A charge of $6.02 per international passenger movement and
$4.93 per domestic passenger movement was introduced at Darwin Airport. A
$5.48 charge per passenger movement was introduced at Alice Springs Airport.
The switch to passenger charges makes comparisons with previous weight-based
charges difficult, but the new charges are more than double the previous
aeronautical charges at both airports.
• Townsville Airport introduced a 60 cent per passenger charge in December
2001, additional to their existing weight-based landing charges.

As at December 2001, Hobart and Launceston airports had not announced any
increases in their charges. A number of airports have indicated that they will review
their charges in the first half of 2002 in light of developments in traffic volumes.

AUSTRALIAN 31
AIRPORTS AND THEIR
MARKETS
International comparisons

User charges at major Australian airports appear to be among the lowest in the
world. The Transport Research Laboratory (TRL) compiles an annual index of user
charges at 40 major international airports.17 The index ranks airports on the basis of
the composite basket of airport services required to land and turn around an
international passenger service, including air traffic control charges, terminal
charges, aircraft parking charges and noise levies (figure 2.8). In 2000, Sydney
Airport (one of Australia’s more expensive airports in terms of international
terminal fees, and the only Australian airport included in the study) was ranked as
the sixth cheapest in the study (TRL 2000b). The ACCC (2001i) noted in its final
decision on the proposed price increases at Sydney Airport that, taking into
consideration the approved price increases, user charges would still be below the
average charges of those airports surveyed by TRL.

The Board of Airline Representatives of Australia (BARA) commented that the use
of the TRL study was of limited assistance as it failed to take into account
adequately the differences in the way in which a number of airports charge. In
particular, BARA noted that:
• several airports in the sample offer large discounts to frequent users, and/or levy
charges which depend on the destination or origin of the aircraft;
• … the impact of different charging policies in relation to transfer passengers is
not taken into account; and
• the study excludes taxes, even though at several airports, taxes are used to
recover costs in the same way as aeronautical charges in other airports.
(sub. DR54, p. 33)

While the study does not account for all differences in pricing policies across
airports, it does provide some indication of the level of user charges at Sydney
Airport relative to the level of charges at a number of major airports worldwide.
Moreover, even if features such as discounts, taxes and transfer passenger charges
were taken into account, it would not necessarily result in an increase in Sydney
Airport’s aeronautical charges relative to the charges levied by other airports in the
study. For example, taxes are not levied at Sydney Airport to recover the costs of
aeronautical services, and Sydney Airport offers discounts to some airport users.

17 Airports are included in the survey to represent the different approaches to airport pricing in
both public and private-sector operating environments. They are not necessarily the 40 largest
airports in the world (TRL 2000b).
32 PRICE REGULATION
OF AIRPORT
SERVICES
Figure 2.8 Airport charges at selected international airports, 2000a,b

120

100
Airport charges index number

80

60

40

20

0
rid

y
Mexic ow
i

r
ey
lulu

bay

en

Berlin

ich

na
- LIN
i

inki

pest
s
cisco

do

burg

Kong

sels

Oslo

ago
ns

ah

o
LHR

- JFK
G
kholm

WR
kfurt
rth

n
LGW

Miam
Duba

ouve

o Cit
ngele

Dubli

Lisbo
apor

Toky
hingto

- CD
terda
Sydn

Athe
Mad

nhag

Vien
Orlan

Jedd
rt Wo

Mun

Mosc
Hels
Hono

Bom

Chic

y-E
Brus

Buda
Fran
nnes
Fran

on -
Milan

Vanc
Sing
on -

Hong

York
Stoc
Los A

Paris
Ams
Was

Cope
s/Fo

Jerse
Lond
Joha
Lond
San

New
Dalla

New
a Based on the charges for a composite basket of airport services required to land and turn around an
international passenger service. b To derive the airport charges index, airport charges in local currencies are
converted to Standard Drawing Rights (SDRs). The SDR is an artificial currency unit calculated using a
weighted basket of five national currencies — the US dollar, the German deutschmark, the UK pound, the
French franc and the Japanese yen. It reflects actual exchange rates, not purchasing power parity.
Data source: TRL 2000b, Review of Airport Charges 2000, Crowthorne, UK.

Another TRL study, reported by the ACCC (2001i), compared 40 major airports
and airport groups on the basis of aeronautical revenue per aircraft movement and
per passenger. The five Australian airports covered in the survey were ranked
among the lowest in the world in terms of revenue per passenger (in the bottom 10
— figure 2.9) and in terms of revenue per aircraft movement (in the bottom 15).
The ACCC (2001i) noted that, allowing for the approved increase in charges at
Sydney Airport, charges at that airport would remain below the international
average for both measures. However, because the bundle of aeronautical services
provided by airport operators can differ substantially between airports, the
conclusions that can be drawn from these rankings are limited. For example, unlike
many overseas airports, the major domestic airlines operate their own domestic
terminals under long-term leases at the Australian airports included in the study. In
addition, a small number of airports included in the study (such as Frankfurt and
Munich) earn aeronautical revenue through the provision of ground-handling
services, which tend to be provided by airlines or specialist providers at other
airports.

AUSTRALIAN 33
AIRPORTS AND THEIR
MARKETS
Figure 2.9 Aeronautical revenue per passenger at selected international
airports, 2000a,b

12

10

8
SDR

0
e

ey

ich
row
i
hingto pain)

ter
lulu
oup
ide
ane

ianta
town

cisco
ary

en
ick

roup

na
frica)

ugal)
Perth

rio

burg
land

roup

apore

aris

Kong
kholm
oup
l

kfurt
lles

Miam
tiona

ouve
ngele
ourn

Maca
Grou

Grou
Sydn

nhag

Vien
Onta

ches
Gatw

Mun
Adela

Calg
Hono

n Du

eath
de P
ts Gr

Brisb

m Gr
Auck

Fran
rts G
nnes

rts G
Cape
Aer R

A (S

(Port
uth A
Fran
n Na

Vanc
Melb

Sing

Hong
Stoc
Los A

Berlin

BAA

Man
Cope

on-H
-
irpor

terda
ports

Airpo
AEN

Joha

Airpo
o
hingto

A (So
San

ANA
Lond

Lond
an A

Aero
Ams
Was

egian

dish
ACS
Was
aii

Swe
Haw

Norw

a Based on total passenger movements (international, domestic and, where applicable, regional). b The SDR
is an artificial currency unit calculated using a weighted basket of five national currencies — the US dollar, the
German deutschmark, the UK pound, the French franc and the Japanese yen. It reflects actual exchange
rates, not purchasing power parity. SDR Standard Drawing Right.
Data source: TRL 2000a, Airport Performance Indicators 2000, Crowthorne, UK.

Airport expenditure

Australia’s core-regulated airport operators incurred total expenses of

approximately $810 million in 1999-00 (including interest and depreciation). The
level of expenses varied across Australian airports, with Sydney, Melbourne and
Brisbane airports accounting for over 70 per cent of the total (ACCC 2001b–f).

In terms of expenses per passenger movement, there are significant variations

across airports (figure 2.10). Differences in average expenses are influenced by a
number of factors, including airport size (reflecting possible economies of scale),
the composition of airport traffic (for example, airports servicing international
traffic tend to incur higher costs than purely domestic airports), and genuine
efficiency differences. The extent to which differences may indicate economies of
scale is discussed in chapter 5.

34 PRICE REGULATION
OF AIRPORT
SERVICES
Figure 2.10 Expenses per passenger movement at core-regulated airports,
1999-00

Total expenses per passenger movement, including interest Total expenses per passenger movement, excluding interest

25

20

15
$A

10

0
ne

in

rth

lle
gs

ey
t
de

rra

ta

ne
ar

o
w
at

vi
rin

dn
ba
ai

ur
st
ob

Pe
be

ar
ng

ns
ce
el

bo

Sy
Sp

is

H
an

D
Ad

w
la
Br

un

el
C

To
oo
e

M
La
ic

C
Al

Data sources: ACCC (2001b–d); DoTRS (unpublished).

Cost structures by expense category

On average, capital costs (defined as depreciation plus interest expense) is the main
expense category for Australian core-regulated airport operators (figure 2.11).
However, given that financing structures (equity and loans) vary significantly
between airports and, in some instances, the distinction appears to be rather
arbitrary, the figure for the interest category is of limited value. Salaries and wages
also tend to be significant on average, although they are a much smaller component
of total costs than interest and depreciation. This in part reflects the fact that labour-
intensive components of airport services do not tend to be provided directly by core-
regulated airports. Land also is an important component of their operations but the
classification of expenses in the regulatory accounts does not provide consistent and
separate measures of the land expenses incurred. The value that should be placed on
land is contentious (appendix F).

AUSTRALIAN 35
AIRPORTS AND THEIR
MARKETS
Figure 2.11 Expenses of core-regulated airports, by category, 1999-00a

Salaries and wages

10%

Depreciation
18%
Interest
44%

Amortisation
5%

Services and utilities

8%
Property maintenance
Other (b) 4%
9%
APS costs
2%

a Expense categories used are those contained in the ACCC regulatory reports.b ‘Other’ includes the ‘other’
category contained in the ACCC regulatory reports plus consultants and advisers, general administration and
leasing for Adelaide Airport; passenger screening for Alice Springs and Darwin airports; and passenger and
checked-baggage screening for Perth Airport. APS Australian Protective Services.
Data sources: ACCC (2001b–d).

In broad terms, the cost pattern illustrated in figure 2.11 is indicative of the situation
at individual airports. However, there is (sometimes significant) variation across
airports. For instance:
• labour costs tend to be a more significant proportion of costs at smaller airports
than they are at the larger ones; and
• not all expense categories are relevant to all airports. For example in 1999-00,
only seven of the 12 core-regulated airports incurred Australian Protective
Services expenses and only eight included amortisation expenses.

Cost structures by aeronautical and non-aeronautical sources

Aeronautical expenditure tends to comprise the bulk of Australian airport operator’s

expenses (figure 2.12), accounting for 60 per cent of total expenses on average in

36 PRICE REGULATION
OF AIRPORT
SERVICES
1999-00.18 Non-aeronautical expenditure exceeded aeronautical expenditure at
three airports only in 1999-00 — Adelaide, Launceston and Melbourne.

Figure 2.12 Aeronautical and non-aeronautical expenses of core-regulated

airports, 1999-00a

Aeronautical Non-aeronautical

100

90

80

70

60
Per cent

50

40

30

20

10

0
in

lle
gs

ey
t
de

ne

rra

ta

ne

rth
on
ar
w
at

vi
rin

dn
ba
ai

ur
st
ob

Pe
be

ar
ng

ns
el

ce

bo

Sy
Sp

is

H
an

D
Ad

w
la
Br

un

el
C

To
oo
e

M
La
ic

C
Al

a Classification of expenses as aeronautical and non-aeronautical is based on ACCC regulatory reports. Total
costs exclude costs that have not been classified as either aeronautical or non-aeronautical.
Data sources: ACCC (2001b–d).

The high proportion of aeronautical expenses possibly reflects the fact that airport
operators are more likely to be directly involved in operations on the aeronautical
side than the non-aeronautical side of the business, where they typically act as
landlords. The relatively low proportion of aeronautical expenditure for Adelaide
Airport is affected by its use of contractors to provide many of its day-to-day
operations. This is because contracting expenses at Adelaide Airport are recorded as
non-aeronautical expenses in the regulatory accounts even though some contracted
labour is used to provide aeronautical services.19

The data presented here rely on information provided in the ACCC regulatory
reports by the airports. Thus, measured differences across airports also are likely to

18 Total expenses in this case exclude interest and other expenses that have not been classified as
either aeronautical or non-aeronautical in the ACCC regulatory reports.
19 Adelaide Airport contracts out its day-to-day operations to Serco Australia, which provides the
staff required for these operations (AAL 2000).
AUSTRALIAN 37
AIRPORTS AND THEIR
MARKETS
be influenced by different accounting and cost allocation methods. According to
WAC:
Regulated airports lodge annual accounts with the ACCC, and while the form of the
final published accounts is consistent between airports, there is no way of determining
whether each airport has applied the same methodology in reporting aeronautical and
non-aeronautical costs. WAC understands that each airport employs different cost
accounting tools to achieve the split. (sub. 21, p. 25)

Investment expenditure

Airport operators invest to increase capacity and/or to maintain or improve the

quality of service (see box 2.2, for examples of airport investment). Overall
investment levels for airports vary depending on the size and nature of the airport,
the extent of previous investment in the airport and, for the privatised airports in
Australia, the amount of investment required under their lease agreements
(chapter 3).

Box 2.2 Examples of investment by Australian airport operators since

privatisation
• At Melbourne Airport, approximately $100 million was invested in capital projects in
the first three years following privatisation (around 75 per cent of which related to
terminal facilities), in addition to $130 million invested by its business partners.
Around $100 million is expected to spent by Melbourne Airport on a range of
projects in 2000-01 and 2001-02, including an expansion of aircraft parking areas,
domestic and international terminal infrastructure projects, retail and car-parking
projects, and cargo facilities. (Melbourne Airport, sub. 7, appendix 1)
• At Mount Isa Airport, approximately $1.45 million was invested in the two years after
11 June 1998, with investments relating to the erection of fencing to protect the
airfield from animal hazards, the treatment of airport pavements and a terminal
upgrade. (Australian Airports (Townsville), sub. 14)
• At Perth Airport, $6.1 million of capital investment in aeronautical infrastructure in
three and a half years of private ownership. (WAC, sub. 21)
• At Canberra Airport, over $35 million of capital expenditure has been made since
privatisation, with a further $100 million planned in the next five years. (Capital
Airport Group, sub. 32)

Information provided by participants shows that total investment by airport

operators at individual airports since privatisation has ranged from several hundred

38 PRICE REGULATION
OF AIRPORT
SERVICES
thousand dollars per year to over $30 million per year,20 with further investment
planned. Terminal upgrades, apron and runway enhancements, the erection of
fencing and retail development are among the types of investment already
undertaken or planned for the next few years.

Productivity

There have been few recent studies on the productivity of core-regulated airports.
One recent study, by Abbott and Wu (2001) measured the (total factor) productivity
growth21 of core-regulated airports over the period 1990-91 to 1999-00 and
benchmarked these airports against a selection of international airports using data
envelopment analysis. Over the period of the study, core-regulated airports
improved their performance in terms of total factor productivity, with productivity
growth averaging 7.7 per cent per year, well above that of the rest of the economy
(which had an average growth rate of 1.36 per cent per year) (Abbott and Wu
2001, p. 17). The high growth rates in productivity were driven by rapid
technological change, particularly over the period 1990-91 to 1994-95. Abbott and
Wu (2001) suggested that improvements came in the form of the introduction of
more advanced computer and air traffic systems, which allowed for the
accommodation of greater traffic flows through the airports. This process helped the
airports to overcome delays and accommodate an expansion of output without the
construction of new and enlarged terminals and runways.

Abbott and Wu (2001) also examined the productivity performance of the 12

Australian airports against a number of overseas airports including two British, two
Canadian, three New Zealand and five US airports for the year 1998-99. When rated
against overseas airports, Australia’s largest airports ‘rate fairly well in terms of
their levels of efficiency’ (Abbott and Wu 2001, p. 27).

Profitability

In 1999-00, the operators of all privatised core-regulated airports earned profits and
positive returns on non-current tangible assets. However, with the exception of the
operators of Townsville Airport, all operators reported a loss after interest and tax
(table 2.3). All Phase 1 airports are highly geared (ACCC, sub. 36, attachment A)

20 This does not include Sydney Airport, where SACL (sub. 27) invested in excess of
$800 million over two years leading up to the 2000 Olympics.
21 Productivity refers to the relationship between inputs and outputs. Productivity growth implies
an increase in the ratio of outputs to inputs. Total factor productivity is a measure of overall
productivity, capturing all sources of productivity.
AUSTRALIAN 39
AIRPORTS AND THEIR
MARKETS
and, in general, high interest payments at all privatised core-regulated airports have
been an important factor contributing to the low returns. Sydney Airport is less
highly geared than other core-regulated airports (although it has substantial
borrowings) and was able to generate sufficient operating profit to cover interest
payments and generate net profits — approximately $43 million in 1999-00.

Table 2.3 Selected financial results for core-regulated airports, 1999-00

Profit/(loss) after
Operating profitsa Operating returnb interest and tax
$m % $m
Adelaide 29.541 14.0 (7.131)
Alice Springs 0.417 1.7 (2.692)
Brisbane 68.274 8.0 (112.650)
Canberra 4.030 4.8 (2.196)
Coolangatta 3.243 10.6 (2.009)
Darwin 1.054 1.6 (10.130)
Hobart 2.082 17.1 (0.216)
Launceston 1.264 6.0 (0.188)
Melbourne 94.154 13.0 (20.290)
Perth 37.652 18.2 (22.398)
Sydney 120.159 3.9c 42.842
Townsville 1.543 10.9 0.117
a Before abnormal items, interest, tax and amortisation. b On tangible non-current assets. c Based on
earnings before interest and tax (EBIT) on tangible non-current assets.
Source: ACCC (sub. 36, attachment A).

The ACCC regulatory accounts (ACCC 2001b–f) indicate that the non-aeronautical
segment of the airport business is much more profitable than the aeronautical
business. In 1999-00, non-aeronautical businesses generated high operating profit
margins at nearly all core-regulated airports, while aeronautical businesses mostly
made small profits or had losses — around half of the core-regulated airports made
a loss on the aeronautical side of their businesses in 1999-00 (tables 2.4 and 2.5).

40 PRICE REGULATION
OF AIRPORT
SERVICES
Table 2.4 Financial indicators for aeronautical and non-aeronautical
business segments at Phase 1 and Sydney airports, 1999-00
Airport Revenue Costs Operating profita Operating returnb
$m $m $m %
Brisbane
Aeronautical 38.211 38.016 0.195 0.0
Non-aeronautical 91.320 23.241 68.079 21.0
Melbourne
Aeronautical 55.160 39.277 15.883 3.7
Non-aeronautical 111.823 33.700 78.123 26.5
Perth
Aeronautical 20.471 17.595 2.876 3.1
Non-aeronautical 49.207 14.431 34.776 32.5
Sydney
Aeronautical 120.052 126.003 (5.951) (0.4)
Non-aeronautical 190.001 66.540 123.461c 8.2
a Before abnormal items, interest, tax and amortisation. b On tangible non-current assets. c Approximately
$3 million of Sydney Airport’s revenue has not been allocated between aeronautical and non-aeronautical
sources.
Source: ACCC (sub. 36, attachment A).

As noted previously, the disaggregation of costs and assets into aeronautical and
non-aeronautical components is determined by airport operators’ interpretation of
the regulatory guidelines. Consequently, there may be some variation in how costs
and assets are allocated across core-regulated airports. Even allowing for variation
in the disaggregation of costs and assets, it seems likely that returns from non-
aeronautical businesses are greater than those from aeronautical businesses at core-
regulated airports. As KPMG observed:
On the assumption that revenue allocations are fairly stated and not at risk of
significant subjectivity, the cost and/or asset allocations disclosed in the regulatory
accounts would need to be materially misstated to change the view that aeronautical
returns are less than non-aeronautical returns. (ACCC, sub. 36, attachment A, p. 6)

AUSTRALIAN 41
AIRPORTS AND THEIR
MARKETS
Table 2.5 Financial indicators for aeronautical and non-aeronautical
business segments at Phase 2 airports, 1999-00
Airport Revenue Costs Operating profita
$m $m $m
Adelaide
Aeronautical 9.867 8.681 1.186
Non-aeronautical 39.442 11.087 28.355
Alice Springs
Aeronautical 1.812 2.724 (0.912)
Non-aeronautical 2.795 1.466 1.329
Canberra
Aeronautical 4.372 5.178 (0.806)
Non-aeronautical 6.255 1.419 4.836
Coolangatta
Aeronautical 5.804 5.447 0.357
Non-aeronautical 3.593 3.703 na
Darwin
Aeronautical 5.334 8.357 (3.023)
Non-aeronautical 6.341 2.264 4.077
Hobart
Aeronautical 3.311 2.375 0.936
Non-aeronautical 2.184 1.038 1.146
Launceston
Aeronautical 1.525 1.617 (0.092)
Non-aeronautical 2.914 1.558 1.356
Townsville
Aeronautical 2.188 2.747 (0.559)
Non-aeronautical 3.611 1.509 2.102
a Before abnormal items, interest, tax and amortisation. na Not available, due to non-allocated revenues.

Sources: ACCC (2001b–f; sub. 36, attachment A).

2.4 The airport business at other Australian airports

Cairns Airport is Australia’s largest non-core-regulated airport (and, overall, is
Australia’s sixth largest in terms of total passenger movements, and fifth largest for
international passengers). In 1999-00, Cairns Airport’s operating revenue was
approximately $41.5 million. Cairns Airport is owned and operated by the Cairns
Port Authority, which assumed control of the airport in 1981 under the Aerodrome
Local Ownership Plan (ALOP).22

22 A Commonwealth Government program where ownership and responsibility for around

230 Commonwealth airports was transferred to local councils.
42 PRICE REGULATION
OF AIRPORT
SERVICES
The organisational structure for the provision of services at Cairns Airport is
different from that at core-regulated airports. Unlike the situation at core-regulated
airports, at Cairns Airport the major domestic airlines do not hold exclusive leases
for their own domestic terminals, and are required to pay a user charge based on
passenger numbers. The difference in charging structures — whereby revenue from
domestic terminals at Cairns Airport is included in aeronautical revenue, while
core-regulated airport revenue from the domestic airlines’ terminals is classified as
property income — is reflected in the relatively high proportion of ‘aeronautical
revenue’ at Cairns Airport, which was around 55 per cent of total airport revenue in
1999-00 (CPA 2000).23

Of the remaining non-core-regulated airports that receive RPT services, some are
privately owned (for example, Hamilton Island and Mount Isa). However, most are
owned and operated by local councils which, in most cases, assumed the running of
the airports under the ALOP.

While there is little publicly-available information on the user charges levied on

RPT at regional airports, anecdotal information suggests there is significant
variation in user charges across these airports. This may reflect the fact that most of
these airports are not managed as separate business entities, but as a component of
the councils’ overall activities (chapter 5).

23 Refers to revenue directly related to the aircraft operations, and comprises charges on aircraft
(33 per cent) and charges on goods and passengers (22 per cent).
AUSTRALIAN 43
AIRPORTS AND THEIR
MARKETS
3 The regulatory environment

In 1997 and 1998 the Commonwealth Government negotiated long-term leases for
22 Commonwealth-owned airports, with 17 being privatised (chapter 1). The
objectives were to:
… improve the efficiency of airport investment and operations in the interests of users
and the general community, and to facilitate innovative management. (Harris 1997,
p. 2)

Many of these airports are subject to a comprehensive Commonwealth regulatory

framework, major economic components of which are the Airports Act 1996 and
Trade Practices Act 1974 (TP Act), and price regulation under the Prices
Surveillance Act 1983 (PS Act). They are also subject to State and Territory
regulations and international agreements, which directly or indirectly affect the
operation of airports, and ultimately the prices that airports charge for services.

This chapter describes the regulatory environment affecting the operation of these
and other airports in Australia. It also describes the changes to prices oversight
arrangements for several airports, as announced by the Commonwealth Government
in October 2001.

Where relevant, the terms and conditions of lease and sale agreements are also
described. Although the inquiry’s terms of reference embrace all airports, this
chapter focuses on the regulation, particularly price regulation and access
arrangements, of airports leased from the Commonwealth given their importance
and the emphasis in the terms of reference.

Chapters 8 and 9 assess price regulation for the period from the granting of leases
against the criteria for efficient regulation, and weigh the costs and benefits of the
regulation. Chapter 9 also examines the effect of other regulations, such as s. 192 of
the Airports Act.

3.1 Price regulation until October 2001

This section describes prices oversight arrangements for the period from the
granting of leases for Phase 1 and 2 airports until October 2001 when prices

THE REGULATORY 45
ENVIRONMENT
oversight arrangements were changed for most airports. Section 3.2 describes the
regulatory changes announced in October 2001.

Prices oversight arrangements, introduced when Phase 1 and 2 airport leases were
sold, were designed to complement the objectives (see above) of leasing the
Commonwealth-owned airports.

According to a Commonwealth Department of Transport and Regional

Development (DoTRD)1 Pricing Policy Paper provided to potential airport lessees
as part of their obligations if they were successful bidders, prices oversight
arrangements were intended to:
… promote operation of the airports in as an efficient and commercial a manner as
possible. Pricing is fundamental to the efficient use of airport infrastructure. It is in the
interests of airport users in particular, and the national economy in general, that
commercially-driven decisions be made about maintaining existing airport
infrastructure, and building new infrastructure. (DoTRD 1996, p. 1)

Another objective was the protection of airport users. The paper noted that ‘the
arrangements should also aim to protect airport users from any potential abuse of
market power by airport operators’ (DoTRD 1996, p. 1).

Finally, the Government intended to encourage more direct commercial

relationships between airport operators and their customers. The paper stated:
It is the Government’s intention to step back from setting prices at individual,
privately-leased airports; and to provide a framework in which — over time — airport
operators and their customers are encouraged to negotiate directly, and resolve prices
rather than involve the Government of the day. (DoTRD 1996, p. 1).

In early 2001, these objectives were re-stated by the Minister for Transport and
Regional Services in correspondence with the ACCC. The Minister also noted that
implicit in the prices oversight arrangements:
… was that regulatory intervention would only occur where it was apparent that airport
behaviour was adversely impacting on consumers or competition between airlines
through the exercise of a real or perceived monopoly power. (Anderson 2001b, p. 1)

The Commonwealth Government at the time of leasing the airports recognised that
a period of adjustment to prices oversight might be necessary in the new regulatory
environment. The first five years were therefore viewed as a transition period, with
a review of prices oversight arrangements to be completed prior to the end of this
period (chapter 1).

1 Now the Department of Transport and Regional Services (DoTRS).

46 PRICE REGULATION
OF AIRPORT
SERVICES
Twelve of the 22 airports were designated as ‘core-regulated’ airports under the
Airports Act and were subject to price regulation under the PS Act. The core-
regulated airports comprise 11 privatised airports and Sydney Kingsford Smith
Airport (Sydney Airport), which is to be sold in early 2002. Privatised core-
regulated airports were those airports with significant ‘regular public transport’
(RPT) and included the three Phase 1 airports (Brisbane, Melbourne and Perth) and
eight of the Phase 2 airports (Adelaide, Alice Springs, Canberra, Coolangatta,
Darwin, Hobart, Launceston and Townsville).2

In essence, price regulation of the Phase 1 and 2 airports, in the period from the sale
of leases until October 2001, comprised:
• prices notification for aeronautical services;
• a CPI-X price cap on aeronautical services at privatised core-regulated airports;
• price monitoring of aeronautical-related services; and
• provisions for necessary new investment (NNI) at airports.3

These arrangements were administered by the ACCC under the PS Act. In carrying
out its functions, the ACCC was to have regard to the following, among other
factors:
(a) the need to maintain investment and employment, including the influence of
profitability on investment and employment; [and]
(b) the need to discourage a person who is in a position substantially to influence a
market for goods or services from taking advantage of that power in setting prices.
(PS Act, s. 17(3))

Price regulation of airport services until October 2001

A range of services provided by core-regulated airports, including Sydney, was

subject to prices notification or monitoring under the PS Act for the period
following the granting of leases until October 2001 (box 3.1). The basket of
aeronautical services — as defined in the now rescinded Federal Airports
Corporation Act 1986 — to which prices notification applied remained unchanged
from when the Federal Airports Corporation (FAC) operated airports.

2 The regulatory environment for airports leased from the Commonwealth and not subject to
price regulation is described in section 3.4. These airports were: Mount Isa, Tennant Creek,
Archerfield, Jandakot, Moorabbin, Parafield, three Sydney basin airports (Bankstown, Camden
and Hoxton Park) and Essendon.
3 All of these regulations, except price caps, also applied to Sydney Airport.
THE REGULATORY 47
ENVIRONMENT
 Box 3.1 Regulation of facilities and activities
Aeronautical services (those notified for the purposes of s. 21(1) of the PS Act)4 at all
core-regulated airports, including Sydney Airport, were grouped in two main categories:
aircraft movement facilities and activities, and passenger processing facilities and
activities.
Aircraft movement facilities and activities comprised: airside grounds, runways,
taxiways and aprons; airfield lighting, airside roads and airside lighting; airside safety;
nose-in guidance; aircraft parking areas; and visual navigation aids.
Passenger processing facilities and activities included: forward airline support area
services; aerobridges and airside buses; departure lounges; immigration and customs
services areas; public address systems and closed circuit surveillance systems;
baggage make-up, handling and reclaim; public areas in terminals; flight information
display systems; landside road and lighting; and covered walkways.
Some aeronautical-related services provided by airport operators were not subject to
notification but were subject to price monitoring under s. 27A of the PS Act.5 These
were: aircraft refuelling; aircraft maintenance sites and buildings; freight equipment
storage sites; freight facility sites and buildings; ground support equipment sites;
check-in counters and related facilities; and public and staff car parks.
Not all services provided by airport operators were subject to prices oversight. For
example, there was no prices oversight of airport operators’ revenue from rents or
leases for retail shops and cafes, administration and office space, catering facilities,
valet parking services and VIP lounges.
Moreover, airport operators did not have responsibility for all aeronautical services
provided at airports. Aeronautical services outside airport operators’ areas of
responsibility included: en-route navigation and terminal navigation (air traffic control
and airspace management); aeronautical information; communications; and firefighting
and rescue services.
Sources: Prices Surveillance Act 1983; ACCC (2000a).

Services and facilities provided by domestic terminals that were leased for the long-
term to airlines were not subject to price regulation. However, domestic common-
user terminals and international terminals were covered by price regulation.

4 Declaration No. 87, June 2000, covered Brisbane, Melbourne and Perth airports, Declaration
No. 88, June 2000 (Adelaide, Alice Springs, Canberra, Coolangatta, Darwin, Hobart,
Launceston and Townsville airports), and Declaration No. 89, June 2000 (Sydney Airport).
These Declarations replaced previous Declarations.
5 Declaration Nos 87, 88 and 89. The Minister for Financial Services and Regulation (pursuant
to s. 27A of the PS Act) directed the ACCC to undertake formal monitoring of prices, costs
and profits related to the supply of specified aeronautical-related services at all core-regulated
airports (Direction No. 21, October 2000, replacing previous Directions).
48 PRICE REGULATION
OF AIRPORT
SERVICES
Aeronautical services were subject to prices notification under the PS Act. This
required specified companies to notify the ACCC of all proposed price increases for
these services. Financial penalties applied for failure to notify price increases. The
ACCC was required to make a determination about the notified price increase
within 21 days unless the company agreed to an extension. Although any such
determinations were not enforceable, failure to comply could, under the PS Act,
trigger a public inquiry (upon direction by the Minister) and the freezing of charges.
Moreover, the Commonwealth Government reserved the right to consider advice
from the ACCC regarding the need for stronger regulation (DoTRD 1996).

The Ministerial direction under the PS Act required the ACCC to monitor and
report publicly on prices of aeronautical-related services and their relationship to
costs. The ACCC could make no determination, but it could recommend whether
more or less rigorous prices oversight of aeronautical-related services was needed.

Price-cap arrangements

Aeronautical services were subject to an annual CPI-X price cap at all core-
regulated airports, except Sydney Airport, for a five-year period following
privatisation (pursuant to s. 20 of the PS Act). Direction No. 20, by the Minister for
Financial Services and Regulation, described the price-cap arrangements.6

Aeronautical-related services and non-aeronautical services were excluded from the

price cap (box 3.1). Only revenue from those services within the cap was taken into
account in assessing airport operator compliance with the cap — a so-called ‘dual-
till’ approach to price regulation.

The ACCC administered the price-cap arrangements. It assessed any individual

notified price increases (as mentioned above), focussing on whether they were
likely to comply with the price cap, and also undertook a formal annual assessment
of each airport operator’s compliance with the price cap. Although compliance with
the overall price cap was not directly enforceable, non-compliance might have
resulted in a response from the ACCC similar to that for individual notified price
increases.

Airport operators were required to provide the ACCC with sufficient information to
enable these assessments to be made. For individual price notifications, this
included information on initial and new prices and previous years revenue shares of
the relevant component. Information to enable an assessment of price-cap
compliance included revenue from each of the charging components for the year of

6 Direction No. 20, October 2000, replaced Direction No. 17, and previously, Direction No. 13.
THE REGULATORY 49
ENVIRONMENT
assessment and the previous year, and volume and output for each component
(ACCC 1997a).

CPI-X

The CPI measure for the price cap was the Treasury Underlying Rate of Consumer
Price Inflation. Each airport’s X value was set by the Commonwealth Government,
on advice from the ACCC, and ‘reflects expected productivity improvements that
the Government considers can be made in the provision of aeronautical services at
each airport’ (Direction No. 12, June 1997, p. 1). The X values varied substantially
between airports, ranging from 1 per cent for Canberra and Townsville airports to
5.5 per cent for Perth Airport (table 3.1). The starting prices for the price cap were
the FAC prices at the time of privatisation — shortly before privatisation (in
January 1997), the FAC increased charges for the use of terminals at Sydney,
Melbourne, Brisbane, Adelaide and Perth airports by varying amounts for each
airport.

Table 3.1 X values applying in each year of the price cap at core-
regulated airports
Airport X values (%)

Adelaide 4.0
Alice Springs 3.0
Brisbane 4.5
Canberra 1.0
Coolangatta 4.5
Darwin 3.0
Hobart 3.0
Launceston 2.5
Melbourne 4.0
Perth 5.5
Townsville 1.0
Sources: Direction No. 12, June 1997; Direction No. 13, May 1998.

The formula

The price-cap formula measured annual changes in the prices of the aeronautical
services for each airport. It was a modified revenue-weighted average price. The
formula took the percentage change in price of each charging component (for
example, landing charges), weighted this by the revenue share in the previous

50 PRICE REGULATION
OF AIRPORT
SERVICES
period, and then summed over all components.7 This was referred to as the ‘tariff
basket’ approach (chapter 10).

Under-recovery and over-recovery

According to the DoTRD Pricing Policy Paper, the ACCC would not object to price
changes unless they breached the price cap. Airport operators therefore could
continue to ‘rebalance’ charges within their airport’s overall price cap (DoTRD
1996).

Under-recoveries, where prices were below the cap in a particular year, could be
carried over between years within the five-year price-cap period. Over-recoveries,
where prices exceeded the price cap, were to be passed back to customers within
two years from the period of over-recovery (except in the case of year 4, where it
was to be passed back fully in year 5).

New or varied charges

New or varied charges on existing aeronautical services, and charges on new or

varied services, were factored into the price cap. Compensating reductions in other
charges could be required to ensure compliance with the cap. In the case of new
charges (such as vehicle access charges at Perth Airport), in the transition year,
revenue from the new charge was allocated to other charging components (for
example, landing charges) in proportion to their previous period revenue share
(ACCC 2001d).

Cost pass-through

Under Direction No. 20, the ACCC was to allow 100 per cent pass-through in the
price cap of mandated security requirements and congestion charges employed as
part of an airport demand management scheme under the Airports Act (described in
section 3.3). Pass-through provisions also existed for NNI.

Necessary new investment

The price-cap arrangements allowed airport operators to seek ACCC approval for
charges in excess of the price cap, to recoup costs associated with NNI.

7 For more information on the price cap formula, see Direction No. 20, October 2000, and
ACCC (1997a).
THE REGULATORY 51
ENVIRONMENT
The DoTRD Pricing Policy Paper explained:
While price caps on aeronautical charges are directed towards ensuring that there is no
abuse of the potential market power of the airport operators, price oversight
arrangements dependent solely on these price caps may restrict the timely development
of necessary new aeronautical infrastructure.
As a result, some flexibility is afforded airport operators to rebalance charges outside
the price cap. In particular, operators need sufficient incentive to invest in new
infrastructure, and the ability to meet the costs of necessary new investment. (DoTRD
1996, p. 4)

The operator could seek to recoup these costs where price rises were required to
fund the new investment and ‘users with a significant interest in the new investment
support the investment, including the associated charges’ (DoTRD 1996, p. 4).

ACCC assessments of proposals by airport operators to increase charges to recover

the costs of investment were undertaken in the context of specified guidelines
(box 3.2).8 The price-cap arrangements did not define the term ‘necessary new
investment’. Moreover, according to the ACCC, although an amount of investment
was factored into the X value for each airport, specific investments were not
detailed (ACCC 2000b). In this regard the Department of Transport and Regional
Services (DoTRS) stated:
There is no mandated or required level of capex [capital expenditure] related to the
revenue airports receive under the pricing cap, thus no basis for revisiting which
investments are paid for and which are not. The slate must be considered clean, or else
every other element used in setting the price cap will become equally open to debate.
(DoTRS letter to Ansett, 30 April 1999, p. 2, cited in ACCC 2000b)

Following industry consultation, the ACCC, as administrator of the price cap and
assessor of NNI proposals, defined the words ‘necessary’, ‘new’ and ‘investment’ to
be able to distinguish between NNI and other forms of airport expenditure
(ACCC 2000b). For example, the ACCC assessed several projects proposed by
Perth Airport and decided that some projects did not fall within its definition of new
investment (ACCC 2000c).

The ACCC’s implementation of the NNI provisions of the price-cap arrangements

is discussed in chapter 8.

8 These guidelines are set out in the DoTRD (1996) Pricing Policy Paper, and replicated in
Direction No. 20, October 2000.
52 PRICE REGULATION
OF AIRPORT
SERVICES
 Box 3.2 Guidelines for the assessment of proposals for price
increases related to necessary new investment
These criteria will, where relevant for its purposes, guide the ACCC in its assessment of
proposals related to necessary new investment to increase aeronautical charges at a rate in
excess of the CPI-X cap:
(a) the operator’s plans for new investment or service innovation and the associated
costs;
(b) the relationship between the proposed increases in aeronautical charges and the
costs (including the level of rate of return) of the new investment or service;
(c) support from airport users with a significant interest in the investment for the
operator’s proposals, including in relation to charging changes;
(d) contribution of the new investment/service to productivity improvements at the
airport;
(e) overall efficiency of the airport’s operation;
(f) the particular demand management characteristics of individual airports, including
any demand management schemes in place, capacity constraints and any
underutilisation of airport infrastructure;
(g) airport performance against quality of service measures, including services under the
control of the airport operator;
(h) airport performance vis á vis other Australian airports and any comparable
international airports; and
(i) the extent to which the proposed investment will facilitate the operations of new
entrants to domestic or international aviation.
While the ACCC must take the above into account in deciding whether to approve a
proposal to increase charges outside the cap, each proposal will be considered on its merits
having regard to the information available to the ACCC. The weight provided by the ACCC
to each of the criteria (a) to (i) may vary on a case by case basis.
Consistent with the provisions of the Prices Surveillance Act 1983, where the ACCC does
not approve a proposal to increase charges outside the price cap, it will provide a statement
of reasons for its determination. (DoTRD 1996, attachment B)

These guidelines were incorporated in Government directions to the ACCC under the
PS Act.

Sydney Airport

Price regulation for Sydney Airport, which was leased but not privatised in 1997,
differed in some respects from that of other core-regulated airports. Aeronautical
services at Sydney Airport were subject to prices notification under the PS Act but a

THE REGULATORY 53
ENVIRONMENT
price cap did not apply to those services. Aeronautical-related services were subject
to price monitoring.9

Because Sydney Airport was not subject to the price-cap arrangements set out in
Direction No. 20, the NNI provisions of this Direction did not apply. However,
Direction (No. 18) provided for Sydney Airport to seek an increase in charges for
aeronautical services to cover NNI.10 The criteria used by the ACCC in assessing
Sydney Airport proposals were the same as those for the core-regulated airports
subject to a price cap (box 3.2).

Quality of service monitoring at Sydney Airport took place under the Airports Act.
Moreover, the Minister for Financial Services and Regulation directed the ACCC
(Direction No. 18) to take account of quality of service information obtained from
Sydney Airport in considering notifications to increase prices for aeronautical
services.

The Minister also directed the ACCC to give consideration to the following when
considering price notifications for Sydney Airport (in effect, a dual-till approach):
In assessing prices for aeronautical services, the Commission should not take into
account the revenues generated, or costs incurred, in the provision of services other
than aeronautical services. (Direction No. 22)

3.2 Price regulation since October 2001

As noted above, the Commonwealth Government changed prices oversight
arrangements at core-regulated airports in October 2001. In so doing, the
Government commented that:
Some airports have indicated that they are facing financial pressures as a result of the
suspension of Ansett’s operations and reduced global demand for aviation services …
The Government accepts that it would be difficult for airports to adjust aeronautical
prices to compensate for reduced traffic flows, and still comply with the current price
oversight arrangements administered by the ACCC. (Hockey 2001)

The changes, pursuant to s. 20 of the PS Act, were as follows.

• Melbourne, Brisbane and Perth airports were allowed a once-only price increase,
as a pass-through in the price cap, of up to 6.2, 6.7 and 7.2 per cent of starting

9 Declaration No. 89, June 2000.

10 Direction No. 18, June 2000. Replaced Direction No. 15.

54 PRICE REGULATION
OF AIRPORT
SERVICES
 point prices at privatisation respectively.11 The basis of calculation underlying
these price increases has not been made public. In all other respects, price
regulation at these airports remained unchanged; aeronautical services continued
to be subject to prices notification with price caps.12 Airport prices notification
arrangements, the basket of aeronautical services and price-cap arrangements,
including the X values and NNI provisions, remained unchanged from those
described in section 3.1.

Also unchanged for Melbourne, Brisbane and Perth airports were price
monitoring arrangements for aeronautical-related services.13

These airports remain subject to all of the requirements of core-regulated

airports under the Airports Act, such as s.192 and quality of service monitoring
(sections 3.3 and 3.4). They also must comply with Commonwealth, State and
Territory regulation, sale and lease agreements, and international agreements
described in section 3.5.
• Price caps on aeronautical services at Adelaide, Canberra and Darwin airports
were replaced by price monitoring of those services under the PS Act.14
Aeronautical-related services which were subject to price monitoring prior to
October 2001 (section 3.1), continue to be monitored.15

As with Melbourne, Brisbane and Perth airports, these airports must continue to
comply with all of the regulatory requirements of core-regulated airports
described in section 3.5, including the Airports Act.
• Coolangatta, Alice Springs, Hobart, Launceston and Townsville airports are no
longer subject to any price regulation — that is, both the price caps and price
monitoring have been removed.16 These airports remain, however, designated as
core-regulated airports and therefore must comply with all of the relevant
regulatory requirements described in section 3.5, including the Airports Act.

11 Direction No. 24, October 2001. Replaced Direction No. 20. According to the Direction,
starting point prices are the ‘Federal Airports Corporation prices introduced on 1 January
1997, as adjusted in accordance with the price cap arrangements applying since that time, and
accounting for under-recoveries or over-recoveries under those price cap arrangements’. The
ACCC has interpreted this to mean starting point prices at 1 January 1997 with no allowance
for adjustments under the price caps. The ACCC (2001k) has treated the price increases as
GST inclusive.
12 Direction No. 24, October 2001. Replaced Direction No. 20.
13 Direction No. 25, October 2001. Replaced Direction No. 21.
14 Direction No. 26, October 2001. Replaced Direction No. 21.
15 Direction No. 26, October 2001. Replaced Direction No. 21.
16 Revocation No. 28, October 2001.

THE REGULATORY 55
ENVIRONMENT
• No changes were made to price regulation at Sydney Airport.17 Sydney Airport
remains subject to prices notification of aeronautical services and price
monitoring of aeronautical-related services (section 3.1). Other regulations,
outlined in section 3.5, continue to apply.

The Government noted that these changes were consistent with the findings of the
Commission’s draft report and that it would give further consideration to price
regulation of airport services ‘in the light of developments in airport prices and the
Productivity Commission’s final report’ (Hockey 2001).

3.3 Quality of service monitoring

The Airports Act (described in section 3.5) provides for the ACCC to undertake
monitoring of quality of service provision by airport operators at core-regulated
airports as a complement to price regulation. Quality of service ‘will be taken into
account by the ACCC in its deliberations on pricing proposals’ (DoTRD
1996, p. 6), including in its assessment of airport compliance with the price cap and
its consideration of prices notification by Sydney Airport.18

Prior to October 2001, quality of service monitoring applied to all airports subject to
price regulation (that is, all core-regulated airports). As noted above, since the
October 2001 regulatory changes, quality of service monitoring has continued to
apply to all core-regulated airports, even though several are no longer subject to
price regulation (Coolangatta, Alice Springs, Hobart, Launceston and Townsville).

Quality of service monitoring, together with the provision of airport company

accounts to the ACCC, is intended to assist in improving the transparency of airport
performance. In this capacity, monitoring is intended to: discourage airport
operators from providing unsatisfactory service quality; encourage them to provide
information as a basis for improved negotiation between them and airport users; and
to assist the Government to address other public interest matters relating to the
regulation of airports (ACCC 1997b).

Under the Airports Act, the ACCC is limited to monitoring the services and
facilities provided by, or which could be influenced by, airport operators. Thus the
ACCC does not directly monitor service quality of other key organisations
providing services at the airport — for example, airlines, Airservices Australia
(ASA), Australian Customs Service, Australian Quarantine and Inspection Service,
Department of Immigration and Multicultural Affairs, and Australian Federal

17 Direction No. 25, October 2001. Replaced Direction No. 21.

18 Direction No. 20, October 2000, and Direction No. 18, June 2000, respectively.

56 PRICE REGULATION
OF AIRPORT
SERVICES
Police. Consequently, key aspects of an airport’s operations, such as domestic
terminals leased to airlines, are not subject to ACCC monitoring.

Airports Regulations 1997 (Airports Regulations), pursuant to the Airports Act,

specify the performance indicators to be used by the ACCC in monitoring and
evaluating the quality of airport services.19 The ACCC uses approximately 50
indicators that cover the following services and facilities:
• airside services (for example, taxiways and aerobridges);
• terminal services (for example, gate lounges and baggage handling); and
• ground access services (for example, car parking).

There are usually several indicators for each service. The services covered, and the
indicators, for Phase 1 airports and Sydney Airport, are different in some respects
from those for Phase 2 airports.

The indicators specified for gates and aircraft parking services for Phase 2 airports
are illustrative of the indicators used. They are:
• any change over time in the number of aircraft parking bays;
• any change over time in the total area of designated bay area; and
• the change over time in satisfaction with the system, according to a
questionnaire of airlines (Airports Regulations, Schedule 2, Part 2).

Some indicators are ‘static’ and objective (for example, the number of aerobridges)
and others are subjective (for example, satisfaction with check-in waiting times).20
Information on the indicators is sought from a variety of sources, including airport
operators (information for static indicators), passengers (survey), airlines (survey),
ASA and Australian Customs Service (survey).

Records to be kept by the airport operators, to be provided to the ACCC, are also set
out in the regulations.21 These vary between Phase 1 and Sydney airports, and
Phase 2 airports.

19 Airports Regulations, Schedule 2, Parts 1 and 2.

20 ‘Static’ indicators are presented by the ACCC in annual regulatory reports. They represent a
snap-shot at a point in time. For example, Brisbane Airport had 11 international aircraft
parking bays as at 30 June 2000 (ACCC 2001b).
21 Airports Regulations, Schedule 3, Parts 1 and 2.

THE REGULATORY 57
ENVIRONMENT
3.4 Access regulation
The access framework applying to airports determines the airport services or
facilities to be subject to access regulation, and sets out procedures for the terms and
conditions (including prices) of access to be determined. There are two separate
legislative instruments providing for access to airports: an airports-specific
instrument (s. 192 of the Airports Act); and a general instrument (Part IIIA of the
TP Act).

Other aspects of these acts that affect airport operations and have the potential to
affect the prices of airport services are described in section 3.5.

Section 192 of the Airports Act

Section 192 sets out an access regime for all privatised airports designated as core-
regulated airports under the Airports Act. Section 192 continues to apply to those
core-regulated airports no longer subject to price regulation since the regulatory
changes of October 2001 (Coolangatta, Alice Springs, Hobart, Launceston and
Townsville). The access regime allows airport operators 12 months after an airport
has been privatised (with a possible 12-month extension for Phase 2 airports) to
have an access undertaking accepted by the ACCC. An access undertaking is a
legally binding document setting out the terms and conditions under which access to
the services provided by essential airport facilities will be made available to access
seekers. If an undertaking is accepted by the ACCC, the services covered by the
undertaking cannot be declared for the purposes of Part IIIA of the TP Act (see
below).

If an undertaking is not accepted by the ACCC within the designated period, the
Minister must determine that each ‘airport service’ at the airport is a declared
service for the purposes of Part IIIA. There is no scope to appeal an ACCC decision
not to accept an undertaking, and an undertaking cannot be lodged once a service
has been declared. The Minister is required to specify the expiry date of the
determination, and no power is conferred on the Minister under s.192 to renew the
declaration once it has expired (ACCC, pers. comm., 5 July 2001).

Rather than listing declared services, the Airports Act sets out specified criteria for
declaration. The criteria are that the service:
(a) is necessary for the purposes of operating and/or maintaining civil aviation services
at the airport; and

58 PRICE REGULATION
OF AIRPORT
SERVICES
 (b) is provided by means of significant facilities at the airport, being facilities that
cannot be economically duplicated;
and includes the use of those facilities for those purposes. (s. 192(5))

Section 192 provides for the ACCC to determine whether a specified service
satisfies the criteria and, therefore, whether it is an airport service covered by the
Minister’s declaration. Amendments were made to s. 192 in 1998 (s. 192(4A–D)),
which gave the ACCC the power to make a written determination that a service is,
or is not, an airport service without reference to the criteria in s. 192(5). However,
the ACCC has stated that a determination always will involve an assessment of the
criteria in s. 192(5), and only where application of the criteria will result in perverse
outcomes will the ACCC make a decision that appears inconsistent with the criteria
(ACCC 1998g). A determination under the amendments can be disallowed by the
Parliament under s. 48 of the Acts Interpretation Act 1901, and can be reviewed by
the Federal Court under s. 39B of the Judiciary Act 1903 in limited
circumstances.22

The ACCC may determine that a service is an airport service on its own volition,
but generally will do so only on the application of an interested party. It will
generally conduct a public inquiry, except where it is clear the service satisfies the
declaration criteria (ACCC 2000a). If a service is determined to be an airport
service, then the airport operator must negotiate commercial terms with access
seekers, or submit to arbitration.

The Airports Act does not provide for appeal against a decision to declare a service
as an airport service.23 However, s. 44K of the TP Act provides limited scope for
appeals to be lodged with the Australian Competition Tribunal within 21 days of the
Minister’s automatic declaration of airport services at an airport.24 There is
currently no scope for appeal against undertaking decisions. Appeal rights for
arbitrated terms and conditions of access exist under Part IIIA.

22 Section 192(4A–B) is delegated legislation, which may be declared to be invalid by the

Federal Court on the grounds of unreasonableness if it leads to manifest arbitrariness, injustice
or partiality (ACCC, pers. comm., 29 June 2001).
23 Applications may be made to the Administrative Appeals Tribunal for review of decisions
made by the Minister under the Airports Act (s. 242). However, this section is not applicable to
access decisions.
24 Because the automatic declaration of ‘airport services’ at an airport does not refer to specific
services, whether a specific service is declared will only be determined following an ACCC
determination that the service is an ‘airport service’ as defined in s. 192(5). Unless such
determinations are made within 21 days of the automatic declaration, the appeals processes
prescribed in the TP Act are not available.
THE REGULATORY 59
ENVIRONMENT
Melbourne and Perth airports lodged access undertakings with the ACCC in early
1998, covering a range of services within and outside the price cap. In its draft
determinations, the ACCC stated that it did not accept either undertaking because,
in its view, the dispute resolution procedures were inadequate and the provisions of
the undertakings were not of sufficient clarity to be enforceable (ACCC 2000a).
Neither airport pursued lodgement of an undertaking beyond this point. (The
lodgement of undertakings by Melbourne and Perth airports is discussed further in
chapter 9.)

Given that no undertakings had been accepted by the ACCC prior to expiry of the
designated period, the Minister determined that airport services at all privatised
core-regulated airports were declared. The Minister also determined that the
declarations should expire on 30 June 2002 for Phase 1 airports and 30 June 2003
for Phase 2 airports. Included are services provided by those other than airport
operators, such as freight operators and terminals leased by airlines. As noted
above, there is no provision for the Minister to renew these declarations.

The ACCC has received two applications for a determination on whether a service
is a declared airport service under s. 192. The applications — from Delta Car
Rentals and Virgin Blue — both related to services at Melbourne Airport. (ACCC
decisions relating to these applications are discussed in chapter 9.)

As noted above, s. 192 sets out an access regime for the privatised core-regulated
airports. Airport services at Sydney Airport are currently not declared under s. 192.
However, the Part IIIA access regime applies to Sydney Airport (and non-core-
regulated airports — section 3.6).

Part IIIA of the Trade Practices Act

Part IIIA of the TP Act establishes a legislative regime to facilitate third-party

access under agreed or determined terms and conditions (including prices) to the
services of essential facilities25 of national significance in prescribed circumstances.
Part IIIA is intended to apply when commercial negotiations fail to secure access to
the services provided by an essential facility or when parties cannot reach
agreement concerning the terms and conditions (including prices) of access. It
applies to all incorporated and unincorporated businesses and Commonwealth, State
and Territory government businesses, and thus applies, in principle, to all Australian

25 Facilities that are the source of intermediate services essential to upstream or downstream
service provision.
60 PRICE REGULATION
OF AIRPORT
SERVICES
airports, including core-regulated airports and airports operated by local
government.26

In providing for access to the services of an essential facility, Part IIIA allows
regulation of the terms and conditions for use of the facility. This is considered
necessary because imposing an obligation to supply could be circumvented by the
terms and conditions of supply, especially price.

Under Part IIIA, a third party can seek access to eligible services through one of
three avenues:
• through a request that the National Competition Council (NCC) recommend that
the responsible Minister declare access to those services;
• through a legally binding undertaking from the facility operator approved by and
registered with the ACCC; or
• through a State or Territory access regime certified by the responsible Minister
as being effective following a recommendation by the NCC.

Specific criteria must be satisfied before access can be obtained through any of
these avenues. While the exact criteria differ among the three avenues, the criteria
generally refer only to access to the services provided by nationally significant
natural monopoly facilities.

Given that no airport undertakings have been accepted by the ACCC and that
certification is not relevant in the case of access to airport services (Commonwealth
access regimes currently are not covered by the certification provisions), parties
seeking access to airport services under Part IIIA must request the NCC to
recommend that the responsible Minister declare the service.

When assessing declaration applications, the NCC and the designated Minister must
consider six criteria specified in the TP Act:
(a) that access (or increased access) to the service would promote competition in at
least one market (whether or not in Australia), other than the market for the service;
(b) that it would be uneconomical for anyone to develop another facility to provide the
service;
(c) that the facility is of national significance, having regard to:
(i) the size of the facility; or

26 While all airports are subject to declaration, there may be limits on arbitration of access
disputes. Under s. 44R, access disputes can only be dealt with if either the provider or access
seeker is a corporation, or if access is required for interstate trade or commerce (ACCC, pers.
comm., 3 May 2001).
THE REGULATORY 61
ENVIRONMENT
 (ii) the importance of the facility to constitutional trade or commerce; or
(iii) the importance of the facility to the national economy;
(d) that access to the service can be provided without undue risk to human health or
safety;
(e) that access to the service is not already the subject of an effective access regime;
(f) that access (or increased access) to the service would not be contrary to the public
interest. (ss 44G(2) and 44H(4))

If a service is declared, the service provider must attempt to negotiate mutually

acceptable terms of access with an access seeker. If negotiations fail, the parties can
appoint a private arbitrator to determine terms and conditions of access. If the
parties cannot agree on an arbitrator, the dispute can be referred to the ACCC to
determine whether, and on what terms and conditions, access should be provided.
Access determinations are enforced via the Federal Court.

As noted above, s. 44K of the TP Act provides for the service provider (if the
Minister declares a service) or the access seeker (if the Minister decides not to
declare a service) to apply in writing to the Australian Competition Tribunal for a
review of the declaration decision. The application for review must be made within
21 days after publication of the Minister’s decision. The review by the Tribunal is a
re-arbitration of the access dispute, and the Tribunal may affirm, vary or set aside
the Minister’s decision.

There has been only one case seeking declaration of an airport service under
Part IIIA. In November 1996, Australian Cargo Terminal Operators Pty Ltd
(ACTO) — a small business which provides cargo terminal services to international
airlines — sought access to various freight handling and related services provided
by facilities owned by the FAC at Melbourne and Sydney airports (chapter 9).

Part IIIA of the TP Act and the associated Clause 6 of the Competition Principles
Agreement (CPA) were the subject of a Productivity Commission inquiry in 2001.
(For a more detailed discussion of Part IIIA, see PC (2001a).)

Other access arrangements

Access arrangements for privatised, core-regulated (and non-core-regulated)

airports are also contained in the individual airport leases. The publicly available
leases for Phase 1 and 2 airports contain similar access provisions, and require the
lessee at all times to provide for the use of the airport site as an airport, and to
provide for access to the airport by intrastate, interstate and international (where

62 PRICE REGULATION
OF AIRPORT
SERVICES
applicable) air transport.27 The leases provide for the lessee to comply with any
demand management scheme under the Airports Act, and to refuse to give access to
aircraft where an owner or operator of the aircraft has failed to pay any amount due
for the use of the airport (and where the lessee has provided the Government with
notice of its intention to refuse access), without being in default of these obligations.

While the access provisions in airport leases provide for access to airports by
aircraft operators, they do not provide for determination of the terms and conditions
of access. Further, the access provisions do not provide for access by other airport
users.

In some cases, access arrangements are also contained in airport sale agreements.
For example, the sale agreement for Canberra Airport requires the airport operator
to negotiate in good faith on access arrangements for the Very High Speed Train
project (Commonwealth of Australia 1998b).

Some years prior to privatisation, long-term domestic terminal lease agreements

between the FAC and the incumbent airlines, Qantas Airways and Ansett, were re-
negotiated. Among other things, the re-negotiated lease agreements required the
incumbents, on application, to make gates at their terminals at some core-regulated
airports available to new entrants — two gates each in Sydney and Melbourne and
one gate each in Adelaide, Perth and Coolangatta. Ansett was also required to make
a gate available at Launceston Airport (IC 1992).

Access to ground-handling services by international airlines is covered in

Australia’s bilateral air services arrangements (section 3.5), which state that ground
services shall be available on an equal basis to all airlines (DoTRS 2000c).

Most of the larger Australian airports have developed a general ‘conditions of use’
document, which sets out physical arrangements, and security and other
requirements, together with commercial conditions such as facilities provided, fees,
services and indemnities that would apply for potential new airlines to obtain access
to the airport. Similar documents exist or are being developed by airports to
facilitate access by operators wishing to provide ground-handling services at an
airport (DoTRS 2000c).

27 Subject to a force majeure clause relating to the lessee’s responsibilities in case of an event
that is beyond the reasonable control of the lessee and prevents the lessee from providing for
access.
THE REGULATORY 63
ENVIRONMENT
3.5 Other relevant regulation for core-regulated
airports
All core-regulated airports were subject to price regulation prior to changes in
October 2001. Since then, price regulation has been removed from several of these
airports (Coolangatta, Alice Springs, Hobart, Launceston and Townsville).
However, they all continue to operate within a comprehensive and complex (non-
price) regulatory environment that includes both domestic (Commonwealth, and
State and Territory) and international agreements. These airports are also subject to
the terms and conditions set out in airport lease and sale agreements.

All of these regulations affect the operations of airports (directly or indirectly) and
therefore have the potential to affect the prices of airport services. Moreover, the
domestic regulations are administered independently by different regulatory
agencies, yet often the effects of these regulations are related. For example, the
ability of airports to comply with price-cap arrangements under the PS Act may be
affected by their compliance with other (non-price) regulatory requirements.

Commonwealth regulation

Trade Practices Act

All core-regulated airports are corporations and therefore are subject to the
provisions of the TP Act, administered by the ACCC. Provisions that may be of
relevance to airport operations include those relating to access to services
(Part IIIA — as discussed in section 3.4), and anti-competitive practices (Part IV).

Part IV prohibits a range of anti-competitive practices. Arrangements or contracts

that contain exclusionary provisions or have the purpose, or effect, of substantially
lessening competition — for example, price fixing, secondary boycotts and
restricting supply — are prohibited (ss 45–45EB).

Misuse of market power is also prohibited. A corporation that has substantial

market power is prohibited from taking advantage of that power for the purpose of
eliminating or damaging a competitor, preventing entry into any market and
deterring or preventing a person from engaging in competition (s. 46).

Exclusive dealing and resale price maintenance are prohibited under ss 47 and 48
respectively.

The TP Act also prohibits mergers and acquisitions that would result in substantial
lessening of competition in a substantial market (s. 50). With this in mind, the

64 PRICE REGULATION
OF AIRPORT
SERVICES
ACCC, which assesses proposed mergers and acquisitions, examined all bids for
airport leases for potential breaches. In particular, the ACCC considered whether
the operator of a major airport (Adelaide) would be permitted to lease a smaller,
nearby airport (Parafield). The ACCC (2000a) concluded that the acquisition would
not substantially lessen competition.

Airports Act

One of the key pieces of economic regulation of core-regulated airports, including

Sydney Airport, is the Airports Act, along with the associated Airports
Regulations.28 The Airports Act has a number of stated objectives, including:
• to promote the sound development of civil aviation in Australia;
• to establish a system for the regulation of airports that has due regard to the
interests of airport users and the general community;
• to promote the efficient and economic development and operation of airports;
and
• to ensure diversity of ownership and control of certain major airports.
(Part 1, s. 3)

Some of the key features of the Airports Act relate to access, ownership controls,
environmental management, building and construction controls, demand
management and quality monitoring (box 3.3). Regulation under the Airports Act
relating to access arrangements and monitoring of the quality of service at core-
regulated airports is discussed in sections 3.4 and 3.3 respectively.

In general, provisions of the Airports Act apply to Sydney Airport as they do to

other core-regulated airports. However, in some instances, the Act regulates Sydney
Airport to a greater extent — for example, in relation to cross-ownership and
demand management. In other areas, Sydney Airport is subject to less regulation; it
has been granted an extension on the development of a master plan because of the
proposed sale of the airport, and access provisions (s. 192) are not applicable until
the airport is sold.

28 The Airports Act was amended by the Aviation Legislation Amendment Act 1997 and the
Airports Amendment Act 1999, which extended the period available for lessees to negotiate
access undertakings. Neither made substantial changes to the key features of the Airports Act
(box 3.3).
THE REGULATORY 65
ENVIRONMENT
 Box 3.3 Key features of the Airports Act
The Airports Act sets out a regime for the regulation of airports, including:
• airport operators are subject to a 49 per cent limit on foreign ownership, a 5 per cent
limit on airline ownership and a 15 per cent limit on cross-ownership for
Sydney/Melbourne, Sydney/Brisbane and Sydney/Perth airports (Part 3). In early
2002 the Government intends to introduce legislation to change the 5 per cent
airline ownership rule so that general aviation airports are exempted (Anderson
2001a);
• each airport must have an airport master plan approved by the relevant Minister.
This 20-year forward plan identifies, among other things, development objectives,
assesses the future needs of aviation users, forecasts noise exposure levels and
includes proposals for land use and related development (Part 5, Div. 3);
• a major development plan (approved by the relevant Minister) is required for each
major airport development — for example, constructing a new runway, constructing
a new building or extending a taxiway or road (Part 5, Div. 4). The Airports (Building
Control) Regulations 1996 establish a system for approval of building and
construction activity on airports;
• an environment strategy, also subject to approval by the relevant Minister, must
specify, among other things, environmental management objectives and sources of
environmental effects associated with airport operations (Part 6, Div. 2);
• an airport operator may be required to give accounts and reports to the ACCC
(Part 7, Div. 3). In addition, the ACCC will monitor the quality of airport services and
facilities (Part 8, Div. 1);
• regulations may implement certain international agreements (Part 10, Div. 8);
• an airport service will be a declared service for the purposes of the access regime
under Part IIIA of the TP ACT unless an access undertaking is given within 12
months of privatisation (Part 13, Div. 2, s. 192). An airport may also be required to
provide access for defence, emergency or disaster relief purposes (Part 9, Div. 10);
• the relevant Minister may formulate a demand management scheme for an airport,
being a category exclusion scheme, a slot allocation scheme, or a movement
limitation scheme (Part 13, Div. 1). The Minister may also declare the capacity of an
airport, in terms of the maximum number of aircraft movements an airport is capable
of handling in a specified time period (Part 13, Div. 4); and
• ASA will oversee the provision of air traffic services and rescue and firefighting
services at airports (Part 14, Div. 10).
Source: Airports Act 1996.

66 PRICE REGULATION
OF AIRPORT
SERVICES
Other regulation

The main other Commonwealth regulations affecting core-regulated airports’

operations, costs and pricing include:
• The Air Services Act 1995 and Air Services Regulations established the
structure, functions and powers of ASA. ASA’s functions include the provision
of facilities to permit safe navigation of aircraft within Australian airspace and
the provision of services, such as air traffic services, aeronautical information
and radio navigation services, and rescue and firefighting services
(Part 2, Div. 2, s. 8). Although ASA directly charges airlines for its services,
smaller core-regulated airports, such as Hobart Airport (sub. 11), contended that
relatively high ASA charges precluded the airport from increasing its charges.
• The Civil Aviation Safety Authority (CASA), which conducts the safety
regulation of civil air operations, was established in 1995 under an amendment
to the Civil Aviation Act 1988. Among other things, CASA is responsible for
establishing and enforcing aviation safety standards, and issuing aerodrome
licences. An airport will incur expenses associated with maintaining the airport
to comply with licence requirements.
• The Environment Protection and Biodiversity Conservation Act 1999 ensures
that matters significantly affecting the environment are fully considered through,
for example, environmental impact statements (EIS). Airport activities that
generate, or have the potential to generate, pollution or excessive noise are
regulated by the Airports (Environment Protection) Regulations 1976, made
under the Airports Act. Any compliance costs (for example, the development of
an EIS) must be met by the airport.
• All passengers, crews and freight (plants, animals and other goods) entering and
departing Australia via airports are subject to customs control procedures
(Customs Act 1901 and regulations). Passengers, freight and the aircraft may
also be subject to quarantine procedures under the Quarantine Act 1908,
Imported Food Control Act 1992 and Export Control Act 1982 (and regulations).
Passengers also are subject to immigration controls under the Migration Act
1958. Outbreaks of contagious diseases in other countries, such as foot and
mouth disease in Britain, can have a substantial effect on the processing of
passengers and freight at airports. Various Commonwealth agencies may be
involved in the above controls, including the Australian Customs Service,
Australian Quarantine and Inspection Service, Department of Immigration and
Multicultural Affairs, and Australian Federal Police. These services generally
require substantial floor space, which is provided by airport operators for
nominal rent.

THE REGULATORY 67
ENVIRONMENT
There are a number of other Commonwealth regulations that affect the operation of
airports, and thus indirectly may affect pricing.
• The Air Navigation Act 1920 and Air Navigation Regulations 1947 regulate air
navigation, such as international aircraft, international airlines, non-scheduled
flights, aircraft on international flights, investigation of accidents, and aviation
and airport security (including passenger and freight). In particular, the Act
approves ratification of the Convention on International Civil Aviation (Chicago
Convention) (see below). Pursuant to this Act, the Air Navigation (Coolangatta
Airport Curfew) Regulations 1999 restrict certain aircraft movements at that
airport between 11pm and 6am.
• The Adelaide Airport Curfew Act 2000 and associated regulations impose curfew
restrictions, similar to those at Coolangatta, on aircraft movements at Adelaide
Airport.
• Curfew arrangements at Sydney Airport are prescribed under the Sydney Airport
Curfew Act 1995 and associated regulations. As at Adelaide and Coolangatta
airports, a curfew applies between 11pm and 6am. However, there are also
several related restrictions controlling use of runways during shoulder times on
week days and weekends, international passenger aircraft movements and quota
requirements applying to take-offs and landings. The Commonwealth
Government has confirmed that the curfew will remain in place following the
sale of the airport (Anderson and Kemp 2001).
• Slot management for aircraft arrivals and departures at Sydney Airport is
regulated separately from other core-regulated airports (box 3.4).
• A ‘noise’ levy is imposed on the operators of jet aircraft landing at leviable
airports under the Aircraft Noise Levy Act 1995 and the Aircraft Noise Levy
Collection Act 1995. An airport is deemed to be leviable if it meets certain
criteria regarding residential exposure to particular noise levels. There are two
leviable airports: Sydney and Adelaide. The levy is not imposed on aircraft with
noise assessed to be below a specified level.

68 PRICE REGULATION
OF AIRPORT
SERVICES
 Box 3.4 Sydney Airport demand management
The Sydney Airport Demand Management Act 1997 prescribes a maximum of 80
aircraft movements in any hour and establishes a framework for a slot management
scheme.29 Details of the system for allocating slots are set out in the Slot Management
Scheme 1998 (made under the Airports Act), including grandfather rights to slots, the
‘use it or lose it’ test, how to apply for a slot, how slots are allocated and slot swaps.
The slots are allocated and administered by the Slot Manager, Airport Coordination
Australia.
One feature of the slot management scheme is the ‘regional ring fence’ which
effectively creates a separate pool for regional slots. Under these special rules, a
regional service operator may gain a permanent slot if a service is operated for two
consecutive seasons (s. 10). Moreover, regional slots cannot be transferred to non-
regional services (s. 19).
Enforcement arrangements are set out in the Sydney Airport Compliance Scheme
1998, also made under the Airports Act. The Commonwealth Government has
confirmed that the 80 movements per hour cap and regional ring fence will continue
after the sale of the airport (Anderson and Kemp 2001).

Airport lease and sale agreements

The lessee of each core-regulated airport (including Sydney Airport even though it
is not privatised) has signed a lease agreement and a sale agreement with the
Commonwealth of Australia (the lessor).

Lease agreements

The lease agreements set out various obligations on the part of lessees.30 For
example, Australian Airports (Townsville), as the lessee:
• must comply with certain access conditions (described in section 3.2);
• must maintain the airport in good repair;
• must develop the airport at its own cost, having regard to the actual and
anticipated future growth in traffic demand, ‘reasonably expected’ quality
standards and good business practice; and

29 A slot is defined under s. 34 of the Act as ‘a permission for an aircraft movement’. A slot
allocated under the scheme permits a specified aircraft movement at a specified time on a
specified day.
30 These agreements also include obligations on the part of the lessor.

THE REGULATORY 69
ENVIRONMENT
• must, if the option to renew the lease has not been exercised, rebuild any
structure damaged or destroyed during the last 10 years of the lease to an agreed
standard. In addition, any removal or demolition of structures during the last 10
years must be approved by the lessor (Commonwealth of Australia 1998a).

Perusal of other publicly-available leases for Phase 1 and other Phase 2 core-
regulated airports indicates that similar conditions apply across these leases.
However, there are some variations in leases to account for specific characteristics
of airports. For example, the Canberra Airport lease includes reference to the
Fairbairn defence base sub-lease (Commonwealth of Australia 1998b).

Sale agreements

Publicly-available airport sale agreements also do not vary significantly across core-
regulated airports in terms of the general contents. Each core-regulated airport’s
sale agreement contains, among other things, provisions relating to the transfer
(sale) price, sale completion process, resale restrictions and superannuation.
However, there are airport-specific variations to the agreements. For example,
unlike other airports, the Canberra Airport sale agreement contains a provision
obliging the airport to negotiate in good faith concerning access arrangements for a
potential very high speed train terminal (Commonwealth of Australia 1998b).
Moreover, clearly the agreements will differ to the extent that sale prices, deposits
and so on vary across airports.

The sale agreements commit core-regulated airports to spend a specified amount of

money on airport development, which could include constructing a new runway,
constructing a new passenger terminal or extending an apron. The airport is
committed to outlay a specified sum of money in each of two time periods — the
first period being within five years of the lease, and the second period being the five
years thereafter. However, there are provisions for non-compliance with the
development obligations, including force majeure31 and other delay events, and
mitigation due to economic factors, such as a shortfall in the targeted increase in
passenger numbers.

State and Territory regulation

In the absence of Commonwealth regulation, core-regulated airports and their

business activities are subject to State and Territory regulation. Even though
environmental, planning and building controls are predominantly Commonwealth,

31 Events beyond the reasonable control of the lessee.

70 PRICE REGULATION
OF AIRPORT
SERVICES
State and Territory legislation applies, for example, in the areas of waste
management and occupational health and safety.

State and Territory regulation, such as fair trading legislation, also relates to on-site
commercial trading, gambling, liquor licensing and vehicle parking at airports.32
The transition to State and Territory regulation in these areas was aided by the
promulgation of the Airports (Control of On-airport Activities) Regulations 1997.
Businesses operating at the airports are also subject to State and Territory
regulation.

Capital Airport Group (Canberra Airport) commented on the considerable degree of

Territory (box 3.5), Commonwealth and even State regulation with which it has to
comply and, in particular, that on occasions the regulation between these
jurisdictions appeared to overlap. For example, in the planning area, the National
Capital Authority overlaps with the Commonwealth airport building controller and
airport environment officer (Capital Airport Group, pers. comm., 30 April 2001).

Box 3.5 Territory and State regulation (and regulators) affecting the
operation of Canberra Airport
Some of the Territory and State regulation (and regulators) affecting the planning, and
development and construction, at Canberra Airport are:
Planning
• National Capital Authority;
• ACT Planning and Land Management;
• Environment ACT;
• ACT Retail Leases Code;
• ACT Tree Preservation legislation; and
• NSW Department of Urban Affairs and Planning (land use planning in flight path
corridors).
Development and Construction
• Building, Electrical and Plumbing Control Regional Office, ACT Government;
• ACT Electricity and Water;
• ACT Stormwater;
• ACT Fire Brigade; and
• ACT Public Health.

Source: Capital Airport Group (pers. comm., 30 April 2001).

There are few State and Territory regulations that are specific to airports or aviation.
Most relevant is the Aerodrome Fees Act 1998 in South Australia, which gives
private and public owners of former Commonwealth airports the statutory authority

32 Fair trading legislation substantially mirrors the consumer protection provisions of the TP Act.

THE REGULATORY 71
ENVIRONMENT
to recover fees from users who endeavour to avoid payment. Other States that have
addressed this issue have done so through the amendment of existing legislation; for
example, New South Wales amended its local government legislation.

International agreements

International agreements have an effect, both directly and indirectly, on the

operation of core-regulated airports. These airports are ‘designated’ international
airports under the Air Navigation Act.33 Perhaps of most significance is the
Convention on International Civil Aviation (Chicago Convention) — a multilateral
agreement signed in 1944 by Australia and other countries at the Chicago
Conference. As noted above, the Air Navigation Act approves the ratification by
Australia of the Chicago Convention (including the Protocols) and reproduces the
Convention as Schedule 1 of the Act.34

The Chicago Convention is the legal foundation for the regulation of world civil
aviation and includes several Articles that bear on economic regulation of aviation
directly, and thus affect airport operations and pricing (box 3.6). The Chicago
Convention (Article 44) also established the International Civil Aviation
Organization (ICAO) — a worldwide intergovernmental organisation seeking to
promote the safe and orderly development of international civil aviation. As a
United Nations agency, it ‘sets international standards and regulations necessary for
safe, regular, efficient and economical air transport and serves as the medium for
co-operation in all fields of civil aviation’ (ICAO 1996, p. 3.4-1).

33 Brisbane, Melbourne, Perth, Adelaide, Sydney and Darwin airports are designated ‘major’
international airports. Designated ‘restricted use’ and ‘alternate’ international airports are
Canberra, Coolangatta and Townsville. Hobart is a designated ‘restricted use’ international
airport. Launceston and Alice Springs are designated ‘alternate’ international airports.
Currently, there are no international RPT operations at several of these airports (for example,
Hobart, Launceston and Townsville). However, there have been international RPT operations
in the past at some of these airports (for example, Hobart and Townsville). The designated
international status of leased non-core-regulated airports and other airports is described in
section 3.4.
34 Several Protocols amending or supplementing the Chicago Convention have been approved by
ICAO over the ensuing years.
72 PRICE REGULATION
OF AIRPORT
SERVICES
 Box 3.6 The Convention on International Civil Aviation
Key economic regulatory features of the Chicago Convention include:
• state sovereignty over airspace (Article 1);
• rules governing permission for international non-scheduled and scheduled air
services in contracting states (Articles 5 and 6);
• cabotage — that is, contracting states may refuse permission for aircraft from other
contracting states to take on its passengers, mail and so on (Article 7);
• the requirement that regulations of the contracting state regarding entry, customs,
quarantine and so on must be complied with by other contracting states (Article 13);
• the requirement that airports in contracting states that are open for public use by
national aircraft must be open under uniform conditions to aircraft from other
contracting states. Airport charges for aircraft from other contracting states should
be no higher than those for national aircraft (Article 15);
• the requirement that air navigation be expedited and that unnecessary (particularly
administrative) delays to aircraft, passengers, crews and cargo be prevented (Article
22); and
• the requirement that contracting states agree to provide airports and navigation
facilities which facilitate international air navigation and are in accordance with the
Convention. Operational systems should also be in accordance with the Convention
(Article 28).
Source: Convention on International Civil Aviation.

ICAO develops and releases various policies and guidelines on the regulation of air
transport (for example, on airport charges, capacity and tariffs), manuals (including
one to provide guidance to airport managers) and technical annexes to the Chicago
Convention dealing with areas such as aeronautical communications, ground
handling, meteorology, operations and security.

Another major agreement signed at the Chicago Conference was the International
Air Services Transit Agreement (known as the Two Freedoms Agreement), which
provides for the multilateral exchange of rights of overflight and non-traffic landing
for scheduled international air services among contracting states. Australia’s
ratification of this agreement has been approved under the Air Navigation Act.

The General Agreement on Trade in Services (GATS) has the potential to affect
airport operations through the liberalisation of international trade in air services. As
a member of the World Trade Organization, Australia is required to accept the
GATS as part of the outcome of the Uruguay Round of multilateral trade
negotiations concluded in 1994.

THE REGULATORY 73
ENVIRONMENT
GATS coverage of air services is limited. The Annex on Air Transport Services to
GATS excludes aviation with three exceptions where GATS commitments apply:
aircraft repair and maintenance services; selling and marketing of air transport
services; and computer reservation system services. However, these services are
covered only if member countries schedule them as specific commitments. Progress
generally has been slow, and Australia has not yet scheduled all three.

Bilateral, rather than multilateral, agreements (air service arrangements) are a

feature of the regulation of international air services. Australia has over 50 bilateral
agreements. Essentially, they define the terms and conditions under which airlines
of either party will have access to the airspace of the other party. Provisions
typically include the specification of capacity, frequency, routes, cities, ownership
provisions and price approval processes (PC 1998a). For example, Australia’s
agreements with Japan and Italy guarantee the designated airlines of these countries
access to particular Australian airports and specify the capacity on those routes.
Such agreements may directly affect airports by placing limits on international
traffic.35 Airports therefore may be limited in the amount of business they can
attract.

However, the nature of bilateral agreements involving Australia is changing. The

Productivity Commission examined bilateral agreements, among other things, in its
inquiry into International Air Services (PC 1998a). It recommended that Australia,
as a step towards further liberalisation of international air services, should seek to
negotiate reciprocal ‘open skies’ agreements on a bilateral basis which would
remove various restrictions, including those on passenger and freight capacity and
frequency, routes, multiple destinations, prices and cabotage. The Commission also
recommended that ‘Australia should offer unlimited capacity to fly to all airports
other than Sydney, provided that Australian carriers are offered the same routes on a
reciprocal basis by their bilateral partners’ (PC 1998a, p. 233).

The Commonwealth Government has implemented most of these recommendations

(Costello and Anderson 1999) and is currently re-negotiating bilateral agreements
with other countries (for example, Singapore, the United Kingdom and the United
States) in moving towards a target of open skies.36 In November 2000, Australia’s
first open skies agreement — with New Zealand — was signed (Anderson and
Gosche 2000). This move towards liberalisation of international air services is

35 Australian agreements relate to RPT, not charter flights.

36 The Commonwealth Government has not implemented the Commission’s recommendations in
full. For example, it has not removed restrictions on cabotage, and unlimited capacity will be
offered only for international airports other than Melbourne, Brisbane and Perth, as well as for
Sydney (Costello and Anderson 1999).
74 PRICE REGULATION
OF AIRPORT
SERVICES
likely to assist airports, where capacity exists, to gain increased international
passenger traffic.

International carriers with access must be treated in accordance with GATS, the
Chicago Convention and bilateral agreements. Airports have an obligation to abide
by the intent and spirit of the agreements negotiated by their governments. If an
international carrier considers that a breach might have occurred, it may ask its
government to take up the complaint with the Australian Government. As noted
above, the Airports Act (Part 10) provides for the introduction of regulations to
require an airport operator to comply with particular international agreements.
Penalties apply for non-compliance. An example of such a complaint occurred in
the 1980s when US carriers alleged that the UK Government (because Heathrow
Airport is operated by the British Airports Authority) had breached the user charges
article in the Bermuda II agreement — a bilateral agreement between the United
Kingdom and United States. The US Government took the complaint of the carriers
to the UK Government. An arbitral tribunal was established to adjudicate the issue
(Toms 1994).

Apart from ICAO, the other major international body to influence the regulation of
international air services is the International Air Transport Association (IATA).
IATA, established in 1945, is a trade association currently representing over 230
airlines. Its stated goals include:
• providing safe, reliable and secure air services;
• developing cost-effective, environmentally friendly standards and procedures to
facilitate the operation of international air services; and
• identifying and articulating common industry positions and supporting the
resolution of key industry issues (IATA 2001).

IATA coordinates and standardises technical, operational and commercial aspects

of airline operations through its committee work, and it acts as a clearing house for
inter-airline accounts. For example, voluntary scheduling conferences provide a
forum for reaching consensus on schedule adjustments necessary to meet airport
capacity constraints, and tariff coordination conferences enable the discussion of,
and agreement on, ‘reference fares’ for passenger fares and cargo rates to facilitate
interlining and the division of revenue between interlining airlines.37

37 Interlining involves connecting passengers between two airlines on the same ticket.

THE REGULATORY 75
ENVIRONMENT
3.6 Regulation of airports not subject to price
regulation
Most airports in Australia are not subject to price regulation; they cater for general
aviation traffic rather than RPT (chapter 2). The regulatory environment differs
depending on whether these airports are leased from the Commonwealth
Government.

Leased core-regulated airports

As noted in section 3.2, in October 2001, the Government announced the removal
of all price regulation for five core-regulated airports; Coolangatta, Alice Springs,
Hobart, Launceston and Townsville. The non-price regulatory framework within
which these airports operate remains unchanged, and is described in section 3.5.

Leased non-core-regulated airports

Seven non-core-regulated airports have been leased by the Government and

privatised: Mount Isa, Tennant Creek, Archerfield, Jandakot, Moorabbin, Parafield
and Essendon.38 In addition, three Sydney basin airports (Bankstown, Camden and
Hoxton Park) have been leased but are yet to be privatised. (Sydney Airport — a
core-regulated airport — was discussed in previous sections of this chapter.) These
10 airports, which are not subject to price regulation, must, however, comply with
other Commonwealth legislation applying to core-regulated airports, including the
Airports Act.

Some of the provisions of the Airports Act, such as the access provisions (s. 192)
and the provisions on quality of service monitoring (Part 8), do not apply to these
10 airports because they are not core-regulated airports. However, the Act does
apply where airports are specified in the regulations. For example, all of these
airports, except for Mount Isa and Tennant Creek, are specified in the Airports
Regulations for the purposes of land use, planning and building controls. Thus, they
must meet the requirements of Part 5 of the Act, including the development of a
master plan.

Airport lease agreements for these leased non-core-regulated airports are similar to
Phase 2 core-regulated airport agreements, but with some airport-specific variation.

38 As noted in chapter 1, the sale of Essendon Airport was completed in September 2001 and the
Sydney basin airports are to be sold in the second half of 2002.
76 PRICE REGULATION
OF AIRPORT
SERVICES
For example, unlike Phase 2 core-regulated airports, Mount Isa Airport has no
provisions in its lease for the costs of an airport environment officer.39

Sale agreements for the airports that have been privatised are, in general, similar to
those for core-regulated airports. However, one important difference is that these
airports are not subject to airport development obligations under the agreement.

As with core-regulated airports, leased non-core-regulated airports are subject to

State and Territory regulation where Commonwealth regulation does not apply.

Unlike the core-regulated airports, none of these airports are designated

international airports. In the absence of international RPT services, international
agreements are of substantially less relevance.

Other airports

Other non-core-regulated airports include regional airports with both RPT and
general aviation such as Orange, Mildura and Broome, and numerous smaller
general aviation airports such as Lilydale and Pine Creek (chapter 2). Also included
in this ‘other’ group is Cairns Airport. Over 200 airports are owned and operated by
local government. Other airports are owned and operated by statutory government
authorities (for example, Cairns) and the remainder are under private control.

These other airports are not subject to Commonwealth price regulation or the
provisions of the Airports Act. They are, however, subject to some other
Commonwealth legislation — for example, the Air Navigation Act and Part IIIA of
the TP Act.40 Part IV of the TP Act applies to airports that are corporations.
Airports that are unincorporated businesses are subject to State and Territory
competition codes (a modified version of Part IV) enforceable by the ACCC.

Without the regulation of the Airports Act, State and Territory government
regulation applies to this group of airports to a greater extent than it applies to
airports leased from the Commonwealth. For example, planning and approval of
building and construction at Commonwealth leased airports is regulated by
Commonwealth legislation; State and Territory legislation applies to building and
construction activity at these other airports.

39 Pursuant to the Airports Act, the Airports (Environment Protection) Regulations 1997 exempt
Mount Isa Airport from the environment provisions of Part 6 of the Act.
40 All Australian airports are subject to declaration under Part IIIA of the TP Act. However,
under s. 44R, there may be limits on arbitration of access disputes for some regional airports,
including those operated by local municipalities (section 3.2).
THE REGULATORY 77
ENVIRONMENT
International agreements are of little relevance to these airports because few are
designated international airports, and when they are designated, it is generally on a
limited basis.41 An exception is Cairns Airport, which has a considerable number of
international RPT services. Some others, such as Broome and Port Hedland airports,
have international RPT services from time to time (DoTRS 2000b). International
agreements affect the costs and pricing of these airports in a similar manner to core-
regulated airports.

41 Designated international airports (either ‘restricted use’ and/or ‘alternate’, ‘non-scheduled’ or

‘external territory’) in this ‘other’ airports category are Cairns, Avalon, Broome, Coffs
Harbour, Learmouth, Lord Howe Island, Port Hedland, Dubbo, Kalgoorlie, Rockhampton,
Tindal, Horn Island, Norfolk Island, Christmas Island and Cocos (Keeling) Island.
78 PRICE REGULATION
OF AIRPORT
SERVICES
4 Assessing the need for price
regulation of airport services: some
preliminary issues

Possible rationales for future price regulation of airport services are discussed in
this chapter. Principles for efficient pricing of airport services and criteria for
efficient regulation are also outlined.

4.1 Why regulate prices of airport services?

The terms of reference ask the Commission, in making its recommendations, to
‘identify the rationale for any future prices regulation at airports’ (para. 8(b)). Most
rationales relate to potential abuses of market power by airports — in short, their
potential to raise prices substantially above costs and the associated inefficient
behaviour this may allow or promote.

Pricing with market power

The prima facie rationale for price regulation of certain airports is their perceived
market power — that is, an ability to raise prices above efficient levels — and their
perceived incentive to use it. (What is meant by efficient pricing is discussed
below.) This concern is reflected in the terms of reference, which state that:
… the purpose of this inquiry is to examine whether new regulatory arrangements,
targeted at those charges for airport services or products where the airport operator has
been identified as having most potential to abuse market power, are needed to ensure
that the exercise of any such power may be appropriately counteracted. (para. 6)

Airports have natural monopoly characteristics (chapter 5). Though the tendency
towards natural monopoly arises from efficiency benefits in the provision of airport
services (one airport in a particular location can provide services more efficiently
than two), it inevitably reduces potential competitive pressures on airports.
However, the extent of market power of a particular airport in practice will depend
on a range of factors, including the price responsiveness of air travellers to that
destination, the share of airport costs in the airfare, airport substitution possibilities
and the cost structures of other input suppliers (chapter 5). Other factors, such as the
PRELIMINARY ISSUES 79
influence of potential profits from commercial activities at airports and
countervailing power of airlines, may affect the incentive for, and ability of, airports
to exercise fully any market power they possess (chapter 7).

Nonetheless, the concern remains that airports with market power will increase
aeronautical charges above (efficient) costs, thus increasing airfares and reducing
consumption of air travel. Persistently-high airport charges normally would also
yield monopoly profits for airports. In these circumstances, price regulation that
promoted efficient aeronautical prices could deliver economic gains to the general
community — provided such regulation did not create offsetting distortions and
costs — and bring benefits to passengers and airlines at the expense of airports.

Other potential effects of market power

It has been suggested by the Board of Airline Representatives of Australia (BARA,

sub. 26) that airports with market power are likely to be inefficient in production
(that is, that they will allow their production costs to rise), and that they may under-
supply quality. Airports with market power also may not invest efficiently to
maintain monopoly profits over time.

The likelihood of such inefficiencies occurring at privatised airports with market

power is discussed in chapter 7. To the extent that price regulation brought about
more efficient pricing, it may also reduce these ‘by-product’ inefficiencies.
However, in practice, some forms of price regulation may promote such forms of
inefficiency by, for example, encouraging firms to ‘pad’ costs (chapter 10).

Airport access

In addition to pricing above (efficient) cost, the market power of airports could also
be used to discourage competition in downstream (or upstream) markets (BARA,
sub. 26; ACCC, sub. 38). This is more likely to occur where airports actively
participate in those markets or can control competition in them. An airport with
market power, for example, may restrict ‘front-door’ access to competing, off-
airport car-park providers. This could be achieved by levying a prohibitively-high
charge or by denying the off-airport provider access to a conveniently-located
contact point at the airport. This is considered further in chapter 7.

However, because of strict limits on vertical integration of airports and airlines in

Australia (chapter 3), airports appear unlikely to have a strong incentive to deny

80 PRICE REGULATION
OF AIRPORT
SERVICES
access to new airline entrants, provided that the latter at least pay marginal cost and
that the price does not undermine contracts with incumbent airlines.1

Rather than being a rationale for price regulation, denial of access to competing
providers of ancillary services, or collusive actions between an airline and airport,
could be addressed more directly and satisfactorily by anti-competitive provisions
of the Trade Practices Act 1974, or by access regulation. Price regulation is unlikely
to be adequate to address access issues because access can be denied by employing
means other than high prices (for example, by restricting access to conveniently-
located apron and/or terminal facilities). Nonetheless, price regulation that prevents
excessive across-the-board price levels could help prevent marginal airlines
effectively being denied access, whether this was intended or not.

Airports as public infrastructure

Another suggested rationale for price regulation, not directly related to market
power as such, is that airports provide essential infrastructure that give benefits to
industries other than direct users, particularly the tourism industry. It is argued that
airlines should not be expected to pay the full costs of providing these facilities
when other industries benefit. However, though marginal-cost pricing will promote
efficient use of existing airport assets, such pricing may not provide adequate
returns on new airport investment and therefore may have perverse long-run
implications for airport users (section 4.3). Thus, price regulation that targets a price
equal to short-run marginal cost is unlikely to be consistent with efficient long-term
provision of airport services, unless it is accompanied by some form of
subsidisation of those services.

Airlines and their representative bodies (International Air Transport Association,

subs 9 and DR56; BARA, subs 26, 41 and DR54; Qantas, sub. 48) suggested that
aeronautical charges should be calculated on a residual basis, after taking into
account non-aeronautical net revenues earned by airports (the so-called single till).
As discussed in chapter 10 and appendix C, however, a single-till approach, in
practice, is unlikely to promote efficiency and may generate efficiency losses.
Alternatively, lower airport charges could be subsidised by taxpayers, though
general taxation also involves deadweight efficiency losses. It is a matter of
weighing up the efficiency effects of each financing option, including the option of
airport users paying the full costs of the aeronautical services provided. The
Commonwealth Government’s decision to privatise Australian airports without

1 As noted in chapter 3, airports are not permitted to deny landing access to planes (for safety
reasons) but effectively could deny access by imposing a very high landing charge or by
restricting access to other necessary ground infrastructure such as terminals or aprons.
PRELIMINARY ISSUES 81
mandating a single till suggests that it favours the full-cost approach. This was
confirmed in April 2001, in relation to future regulation of Sydney Airport, by the
Minister for Financial Services and Regulation in Direction No. 22 to the ACCC.

Regulating airport profits

It has been argued by Professor Forsyth that the major rationale for price regulation
is to limit the ability of airports to earn profits in excess of normal rates of return,
even though such regulation is likely to have a net efficiency cost. Thus, he argued
that:
Price regulation lessens, rather than increases, the overall efficiency of airports. While
not perfect, a private, unregulated airport is likely to perform better in pure efficiency
terms than a regulated airport because of the efficiency costs of regulation. (sub. 5, p. 5)

But that:
The dominant rationale in Australia for airport price regulation is one of eliminating
excess profits from the use of market power. (sub. 5, p. 5)

While governments may see a need to regulate some firms to reduce perceived
excess profits, the Commission has focused its attention on the likely efficiency
outcomes of various options, including no price regulation of airports.

4.2 Potential efficiency and distributional effects of

market power
As noted above, a major concern about the exercise of market power by airports —
and the key rationale for price regulation — is that prices will exceed efficient
levels, thereby reducing consumption. Principles of efficient pricing, given the
nature of the airport business, are explored in section 4.3. Discussion in this section
focuses on the economic cost of inefficient price levels, distinguishing between
efficiency and distributional effects.

This latter distinction is drawn because it is the impact on efficiency that determines
the net effect of market power on the size of the national economic ‘pie’.
Redistribution of income, of itself, as the result of higher (or lower) prices — at
least to the extent that income is transferred between Australian residents — does
not involve efficiency effects, though it may generate concerns in the community
about fairness. Typically, inefficient pricing involves both efficiency losses and
income transfers.

82 PRICE REGULATION
OF AIRPORT
SERVICES
Efficiency effects of market power

Consumption below its efficient level

In essence, a firm with market power (and assuming that price discrimination is not
feasible) will restrict the amount supplied and raise the price in order to increase its
profits at the expense of consumers. The source of efficiency loss is the reduction in
production and consumption of the good or service below the efficient level — the
so-called monopoly deadweight loss. This is depicted in a simple framework in
box 4.1.

How large this efficiency loss will be depends on the firm’s cost structure and the
price elasticity of demand. For a given marginal cost, and assuming no price
discrimination, the lower the price responsiveness (elasticity) of demand, the greater
the optimal price mark-up over cost.2

Several airports suggested, if elasticities of demand for airport services are very
low, that potential efficiency losses from airport market power also will be low
(compared with efficiency losses if demand were more responsive to price
increases). However, this result does not necessarily hold — it assumes a price
increase of a given amount, not price increases that accord with profit-maximising
behaviour. In other words, with unconstrained pricing, low demand elasticities
could be consistent with large efficiency losses because low elasticities imply high
prices.3 Therefore, in order to make an assessment of likely efficiency losses from
pricing consistent with market power, knowledge is required about the shape and
slope of the demand curve.

If a firm with market power can discriminate in pricing — that is, set different
prices for different consumers and/or units sold — then efficiency costs will be
reduced to the extent that consumption is not discouraged. In the limit, it is feasible
that the quantity provided could be equal to the competitive level (Qc in figure 4.1),
but with some consumers (with a higher willingness to pay) paying prices above
marginal cost.

2 The optimal mark-up is the inverse of the price elasticity of demand for the firm’s output,
where the relevant elasticity is measured at the mark-up inclusive price. Therefore some
caution is required in applying the inverse elasticity formula to an elasticity measured at a point
on the demand curve before a mark-up is imposed. Even if that measured elasticity is very low,
as the price is increased, the price elasticity typically will increase, with the rate of increase
depending on the slope of the demand curve. The flatter the demand curve, the more rapidly
will elasticity increase and therefore the smaller the optimal mark-up over costs.
3 Unless the elasticity of demand is zero — in which case, the optimal price mark-up is infinite
but the efficiency loss is zero.
PRELIMINARY ISSUES 83
 Box 4.1 Pricing with market power
Figure 4.1 depicts the cost and demand curves for a firm with market power. For
simplicity, it is first assumed that marginal costs (MC) are constant and therefore equal
to average costs (AC). (D is the demand curve; MR the marginal revenue curve.)
Pricing with decreasing costs is discussed in box 4.2. The framework is static in the
sense that it is concerned with the efficient use of existing airport infrastructure. Over
time, a firm with market power will invest to increase profits.

Figure 4.1

Pm a

Pc MC=AC
b
D
MR

Qm Qc Q

A profit-maximising firm with market power (assuming price discrimination is not

feasible) will set a price, Pm, at which marginal revenue and marginal cost are equal.
This price lies above the competitive, or efficient, price of Pc where price equals
marginal cost and at which price potential economic surplus (the triangle below the
demand curve and above costs) is greatest. At price Pm a lower quantity is consumed
and supplied (Qm) than if price were set equal to marginal cost (Qc). The shaded
triangle indicates the loss of allocative efficiency from the exercise of market power. It
shows the loss of consumers’ value of the good or service (over the costs of provision)
or the loss of consumer surplus less producer profits attributable to restricted
consumption and supply.

Other potential efficiency effects

Market power could lead to other efficiency losses. Lack of competitive pressure
could enable a firm to operate inefficiently by allowing its costs to rise or by it not
adopting cost-saving or innovative technologies. Such inefficiency could be at the
expense of the airport’s profits but may yield a ‘quiet life’ for managers. Firms with
market power may also seek to increase profits by allowing quality to fall, for
84 PRICE REGULATION
OF AIRPORT
SERVICES
example, by reducing staff and investment levels. As discussed in chapter 7, the risk
of such inefficiency at privatised airports appears to be constrained by a
combination of competitive pressures, the incentive of airport owners to increase
profits (especially from non-aeronautical activities) and the ability of airports to
discriminate in pricing. Firms with market power also may seek to keep out
potential competitors by lobbying government — engaging in so-called rent-seeking
activities — or by investing pre-emptively in additional capacity. In other words,
firms with market power may use up real resources (and profits) in a bid to retain
their position in the market.

Some participants argued that the efficiency costs of higher aeronautical charges
would extend beyond the immediate effects in the market for the intermediate
service (that is, aeronautical services). BARA observed that the exercise of airport
monopoly power would generate ‘flow-on inefficiency and costs in the broader
economy’ (sub. 26, p. 10). The ACCC stated that:
There are further allocative efficiency implications where the service is also an
intermediate input. High prices can distort production and consumption patterns of the
goods and services using air travel as an input. For example:
• Air travel is a business input for many companies. Higher prices can affect business
input costs and the ability of such companies to compete in Australia and overseas.
• Air travel is critical to the development of the tourism industry. Tourism is a major
contributor to the Australian economy. High airport charges have the potential to
damage both domestic and international tourism. (sub. 36, p. 6)

It is almost certain that activity levels and prices in user industries will be affected
by the exercise of any airport market power. The extent of the impact generally will
depend on the price responsiveness of demand and supply in those markets and the
degree of competition in them. However, measures of efficiency losses in the
(intermediate) market for aeronautical services resulting from airport market power
(as depicted in figure 4.1) generally will capture allocative efficiency losses in
downstream markets. This is because the derived demand for airport services distils
all the demands and supplies in those markets.

The result holds provided demand and supply in downstream markets (and, indeed,
in the market for aeronautical services itself) are not distorted. If there are no
(positive or negative) spillover effects in downstream markets that are not already
accounted for in prices, then analysis need only focus on efficiency effects in the
market for services provided by airports. On the other hand, if airlines have market
power, the detrimental impact of excessive airport charges on air travel could be
amplified.

PRELIMINARY ISSUES 85
Distributional effects of market power

Higher prices generally mean that there is a transfer of income from consumers to
the producer. (In figure 4.1, at the price Pm, there is a transfer equal to rectangle
PcPmab from consumers to the firm exercising market power.) As noted above,
however, provided the transfer occurs among Australian residents, and there is no
leakage of economic surplus abroad, it does not constitute economic loss to the
nation. Nonetheless, in the case of airports, there may be concerns about any impact
on the distribution of income among passengers, airline shareholders and airport
shareholders. Possible distributional effects of higher airport charges are discussed
further in chapter 7.

As noted above, the Commission has focused on likely efficiency effects to guide
its assessment of regulatory options. This approach assumes that ‘a dollar is a
dollar’ to whomever it accrues, provided the recipient is an Australian resident. The
Commission considers this to be an appropriate analytical approach for two main
reasons.
• First, as discussed in chapter 7, there do not seem to be large differences in
incomes of the broad groups affected by airport charging.
• Second, even if there were identifiable income disparities between affected
groups, attempting to bring about some desired income redistribution through
the restraint of airport charges would be very blunt and inefficient. Income
redistribution generally is more effectively and efficiently targeted through
taxation policy which takes into account the overall income or consumption of
individuals (and, therefore, a measure of their overall economic welfare), and not
just their purchases of airport services and/or their airport shareholdings. (See
PC (2001e) for a discussion of the policy implications of structural adjustment
and related distributional issues.)

Efficiency effects of constraining prices below efficient levels

While prices above marginal cost generate efficiency losses, prices below the
market-clearing level can also create problems. This may occur when a price-
regulated facility is approaching, or has reached, its capacity limit. (In figure 4.1, if
capacity were restricted to quantity Qm, the market-clearing price would be Pm.
However, if the price were capped at Pc, then demand would exceed the amount
being supplied (by the amount QmQc).)

Without price rationing, this excess demand necessarily would result in some other
form of formal or informal rationing (for example, queuing or a quantity allocation
scheme), which generally incurs efficiency costs. The magnitude of these costs

86 PRICE REGULATION
OF AIRPORT
SERVICES
depends largely on the type of non-price rationing. Under certain conditions, the
costs incurred by queuing, for example, could dissipate the entire surplus. Queuing
at an airport may involve aircraft being placed in holding patterns or being made to
wait on the apron/taxiways for clearance to depart, imposing large costs on airlines
(and passengers).

Even with a more orderly rationing scheme, implementation costs of the system,
waiting periods for slot allocation, and the likelihood that some airlines and
consumers with high slot valuations will miss out, will generate efficiency losses. In
other words, a regulated price below the efficient price may incur efficiency costs
larger than the efficiency costs incurred when a firm prices above it. (The potential
for asymmetric outcomes is amplified where investments are ‘lumpy’ and costs are
decreasing. In this case, a regulated price set below average cost may mean that
efficient investment is not undertaken. See PC (2001a) for a discussion.)

Moreover, where rationed slots are allocated (not sold) to airlines, the airlines
effectively hold quotas, the value of which reflects the willingness to pay of
passengers arriving or departing from the capacity-constrained airport. In other
words, airlines have an incentive to price-ration seats, charging passengers (up to)
the market-clearing price for accessing the airport, while they pay a lower price to
the airport. In this way, the airlines, rather than their passengers, can capture at least
some of the scarcity rents (equal to area PcPmab in figure 4.1).4 The airlines could
apply this rent in various ways, including distributing it to shareholders or using it
to cover the fixed costs of providing services on more marginal routes. These issues
are discussed further in appendix H.

4.3 Principles of efficient pricing of airport services

The case for price regulation rests on a comparison of the likely efficiency costs of
unregulated airport pricing and the likely outcomes under price regulation. This
requires an understanding of what is meant by efficient airport pricing.

Market prices typically perform several roles: they ration use of existing assets and
scarce resources; they indicate the opportunity cost of using those resources; and
they signal the need for investment/disinvestment in a particular activity.

4 For reasons discussed above, likely inefficiencies in the allocation system will reduce available
surplus somewhat.
PRELIMINARY ISSUES 87
Decreasing costs and efficiency

Allocative efficiency, in the sense of the best use of existing resources, generally
requires prices equal to (short-run) marginal cost. Where marginal costs are equal to
or above average costs, pricing at short-run marginal cost also will ensure, in the
long-run, that average total costs are covered — that is, efficient producers will
receive a ‘normal’ rate of return on their investments, including an appropriate
margin for risk.

However, as shown in box 4.2, where marginal costs lie below average costs for
relevant ranges of output — which may be the case for Australian airports with
spare capacity — marginal-cost pricing (for all units sold) will not provide an
adequate return on existing assets and, of greater relevance for economic efficiency,
will not provide adequate incentives for airports to undertake efficient investment,
replacement or otherwise. In other words, short-run (uniform) marginal-cost pricing
is unlikely to lead to efficient long-term service provision when costs are
decreasing.

It is conceivable, if capacity utilisation is being approached or has been reached,

that short-run marginal (opportunity) cost pricing may generate very high prices.5
Thus, over the life of the asset, if demand increases over time and for long enough
periods, costs could (eventually) be recovered by such pricing. However, this
pricing model probably would not be adopted in practice; in the absence of
regulation, long-term contracts, or longer-term horizons by airport operators, or just
the transaction costs of continually adjusting prices, probably would smooth the
peaks and raise the troughs that otherwise might exist. Moreover, such charging —
which means that prices could reach very high levels in some periods — is often
infeasible for political reasons.

Thus, efficient pricing for firms that exhibit significant economies of scale (and
which, like privatised airports, are required to be self-financing) generally must
depart from all prices being set at short-run marginal cost. Though such pricing
incurs an efficiency loss, compared with marginal-cost pricing (equal to the shaded
area abc in figure 4.2), uniform marginal-cost pricing, in the absence of government
subsidy, is not consistent with sustained provision of the facility and the services it
provides. Of course, raising revenue to pay subsidies itself has efficiency costs.

5 Marginal opportunity costs will incorporate the marginal cost of supplying the service, which
may increase with high utilisation levels plus, for example, congestion costs incurred by users.
88 PRICE REGULATION
OF AIRPORT
SERVICES
 Box 4.2 Pricing with decreasing costs
Figure 4.2 shows a firm for which, over the relevant range, marginal costs (MC) lie
below average costs (AC). (D is the demand curve while MR is marginal revenue.) This
means, if price is set equal to marginal costs (Pc), that the full costs of providing the
service will not be covered. If the firm is required to be self-financing, then average
prices charged must cover average costs. This could be achieved by levying a uniform
price equal to Pe (with a consequent efficiency loss equal to abc) or levying
discriminatory charges, such that those consumers with higher willingness to pay
contribute more to fixed costs than those with lower marginal valuations. To the extent
that marginal units are priced close to or at marginal cost, total costs will be recouped
while minimising any efficiency loss.

Figure 4.2

Pm e

g f

a
Pe
d AC
b c
MC
Pc
D
MR
O
Qm Qe Qc Q

A firm with market power will have the capacity to raise charges beyond cost-recovery
levels. If constrained to set a uniform price, the firm will increase the price to the point
that marginal cost equals marginal revenue (Pm). At this price, efficiency losses equal
all shaded areas in figure 4.2 — that is, the excess of marginal valuation of the service
over marginal cost for all units forgone (from Qc to Qm), with the area edba
representing the efficiency loss attributable to an increase in price from Pe to Pm. At
price Pm the firm earns excess profits equal to efgPm, while consumers are worse off by
an amount equal to the area eaPePm. However, if the firm can discriminate in pricing
(and it will have an incentive to do so to increase profits), then it is feasible that the
marginal consumers will not be priced out of the market.

Therefore, given a requirement to cover total costs, efficient (average) prices

typically will exceed short-run marginal cost but will not be so high as to generate
persistent excess profits. (In figure 4.2, this means a price of Pe rather than the
profit-maximising price of Pm or the short-run marginal-cost price of Pc). Whether
PRELIMINARY ISSUES 89
this price provides an adequate return on existing assets essentially is immaterial to
efficiency considerations. (However, this does not mean that today’s prices should
be set, as a matter of regulatory policy, as though all existing assets are sunk.
Indeed, such a policy may generate efficiency losses over time because firms will
factor sovereign risk into their required rate of return for regulated assets.) Today’s
prices should signal the incremental cost of (efficiently) providing services into the
future. Therefore, current prices may generate adequate returns, windfall gains or
windfall losses for current investors. Kahn summarises:
… partly because of the infeasibility of permitting prices to fluctuate widely along the
SRMC [short-run marginal-cost] function … the practically achievable benchmark for
efficient pricing is more likely to be a type of average long-run incremental cost,
computed for a large, expected incremental block of sales, instead of SRMC, estimated
for a single additional sale. This long-run incremental cost … would be based on
(1) the average incremental variable costs of those added sales and (2) estimated
additional capital costs per unit, for the additional capacity that will have to be
constructed if sales at that price are expected to continue over time or to grow. Both of
these components would be estimated as averages over some period of years extending
into the future. (Kahn 1988, vol. 1, pp. 84–5)

Multi-part pricing and price discrimination

Even in the presence of large fixed costs and a requirement for a natural monopoly
to be (just) self-financing, efficient levels of output may be feasible if the firm is not
constrained to set uniform prices. Typically, this will require some form of multi-
part pricing, such as different prices for different units sold to a customer (for
example, an up-front access fee plus user charge) or different prices for different
customers of the same or different goods and services (according to capacity to
pay), or some combination of the two approaches. In this way, fixed costs can be
allocated fully to customers, but with marginal consumers and/or marginal sales
making little, if any, contribution.

As a result, the efficiency loss arising from uniform average cost pricing could in
principle be reduced or even eliminated; marginal units are sold at prices equal to
short-run marginal cost. The limits of such pricing are set by the transaction costs
(including information costs) of doing so, and the ability of the provider to prevent
arbitrage across market segments.

Airports traditionally have charged according to the maximum take-off weight

(MTOW) of aircraft and this approach has been characterised as roughly amounting
to Ramsey pricing.6 However, privatised Australian airports seem to be moving

6 Ramsey pricing involves setting prices according to consumers’ willingness to pay — more
technically, setting prices that are inversely proportionate to the elasticities of demand of
90 PRICE REGULATION
OF AIRPORT
SERVICES
away from weight-based charges towards passenger-based charges for terminal and
even runway facilities. Passenger-based charging reduces scope for price
discrimination on some margins (for example, aircraft size), but scope for
discrimination remains on others (for example, between passengers travelling on
different airlines or at different times of day, or between international and domestic
passengers). There is solid evidence that airports do in fact discriminate in favour of
marginal airlines and/or flights by offering a range of concessions and rebates
(chapter 7).

Pricing of scarce slot capacity

Price differentiation may occur with respect to the timing of consumption, with
different prices for peak and off-peak users, such that peak users bear a higher share
of an airport’s fixed costs.7 Peak charges also signal the higher opportunity cost of
using the facilities and the need for additional capacity.

At airports experiencing at least some excess demand for landing slots, weight- or
passenger-based charging may send the wrong signals to airlines. A small plane or a
plane with a low passenger load is charged less than a large plane or one with a high
passenger load, even though the former may take up as much, or more, time on
approach and on the runway. Uniform charges for runway use, possibly combined
with differentiated peak/off-peak charges, would be one way of restructuring prices.
Another possible form of multi-part pricing would involve the sale or auction of
landing slots (or sale of leases of slots) combined with a user charge. Options for
the allocation of limited airport slot capacity are discussed in more detail in
appendix H.

Some participants suggested, given the demand management system (slot allocation
scheme) in place at Sydney Airport, that price levels are largely irrelevant for the
efficient use of that facility. Suffice to say here that although a slot management
scheme can ration demand among airlines, it is not necessarily the case that airlines
carrying passengers who value landing at peak periods the most obtain the slots, or
that such schemes operate as efficiently as the price mechanism.

different purchasers, or groups of purchasers, in order to minimise deadweight losses arising

from prices above marginal cost, subject to the requirement to recoup full costs. Weight-based
charging also may to some degree reflect differences in landing costs, though Doganis suggests
the relationship of airport pricing to costs has been indirect at best (Doganis 1992).
7 Strictly speaking, peak charging is a form of price differentiation, not price discrimination if
peak users generate higher capacity costs than generated by off-peak users. Price discrimination
refers to different prices being charged for goods or services produced at the same cost.
PRELIMINARY ISSUES 91
Thus, as discussed in section 4.2, while market clearing is attained with quantity
rationing, some economic inefficiency is likely to result.8 Generally, price rationing
of scarce slots is likely to promote more efficient use of limited capacity as well as
signal more explicitly and efficiently the opportunity cost of using a capacity-
constrained airport, the viability of building a new airport and the need for new
investment at the facility. In these circumstances, a price yielding above-normal
profits (for a limited period) would be appropriate.

Efficient pricing with locational rents and complementary demands

The single-till/dual-till issue pervaded many participants’ views about appropriate

pricing of airport services. Even though core-regulated airports ostensibly were
privatised on a dual-till basis, aeronautical charges largely continue to reflect to a
significant degree historical, single-till pricing (chapter 8).

BARA (sub. 41) and Qantas (sub. 48) submitted that efficient pricing of airport
services requires rents earned in non-aeronautical activities to be used to drive down
aeronautical charges. This outcome, they contended, was consistent with airports
operating in a competitive environment. This is a complex issue, but for reasons
explained in chapter 10 and appendix C, the Commission does not agree with their
contention. In competitive markets, efficiency requires that users (on average) pay
the long-run incremental cost of providing a good or service. Even in competitive
markets, owners of scarce factors (such as land) generally retain scarcity rents (after
general taxation).

It may be that the costs of providing aeronautical services will fall as the result of an
airport exploiting non-aeronautical opportunities and this will affect the efficient
aeronautical price. It also may be that an airport operator rewards passengers
(and/or airlines) that generate additional custom and profits. But efficient
aeronautical pricing does not necessarily require the transfer of locational rents to
reduce aeronautical charges. To do so may well lower aeronautical charges below
the marginal (long-run or even short-run) costs of providing those services and
discourage investment in them, as well as discourage the development of valuable
non-aeronautical activities.

8 Quantity rationing may be preferred to price rationing for distributional reasons. But, price
rationing is likely to promote efficiency objectives better.
92 PRICE REGULATION
OF AIRPORT
SERVICES
4.4 Principles of good regulation
Regulation is not costless. In addition to observable administration and compliance
costs, regulators inevitably must make decisions based on imperfect information.
This creates a risk that the costs of regulating — in terms of creating poor incentives
for investment and high compliance costs — may exceed the costs of inaction.

As discussed above, efficient pricing of airport services has several dimensions.

Addressing one dimension (for example, short-run marginal-cost pricing that
encourages efficient use of existing assets) may distort other dimensions (for
example, incentives to invest or to provide appropriate quality levels). Further, the
virtual impossibility of knowing precisely what the efficient prices are forces
regulators to construct prices based on measurable, often historical costs, even
though these may have little relevance to opportunity costs.

As Starkie observed:
Unfortunately, a little knowledge can be a dangerous thing; the incentive mechanisms
themselves can lead to distortion and unnecessary costs … In turn, this can lead to
further regulatory intervention, to complex regulation (possibly with significant
compliance costs) and to increased regulatory risk that has the effect of increasing the
cost of capital. At the end of the day, therefore, there is a trade-off between living with
imperfect regulation or with imperfect markets. It is only when the market does not
work well, when there is a clear case of natural monopoly and when regulation can
reasonably be expected to improve matters that the regulatory option is worthwhile.
Market imperfections alone are not a sufficient justification for intervention.
(Starkie 2001a)

The overarching objective of regulation should be that it promotes the national

interest by achieving outcomes that are more efficient than those achieved by the
alternatives, including no regulation. Thus, in assessing whether new regulatory
arrangements ‘are needed to ensure that the exercise of any such [market] power
may be appropriately counteracted’, as required by the terms of reference, the
Commission has assessed and compared possible efficiency and distributional
outcomes with and without regulation, as well as possible outcomes under various
regulatory options.

The terms of reference also specify that, to the extent it is called for:
7(b) future prices regulation should be applied to those aeronautical services and those
airports where airport operators have most potential to abuse market power; …
7(d) prices regulation should minimise compliance costs on airport operators and the
Government;
7(e) prices regulation should promote the efficient operation of airports;

PRELIMINARY ISSUES 93
 7(f) prices regulation should facilitate benchmarking comparisons between airports,
competition in the provision of services within airports (especially protecting
against discrimination in relation to small users and new entrants), and
commercially negotiated outcomes in airport operations.

These criteria (which largely mirror the Competition Principles Agreement (CPA))
suggest that any future regulation should be the minimum required to target the
source of the problem and promote efficient outcomes, while being applied in a way
that fosters market outcomes where feasible, imposes minimal compliance costs on
all parties, and promotes transparency and competition.

In addition, in assessing the advantages and disadvantages of the current system of

price regulation of airport services and possible options for future regulation, the
Commission has been guided by the Productivity Commission Act 1998 and the
broad regulation review principles developed by the Office of Regulation Review to
implement the CPA (PC 1998b).

Several of these principles are reflected in the terms of reference, including that
regulation should:
• promote overall economic performance;
• minimise the regulatory burden on industry consistent with efficient outcomes;
and
• be transparent and low cost.

Other desirable principles of regulation are that it, as far as possible, be predictable,
promote certainty, and be open to scrutiny and regular review.

94 PRICE REGULATION
OF AIRPORT
SERVICES
5 Market power of airports

The terms of reference require that, in considering the need for price regulation of
airports, the Commission should take into account that ‘future prices regulation
should be applied to those aeronautical services and those airports where airport
operators have most potential to abuse market power’. This chapter examines
potential sources of market power of airports, and the extent to which market power
exists for particular airports in Australia. Chapter 6 examines market power in
particular services provided at airports in Australia. Chapter 7 considers the
potential for abuse of, and the consequences of, market power.

5.1 Introduction
A firm can be said to have market power if it can profitably sustain prices above the
efficient cost of supply for a significant period of time.

Most firms have some market power. The degree of market power, and the extent to
which it persists, depend broadly on barriers to entry to an industry and the
availability to consumers of reasonably close substitutes (which is reflected in the
price elasticity of demand).

The potential negative effects of the exercise of market power on consumers or

users and on the efficiency of resource allocation are the reasons that market power
is cited as a rationale for price regulation of airports (chapter 4).

Preliminary issues in assessing market power

The next two sections examine the market power of airports in detail by considering
barriers to entry (section 5.2) and the price elasticity of demand for a particular
airport’s services (section 5.3). However, some preliminary matters warrant brief
discussion here.

First, the degree of market power a firm is assessed to have depends critically on
how the market in which it operates is defined. The market should be defined so as
to identify fully any potential sources of substitution for the firm’s products or
services.

MARKET POWER OF 95
AIRPORTS
Market definition requires consideration of what is being demanded (and by whom),
the geographic area of the market, the functional level of the market (that is, the
position of the firm in the overall supply chain), and the timeframe.
• Airports provide a number of services (chapter 2). To assess airport market
power in these services, it is important to identify the services that are essential
to airlines — and must be consumed as a bundle — and those that are optional,
either in the quantity or quality consumed. What constitutes the ‘necessary
bundle’ may vary depending on the airline customer. To the extent that services
do not need to be consumed or supplied as a bundle, the degree of market power
may differ across services. In addition, market power can vary depending on the
main market segment being considered — for example, business or holiday
traffic, and international or domestic traffic.
• The geographic dimension of a particular airport’s market could be defined
narrowly as the city in which it is located, or more broadly as the region in
which it operates. The appropriate definition will vary depending on the market
segment — for example, international or domestic — as well as on the particular
airport being examined.
• At a functional level, consideration may need to be given to whether the market
should be defined as that for transport services, airport services, that particular
airport’s services, or particular services at that particular airport. As noted by the
ACCC (sub. 36), for this inquiry, initially it is worth considering the market for
an airport’s services rather than looking at particular airport services.
• The nature of the supply of, and demand for, some airport services may mean
that substitution possibilities are limited in the very short term. To assess airport
market power, enough time must be allowed for all market responses to be made
to a price change. Professor King (ACCC, sub. 36, attachment C) suggested that
an appropriate timeframe for an airport could be one to five years. If a very long
time is required for the market to respond, then that in itself could indicate
market power.

These elements of market definition are part of the framework for assessing market
power that was recommended by Professor King (ACCC, sub. 36, attachment C).
The analysis of market power that follows is consistent with his suggested
framework, although the formal structure of analysis differs somewhat.

Finally, it should be noted that, even where a firm has market power, the potential
for its abuse can be mitigated by factors such as the countervailing power of users
and complementarities in demand. In addition, price discrimination may reduce
efficiency losses that otherwise might result from the exercise of market power
(chapter 7).

96 PRICE REGULATION
OF AIRPORT
SERVICES
5.2 Airports and barriers to entry
A barrier to entry is something that gives an incumbent firm an advantage over a
potential entrant. Barriers to entry can originate from supply or demand conditions,
or from regulation. They need not be the result of a deliberate action by an
incumbent firm.

In the case of airports, the main potential sources of barriers to entry appear to be
natural monopoly characteristics and regulatory constraints.

Natural monopoly characteristics of airports

A number of conditions can generate natural monopoly (box 5.1), but the essence is
that production of the good or service in question by one firm, for a given market
size, is less costly than production by more than one firm. This may change over
time, however, as demand and technology change.

Because of the weak direct competitive pressures faced by a natural monopolist,

there may be concerns about potential abuses of market power. However, the very
conditions that give rise to natural monopoly mean that there also may be efficiency
benefits from having one supplier. Thus, there is an inherent tension between
efficiency benefits and potential abuses of market power.

In the case of airports, natural monopoly characteristics arise from:

• investment requirements, economies of scale and economies of scope; and
• network benefits, which determine the extent to which airlines are willing to
spread their services across more than one airport in any given location.

These characteristics affect industry structure, performance and the efficient pricing
of airport services. However, the focus in this section is on the extent to which they
create potential barriers to entry and exit, and thus hinder the entry of potentially
competing airports.

Airport costs and investments

On the supply side, natural monopoly is more likely, and hence potential barriers to
entry are more significant, the greater the requirement for large fixed investments
and therefore the likely economies of scale, and the more significant the sunk costs
or economies of scope.

MARKET POWER OF 97
AIRPORTS
 Box 5.1 Conditions for the existence of natural monopoly
A natural monopoly is said to exist if, given the demand for a good, service or facility,
one firm can produce a given set of outputs at a lower cost than two or more firms can.
The basic conditions for natural monopoly generally relate to the nature of costs and
investment — such as investment indivisibilities, economies of scale, sunk costs and,
in multi-product industries, economies of scope.
• Investment is said to be indivisible, or lumpy, when it can be undertaken
economically only in large increments. To the extent that this makes the cost of
establishing a new facility higher than the cost of expanding an existing facility, this
can create an advantage for an incumbent supplier.
• Economies of scale occur when average costs decline as output increases.
Investment indivisibilities can contribute to economies of scale by increasing the
fixed costs of production. Given the existence of scale economies, any potential
competitor would need to capture a large market share, or increase the total market,
to be competitive. However, competition may not be efficient in this setting, to the
extent that one producer can supply the market at a lower cost than two.
• Sunk costs are costs that, once made, cannot be recouped. The extent to which an
investment is sunk cannot necessarily be ascertained at the time an investment is
made. Nonetheless, from the point of view of a potential entrant, the more
significant the amount of immovable, industry-specific investment required to
establish operations, the higher the potential sunk costs if entry is unsuccessful, and
the higher the risk of investment. Exit costs effectively become an entry barrier.
• Economies of scope exist if it is less costly to have one firm supply a number of
products or services than to have each service provided by a different firm. As with
economies of scale, economies of scope can mean that it is more efficient to have
only one supplier of the relevant products.
These factors may be neither necessary nor sufficient for natural monopoly. For
instance, one firm may be able to produce at a lower cost than two or more firms can
produce, given market demand, even if constant returns to scale (where average costs
are static as output increases) or diseconomies of scale (where average costs increase
as output increases) are present at the margin. Further, for a multi-product business,
there could be diseconomies of scope in the production of several outputs — that is,
the cost of one firm supplying all products is greater than what it would cost for firms to
specialise. This could offset the impact of economies of scale in the supply of the
individual product or service.
Even the coexistence of economies of scale and scope does not necessarily imply the
existence of natural monopoly (Baumol, Panzar and Willig 1982), since the concept of
economies of scope does not consider the possibility of two firms producing different
combinations of output. Nonetheless, an appropriate rule of thumb is that natural
monopoly is more likely where fixed costs are large relative to marginal costs (implying
high average costs compared with marginal costs) (King 2000).
Sources: Baumol, Panzar and Willig (1982); King (2000).

98 PRICE REGULATION
OF AIRPORT
SERVICES
The characteristics of investment and cost structures can differ significantly
depending on the type of airport being considered. The discussion below considers
the situation of airports, such as Australia’s core-regulated airports, that are capable
of servicing regular public transport (RPT) traffic in large jet aircraft.

Indivisibility of airport investment

Construction of a new airport requires the purchase of land and the development of
facilities such as runways, aprons, terminals and processing facilities. To the extent
that the cost of establishing facilities at new airports is higher than the cost of
expansion at existing airports, this can create an advantage for an incumbent airport
operator.

The most significant sources of indivisibility appear to be land and runways. The
costs of procuring the large area of land required to build an airport can be very
high, particularly in larger cities where vacant land close to population centres and
amenities is scarce. This means that any new operator potentially would have to
locate in a position that is less convenient than that of an existing operator, or try to
‘create’ a large area of land by buying up smaller parcels from a number of
owners.1 This latter option may lead to problems such as hold-out by existing
owners. Regulatory and environmental constraints also may create barriers.2

The construction of a runway also involves significant indivisible investment. If a

prospective airport operator wants to build an airport capable of servicing jet
aircraft, it must build at least two complete runways capable of handling such
aircraft (to enable landing and take-off in all wind directions). A runway is
indivisible in that it cannot be built for ‘a few aircraft movements’. The costs of
building a runway also are significant.

Other potential sources of indivisibility include terminal navigation and instrument

landing systems and lights, a full set of which is required regardless of the number
of RPT flights in a day or year.

1 Technically, contiguous land is required for infrastructure, such as runways, taxiways, aprons
and some form of passenger or cargo processing facilities. Due to changes in technology,
however, some services, such as check-in, increasingly may be provided at off-airport locations
(chapter 6).
2 Convenience may not necessarily be related directly to distance. A new airport 200 kilometres
south of Madrid, for example, will be only about 50 minutes from the city on a new high-speed
rail link (Starkie 2001a). Similarly, the Eastern Distributor in Sydney has made the airport
‘closer’ to the city by decreasing journey times, and CityLink in Melbourne has decreased
journey times between Melbourne Airport and some suburbs of Melbourne.
MARKET POWER OF 99
AIRPORTS
The indivisibilities involved in constructing other necessary assets for a new airport,
such as terminals and aprons, do not appear to be as significant. Although the
capital expenditure involved may be substantial, these facilities can be designed to
enable incremental expansion when (and if) traffic grows. According to the Motor
Trades Association of Australia Superannuation Fund (MTAA Super Fund):
Runways … come in very large increments. On the other hand, aprons, taxiways,
terminals, car parks are a little less lumpy due to the modular design nature of modern
airports and airports can plan to add on a few extra aerobridges, another baggage
carousel and another acre of car parking every few years. (sub. 22, p. 33)

Where, for various historical reasons (often defence related), there already are a
number of airports in a given location — such as in Melbourne, which has
Essendon, Avalon, Moorabbin and Point Cook airports, in addition to Melbourne
Airport — a potential airport competitor may not need to build a new airport.
Instead, it may be possible to change the capability of an existing airport, say from a
domestic to an international airport, enabling it to compete in a new market
segment.

In this case, the supply-side barrier to entry depends on how extensive the existing
facilities are. For instance, if upgrading the airport would require a significant
expansion of airport land, then entry barriers will be higher. Similarly, barriers to
entry will be higher if a new runway (to accommodate larger aircraft) is required.
On the other hand, the main investment required may be new facilities, such as
passenger processing and customs facilities, in which case the extent of
indivisibility, and hence the entry barrier, will not be as significant.

Economies of scale

Economies of scale (declining average costs as output increases) are more

significant in industries with a relatively high proportion of fixed costs. Economies
of scale can create a barrier to entry because an incumbent supplier, especially one
with excess capacity, can expand production at a lower unit cost than a potential
entrant could produce (unless the potential entrant can capture a significant market
share or increase the size of the total market (box 5.1)).

Airports often are said to have strong economies of scale, driven by large indivisible
investments such as runways. However, although there are economies of scale in
the provision of runways, there are likely to be diseconomies in other areas, such as
terminal facilities (passenger handling) (Betancor and Rendeiro 1999;
Walters 1978). Thus, the extent of scale economies in overall airport operations
depends on which effect dominates. Difficulties in maintaining access between

100 PRICE REGULATION

OF AIRPORT
SERVICES
airside and landside facilities as airports expand also influence the overall degree of
scale economies in airport operations (Starkie 2001b; Starkie and Thompson 1985).

Overall, economies of scale appear to be limited to airports with relatively low

passenger numbers — some analysts claiming that falls in average costs are
exploited fully at about 3–3.5 million passengers per annum (Doganis 1992; Salazar
de la Cruz 1999).

Larger airports may encounter decreasing returns to scale. Salazar de la Cruz

(1999), for example, found that average costs may increase past 12.5 million
passengers per annum (with constant returns between about 3.5 and 12.5 million
passengers). This would imply that natural monopoly could exist beyond
12.5 million passengers since, at passenger numbers just beyond this number, one
airport would still have lower unit costs than two airports serving the same total
market.

Although these results may not hold precisely for Australian airports (given
differences in airport traffic and general economic and regulatory conditions), they
are likely to be indicative of the situation in Australia.

The data on average expenses (excluding interest) presented in chapter 2

(figure 2.10) provide tentative support for this proposition. Of the core-regulated
airports with international RPT services, smaller ones such as Darwin and Perth
have higher average expenses than the larger international airports (Brisbane and
Melbourne). Diseconomies of scale at higher passenger levels partially may explain
the higher average expenses of Sydney compared with those of Brisbane and
Melbourne. The lower average expenses of some of the smaller airports may reflect
the lower costs of operating domestic, compared with international, airports.

Sunk costs

Sunk costs are investment costs that ‘produce a stream of benefits over a long
horizon but can never be recouped’ (Tirole 1990, p. 308). Generally, they relate to
investments specific to a firm, industry or location — which means the assets either
cannot be moved, or cannot be sold to another party unless that party intends putting
them to the same use, without substantial loss.

Sunk costs often are said to be significant for airports. For instance, the Board of
Airline Representatives of Australia (BARA) commented:
A number of the costs of investing in an airport are sunk once incurred. The costs of
land development, runways and taxiways, aprons and so on are sunk once incurred …
the size of the sunk costs are significant. (sub. 26, p. 9)

MARKET POWER OF 101

AIRPORTS
From the point of view of a potential investor in a new airport, the extent to which
costs potentially could be sunk affects the risk of investment and willingness to
undertake it. If an airport cannot generate sufficient traffic to be viable, then much
of the airport-specific investment will not be recouped, so will be sunk. Thus, the
more significant are the immovable airport-specific investments, the greater the
risk. The costs of exit, then, effectively become the barrier to entry.

Most major airport investment, such as runways, taxiways, and apron facilities, is in
immovable assets, so it cannot be sold for use in another location. As the Australian
Council for Infrastructure Development noted:
With an airport, which is a very, very big sunk cost investment, if the demand
characteristics for your airport change, it’s very hard to dig up the runway and take it
somewhere else where it will be used. (trans., p. 288)

Some facilities at airports potentially may have alternative uses. Office space and
some aspects of terminal facilities, for instance, could be used (if in a modified
form) for general purposes. The development of business parks by some core-
regulated airports (chapter 2) shows that, even though leaseholders must continue to
operate airports on their land, some of the land does have alternative non-airport
uses. Furthermore, speculation prior to the sale of Essendon Airport that it would be
worth up to six times more as a residential development than as an airport
(Davidson 2001) suggests that the opportunity cost of airport land (that is, its value
in alternative uses) is not zero. (Appendix F discusses aeronautical land valuation in
detail.)

However, much airport investment is in airport-specific assets, which by definition

cannot be put to alternative non-airport uses.

Economies of scope

Airports provide a number of services, both aeronautical and non-aeronautical

(chapter 2). Economies of scope exist if it is less costly to have one firm provide a
group of services than to have each service provided by a different supplier.

In the case of airports, economies of scope appear to be most significant in the

provision of aeronautical services. Economies of scope do not appear to derive from
joint construction of aeronautical infrastructure, however — since, for example,
aprons can be expanded incrementally, even if no changes are made to runways.
Instead, economies of scope derive from the lower coordination costs of having one
supplier provide access to (or coordinate the use of) several aeronautical services at
the airport — that is, runways, taxiways and aprons. In addition, it may be difficult
to attribute costs to specific aeronautical services.

102 PRICE REGULATION

OF AIRPORT
SERVICES
Another possible source of aeronautical economies of scope may arise from the use
of the same runway(s) by different traffic types — scheduled passenger, charter and
freight services. Thus, for certain levels of overall demand, it may be less costly to
have all traffic types use the same runway than to provide different facilities for
each service.

There also may be economies of scope in the provision of multiple non-aeronautical

services. Terminal buildings, for example, can be used for a number of purposes,
from retail to office space. BARA noted:
… there are economies of scope to the airport in providing many of these services. For
example, there are economies in providing a full range of terminal services from retail
space, office space and check-in areas. (sub. 26, p. 6)

These non-aeronautical economies of scope appear to derive from having one party
— the airport operator — allocate a large terminal space to multiple uses (and from
the benefits to users of having a number of services in the one general area). This
does not mean necessarily that there are benefits from having the airport operator
directly operate all the services provided in the terminal. First, the airport operator
may not have the expertise to operate each component (for example, different types
of retail stores). Second, most non-aeronautical costs are likely to be attributable to
particular uses, so it is feasible to have different parties operate individual services.
According to the MTAA Super Fund:
The costs of non-aeronautical outputs are distinct from, and more closely attributed to,
particular services than aeronautical services costs. That is, their costs can be more
clearly separately identified than are aeronautical services. (sub. 22, p. 33)

Finally, some participants commented that there are economies of scope in the joint
provision of aeronautical and non-aeronautical services. The MTAA Super Fund,
for instance, noted that airports have:
… strong economies of scope due to the co-location of non-aeronautical services
directly relevant to the demand for air travel (car parking and rental facilities) and of
services less directly related (retail outlets and light industrial property) that are
attracted by the passenger flows through the airport or the proximity of customers for
other services (catering for airlines, mechanical workshops for private aircraft).
(sub. 22, p. 33)

Similarly, BARA commented:

… consumers and airlines demand a range of services from an airport and … it is less
costly or simply impractical for more than one provider to provide individual elements
of the package. (sub. 26, p. 6)

There may be some benefits on the supply side of having all these services provided
by, or at least at, the airport. For example, a large area of land is required to operate

MARKET POWER OF 103

AIRPORTS
aeronautical services but only a portion of this land is required directly for
aeronautical purposes. Some of it is needed simply to meet legislative requirements,
such as safety or noise restrictions in residential areas. Hence, there may be cost
savings from using the land more intensively, which can be done by expanding non-
aeronautical uses. However, demand attributes, and consequent revenue benefits to
airports, are likely to contribute more significantly to the benefits of the joint
provision of aeronautical and non-aeronautical services.

Network benefits

The above discussion suggests that some aspects of the supply of airport services
give airports natural monopoly characteristics.

However, as already noted, there are examples in Australia of more than one airport
in a city (due to various historical factors), yet only one is used for domestic and
international passenger traffic in each city. Avalon Airport, for instance, has some
of the necessary aeronautical infrastructure in place (and reasonable access to the
city) to compete with Melbourne Airport. Indeed, currently it is the designated
alternative airport for Boeing 747 aircraft.

This indicates that supply characteristics are not the only factors that can make one
airport dominant in a location. The overall level of demand obviously is another
factor. However, the network benefits that accrue to airlines and their passengers
from using one large facility appear to be particularly important. In effect, there are
economies of scale on the demand side.

The benefits to airlines and passengers of using one airport accrue in a number of
ways. By concentrating services at fewer airports, airlines can use larger (more
economical) aircraft. Passengers transferring between flights benefit from not
having to commute between airports in a city. Thus, the higher the degree of
interconnecting traffic, the greater the preference for using one airport in a location.

Even without significant interconnecting traffic, a particular airline is likely to

prefer operating out of one airport in a city. Spreading services across airports or
moving out of a particular airport could involve significant costs, including sunk
costs, for an airline. These costs would relate to infrastructure (including terminals),
route development, repositioning aircraft, staffing and administration.

Thus, even with a choice of two similar (uncongested) airports serving the same
destination, airlines are unlikely to spread services across both. Furthermore, they
are unlikely to move to another airport in the area unless charges or congestion
costs increased substantially.

104 PRICE REGULATION

OF AIRPORT
SERVICES
In addition, the most efficient way to expand services or capacity, given network
benefits, typically will be through expanding capacity or duplicating facilities at the
existing airport. This, of course, is subject to land availability and environmental
considerations. Thus, network benefits create a more significant barrier to entry than
do airport supply characteristics alone.

The impact of regulation on barriers to entry

Airports in Australia must comply with a number of regulations, both general and
airport specific (chapter 3). Some of these regulations may form a barrier to entry,
interacting with any barriers created by airport costs, investment, and network
benefits.

Examples of the main potential regulatory barriers include:

• planning restrictions in, or near, large cities, which may constrain the
construction and design of a new airport, and inhibit the development of
effective access to the site (through road or rail links, for example);
• noise restrictions, which limit the ability to locate an airport near residential
areas, as well as influencing the amount of land airports require around
aeronautical facilities; and
• environmental legislation, such as the Airports (Environment Protection)
Regulations 1976, made under the Airports Act 1996, which regulates airport
activities that generate, or have the potential to generate, pollution or excessive
noise (chapter 3).

Implications for market power

Clearly, there are significant barriers to entry in the provision of airports. These
arise from natural monopoly characteristics that are reinforced by regulatory
constraints.

That there are natural monopoly elements in the supply of airport services,
combined with preferences on the demand side to use one airport in a given
location, makes it unlikely that a direct competitor RPT airport would emerge in
any city in Australia, given passenger volumes in most airports.3

3 As already noted, although there is more than one airport in a number of capital cities in
Australia, this tends to be the result of historical accident. In any case, only one airport in each
city is used for RPT traffic, despite the availability of the other airports.
MARKET POWER OF 105
AIRPORTS
Considered in isolation, this would appear to give airports in Australia significant
market power.

However, it is important to note that the essence of an airport’s monopoly is spatial

or locational in nature. A direct competitor may not emerge in the same city, but an
airport in another city may provide some competition. The degree to which this type
of competition could emerge can be assessed only by examining the demand
characteristics of particular locations and airports. This is done in the next section.

FINDING 5.1

There are significant barriers to entry in the provision of airports. These arise from
natural monopoly characteristics that are reinforced by regulatory constraints.

5.3 Price elasticity of demand for an airport’s services

The ‘price elasticity of demand’ is a measure of the responsiveness of demand for a
good or service to changes in its price; formally, it refers to the percentage change
in quantity that would result from a one per cent change in price. Where the
percentage change in quantity is greater than one, demand is said to be elastic;
where it is less than one, demand is inelastic. A firm facing inelastic demand can
increase its revenue by increasing the price of its product.

Elasticities can provide an indication of the degree of market power a firm has.
High demand elasticities for a firm’s product reflect competition in a market. (A
firm can face a high price elasticity for its product, even where demand at the
industry level is relatively inelastic.) Firms facing low elasticities of demand for
their product have market power since they can increase their price without
significant substitution away from their product.4

Framework for assessing the elasticity of demand for airport services

As noted in chapter 2, the demand for airport services is a ‘derived’ demand; indeed
it can be considered a ‘derived-derived’ demand. It is derived from the demand for
airline services, which in turn is derived from the demand for travel for business

4 That said, if a firm sets prices to exploit fully its market power, it will increase prices until the
point at which demand becomes elastic. Thus, the demand elasticity measured from a
‘monopoly price’ would give an erroneous indication of the degree of market power held.
Further, if the demand of a firm (that is unregulated) is estimated to be price inelastic, this
could imply that market power is not being fully exercised. Hence, it could suggest that there
may be some constraint on the exercise of market power (assuming the elasticity is estimated
correctly).
106 PRICE REGULATION
OF AIRPORT
SERVICES
meetings, holidays, visiting friends and relatives (VFR), migration, cargo handling
and so on.

The analysis here considers the demand for the services of a particular airport,
which may be more elastic than the demand for airport services in general. The
discussion focuses on passenger services since much of the freight passing through
RPT airports is carried on passenger flights (Melbourne Airport, sub. 7).
Nonetheless, where appropriate, dedicated freight services (where freight is not
carried in passenger aircraft) are discussed separately.

Since the demand for an airport’s services is a derived demand, its elasticity will be
influenced by four factors:
• the elasticity of demand for air travel to that destination;
• alternative sources of supply for a particular airport’s services;
• the proportion of airfares (or freight charges) and airline costs that airport
charges comprise; and
• the elasticity of supply (supply responses) of other input providers, such as
airlines (box 5.2).

The focus with respect to airports generally is on the third factor — that is, the (low)
proportion of airfares and airline costs that airport charges comprise on average
(discussed below). The lower the proportion of costs that a particular input (such as
airport services) comprises, the lower the elasticity of demand could be expected to
be, all other things being equal (box 5.2).

However, to the extent that the elasticity of demand for the final product is high, or
that substitutes for the input (in this case, airport services) exist, the elasticity of
demand for the input will be higher. This does not mean necessarily that demand for
airport services will be highly price elastic, but that it could be higher than would be
suggested by its cost share alone.

In addition, different segments in the market may vary in their price responsiveness.
The overall demand elasticity for the airport’s services will be a composite of these
different market segments, appropriately weighted. The question is which customers
or segments drive the behaviour of the airport and its ability to increase prices or
otherwise exercise market power. Further, estimated elasticities generally will be
measured at a point on the demand curve. Less appears to be known about the shape
and/or slope of the demand curve for air travel. Elasticities could increase
significantly as prices increase, which could constrain the scope for any large
changes in airport prices.

MARKET POWER OF 107

AIRPORTS
 Box 5.2 Conditions for assessing price elasticities of derived demands
Four conditions influence an assessment of the elasticity of a derived demand.
Following Stigler (1969), the service for which there is a derived demand — as is the
demand for airport services — is referred to as a ‘productive service’.

Elasticity of demand for the final product or service

The elasticity of demand for a productive service will be higher the larger the elasticity
of demand for the final product. In this case, any given increase in the cost of the input
(and hence any given impact on final prices) will have a greater effect on final output.

Availability of alternative sources of supply of the service

The demand for a productive service is more elastic the more easily other services
may be substituted for it (that is, the higher the elasticity of substitution). The elasticity
of substitution is defined formally as the percentage change in the ratio of two
productive services consumed resulting from a percentage change in the relative
prices of the services. Given that an airport is essential to operate air services, the only
possible substitute for an airport is another airport. Nonetheless, airlines may have
some scope to vary the amount of airport services consumed by consolidating
services.

Proportion of total cost that the price of the productive service comprises
The lower the price of the service relative to the total cost of the final product (that is,
the smaller the proportion of the total cost it comprises), the lower is its price elasticity.
Intuitively, few people are discouraged from buying a final product if the price of an
input that is a small part of total costs increases (even if its price rises significantly and
the cost is passed on fully). However, this factor can be overshadowed if there are
significant alternative sources of supply for the input. Formally, for this condition to
hold, the elasticity of demand must exceed the elasticity of substitution.

Elasticity of supply of other inputs

The demand for the productive service will be more elastic the more elastic is the
supply of other productive services. The maximum price that a particular input provider
can receive (for a given level of production) is limited by the amount that consumers
are willing to pay for the final good, and the price that other input providers require for
their services. Thus, if the price charged by one input supplier increases, then either
the price of the final good must rise and/or other input suppliers must accept a lower
price for their services. The lower the ability of other input providers to accept a lower
price (because they have a high elasticity of supply, or operate under conditions of
decreasing costs), the greater is the adjustment required by consumers (that is, the
higher is the change in the price and, hence, demand for the final product).
Source: Stigler (1969).

The rest of this section considers the importance of each factor in influencing the
elasticity of demand for a particular airport’s services. As for section 5.1, the
108 PRICE REGULATION
OF AIRPORT
SERVICES
approach adopted here is consistent with the framework outlined by Professor King
(ACCC, sub. 36, attachment C), though its structure differs.

Elasticity of demand for air travel to a particular destination

The higher the elasticity of demand of air travel to a particular destination, the
higher the likely elasticity of demand for a particular airport’s services.

Reliable estimates of price elasticities for passenger air travel are difficult to obtain.
As noted by Oum, Waters and Yong (1992), elasticities differ depending on fare
class (business, economy, discount), as well as distance. Problems obtaining data,
however, mean that only aggregate price elasticities tend to be estimated. (One
exception is a recent Australian study (Battersby and Oczkowski 2001), which
attempts to estimate demand elasticities by market segment and is discussed further
below.) Calculated elasticities also tend to be based on some average estimated fare,
rather than actual prices paid, further distorting estimates.

As a result of these problems, estimates vary widely, ranging from -0.4 to -4.51,
with most falling between -0.8 and -2.0 (Oum, Waters and Yong 1992). Although
these results are not unambiguous, the general conclusion seems to be that business
travel is relatively inelastic, while holiday travel is elastic.

Battersby and Oczkowski (2001) focused on demand elasticities for domestic air
travel in Australia. They found that elasticities differed across the four routes
examined, as well as across airfare classes. The estimated price elasticities tended to
be lower than those found in other studies — and were, in fact, inelastic for most
routes and fare classes — but economy fares tended to be the most elastic.
Counterintuitively, there did not appear to be a significant difference between the
elasticity of demand for business compared with discount fares. As the authors
noted, the relatively fixed supply of discount fares, particularly during the time
under consideration (1992–1998), partly could explain this result.

On the other hand, BARA noted that ‘available evidence suggests that demand for
air travel is relatively elastic’ (sub. 26, p. 11). This point was echoed by Qantas,
which ‘believes that air services have a reasonably high price elasticity’ (sub. 48,
p. 18). Virgin Blue also argued that:
Virgin Blue has proven beyond doubt that more people fly if you lower the price, even
by a few dollars. It’s as simple as that. (Huttner 2001)

The price elasticity of demand for international air travel to Australia seems to
follow broadly the patterns found in the international studies. According to the then
Bureau of Transport and Communications Economics (BTCE 1995), foreign leisure

MARKET POWER OF 109

AIRPORTS
travellers appeared to be more price sensitive than business travellers. The
elasticities also varied depending on the country of origin of the travellers. The
estimated price elasticities (between -0.23 and -1.19 for leisure travellers, and
inelastic or zero price elasticity of demand for business travellers) tended to be
towards the lower end of the range of elasticity estimates found in the other studies
cited in the above discussion.

The available elasticity estimates do not, however, indicate the potential price
responsiveness of travellers to all particular destinations in Australia and are of
somewhat limited use in assessing the elasticity of demand for particular airports.
Although Battersby and Oczkowski (2001) examined elasticities for different
routes, they did so for four routes (covering three airports) only, and the model was
based on data between 1992 and 1998 (that is, before the entrance of Impulse and
Virgin Blue).

Given this, it can be more informative to examine qualitatively the factors that
influence the elasticity of demand for air travel to a particular destination. Broadly
speaking, this elasticity will depend on the relative attractiveness of the destination
and the relative attractiveness of air travel to that destination compared with other
transport modes. The impact of these two factors will vary according to the market
segment being served (domestic or international; business or leisure), and whether
the service is freight or passenger transport.

Demand substitution possibilities: alternative destinations for passenger travel

Given that the demand for air travel is derived from the demand for business trips,
holidays and so on (chapter 2), several factors influence the decision to travel to a
particular destination. These include the tourist attractions and amenities in a region,
business opportunities, friends and relatives, and the cost of travel (including the
cost of the journey and living costs while there).

The primary purpose for travel is likely to influence the relative importance of each
factor and thus the extent to which travellers are willing to substitute one destination
for another, or change their frequency of travel. In particular, holiday makers (both
domestic and international) are more likely to have discretion over their destination
than are those visiting friends and relatives or travelling for business purposes.5
Gold Coast Airport commented:
Competition can come in several ways for a leisure destination such as ours … the
tourism market is reasonably fickle. (trans., p. 373)

5 Conference organisers have much more discretion over destination. Thus, this segment of the
business market is likely to be more price elastic than the general business market.
110 PRICE REGULATION
OF AIRPORT
SERVICES
Other regions within Australia and, in some cases, overseas can be alternative
destinations for Australian and international holiday travellers. As noted by the
ACCC:
… both international and domestic tourists might substitute destination on the basis of
cost differences. For example, a tourist may decide to visit Coolangatta rather than Fiji,
if the cost of doing so is substantially lower. (sub. 36, pp. 62–3)

Reported changes to travel plans following the events of September 2001

(chapter 2) show that travellers are willing to change holiday destinations (although
these changes were not influenced primarily by price factors).

In addition, even if international holiday travellers have made the decision to come
to Australia, they have choices about which regions they visit within the country,
and about their points of arrival and departure.

Though often seen to be fairly inelastic market segments, there also are some
substitution possibilities for travellers in other categories. For instance, people
visiting friends and relatives may have little choice over destination but do have a
choice about frequency (and possibly transport mode) and therefore may be quite
price sensitive. Business travellers also may change the frequency of visits (or
choose alternative modes, as discussed below). If business travellers ultimately are
selling their products into competitive export markets, then they are likely to be
more price sensitive.

The potential price responsiveness of travellers, both tourists and those visiting
family and friends, was highlighted by Tourism Tasmania:
… Tourism Tasmania’s own TVS Special Fares Survey … covered the period in
September and October 2000 when $55 fares were placed on Melbourne to Tasmanian
ports as well as most other city pairs across the country. The survey showed that
34 per cent of the fares to Tasmania were purchased by visitors who had no plans in the
next two years or more to visit Tasmania and that the primary reason for purchasing the
tickets for 92 per cent was the price of the ticket. Visiting friends and relatives was the
motivation of 53 per cent of those travellers, with the remainder being predominantly
leisure focused. (sub. 13, p. 1)

Overall, although each market segment has some substitution possibilities, these
possibilities are higher for holiday travellers. Therefore, destinations with a greater
proportion of leisure traffic are likely to be more susceptible to competition from
other destinations. On the other hand, the higher the proportion of business traffic,
the less price sensitive travellers to a destination are likely to be (all other factors
being equal).

As can be seen in table 5.1, the proportion of interstate holiday travellers varies
greatly across regions serviced by Australia’s core-regulated airports.
MARKET POWER OF 111
AIRPORTS
Table 5.1 Primary purpose of visit, interstate overnight visitors in
Australia, 1999a
Visiting friends
Destination Businessb and relatives Holiday/leisure Other
% % % %
New South Walesc 20 32 43 6
Sydney 33 34 28 6
Victoria 34 30 34 4
Queenslandd 22 30 45 3
Brisbane 32 32 30 6
Gold Coast 11 24 62 3
Tropical North Queensland 21 23 53 3
South Australiae 36 25 30 13
Western Australia 40 26 37 1
Tasmaniaf 18 22 56 4
Northern Territory 24 13 55 6
ACT 30 34 32 4
a Percentage of all interstate overnight travellers to the destination who go there for each purpose. Overnight
visitors are travellers aged 15 years and over who do not arrive and depart within the same day. Unless
otherwise stated, the source is BTR (2000b). Rows may add to more than 100 per cent because some people
reported more than one purpose for their visit. b ‘Business’ includes conferences. c NSW data are from
Tourism New South Wales (2001) for 1999-00, and refer to domestic (intrastate and interstate) visitors. Main
purpose of visit does not appear to differ significantly between domestic and interstate categories (though the
proportion of business travel is higher for interstate visitors). d Queensland data are from Tourism Queensland
(2000a–d), and refer to domestic (intrastate and interstate) visitors. Main purpose of visit does not appear to
differ significantly between domestic and interstate categories. e SA data are taken from SATC (2001) for
2000. ‘Other’ includes a ’not asked’ category. f Tasmanian data are from Tourism Tasmania (2001) for 2000.
Sources: BTR (2000b); SATC (2001); Tourism Queensland (2000a–d); Tourism New South Wales (2001);
Tourism Tasmania (2001).

In general, the eastern mainland States, as well as South Australia and Western
Australia, have higher proportions of business and VFR travellers compared with
the other destinations considered.6 Also, there are differences in the main market
segments going to different regions within these States. For example, the proportion
of business traffic to Brisbane is significantly higher than to the Gold Coast and
Tropical North Queensland. Likewise, there is more business traffic travelling to
Sydney than the NSW average. Similarly, in the Northern Territory, business or
VFR travellers account for a greater proportion of visitors to the ‘Top End’ than of
visitors to the ‘Centre’ region (NTTC 2000).7 It is likely that such patterns (that is, a

6 The data presented relate to overnight travellers (that is, travellers who do not arrive and depart
within a day), so the proportion of business travellers is understated.
7 NTTC (2000) provides data on the main purpose of visit of all visitors to the Northern Territory
(not separated by domestic and international travellers), by region visited in 1999-00. Overall,
of visitors to the ‘Centre’ region, 77 per cent went for holiday/leisure purposes and 7 per cent
112 PRICE REGULATION
OF AIRPORT
SERVICES
lower proportion of leisure traffic in capital cities) would be repeated in the States
for which the data are not available.

This suggests that substitution possibilities and hence demand elasticities for
destinations such as Adelaide, Brisbane, Canberra, Melbourne, Perth and Sydney
will be lower than for the other destinations considered — unless business
frequency is price elastic.

This conclusion effectively assumes that the mean elasticity (that is, an elasticity
based on the main market segments served) is an appropriate measure of price
sensitivity. If, however, most marginal travellers come from elastic market
segments, then an increase in prices will result in a larger quantity response than
suggested by an unweighted average elasticity. That said, if these elastic portions
are a very small part of the market, they are unlikely to affect the overall elasticity
significantly.

International travellers primarily come to Australia for a holiday (table 5.2). The
proportion of foreign business visitors tends to be low, with the highest levels
around 14 per cent for Melbourne and 12 per cent for Sydney in 1999. Melbourne
also had the highest proportion of international VFR visitors in that year, followed
by Sydney, Brisbane and Canberra. Differences in the purpose of visit across
destinations within a State do not appear to be significant. Overall, since holidaying
is the dominant reason for international travellers visiting Australia, international
travellers may be more price sensitive as a group than domestic travellers. However,
the fact that international visitors to Australia have little choice but to fly may
reduce their price elasticity of demand for air travel.

The importance of local population size and density

The importance of the local population in encouraging travel to (and from) a

destination is highlighted by the importance of business and VFR travellers to some
destinations. Where the local population is relatively large, it is more likely that
travel to the destination will comprise predominantly the less price responsive
business and VFR markets (for a given level of modal and airport substitution).
Thus, the Queensland Department of Tourism, Racing and Fair Trading argued:
Sydney serves as a gateway to Australia’s largest catchment area for high yielding
premium fare paying passengers. Sydney also has geographic advantages in that it acts
as a conduit to funnel traffic to and from Australia’s densest population centre. The

each went for VFR and business purposes. Of visitors to the ‘Top End’, 55 per cent went for
holiday/leisure purposes, 17 per cent went for VFR, and 14 per cent went for business
purposes. The proportion of international visitors was higher for the ‘Centre’ than for the ‘Top
End’.
MARKET POWER OF 113
AIRPORTS
 traffic volume, premium revenue mix of business and leisure traffic coupled with
significantly greater frequency and volume of air services gives Sydney significant
advantages over Brisbane in terms of obtaining favourable scale, scope and revenue
efficiencies, for passenger air transport. (sub. 6, p. 2)

Table 5.2 Primary purpose of visit, international visitors to Australia,

1999a
Visiting friends
Destination Businessb and relatives Holiday Other
% % % %
New South Wales 12 16 59 13
Sydney 12 15 59 13
Victoria 13 20 52 14
Melbourne 14 20 53 14
Queensland 5 12 73 10
Brisbane 9 16 62 14
Gold Coast 3 7 84 5
Tropical North Queensland 2 7 82 8
South Australia 8 18 59 15
Adelaide 7 18 61 13
Western Australia 7 27 52 13
Perth 8 26 53 13
Tasmania 10 18 52 20
Hobart 6 20 59 15
Northern Territory 2 9 80 9
Darwin 2 9 76 13
Alice Springs 2 8 81 9
ACT 10 13 64 13
a Percentage of all international travellers (aged 15 years and over) to the destination who go there for each
purpose. These figures differ from those of the ACCC (sub. 36, p. 63). The ACCC data relate to inbound traffic
on international flights at the respective airports. The data in this table are based on BTR (2000a), which is the
result of surveys conducted as visitors left Australia’s international airports. b ‘Business’ includes conferences.
Source: BTR (2000a).

Demand substitution possibilities: alternative modes for passenger travel

People potentially can travel to a given destination in a number of ways — by bus,

train, ferry, or car, as well as by air (although air is really the only practicable way
for international arrivals to travel to their first Australian destination). The potential
for modal substitution should be considered broadly — for example, advances in
technology mean that teleconferencing or videoconferencing provide a good
alternative to some face-to-face business meetings.

114 PRICE REGULATION

OF AIRPORT
SERVICES
The greater the attractiveness of modal substitutes for travel to a particular
destination, the higher will be the price elasticity of demand for air travel (and
hence the higher will be the price elasticity of demand with respect to airport
charges).

Factors that influence the substitutability of travel modes

Factors that influence the decision over mode include the proposed length of stay,
the distance to be travelled, and the relative cost of each mode (both in financial
terms and in terms of convenience and the time taken to travel). As with the
possibility for destination substitution, the primary purpose for a visit is likely to
influence the relative importance of each factor and thus the extent to which
travellers are willing to substitute modes. Of course, the first decision to be made by
potential travellers, regardless of the purpose of travel, is whether to go at all, or
how often to go.

Several inquiry participants argued that the modal alternatives to air travel generally
are weak, mainly due to time savings associated with air travel. The ACCC, for
example, noted the particular importance of time in determining preferences for
business air travel:
For business travellers the convenience of air travel is vastly superior to the
alternatives. It is likely that for this customer group, location is primarily determined by
factors other than airport pricing, and that time constraints are a critical element of the
travel decision. In such circumstances, there are no viable substitutes to flying to a
particular destination; the choice is simply between travelling and not travelling.
(sub. 36, p. 62)

Though not travelling within a particular day is not always a viable alternative for
business travellers, business is not totally unresponsive to costs. It can minimise the
use of air travel in various ways — for example, by scheduling intercity meetings so
fewer trips are required, or by teleconferencing or videoconferencing.

Although videoconferencing can save time and cost less than face-to-face meetings,
it will not always be an adequate alternative. Its adequacy as an alternative will
depend on the nature and expected duration of meetings. It is unlikely to be
adequate if a large group of people who are not well acquainted need to discuss
sensitive issues, for example. On the other hand, a smaller better-acquainted group
discussing routine issues is more likely to find videoconferencing a viable option.

The potential substitutability of videoconferencing for face-to-face meetings has

been highlighted since September 2001. Concerns about the security of travel and
problems obtaining domestic flights saw many businesses turn to videoconference
providers (Heasley 2001; Skulley and Bolton 2001). Although partly a response to
MARKET POWER OF 115
AIRPORTS
non-price factors, it indicates the ability of business to use videoconferencing in
some cases. The long-term impact on demand for videoconferencing is unclear.
However, continuing improvements in the technology and its cost effectiveness
mean that videoconferencing may grow in importance in the future.

Virgin Blue argued that time also was an important factor for leisure travellers:
… the time to travel between the major airline destinations is significantly longer by
car, bus or train than by aircraft (12 hours compared to an hour and [a] half for
example) … For leisure travellers, the current low fares compare favourably with other
forms of transport, and any overnight stays relating to the time taken by other forms of
transport contribute to the total cost of the holiday. Accordingly, other forms of
passenger transport are not readily substitutable for aircraft travel. (sub. 30, p. 12)

Assessing the time taken to travel by air, however, should include the journey time
from the initial departure point to the origin airport, as well as the time taken to
travel between the destination airport and the final destination point. In the case of
travel between Canberra and some suburbs of Sydney (particularly southern
Sydney), for example, it may be more time effective (even for business travellers) to
travel by car.

In any case, leisure travellers are likely to be far more sensitive to price than to
time, especially compared with business travellers.
• Holidays tend to be longer than business trips so time may not be as critical for
some holiday travellers. They may, therefore, place a greater emphasis on the
generally higher direct cost of air travel.
• The journey to a final destination often can provide value to holiday travellers
who visit attractions along the way. This means that the extra time taken to
travel using modes other than air may not be viewed purely as a cost.
• The use of a car when at the destination often is valued by holiday travellers.
Thus, driving to a holiday destination may provide benefits to counteract the
extra time taken to travel.

Nonetheless, the further the distance between destinations, the less attractive are
alternative travel modes. The ACCC noted two reasons for this:
Firstly, the difference in journey times between air transport and the alternatives
increases rapidly. Secondly, the cost differential between air transport and any
alternative mode of transport may narrow as the distance increases. (sub. 36, p. 63)

Data provided in BTR (2000b) and reproduced in table 5.3 show that interstate
overnight business travellers are far more likely than holiday/leisure or VFR
travellers to use air transport.

116 PRICE REGULATION

OF AIRPORT
SERVICES
Table 5.3 Transport used by overnight interstate visitors in Australia, by
purpose of visit, 1999a
Visiting friends
Holiday/leisure and relatives Businessb Other Total
% % % % %
Air 34 35 70 17 48
Private vehicle 58 60 26 72 52
Other 16 11 8 17 13
a Percentage of travellers going for each purpose that use each mode. Columns add to more than
100 per cent because some people reported more than one mode and purpose for visit. b ‘Business’ includes
conferences.
Source: BTR (2000b).

For international travel to Australia, there is little option but to fly. The only real
choice about travel mode for international visitors is how to travel between various
regions within the country.

Alternative travel modes to specific Australian destinations

Overall price elasticities of demand for air travel incorporate factors such as modal
substitution. However, given the variety of factors influencing the choice of travel
modes, the price sensitivity of travellers attributable to modal substitution is likely
to differ across Australian cities.

Since holiday travellers are likely to be more price elastic than business travellers,
one way to assess the potential competition that air faces from other travel modes is
by examining which market segments dominate travel to a particular region. Thus,
using the information in table 5.1, it would appear that air travel is likely to face
more intermodal competition (and greater price elasticity of demand) in destinations
such as the Northern Territory, the Gold Coast and Tropical North Queensland,
where holiday travellers dominate.

However, even for holiday travellers, the viability of modal substitutes is dependent
on factors such as the distance between origin and destination. Thus, examining
modal choice to particular destinations can give an idea of the perceived
substitutability of travel modes (table 5. 4).

As illustrated in table 5.4, air travel comprised no more than around 30 per cent of
total interstate overnight trips made to New South Wales and the ACT. The
dominance of private vehicle travel for those visiting the ACT may reflect the high
proportion of visitors from Sydney (and surrounding areas of New South Wales).
As noted above, for these visitors, the time difference between air and road travel
may be relatively low (and may even favour private vehicle in some cases).
MARKET POWER OF 117
AIRPORTS
Interstate visitors to South Australia were also slightly more likely to travel by
private vehicle than by air in 2000. The use of air travel to Adelaide subsequently
may have increased following the introduction of Virgin Blue services (but before
the reduction in Ansett services). However, given that Virgin Blue’s passengers are
likely to have more price-sensitive demands, this would not have increased the
market power of Adelaide Airport.

Table 5.4 Mode of travel to Australian States by overnight interstate

visitors, 1999a
Destination Air Private vehicle Other
% % %
New South Wales 32 59 10
Victoria 49 44 10
Queensland 55 41 11
South Australiab 40 51 12
Western Australia 82 13 21
Tasmaniac 78 na 22
Northern Territoryd 53 37 9
ACT 20 71 9
a Percentage of total overnight interstate visitors to the State using each travel mode. Overnight visitors are
travellers aged 15 years and over who do not arrive and depart within the same day. Unless otherwise stated,
the data source is BTR (2000b). Rows may add to more than 100 per cent because some people reported
more than one mode. b SA data are taken from SATC (2001) for 2000. ‘Other’ includes a ’not asked’ category.
c Tasmanian data are from Tourism Tasmania (2001), and refer to 2000. d NT data are from NTTC (2000) for
1999-00. ‘Private vehicle’ for the Northern Territory incorporates 2-wheel drive and 4-wheel drive vehicles,
which may include hire cars, and also includes ‘other road transport’ (excludes coach, which is included in
‘other’ in this table). na Not applicable.
Sources: BTR (2000b); SATC (2001); NTTC (2000); Tourism Tasmania (2001).

On a whole-of-State basis, air travel is the dominant mode only to Western

Australia, Tasmania and, to a lesser extent, the Northern Territory and Queensland.
However, as shown in box 5.3, the relative importance of air travel differs
according to the region being considered in Queensland. In particular, the
proportion of visitors travelling by air to Tropical North Queensland is higher than
to Brisbane or the Gold Coast.

For Tasmania, however, sea travel often is seen to provide a viable alternative to air
travel. The ACCC, for instance, argued that:
… ferry services connecting Melbourne and northern Tasmania are an attractive
alternative to flying (for many visitors). In this case the demand side substitution
possibilities are much greater than for other capital city airports. (sub. 36, p. 64)

Further, as discussed below, competition between airports in Tasmania is likely to

be more significant than in other destinations.

118 PRICE REGULATION

OF AIRPORT
SERVICES
Box 5.3 Mode of arrival of visitors to selected Australian destinations
The tourist bureaux in some States publish breakdowns on mode of arrival, either by
domestic and international visitors, or by regions within the States. Examples of these
are provided below.
Northern Territory
In 1999-00, 65 per cent of all visitors to the Territory arrived by air, 24 per cent by
private vehicle, 7 per cent by coach, and 4 per cent by rail. As shown in the table
below, the relative proportions using each mode differed, sometimes significantly,
between domestic and international travellers. In particular, a far greater proportion of
domestic visitors than of international visitors travelled by private vehicle. The overall
proportions appear to be similar for the ‘Centre’ and ‘Top End’ regions. (NTTC 2000)

International
Mode of entry Interstate visitors visitors Total visitors
% % %
Air — domestic 52 59 56
Air — international 1 15 8
Coach 6 7 7
Private vehiclea 37 9 24
Rail 2 6 4
Other 1 1 1
a Private vehicle includes 2-wheel drive vehicles, 4-wheel drive vehicles and ‘other road transport’,
which may include hire cars.
Queensland
In 1999, 71 per cent of domestic (including intrastate) visitors to Queensland travelled
by private vehicle, 21 per cent by air and 4 per cent by rail (Tourism Queensland
2000c). The proportion of interstate visitors who travel by private vehicle is lower than
for the domestic total. However, differences across regions are likely to be similar for
interstate and intrastate travellers. Data for three regions are illustrated below (Tourism
Queensland 2000a, b, d).

Tropical North
Mode of entry Brisbane Gold Coast Queensland
% % %
Air 30 20 43
Private vehicle 58 69 47
Otherb 12 11 10
b ‘Other’ includes rail (7 per cent) for Brisbane; bus/coach (5 per cent) for Gold Coast; and rented
vehicle (5 per cent) for Tropical North Queensland.
Tasmania
Approximately 78 per cent of travellers to Tasmania arrived by air in 2000: 44 per cent
arrived at Hobart Airport, 24.7 per cent at Launceston, 5.7 per cent at Devonport, and
3.1 per cent at Burnie (Wynyard). The remaining 22 per cent arrived by sea (most of
those going to Devonport with TT-Line). (Tourism Tasmania 2001)
Sources: NTTC (2000); Tourism Queensland (2000a–d); Tourism Tasmania (2001).

MARKET POWER OF 119

AIRPORTS
In the case of the Northern Territory, the relative importance of air travel differs
significantly between domestic and international tourists (box 5.3). A far greater
proportion of domestic visitors than of international visitors travelled to the
Northern Territory by private vehicle.

The response of travellers to the reduction of Ansett services in late 2001 may
provide some additional evidence of potential modal substitutability. Anecdotal
evidence suggests that holiday makers were more willing than business travellers to
change transport modes (although business travellers made more use of
videoconferencing) (Heasley 2001). On the other hand, some holiday makers
decided to take holidays closer to home so they could drive to their destination
(indicating destination substitution for leisure travellers) (Innis 2001). Gold Coast
Airport submitted that:
The surveys indicated that because of Gold Coast’s proximity to the major east coast
population centres, many visitors who had booked flights and accommodation chose to
drive or use other forms of transport to complete their holidays …
Surveys undertaken by other State and national tourism organisations indicate that the
experience on the Gold Coast was not evident at the more distant tourism centres such
as Cairns. (sub. DR58, p. 5)

However, what can be inferred from these responses is limited by the sudden and
‘abnormal’ nature of the events.
• In particular, with significantly diminished air capacity, initial traveller
behaviour (particularly in terms of increased rail and coach bookings) was
dominated by those needing to return home. This would tend to overstate the
apparent substitutability of modes compared with what would be observed in
more normal circumstances.
• The large number of holiday cancellations to certain destinations may indicate:
- the inability to change travel modes at short notice — in which case, potential
modal substitutability may be underestimated if, in the medium to longer
term, the length of holidays can be varied to accommodate the use of slower
transport modes; or
- the inherent unattractiveness of using other travel modes to those destinations
— in which case, modal substitutability would be more accurately reflected
in the number of cancellations. Gold Coast Airport noted that ‘as Gold Coast
did receive an increase in new short term bookings during the holiday period
this indicates it could have benefited at the expense of the more distant
destinations because of its proximity to the main population centres’.
(sub. DR58, p. 5)

120 PRICE REGULATION

OF AIRPORT
SERVICES
Demand substitution for dedicated freight traffic

As with passenger traffic, demand substitution for dedicated freight services (that is,
where freight is not carried in passenger aircraft) can be looked at in terms of
destination and modal substitution. And, as with passenger traffic, the substitution
possibilities are likely to differ between international and domestic freight.

For international freight, the only possible modal substitute is shipping. For the
transport of domestic freight, on the other hand, road, rail or sea may be viable
alternatives.

Not all freight is amenable to transport by air; shipping is the only viable alternative
for some types of international cargo, for example. However, the freight that is
transported by air tends to be highly time-sensitive. As Virgin Blue noted:
The means of freight transport is determined by the required delivery time and
characteristics of the freight item. Air transport is higher cost than other forms of
transport, reflecting its short delivery time. Highly time sensitive freight is typically
only suitable for air transport. (sub. 30, p. 12)

Nonetheless, BARA noted:

For some domestic overnight freight services road and rail might provide a viable
alternative. (sub. 26, p. 7)

Apart from the time factor, the relative viability of other modes depends on the
extent of infrastructure for other modes, as well as the location of production of the
goods to be freighted. For example, in Tasmania there are sea freight terminals at
Hobart, Launceston and Burnie. In addition, it was noted by Hobart Airport (and
supported by Launceston Airport (sub. 35)) that:
… Tasmanian salmon producers use both sea and air transport. Our advice is that it
costs $2 per case less to transport salmon by road-sea freight from Hobart to
Melbourne, than by air. The other advantage of road-sea freight is that it is a seamless
service from say the salmon factory at Dover in the south of Tasmania, across Bass
Strait, to the markets of South Australia, Victoria and New South Wales. (sub. 11, p. 5)

Thus, although the substitutability of transport modes for freight is likely to be

relatively low, substitution is possible.

The possibilities for destination substitution appear to be lower for freight than for
passenger traffic. To the extent that domestic air freight is generally time-sensitive,
it needs to be sent by relatively direct routes to a required destination.

In terms of destination substitution of international freight, excessive transport costs

may mean that some goods that otherwise would be transported to (or from)
Australia may be sent elsewhere (or kept within the country). To the extent that
MARKET POWER OF 121
AIRPORTS
freight needs to be sent to Australia, the potential for destination substitution is
limited to the substitution of ports within Australia. That is, freight could be sent to
different airports within Australia, and then freighted by other means (such as road)
within the country, depending on the relative costs of doing so. For example,
Melbourne and Sydney appear to be viable substitutes for transporting exports
originating in Adelaide in some circumstances. A significant proportion of air
freight generated in South Australia — high value and perishable freight — goes to
Sydney and Melbourne airports (SA Government, sub. to PC (1998a)).

The extent to which alternative airports can compete for air freight (both
international and domestic) is discussed below.

Alternative sources of supply for a particular airport’s services

Given that a decision is made to travel (or send freight) to a particular destination,
some form of airport service is required by airlines (and their users). Substitution
possibilities for a particular airport’s services may be available to airlines in terms
of using larger planes or changing the frequency of landings at a particular airport
(and, therefore, shifting some operations to other airports). Virgin Blue highlighted
the potential to substitute airports when, in relation to charge increases at
Coolangatta Airport, it noted that it:
… will have no option but to operate its new Boeing aircraft on routes where the
relevant airports act as responsible members of the community instead of as
opportunistic monopolies. (Huttner 2001)

Nonetheless, the main potential source of input substitution appears to be the

availability of a proximate airport. How close a particular airport must be to be
considered proximate needs to be examined on a case-by-case basis.

Supply substitution possibilities for passenger traffic

For reasons discussed in section 5.2, there is unlikely to be direct competition in a

particular city from another airport for passenger traffic because the core-regulated
airports are the only RPT airports in their city. Nonetheless, airports in nearby
locations may be relatively good substitutes in some cases. The Hon. Ralph Willis,
the then Treasurer, noted that Cabinet set down as a condition of sale that ‘there
would be appropriate cross-ownership controls so that there would be effective
competition between airports and between airport users’ (Willis 1994, p. 3). This
suggests that there is at least some perceived potential for competition among
airports. To the extent that this is the case, the market power of these airports would
be diminished.

122 PRICE REGULATION

OF AIRPORT
SERVICES
Determining the potential substitutability of airports requires consideration of the
geographic dimension of the market (section 5.1). What constitutes a proximate
airport will differ according to the market segment being examined (business
compared with holiday; domestic compared with international).

As noted with respect to modal substitution, although time may be important for
holiday travellers, for many of them it may not be the primary consideration. Thus,
domestic holiday makers are likely to have a more flexible view than business
travellers of what constitutes a proximate airport. This suggests that airports in
destinations with predominantly domestic holiday (or even VFR) travellers are
likely to be more susceptible to competition from nearby airports.8

As discussed above, Tasmania, the Gold Coast, Tropical North Queensland and the
Northern Territory receive high proportions of holiday traffic. Of core-regulated
airports serving these regions, it appears that Launceston, Hobart, Coolangatta and
Alice Springs face effective competition from other ‘nearby’ airports.

Hobart and Launceston face a potentially high degree of competition from each
other. They are just over two hours drive apart, making them relatively close
destinations for holiday — and even VFR — travellers, who are their main market
segments. In addition, since most holiday makers to Tasmania visit more than one
city in the State (Tourism Tasmania 2001), and many travel around Tasmania in
cars, the particular point of arrival is less important than it may be in other States.
Hobart and Launceston airports also face some potential competition for passenger
traffic from Devonport Airport. Indeed, when Kendell Airlines resumed its
Tasmanian service from Devonport Airport, people from around Tasmania,
including Hobart, reportedly travelled to Devonport to take advantage of the low
fares (Barbeliuk 2001).

Coolangatta Airport faces competition from other airports in the southern

Queensland region and northern New South Wales. One hundred kilometres south
of Brisbane (and with a new rail link joining Brisbane and the Gold Coast), visitors
to the Gold Coast have a viable option to go to Brisbane rather than Coolangatta.
There are two indicators of the strength of this potential substitutability. First, even
when no domestic new entrants flew there, it appeared that airfares to Coolangatta
Airport fell when new entrants began flying to Brisbane. Second, airline websites
and brochures issued by the Gold Coast Tourism Bureau note Brisbane or

8 As discussed below, airport charges are a small proportion of airfares. Nonetheless, at the
margin, holiday travellers are likely to be more responsive to any price change. Further, to the
extent that airlines respond to increased charges at one airport by decreasing the frequency of
service, this also may make another proximate airport more attractive to travellers, and
encourage airport substitution.
MARKET POWER OF 123
AIRPORTS
Coolangatta airports as alternative ways of reaching the Gold Coast by air (Gold
Coast Airport, trans., pp. 373–4).

The reaction of Virgin Blue to the proposed 170 per cent increase in aeronautical
charges at Coolangatta Airport (chapter 2), highlighted the potential substitutability
of Brisbane for Coolangatta. For instance, Virgin Blue was reported as saying:
We are bringing thousands and thousands of tourists every week to the Gold Coast
through Brisbane airport. If we can fly into Brisbane for one third of the cost of the
Gold Coast, why would we fly to the Gold Coast? (in Templeton and Mills 2001)

Gold Coast Airport also noted that:

… we are very much open to competition from Brisbane and, apart from the fact that a
lot of passengers want to come to the Gold Coast, the airlines are well established in
Brisbane and it’s not a hard thing for them to just move their whole operation up there.
(trans., p. 499)

The degree to which the announced increase in charges at Coolangatta Airport

could reflect market power is discussed below.

Brisbane, on the other hand, does not appear to face significant reciprocal
competition from Coolangatta Airport. The Qantas Airways website (and in the past
the Ansett website) suggests Coolangatta as an option for some flights to Brisbane.
However, the scale of Brisbane Airport, the extent of business traffic it has, and the
fact that it has a much stronger ability to service international traffic, mean that it is
in a much stronger position than Coolangatta Airport. The ACCC commented:
… while Brisbane is geographically and functionally distinct from Coolangatta, the
reverse does not apply. That is, Coolangatta faces competition from Brisbane, but
Brisbane is not necessarily constrained in its pricing by the proximity of Coolangatta,
given the capacity limitations at the latter. (sub. 36, p. 67)

In the Northern Territory, Alice Springs Airport faces the most substantial
competition from another airport. Many visitors to Alice Springs use the city as a
starting point for their visit to outback areas such as Uluru. The airport at Alice
Springs faces significant competition for this segment of holiday makers from
Yulara Airport, which is much closer to Uluru. Despite higher charges at Yulara
(Northern Territory Airports, sub. 25), Yulara has experienced strong traffic growth
since 1997, in contrast to Alice Springs.

Potential airport substitution for Darwin Airport appears to be lower than for other
NT airports such as Alice Springs. A higher proportion of visitors to Darwin than
other areas of the Territory go for business and VFR purposes (although the
majority still go for holidays). To the extent that these visitors travel by air, the
relative isolation of Darwin means that other NT airports will not be sufficiently

124 PRICE REGULATION

OF AIRPORT
SERVICES
close to be considered proximate. If international tourists want to travel directly to
the ‘Top End’, the potential for airport substitution also will be lower, although the
fact that most international tourists to the Territory arrive on domestic flights
(box 5.3) mitigates this somewhat.

Northern Territory Airports noted that Darwin’s location along the ‘Kangaroo
Route’ means that ‘it is relatively easy for airlines to bypass Darwin if they feel
unhappy with our prices or our service’ (sub. 50, p. 9). It noted further that Darwin
Airport potentially competes for international hubbing traffic with other airports,
particularly in South East Asia (sub. 50). However, it is not clear that these factors
currently act as strong influences on Darwin’s pricing behaviour. The main
potential for airport substitution, then, is from airports within Australia for holiday
visitors who travel to more than one region within the Northern Territory.

It does not appear that the other core-regulated airports face strong competition
from other airports for domestic passenger traffic. As discussed in section 5.2,
although there is more than one airport in some of these cities, such as Melbourne,
the preference of users for using one hub airport means that the others do not
present significant competition.

For international passenger traffic, the ability to substitute points of arrival and
departure within Australia potentially is more significant. According to Bureau of
Tourism Research data, slightly more than half of all international travellers visit
more than one State or Territory in Australia (BTR 2000a, p. 36). Travellers visiting
for VFR purposes are least likely, and holiday travellers most likely, to visit more
than one State or Territory. To the extent that travellers visit more than one State or
Territory, the potential for airport substitution, and hence the elasticity of demand
for international travel to a particular airport, would be higher.

The potential for airport substitution appears to be most significant along the eastern
seaboard. Melbourne Airport noted that:
Whilst Melbourne Airport does not have any concrete evidence, our experience in
dealing with travellers, airlines and tourist authorities suggests … international visitors
tend to be indifferent to their place of arrival and destination but to prefer to arrive and
depart from locations they wish to visit. (sub. 7, p. 10)

Further, it argued that it views its aeronautical market:

… not just as Melbourne or Victoria but for international and long haul domestic
services to encompass Tasmania, South Australia, the ACT and south New South
Wales and potentially New Zealand and even Sydney for Europe and Asia.
(sub. 7, p. 9)

MARKET POWER OF 125

AIRPORTS
On this basis, it has undertaken a survey to identify international segments that
present growth opportunities (sub. 7). In addition, it noted that ‘almost half the
services added to Melbourne Airport since 1997 have been during times when
Sydney Airport has been closed’ (sub. 7, p. 12) — that is, when the curfew prevents
Sydney Airport from operating. This implies that the curfew at Sydney might have
influenced whether some new routes to Australia went to Melbourne rather than
Sydney. Though these decisions were driven by a legislative constraint, it also
indicates an ability and willingness to substitute airports for some international
services.

Nonetheless, Sydney tends to be the dominant point of arrival and departure for
international visitors to Australia. Over 45 per cent of international arrivals and
departures went through Sydney, with about 20 per cent arriving or departing
through Brisbane and about 15 per cent through Melbourne in 1999 (BTR 2000a).
According to the Queensland Department of Tourism, Racing and Fair Trading:
Many of the Queensland destined passengers are now being routed via Sydney for
reasons of airline economics and not travelling public desire. Sydney is therefore a
‘close substitute’ and dilutes Brisbane’s opportunity to be ‘sole supplier’ of airport
facilities or access. Although Brisbane receives a proportion of short haul international
services, consolidation of the international services, particularly from South East Asia,
via Sydney continues to transport a significant proportion of Queensland destination
traffic. (sub. 6, p. 2)

According to some, the rationalisation of airline networks following the events of

September 2001 (chapter 2) may benefit Sydney ‘because carriers are more likely to
trim services to secondary capital city destinations first, maintaining capacity into a
country’s gateway airport’ (Ballantyne 2001, p. 85). This indicates that Sydney may
face less competition for international traffic from other Australian airports than the
other airports face from each other and Sydney.

Supply substitution possibilities for dedicated freight traffic

For dedicated freight (freight that is not carried in passenger aircraft), the potential
substitutability of airports is likely to be greater than for passenger services.

In some cases, there is more than one airport capable of servicing freight traffic in a
particular city. Therefore, potential competition can come from airports within a
city, as well as from airports in another proximate city. In Melbourne, for example,
airports at Avalon, Essendon and Moorabbin all service freight traffic.

Melbourne Airport noted:

… as freight services require less intensive infrastructure than passenger services
(freight terminals are simpler than passenger terminals, freight doesn’t require food and
126 PRICE REGULATION
OF AIRPORT
SERVICES
 beverage outlets or car parks and so on) secondary airports also provide competition. In
the case of Melbourne Airport, Avalon provides services for long haul domestic and
international freight operators whilst Essendon and Moorabbin can accommodate
aircraft carrying small high value consignments (such as courier bags). (sub. 7, p. 16)

In other cases, such as in Tasmania, the proximity of airports in other cities can
make it less costly (and not significantly more inconvenient) to transport goods
between centres by road than by air. As well as indicating intermodal substitution
possibilities, it suggests that the market power of any one airport is constrained by
the potential for its business to be moved to the other centre by other modes. Hobart
Airport commented:
Due to the close proximity of Launceston to Hobart, intrastate and interstate domestic
producers and manufacturers, and exporters, road freight their goods to Launceston
Airport for transhipment by air, due to the cost impact of a 20-minute flight between
Hobart and Launceston, compared to road freight costs. (sub. 11, p. 4)

What constitutes a proximate airport in the case of dedicated freight traffic,

particularly international freight traffic, is likely to be different than for passenger
traffic. For instance, as noted above, Melbourne and Sydney airports are viable
substitutes for Adelaide Airport in certain circumstances.

In addition, the network benefits that lead to a preference for using one airport for
passenger traffic (section 5.2) are not as great in the case of freight in Australia.
This gives airlines more scope to use more than one airport, or shift operations, for
freight services. This seems to be the case particularly for international freight, for
which the most important issue often is to move cargo to (or from) Australia rather
than to a particular city. Melbourne Airport submitted:
Freight service operators seem to be quite willing to change airports, which is a
reflection that their businesses are driven primarily by the task of delivering imports to
Australia rather than to any particular location or the carriage of export freight from a
specific location. This is because what really drives the provision of freight services is
the known availability of high value cargo. It is this preparedness to move coupled with
a general availability of airport capacity that gives airports little market power in
relation to freight services. (sub. 7, pp. 15–16)

Relative importance of airport charges in total airfares and airline

costs

The relative importance of airport charges in total airfares can provide an indication
of the potential responsiveness of (prospective) passengers to changes in airport
charges. The relative proportion of airport charges in airline costs, on the other
hand, may give an indication of the degree to which airlines may be willing to
substitute away from an airport’s services in response to a change in airport charges.
MARKET POWER OF 127
AIRPORTS
Each approach provides a different way of examining the potential price elasticity
of demand for an airport’s services.

The use of price discrimination (‘yield management’) by airlines (discussed below)

makes it difficult to ascertain the proportion of a particular fare that airport charges
comprise. The proportion of the fixed costs of operations (including landing
charges) that airlines allocate to each customer segment differs. Thus, any estimate
of the cost share of airport charges can be based only on an average fare. Applying
the average charge to highest fares, for example, may underestimate the real cost
share of airport charges in that fare. Likewise, applying the average airport charge
to the lowest fare may overestimate the cost share of airport charges in that fare.

On average, airport charges tend to be a small proportion of total airfares. The exact
proportion depends on the airport being considered (because charges vary across
airports), as well as the particular airfare in question. Airfares in turn depend on the
route, carrier and other conditions attached to the ticket, such as discounts for
advance purchase.

For domestic passengers, participants suggested that airport charges comprise 2 to

3 per cent of the average airfare (MTAA Super Fund, sub. 22), while on particular
routes this can be lower. For instance, on the Sydney–Melbourne route (the busiest
route in Australia), airport charges comprise less than 1 per cent of the total full
economy airfare (Hastings Funds Management, sub. 19), and are less than
0.8 per cent of a Canberra–Brisbane full fare (Capital Airport Group, sub. 32).9

However, these figures need to be interpreted carefully. In particular, the airlines

that have leases over their terminals (chapter 2) pay for their leases, which are not
treated as direct airport charges, but incur no other domestic terminal charges. The
absolute level of charges — comprising both landing and terminal charges — will
be higher for new entrant airlines, who do not operate their own terminals.
Combined with their generally lower airfares, airport charges are likely to comprise
a more significant proportion of new entrant airfares.10 Thus, although charges are

9 These calculations were based on charges in place before the ACCC-vetted increase in charges
at Sydney Airport, and before the increases at other airports that followed the Commonwealth
Government’s decision to change price regulation in October 2001 (chapter 3). However, given
that average domestic airfares also rose after September 2001, it is not clear exactly how these
proportions would have changed.
10 For example, the minimum (marginal) cost per passenger incurred by Virgin Blue is the
terminal charge. Terminal charges actually paid by Virgin Blue are difficult to ascertain — for
example because rebates may be provided, offsetting notional terminal charges levied. For
illustrative purposes, assume terminal charges of $1.65 per passenger (at each end, GST
inclusive), as allowed by the ACCC for the multi-user domestic terminal at Melbourne Airport.
In this case, terminal charges would comprise less than 1 per cent of Virgin Blue’s highest full
128 PRICE REGULATION
OF AIRPORT
SERVICES
not likely to be a high proportion of full economy domestic airfares, they are likely
to be more significant for low-fare carriers.

For international passengers, charges at Australian airports are an even smaller

percentage of the total airfare. Though in absolute terms charges are higher (since,
for example, international operators also incur terminal charges), airfares also are
higher. The net effect is that airport charges levied by Australian airports comprise
less than 1 per cent of the total average international airfare (MTAA Super Fund,
sub. 22).

Aeronautical charges also are a relatively small component of airline costs,

comprising about 4 per cent on average (Hastings Funds Management, sub. 19;
Melbourne Airport, subs 7, 37; Westralia Airports Corporation, sub. 21).11 Again,
however, the situation is likely to be different for new entrants, which tend to have
lower per seat cost structures (see footnote 10). As Virgin Blue noted:
Airport charges constitute a significant component of the variable costs per passenger
and are therefore critical to the success of a low fare operator. They are likely to be
significantly less important to a traditional high cost airline. (sub. 30, p. 8)

In the absence of substitution possibilities, the fact that airport charges comprise a
relatively low proportion of airline costs12 and airfares suggests that the price
elasticity of demand for aeronautical services could be relatively low.

Nonetheless, BARA (sub. DR54) noted the importance of aeronautical charges to

passenger numbers, and airline schedules and viability.

airfare on the Melbourne–Perth route (as advertised on its website on 11 December 2001), and
up to almost 4 per cent of its lowest airfare on the Sydney–Melbourne route. The cost for the
‘average’ passenger would include this amount plus a proportion of landing charges. A B737
(as used by Virgin Blue) at typical load factors has around 1.5 passengers per tonne
(MTAA Super Fund, sub. 22). Given this, based on the landing charges in place after the
October changes to price regulation and the increase in charges at Sydney vetted by the ACCC
(chapter 2), landing charges per passenger on the Melbourne–Perth route would be about $8.49
(in total, return, GST inclusive); and on the Sydney–Melbourne route about $9.33 per
passenger (in total, return, GST inclusive). For the average customer on these routes, these
estimates suggest that airport charges (incorporating terminal and landing charges) could
comprise about 1.5 and almost 9 per cent of these fares respectively.
11 Charges levied directly by airports in Australia (that is, airport charges) are only one
component of aeronautical charges. Other aeronautical charges incurred by airlines, such as
terminal navigation and rescue service charges — which are levied in Australia by Airservices
Australia, not airports — may be levied by airports in other countries. In addition, as already
noted, airports do not levy domestic terminal charges on airlines that have domestic terminal
leases. Thus, this figure, based on international estimates, may overstate the proportion of
airport charges in the costs of the domestic operations of airlines in Australia.
12 Though this appears to be low, Ansett (sub. 42) noted that an individual input is considered to
be high cost if its total cost represents more than 1 per cent of an organisation’s total costs.
MARKET POWER OF 129
AIRPORTS
As noted above, new entrant airlines are also likely to be far more price sensitive
than the more established domestic airlines. As a consequence, Melbourne Airport
commented:
… an airport that increases its prices is more likely to lose new entrant volume and,
indeed, if an airport is too aggressive with respect to its prices, new entrants may
bypass it much in the same way that international carriers can … The decision that
airports then face is whether to drop price or not to gain business. (sub. 7, p. 14)

Further, as noted in box 5.2, the relative importance of this factor also depends on
the substitution possibilities available to consumers and airlines (discussed above).
Layard and Walters noted:
… the short-run demand elasticity for any airport’s service is quite low, though in the
longer run, of course, other airports can be substituted, and the number of landings per
trip reduced by larger planes, fewer stopovers, and so on. Proportions are likely to be
variable in the long run. (Layard and Walters 1978, p. 262)

Supply responses of other input providers

As noted in box 5.2, the maximum price that a particular input provider can receive
(for a given level of production) is limited by the amount that consumers are willing
to pay for the final good, and the price that other input providers require for their
own services. Thus, an increase in the price of one input requires an increase in the
price of the final good and/or a lower price being accepted by other input providers.
The lower the ability of other input suppliers to accept a lower price, the greater is
the required increase in the price of the final good and, hence, the greater is the fall
in quantity demanded.

In other words, the elasticity of demand for airport services will be influenced to
some extent by the elasticity of supply of airline services and other input suppliers
to air travel, such as Airservices Australia (ASA). The Commission has been unable
to find literature discussing the interrelationship between airport pricing and the cost
structures of other input suppliers. Therefore, the discussion in this section is
tentative. Nonetheless, it seems clear that the pricing decision of an airport will tend
to induce a supply and/or price response from other input suppliers. This section
briefly examines airline cost and pricing structures, and the charges of ASA.

Airline costs and pricing

Broadly speaking, it appears that airlines have decreasing costs, though some
studies have found this to be driven by economies of scope, networks and density,
rather than economies of scale per se (Reed 1999).

130 PRICE REGULATION

OF AIRPORT
SERVICES
Button and Stough noted:
… the evidence indicates that within any city pair markets there are rapidly declining
costs of service but that there are approximately constant returns to scale for airline
systems that have reached the size of the US trunk carriers. Savings come from
attracting more traffic rather than expanding the network to cover additional
origin/destinations; any additional routes increase the operator’s quasi-fixed costs
which may negate the benefits derived from more traffic. (Button and Stough 2000,
p. 22)

It also appears that the cost of carrying an extra passenger on a flight is minimal
(OECD 2000), assuming there is spare capacity.

To the extent that airlines have decreasing costs, the ability of airports to increase
profits by increasing their charges may be constrained. If the consumption of air
travel falls in response to an increase in airport charges, then the unit costs of
airlines also increase, exacerbating the effect on price (and hence quantity
consumed) of air travel.

Further, it appears that price discrimination by airlines (charging different prices to

different users with a different willingness to pay) is pervasive and increasingly
sophisticated. Network Economics Consulting Group argued that, although airlines
are limited in their ability to price discriminate, there is likely to be ‘a considerable
amount of price discrimination in airline services’ (NECG 2000a, p. 3). And,
according to the MTAA Super Fund:
Business class and full economy fares cover much of the fixed costs of airline
operations, with discount leisure fares covering marginal costs and only a small
contribution to fixed costs … the airline faces separable demand for capacity on each
flight and sets prices to raise revenue in the most efficient manner. Airlines recover
proportionally more from the relatively more price-inelastic demands. (sub. 22, p. 31)

To the extent that price discrimination is needed to cover the fixed costs of airlines,
the ability of airports to increase charges (or impose other costs on airlines) may be
constrained somewhat. The entry of Impulse and Virgin Blue appears to highlight
the fact that airlines in Australia recently have operated in a competitive
environment and have not been earning returns greater than those of other
competitive parts of the economy. Thus, the price discrimination undertaken by
airlines may indeed be necessary to cover their fixed costs.

Airservices Australia

The other major input supplier to air travel is ASA, the Government-owned
commercial authority that levies charges such as those for terminal navigation,
rescue services and firefighting services (chapter 2).

MARKET POWER OF 131

AIRPORTS
ASA charges tend to be significantly higher than the charges levied by airports and,
because of the way in which the charges are calculated, they are particularly high at
smaller airports (chapter 2). Hobart Airport highlighted the relative importance of
these charges for airports like itself, noting that they are about ‘three times dearer
than we [Hobart Airport] are’ (trans., p. 329).

ASA determines whether the services for which they charge are provided, so neither
airports nor airlines have discretion about whether they are provided at a particular
airport. In addition, the fact that ASA charges are based on projected passenger
throughput in a period means they are effectively fixed charges for that period.
Therefore, any increase in charges (or other costs) imposed by an airport that leads
to a fall in passenger throughput (below the projected levels) also leads to an
increase in the average ASA charge for air travel to the airport. This suggests that
airport operators, particularly at smaller airports, may be restrained somewhat in
their pricing decisions by the fixed quantum of ASA charges.

Summing up: implications for market power of particular airports

Given that barriers to entry are likely to be of similar relevance to all RPT airports
(section 5.2), other factors that influence the elasticity of demand for a particular
airport’s services are likely to be the dominant influence on the extent to which
market power differs across airports. A common view is that the elasticity of
demand for airport services is very low, mainly because airport charges comprise a
relatively small proportion of airline costs and airfares. When combined with
natural monopoly characteristics in the provision of airports, this is seen to give
airports a significant degree of market power.

The above analysis suggests that factors other than the proportion of airport charges
in average airfares or airline costs can also be important in determining the elasticity
of demand. The potential for destination, modal and airport substitution has
ramifications for the potential market power of particular airports.

FINDING 5.2

The price elasticity of demand for the services of a particular airport is influenced
by a number of factors. Although the typically low proportion of airport charges in
airfares and airline costs suggests low price sensitivity, this will be mitigated by any
potential for destination, modal and airport substitution, and the supply responses
of other input providers to changes in airport charges.
Airports that face more significant substitution possibilities will face more
price-sensitive demand (and hence have lower market power).

132 PRICE REGULATION

OF AIRPORT
SERVICES
A summary of the Commission’s assessment of these demand factors in relation to
each of the core-regulated airports is provided in table 5.5, and elaborated on in
appendix D. (The supply responses of other input providers are not explicitly
accounted for in this table.)

Table 5.5 Summary of demand and competition characteristics of core-

regulated airports for domestic passenger traffic
Main market
segment to Potential for Potential for Potential for Overall degree
destination for destination modal airport of market
Airport interstate travel substitution substitution substitution power

Adelaide Business/VFR Low Moderate Low Moderate

Alice Springs Holiday High Moderate High Low
Brisbane Business/VFR Low Moderate Low High
Canberra Business/VFR Low High Low Low/Moderate
Coolangatta Holiday High Moderate High Low
Darwin Holiday High Moderate Low Low/Moderate
Hobart Holiday High Moderate High Low
Launceston Holiday High Moderate High Low
Melbourne Business/VFR Low Moderate Low High
Perth Business/VFR Low Low Low High
Sydney Business/VFR Low Moderate Low High
Townsville Holiday High Moderate Low Low

From this analysis, it appears that Alice Springs, Coolangatta, Hobart, Launceston,
and Townsville airports are likely to possess least market power.
• They all serve predominantly tourist markets that face a high degree of
destination substitution.
• Though all these airports are the only RPT airports in their respective cities,
Alice Springs, Coolangatta, Hobart and Launceston appear to face effective
competition from airports in proximate locations:
- Alice Springs from Yulara;
- Coolangatta from Brisbane; and
- Hobart and Launceston from each other.
• Potential competition from other airports does not appear to be significant for
Townsville. However, there appear to be some modal substitution possibilities.
As noted earlier, this does not imply necessarily that the elasticity is high at the
margin. Nonetheless, the fact that it is predominantly a tourist market suggests
that market power would be relatively low.

MARKET POWER OF 133

AIRPORTS
Participants generally agreed that these airports did not have significant market
power. This included the ACCC (sub. 36), but not some airlines that felt that all
airports potentially have significant market power.

On 5 October 2001, the Commonwealth Government announced changes to prices

regulation for core-regulated airports (chapter 3). Under Revocation No.28, Alice
Springs, Coolangatta, Hobart, Launceston and Townsville airports are no longer
subject to price regulation.

Of this group, Coolangatta, Alice Springs and Townsville increased charges

following the announced changes to price regulation (chapter 2). Some have argued
that this shows that these airports can exercise market power. The airports, on the
other hand, have argued that these increases are necessary, given the impact of the
reduction of Ansett services on traffic levels.

It is unclear at this stage whether the increase in charges represents the exercise of
market power. Given airport cost structures (particularly at smaller airports), a
significant fall in traffic would increase unit costs. Most airports that have increased
their charges have indicated that the charges will be reviewed in 2002, in the light
of market developments.

The extent to which the price rises might reflect the exercise of market power can
be assessed only in the medium to longer term, when the response of airlines and
their passengers, and the consequent impact on revenues and profits, become
apparent. Ultimately, assessing whether the price increase is efficient requires
examination of whether another operator, charging lower prices, could efficiently
operate the airport on a continuing long-term basis. This cannot be assessed with the
information that is currently available.

It appears that, although they are not immune to substitution possibilities, Brisbane,
Melbourne, Perth and Sydney airports possess substantial market power. In terms of
the eastern capital cities:
• They have high proportions of business and VFR travellers, who tend to have
more price inelastic (and time-sensitive) demands. Business travellers also tend
to be more significant users of air travel than other groups (table 5.3).
- Nonetheless, a fairly high proportion of interstate visitors to eastern States
travels by modes other than air (table 5.4), indicating that modal competition
is strong for at least some market segments.
• They do not appear to face significant competition for domestic passenger traffic
from other airports.

134 PRICE REGULATION

OF AIRPORT
SERVICES
• Their status as the main international ports of arrival and departure in the
country may add to their potential market power (and leverage) in the domestic
market. Competition among these airports for international traffic may decrease,
though not eliminate, this effect.

With respect to Perth Airport, the isolation of Perth (and the fact that most visitors
go there for business or VFR) is a significant source of its market power. This is
highlighted by the fact that a very high proportion of visitors arrives by plane
(table 5.4). Westralia Airports Corporation itself accepted that, on the basis of the
high barriers to entry and limited modal substitutes to Perth, it:
… possesses market power in the provision of aeronautical services, but is strongly of
the view that it does not abuse this market power. (sub. 21, p. 30)

Most participants agreed that, of all the core-regulated airports, Brisbane,

Melbourne, Perth and Sydney have most market power. The Queensland
Department of Tourism, Racing and Fair Trading (sub. 6) argued, however, that
Sydney and Melbourne airports had more market power than Brisbane Airport.

In general, any claims of abuse of market power for these airports have related to
specific services. These issues are addressed in chapter 6. Sydney Airport has been
accused of a more general abuse of market power in relation to its 2000 pricing
proposal and, in particular, the size of the proposed increase in its charges. Of itself,
the magnitude of the proposed increase — 75 per cent of which was approved by
the ACCC in May 2001 — does not indicate an abuse of market power, particularly
given the special circumstances of Sydney Airport. Prior to the increase, landing
charges at Sydney were significantly lower than those at any other core-regulated
airport, despite excess demand for several hours a day (and the convenient location
of the facility) (chapter 8). The price increases also reflected a switch from single-
till to dual-till pricing.

For Brisbane, Perth and Melbourne airports, the changes to price regulation
announced in October 2001 allowed a one-off increase to be passed through the cap
(chapter 3). All three applied the allowed increases. Melbourne Airport, for
example, implemented the price increase by rebalancing charges. While domestic
charges increased, there was no net change to international charges (chapter 2).
Given the different natures of the shocks in the two markets — supply side in the
domestic market, and demand side in the international market — this may be an
efficient price response.

The degree of market power held by Adelaide, Canberra and Darwin airports is less
clear. The ACCC (sub. 36) submitted that all three possess sufficient market power
to warrant continued regulation.

MARKET POWER OF 135

AIRPORTS
Darwin is the major air entry point for the Northern Territory (NTTC 2000), and the
only direct port of arrival to the Territory for international travellers. It also is
relatively isolated. Nonetheless, although it has a higher proportion of business and
VFR travellers compared with other NT regions, its dominant market segment is
holiday travel. Thus, it would seem to face a potentially significant degree of
destination substitution. In addition, most international travellers who arrive in (and
leave) the Northern Territory by air do so on domestic flights. This decreases any
market power that otherwise may accrue to Darwin through its ability to service
international traffic. On balance, it does not appear that Darwin Airport has a high
degree of market power, although its market power is likely to be more significant
than that of Alice Springs.

Both Adelaide and Canberra have high proportions of business and VFR travellers.
They also do not face significant potential for airport substitution. Nonetheless,
they, and some others such as Melbourne Airport (trans., p. 177), suggested that
Canberra and Adelaide airports do not hold significant market power. Adelaide
Airport, for example, argued:
… we’re not actually a strategic hub base for at least one national carrier, nor is our city
a major economic or cultural capital for our nation here in Australia. Adelaide is very
much of secondary importance to our largest customer groups. These are very
important drivers of market power in our perception. (trans., p. 181)

It also argued that it faces a high degree of modal substitution. BARA (sub. DR54)
and the ACCC (sub. DR55) disagreed, however. BARA argued that:
… for business travellers such substitution is limited (given the time difference
between air travel and road travel between the cities). For the majority of travel
between Adelaide and other cities the substitution is likely to be even more limited. To
assess the degree of modal substitution for Adelaide Airport requires information on
the origin and destination of persons using Adelaide Airport. (sub. DR54, pp. 30–1)

Tourism data suggest that 75 per cent of interstate overnight visitors to South
Australia were from Victoria and New South Wales in 2000 (SATC 2001).
Adelaide Airport Limited (sub. DR77) also noted that the majority of visitors from
these States arrive by surface transport.

Thus, for the major visitor markets to Adelaide — holiday and VFR from Victoria
and New South Wales — the degree of modal substitution may not be insignificant.
On the other hand, modal substitution for business travellers, and those from other
areas, would be fairly low. On balance, the degree of modal substitution for
Adelaide Airport appears to be moderate, which would tend to constrain the
airport’s overall market power.

136 PRICE REGULATION

OF AIRPORT
SERVICES
With regard to Canberra Airport, a large proportion of visitors to Canberra arrives
in private vehicles (table 5.4). This may reflect the high proportion of Sydney–
Canberra traffic in the total arrivals in Canberra.

Capital Airport Group noted that the Canberra–Sydney route comprises about
50 per cent of the airport’s market (trans., p. 463). Although this is likely to be
predominantly a business market, other modes are likely to provide relatively strong
transport options even for this market segment. As already noted, on this route, the
distance between origin and destination is such that driving may be less time-
consuming than flying (taking into account travel time to and from the airports at
both ends). Thus, even some business travellers may find this a close substitute for
air travel. Capital Airport Group contended that:
… the impact of modal substitution on Canberra Airport is at least as great as that
which results from airport substitution possibilities on airports such as Coolangatta and
Launceston. (sub. DR75, appendix 1, p. 1)

That its major market potentially has strong modal substitutes would act to
moderate somewhat the market power of Canberra Airport. This is so even though
its primary market — business — normally is considered a relatively less elastic
market segment. Nonetheless, the overall market composition of Canberra Airport
suggests that it may have more market power than airports such as Coolangatta and
Launceston, which face strong airport substitution possibilities.

On balance, although the market power of Canberra and Adelaide airports does not
appear to be as significant as Brisbane, Melbourne, Perth or Sydney, both airports
appear to have a moderate degree of market power.

Following the Government’s issue of Revocation No. 28 and Direction No. 26,
which together replaced the price cap with price monitoring for Darwin, Adelaide
and Canberra airports (chapter 3), all three airports increased their aeronautical
charges (chapter 2).

As noted earlier, such increases may not, of themselves, reflect the existence and
exercise of market power. Airport cost structures may justify this response to falling
volumes. Only in the medium to longer term will demand responses provide some
indication of the profitability of the change for airports (and, thus, whether the
changes might reflect an abuse of market power).

Furthermore, Canberra Airport also provided incentives to attract flights following

the reduction of Ansett services. It commented:
So we’ve got back 6000 out of the 28,000 seats we lost. We did that by offering
substantial incentives to these businesses. We assisted with their marketing, we worked

MARKET POWER OF 137

AIRPORTS
 with the governments to provide them with assistance. We worked with hotel groups to
provide them with accommodation, [and with] the taxi industry … (trans., p. 461)

FINDING 5.3

Of the core-regulated airports, Sydney, Melbourne, Brisbane and Perth have most
market power. Adelaide Airport is likely to have a moderate degree of market
power. Canberra and Darwin airports also are likely to have a moderate degree of
market power, though they appear to face stronger substitution possibilities than
Adelaide Airport.
Core-regulated airports that do not appear to have significant market power (due
mainly — except for Townsville — to the scope for effective inter-airport
competition) are Alice Springs, Coolangatta, Hobart, Launceston and Townsville.

Market power of core-regulated airports in providing international services

The market power of core-regulated airports generally is not likely to be significant

in the international market, even for Australia’s major international airports
(Sydney, Melbourne, Brisbane and Perth).
• For international passengers, the ability to substitute points of arrival and
departure within Australia potentially is more significant than for domestic
travellers. In particular, there is potential substitutability of airports along the
eastern seaboard.
- The apparent impact of the curfew at Sydney Airport — in terms of allowing
Melbourne Airport to attract new services — highlights this potential
substitutability.
- Airports appear to have competed strongly to attract new international
services.
• International air carriers may be quite sensitive to increased airport charges.
Brisbane Airport Corporation noted that, prior to the change in price regulation
(chapter 3), it ‘had already approached the airlines seeking a 25 per cent increase
in charges for the remaining period of the price cap’ (trans., p. 513). BARA
commented that:
… a number of our members who operate to Brisbane Airport said that that sort of level
of increase in prices would under the current circumstances result in their reducing
services to Brisbane Airport quite markedly. It had a very significant effect on the
returns for the likes of Singapore Airlines, for instance. (trans., p. 540)

138 PRICE REGULATION

OF AIRPORT
SERVICES
Nonetheless, Sydney, as the main international gateway airport in Australia, is
likely to have a more significant degree of market power for international traffic
compared with the other core-regulated airports.

Market power of non-core-regulated airports

The focus of the above discussion has been on market power at core-regulated
airports. However, a number of participants discussed market power in relation to
other airports — mainly with reference to regional airports in general, though some
airports have been discussed specifically.

Given the number of non-core-regulated airports that cater for RPT traffic in
Australia, this section will focus on general issues relating to regional airports.
Some particular cases mentioned by participants (Cairns and Yulara) are treated
separately.

Market power and non-core-regulated regional airports

Regional airports in Australia are not a homogeneous group; a relatively small

number service RPT traffic, with most catering for general aviation movements
only (chapter 2). Facilities provided at these airports also differ widely, ranging
from little more than an airstrip, to international terminal facilities.

Despite these differences, it is possible to make some general comments about the
potential market power of regional airports. Some factors may imply the existence
of market power.
• It is unlikely that there would be more than one airport in any particular regional
centre. Although capital investment requirements at regional airports are
significantly lower than those for airports in the major capital cities, the potential
traffic volume through the airport is unlikely to be high in most cases. There also
may be issues of obtaining approval to construct a new airport.
• Most remote locations depend on air transport. As Australian Airports
(Townsville) noted in the case of Mount Isa: ‘We’re talking about a community
of 19 000 people. It relies on the airport as basically its only means of quick
communication’ (trans., p. 7).

However, a number of factors may mitigate the extent and exercise of market power
by regional airports. The importance of air transport to many regional communities
means that the community is likely to take a great interest in the decisions of the
airport and try to exert its collective influence over those decisions.

MARKET POWER OF 139

AIRPORTS
Australian Airports (Townsville) noted:
The local community also plays a major role in terms of market power restraint of
airports. The community has expectations and subsequent demands as to the level and
quality of services to be provided by the airport as well as the fee that they are willing
to pay for these. The community is able to exert pressure on the airport operator to
ensure that service levels are appropriate and that prices are at a level that supports
economic goals and development within the region the airport serves. (sub. 14, p. 10)

In addition, with some exceptions, regional airports do not tend to be critical to the
overall operations of airlines, so competitive pressures are unlikely to compel
airlines to fly to many of these airports. In this situation, it is not clear that it is the
airports that have the market power. As noted by the Department of Industry,
Science and Resources:
Airlines can more easily reduce services at smaller regional airports but the commercial
risks of sacrificing access at major city airports would be much higher. (sub. 40, p. 3)

Further, the larger regional airports (including some regional core-regulated

airports) tend to serve predominantly tourist traffic. As noted above, this market
segment is likely to be more price sensitive than other market segments, decreasing
somewhat the market power of airports in these locations.

Finally, many general aviation and regional airports in Australia are local
government-owned (chapter 2). Therefore, the users and (direct or indirect)
beneficiaries of the airport’s services are the residents who also indirectly own and
operate the airport through their elected council representatives. Given the
importance of the local airport in many regional communities, and because the
impacts of airport policy are relatively localised, discipline on the airport operator
— the council — to represent the best interests of users could be provided through
the electoral process.

Nonetheless, some participants expressed concern about abuses of market power at

regional airports. The Western Australian Government, for instance, commented:
We are concerned that some airports … are using their monopoly position to maintain
and increase charges. We are aware, certainly from the airlines and other sectors of the
industry, there is little consultation when this takes place. We do have examples in
Western Australia where … key infrastructure has gone in ahead of time, as to when it
was needed. (trans., p. 389)

Concerns also have been raised about the extent to which councils use airport
charges to fund general community projects rather than aeronautical uses. Again,
the Western Australian Government stated:
An area of concern to us — as has been shown in other parts of Australia — is that
airport revenue is being used to fund non-airport related plant and infrastructure.
(trans., p. 389)
140 PRICE REGULATION
OF AIRPORT
SERVICES
Anecdotal evidence suggests that charges at regional airports vary significantly. No
charges are imposed at some regional airports, while charges can be very high at
others. Ansett suggested that ‘a lot of the regional ports have exceptionally high
charges’ (trans., p. 125). Of themselves, high charges do not necessarily indicate an
abuse of market power. The small traffic volumes of most regional airports suggest
that efficient unit costs may be high.

Given the ownership structure of most regional airports, the variation in charges
also reflects local government objectives, which Professor Forsyth noted can be
diverse:
Some local government owned firms may indulge in monument building. On the other
hand, a local government owner may use its airport to attract visitors to the region; if
this is the objective, it will induce the airport to keep costs low and to keep prices in
line with costs. It is also possible that some local governments will require their airports
to maximise profits, so that they can use these to cross subsidise land rates.
(sub. 5, p. 17)

However, if the airport policy of regional councils does not reflect local community
wishes, then this arguably is more properly a local governance issue, not a market
power issue to be addressed by the national economic regulation of regional
airports.

Objectives other than the exploitation of market power also may explain seemingly
high charges at airports located at resorts, for instance. Scott-Bloxam (sub. 2)
submitted that landing charges at Lizard Island were ‘astronomically high’.
Although acknowledging that the maintenance of the sealed strip would be
substantial, he added: ‘I fail to see where a 3 seat Cessna 172 or 5 seat Cessna 206
can attract such high fees other than to discourage their visit’ (sub. 2, p. 1).

If landing charges reflect the high costs of the strip, an objective to keep visitor
numbers low to cater for a particular market segment, or environmental objectives,
then it is not clear that this constitutes an abuse of market power. Further, although
resorts are differentiated from each other, potential visitors have a variety of resorts
from which to choose. Thus, the resort market can be seen as competitive. In this
case, the overall pricing behaviour of a particular resort, including the charges of its
airport (an input to the resort product), would be constrained somewhat by
competition among resorts.

Does Cairns Airport have market power?

Given that Cairns is a regional airport, much of the general discussion above is
relevant to it. However, Cairns has unique characteristics that warrant further
discussion. In addition, Ansett and BARA pointed to issues in their dealings with
MARKET POWER OF 141
AIRPORTS
Cairns that they argued resulted from its market power. These issues related to the
level of charges, efficiency and degree of consultation. On the other hand,
Australian Airports (Townsville) (sub. 14) argued that it did not believe that
unregulated Queensland airports, including Cairns, had abused market power.

Cairns potentially may have more market power than most other regional airports. It
is the sixth largest RPT airport in the country in terms of passenger movements
(larger than many core-regulated airports (chapter 2)), and is a designated
international airport, suggesting that it has a significant demand base. Indeed,
Australian Airports (Townsville) noted that Cairns is relatively attractive as a tourist
destination compared with Townsville, for instance, ‘which is not as well developed
as a tourist attraction’ (sub. 14, p. 14).

The relative popularity of Cairns as a tourist attraction may give it an advantage

over other destinations and mean that it is a more important part of airline
operations compared with some other regional airports. Nonetheless, like any
primarily tourist-based destination, it is relatively vulnerable to competition from
other destinations. The recent withdrawal of a number of international services from
Cairns highlights this vulnerability and, according to Gold Coast Airport, reflects
that ‘a full-service airline finds it difficult to make money on a tourism route’
(trans., p. 378). In addition, the airport faces some competition from other modes
(box 5.3).

In this situation, the market power of Cairns does not appear to be significant. As
noted by the Department of Industry, Science and Resources:
Airports which primarily service seasonal tourist destinations such as Cairns and
Maroochydore have the least amount of market power. (sub. 40, p. 3)

Given this, to the extent that charges are higher at Cairns than other airports, they
appear unlikely to reflect an abuse of market power. Since the major airlines incur
domestic terminal charges at Cairns, which they do not incur at the core-regulated
airports where they provide these services themselves, the charges at Cairns could
be expected to be higher than those at core-regulated airports.

Further, the Commission understands that landing and terminal charges at Cairns
Airport (exclusive of GST) have not increased since new charges were negotiated in
1990 (although some new charges, relating to Government-mandated security
requirements, for instance, have been introduced). Including GST, current domestic
landing charges are $3.84 per tonne (based on maximum take-off weight (MTOW),
which is lower than the core-regulated airports (chapter 2)), and domestic terminal
charges are $3.84 per passenger. Though terminal charges are higher than those for
some common-user terminals, a more relevant comparison may be the per

142 PRICE REGULATION

OF AIRPORT
SERVICES
passenger costs of the domestic terminals operated by the major domestic airlines.
However, the Commission does not have these data.

Market power of Yulara (Ayers Rock) Airport

Ansett cited its experiences with Yulara Airport as an example of the fact that
‘smaller non-regulated airports are also not averse to exercising their market power’
(sub. 42, p. 27).

However, it is not clear that Yulara Airport has significant market power.
• It may have an advantage as a tourist destination. However, it competes for
visitors with other ‘unique’ destinations.
• The airport is owned and managed by the owners of the Yulara Resort. It is thus
part of an integrated resort operation. Managers may seek to make profits from
the operation as a whole. However, it is likely that this would be done by
attracting visitors to the resort (subject to capacity constraints). In addition, as
noted above, the resort market can be seen as competitive. Competition would
tend to constrain somewhat the pricing of a particular resort, including the
charges of its airport (an input to the resort product).
• The costs of establishing and running an airport in that location are likely to be
very high, especially given the runway requirements for jet aircraft and
comparatively small volumes of planes and passengers.
• To the extent that there is a type of locational rent associated with its location,
this would have been captured by the owner of the site when the lease was
purchased.

5.4 Summing up
The overall market power of a particular airport is encapsulated in the price
elasticity of demand for its services, which will reflect substitutes available to users,
and the share of airport charges in airline costs and airfares.

In the absence of substitution possibilities, the low proportion of airport charges in

airline costs and airfares might suggest low price sensitivity and strong market
power. Barriers to entry in the provision of airport services also suggest that the
scope for airport substitution is likely to be low. However, substitution possibilities
exist — destination, modal and even airport substitution (an airport in another
location may provide competition in some cases) — which may be quite strong for

MARKET POWER OF 143

AIRPORTS
some airports. Hence, the sensitivity of some travellers to price changes at particular
airports may be quite high. Thus:
• Of the core-regulated airports:
- Hobart and Launceston, as well as Coolangatta and Alice Springs, appear to
have little market power because of their reliance on the tourism market and,
particularly for this market segment, scope for competition from ‘nearby’
airports. Townsville Airport also has limited market power because of its
reliance on the holiday market (and scope for competition between holiday
destinations and from other transport modes).
- Adelaide, and to a lesser extent, Canberra and Darwin have a moderate
degree of market power. Although Adelaide and Canberra have high
proportions of business and VFR travellers and do not face significant
potential for airport substitution, they (especially Canberra) do face modal
substitution for visitors from their major markets.
- Brisbane, Melbourne, Perth and Sydney airports possess a high degree of
market power in domestic markets due to high proportions of business and
VFR travellers, and their status as the main international ports of arrival and
departure in the country. Competition among those airports for international
traffic may moderate, though not eliminate, this latter effect.
• With respect to non-core-regulated airports:
- No charges are imposed at some regional airports, while at others charges can
be very high. Of themselves, relatively high charges need not reflect an abuse
of market power. The small traffic volumes of most regional airports suggest
that efficient unit costs typically may be high. Charging at regional airports
(which in many cases are owned by local municipalities) also can reflect a
range of objectives. Privately-owned airports servicing resorts generally
operate in the highly competitive tourist market.

144 PRICE REGULATION

OF AIRPORT
SERVICES
6 Market power in particular airport
services

In addressing the requirement of the terms of reference that ‘future prices regulation
should be applied to those aeronautical services and those airports where airport
operators have most potential to abuse market power’, chapter 5 examined potential
market power of particular airports in Australia. This chapter considers the extent of
an airport’s market power in particular services provided at airports.

6.1 Introduction
It was found in chapter 5 that Sydney, Melbourne, Brisbane and Perth are likely to
have the highest degree of market power of Australia’s core-regulated airports,
followed by Adelaide, Canberra and Darwin, which appear to have moderate market
power. The others — Townsville, Alice Springs, Hobart, Launceston and
Coolangatta — possess little, if any, market power.

The market power of an airport depends on its market power in providing particular
airport services. Although an airport that has market power in the overall market for
airport services must have market power in some of the individual services, this
does not mean necessarily that it has market power in all services provided at the
airport.

The extent to which it is meaningful (from a regulatory perspective) to distinguish

market power by service depends on the degree to which airport users, having made
a decision to use an airport, have alternatives to the consumption of particular
airport services. This in turn depends on the availability of off-airport substitutes,
the availability of alternative on-airport suppliers and/or the strength of the demand
preference for that service (that is, the extent to which the service is required by
users). The Board of Airline Representatives of Australia (BARA) considered that
all airport services form part of the required bundle for airlines:
… airports compete in a cluster market for the supply of a range of services required by
airport users and particularly airlines. As a result, the airport’s market power extends
across the package of services provided … As these services are purchased as a
package, constraining the price of one of the elements will simply result in a higher
price for one or more of the other elements. In order to address the market power (and

MARKET POWER IN 145

PARTICULAR
AIRPORT SERVICES
 associated efficiency loss) the regulation must encompass the package … (sub. 26,
pp. 13–14)

The ACCC noted the importance of considering non-discretionary bundles of

services, but added:
Determining the services which should be included in the cluster of ‘airport services’
… rests primarily on an assessment of the scope for supply side substitution.
(sub. 36, p. 78)

Virgin Blue argued:

A firm is likely to have market power when its customers have no alternative source of
supply, albeit that the customers may choose not to take supply at all. (sub. DR74, p. 3)

However, both demand-side and supply-side conditions are important in

determining airport market power in particular services, and both dimensions are
considered here.

In addition, in assessing any evidence of (the use of) market power in particular
airport services, it is important to distinguish between ‘monopoly profits’ on the one
hand and ‘locational rents’ on the other.
• Monopoly profits are profits in excess of the rates of return required to maintain
supply of the good or service.
• Locational rents are returns accruing to a scarce factor rather than returns
deriving from the exercise of market power as such.

To the extent that profits reflect locational rather than monopoly rents, they do not
involve an efficiency loss through distorted supply and demand (appendix C), but
rather reflect the value placed on scarce resources by consumers and provide signals
for their efficient use.

Different classifications of airport services are possible. Those used below broadly
follow the ACCC (sub. 36) and Department of Transport and Regional Services
(sub. 39) approaches.

6.2 Aircraft movement facilities

Aircraft movement facilities include runways, taxiways, aprons and aircraft parking,
as well as airside grounds, airfield lighting, airside roads and lighting, airside safety,
nose-in guidance, and visual navigation aids.

These facilities are essential to the operation of the airport. Planes arriving at an
airport need a runway to land, taxiways are needed to move aircraft from the
146 PRICE REGULATION
OF AIRPORT
SERVICES
runway to aprons or parking areas and so on. Airline discretion over aircraft parking
may be higher than for the other components of aircraft movement facilities, to the
extent that airlines have a choice about where to end their services for the day.
Nonetheless, it is likely to be infeasible for airlines to move empty aircraft and
flight crew to another airport simply to minimise parking charges, given the costs
involved. Aircraft parking charges may, however, influence whether airlines choose
to park on the apron or elsewhere at the airport.

As well as being essential to users of an airport, these facilities, particularly

runways and taxiways, are likely to be the most significant source of natural
monopoly for airports. Indeed, as noted in chapter 5, to the extent that there are
economies of scale in the provision of airport services, they derive mainly from the
provision of runways. Economies of scope arise in that different types of aircraft
can land at the one facility (chapter 5). Hence, the supply characteristics are such
that there generally will be only one supplier of these facilities in a location.

Thus, where an airport has market power, its market power in relation to aircraft
movement facilities is high (and effectively equal to its overall market power).

6.3 Passenger processing facilities

In broad terms, passenger processing facilities refer to all facilities that provide
services to passengers, up to the point that passengers enter, or from the time they
exit, the plane. They can include check-in desks, aerobridges, airside buses,
departure and holding lounges (but not VIP lounges, which are discussed separately
below), immigration and customs service areas, public address systems, closed
circuit surveillance systems, security systems, baggage handling and reclaim areas,
public amenities, other public areas in terminals, lifts, escalators, moving walkways
and flight information display systems (ACCC, sub. 36).

Some forms of passenger processing facilities — such as check-in desks and

baggage handling and reclaim areas — are required for both domestic and
international traffic. However, there also are inherent differences in the nature and
extent of processing facilities required for international, compared with domestic,
services. In particular, immigration and customs service areas are required (by law)
for international airlines, but are not necessary for domestic travel.

This section examines in two stages the market power held by airports in passenger
processing facilities. The first part explores general issues relating to terminals and
passenger facilities. The second looks specifically at check-in counters, which, for
the airports subject to the price cap, are not included in the cap but are monitored
(chapter 3).
MARKET POWER IN 147
PARTICULAR
AIRPORT SERVICES
General terminal facilities

For the airlines that hold long-term leases over their domestic terminals (chapter 2),
the airlines, rather than the airports, provide many domestic terminal facilities and
services. In this case, BARA noted:
… the service provided by the airport is access to space adjacent to the runways and
taxiways upon which the terminal is located. (sub. 26, p. 4)

On the other hand, domestic new entrant airlines have relied on airports for direct
provision (and construction) of common-user terminal facilities.

The extent of airport market power in these services will be determined by the
degree of discretion that airlines have over what is to be provided, the ability to
build new facilities, and opportunities for off-airport provision of the services.
Australian Airports (Townsville) argued:
Passenger processing facilities … may be used selectively by the customers and can be
provided to the level of quality to meet the customers and end consumers needs.
(sub. 14, p. 17)

The ability and willingness of airlines to tailor passenger processing facilities to the
needs of their customers has been highlighted in Australia by the domestic new
entrants. Terminals built to accommodate them have been more spartan than those
of the established airlines. In addition, the preference of Virgin Blue not to use
aerobridges shows that some of the facilities do not need to be used at all.1 Airline
discretion about the extent of these facilities will vary depending on the main
market segments they serve.

On the supply side, unlike runways, on-site incremental development of passenger

processing facilities is feasible, so the barriers to entry may not be significant
(especially if the extent of facilities required is small). Nonetheless, as the ACCC
(sub. 36) noted, space constraints may limit the future ability to construct new
terminals.

Where on-site expansion is feasible, issues appear to relate to access to airport land
to build facilities, or disagreements between the airline and the airport about the
type of facilities that should be built. For instance, Impulse Airlines noted that
‘there are issues in terms of what we think should be a standard of terminal
supplied’ (trans., p. 40). Virgin Blue also argued that:
Where new facilities are required, in Virgin Blue’s experience, some airport operators
are not willing to be flexible about the standard of terminal facility which Virgin Blue
is able to access. (sub. 30, p. 9)

1 However, where there is apron congestion, the choice of whether to use aerobridges is limited.
148 PRICE REGULATION
OF AIRPORT
SERVICES
In addition, Virgin Blue (sub. 30) submitted that it offered to construct a terminal at
Melbourne Airport, and provide access to other entrants, but was refused by
Australia Pacific Airports (Melbourne) (APAM, the operator of Melbourne
Airport), which decided to build its own common-user terminal. Following
negotiation over the terms on which access would be granted to the terminal on a
long-term basis, Virgin Blue made an application for an access determination under
s. 192 of the Airports Act 1996 (chapter 9).

In its report for the ACCC in relation to the Virgin Blue determination, Network
Economics Consulting Group (NECG) commented that:
… the fact that the owner of the existing terminal is also the owner of all the land in the
surrounding vicinity is highly relevant. We believe that Melbourne Airport would have
strong incentives to deter entry by a potential competitor in the market for the provision
of terminal facilities … the new entrant could easily find that it becomes uneconomic to
duplicate the existing terminals. (NECG 2001b, p. 46)

However, it is not clear that the difficulties in the Virgin Blue–Melbourne Airport
negotiation process demonstrate an abuse of market power by the airport. Airports,
including Melbourne, appear to have been eager to attract the business of new
entrant airlines (who have sought new facilities from airports, rather than seek
access to existing terminals operated by other airlines).

The problems that arose in this case seem to have stemmed primarily from a
difference of opinion about what was appropriate and required. This seems to be an
inherent problem of common-user facilities, rather than an issue of market power. It
also must be noted that negotiations with Impulse over the same terminal were
successful. Impulse Airlines noted that:
… we signed a commercial agreement with Melbourne Airport … on a struck
passenger charge, which we thought was fair to get the facility built. (trans., p. 40)

If off-site passenger processing is possible, then the market power of airports in

providing these facilities would be diminished. Participants noted that the
possibility for off-site provision is limited at present but may grow as technology
improves. The ACCC noted in its draft decision on the Virgin Blue access
determination that:
It is difficult to conceive how a number of terminal services (such as the loading and
unloading of passengers) could be provided anywhere other than the airport. Certainly
some functions can, in principle, already be provided off-site. Indeed in future years,
off-site facilities may become more feasible and common. However in the time period
relevant to the current analysis, existing or potential off-site terminal facilities do not
appear likely to significantly constrain the behaviour of the owners of on-site terminal
facilities. (ACCC 2001l, p. 15)

MARKET POWER IN 149

PARTICULAR
AIRPORT SERVICES
Some inquiry participants have argued that, of the passenger processing facilities,
off-site provision is most likely to be feasible for check-in counters. The market
power of airports in providing check-in counters is discussed below.

Check-in counters

As already noted, for the airports subject to the price cap, check-in counters are not
included in the cap but are monitored (chapter 3).

The degree of market power held specifically in check-in counters depends on the
degree of discretion airlines have over what is to be provided, the ability to use
other facilities on-airport, and opportunities for off-airport provision of the
facilities.

Sydney Airports Corporation Limited (SACL) argued that airlines have a choice
about the quantity of check-in facilities used, limiting the market power of airports:
If charges for use of check-in counters were considered to be too high, airlines would
choose to manage their passengers through a smaller number of counters. (sub. DR62,
attachment A, p. 1)

If, however, decreasing the number of check-in counters increased passenger

processing times significantly, then the willingness of airlines to take this action
would be diminished.

Slower passenger processing also could be against the airport’s interests. SACL
noted that the use of fewer check-in counters ‘would clearly also be a concern for
airport operators in managing terminal congestion’ (sub. DR62, attachment A, p. 1).
In addition, slower processing times would decrease the time that passengers have
to shop in airport retail outlets. These factors could, therefore, offset incentives the
airport operator otherwise may have to increase, above efficient levels, the charges
imposed on airlines for check-in counters.

In addition, in Australia, domestic airlines appear to have some on-airport

alternatives for check-in. SACL noted that one factor limiting the market power of
airports was the ‘ability of domestic carriers to check-in passengers at their leased
terminals where those passengers are transferring to an international service’
(sub. DR62, attachment A, p. 1).

In terms of off-site check-in, this already exists in a number of places overseas

(sometimes for international, as well as domestic, passengers) — at Gatwick
(Victoria Station), Heathrow (Paddington Station), Hong Kong (Hong Kong Island
Station) and Geneva (Cornavin Station), for example.

150 PRICE REGULATION

OF AIRPORT
SERVICES
According to Brisbane Airport Corporation (BAC):
New technologies and processes are likely to provide alternatives to aspects of the
current methods of international passenger processing, eg: electronic check-in, off-
airport processing at hotels, in the CBD, on other transport modes (trains), etc. As these
technologies and processes evolve, these services are unlikely to continue to satisfy the
tests for monopoly infrastructure. (sub. 8, p. 13)

Nonetheless, some participants argued that the potential for off-site check-in
currently is limited in Australia. BARA, for instance, commented that:
It is possible that a few of the larger hotels may establish these facilities. However, they
would only cater for a small proportion of passengers. There is no central point where a
large number of passengers congregate (such as a train station). Electronic check-in is
unlikely to be effective as most international passengers have baggage. The scope for
supply-side substitution is extremely limited. It is highly unlikely that a significant non-
transitory increase in charges for check-in counters would have much effect on the
supply of off-site facilities. (sub. DR54, p. 28)

Similarly, NECG considered that:

… in some major world cities, a growing range of terminal facilities is being provided
at city-centre locations and at connecting railway stations. However, in this specific
[Virgin Blue–Melbourne Airport] case we do not believe that an airline would find it
profitable to provide all services to all passengers in this manner, and, therefore,
(potential) off-airport terminal facilities cannot credibly be considered to compete in
the same market as terminal facilities at Melbourne Airport. (NECG 2001b, pp. 38–9)

Although off-site provision of check-in counters is not likely to be feasible in all

cases, to the extent that it is available for some passengers, this may decrease the
market power of airports. Nonetheless, the current feasibility of this is likely to be
limited in Australia.

NECG (2001b) also considered substitution in terms of location at a nearby airport.

However, the lack of nearby airports in most instances means that this is not likely
to be a significant source of potential substitution. Moreover, it would appear to be
infeasible to locate check-ins at one airport and other passenger processing facilities
at another. Finally, once a decision is made to locate at a particular airport in a city,
the costs of moving are likely to be prohibitive. NECG (2001b) also concluded that
this was not a viable option.

Summing up of market power in passenger processing facilities

Overall, although the market power of airports in relation to passenger processing

facilities is not as high as for aircraft movement facilities, airports with market

MARKET POWER IN 151

PARTICULAR
AIRPORT SERVICES
power will have some market power in the provision of passenger processing
facilities.

Of the passenger processing facilities, market power is likely to be least significant

for check-in counters. Though counters are necessary, airlines may have some scope
to reduce the number of counters used in response to a significant increase in
charges. Such a reduction could be expected to have negative consequences for the
airport, including terminal congestion and reduced shopping time for passengers.
Furthermore, domestic airlines may be able to process some international
passengers through their domestic terminals. Finally, check-in counters also are the
most likely of the passenger processing facilities to be amenable to off-site
provision, although it appears that this potential currently is limited for international
passengers in Australia.

6.4 Lounge space (VIP and business)

Lounge facilities refer to business and VIP lounges that airlines provide for their
customers. They do not include departure lounges (which comprise part of
passenger processing facilities discussed in section 6.3), or the domestic lounges of
the airlines that have leases over their domestic terminals.

Lounges are not necessary to airlines or their passengers in the way that aircraft
movement and some passenger processing facilities are. Nonetheless, they are part
of the service offered by many airlines. The degree of market power that airports
hold in the provision of these services depends largely on the extent to which
airlines have discretion over the quantity and quality of lounge space that they use.

The behaviour of new entrants in the domestic market appears to show that lounges
are largely discretionary. Hobart Airport noted that:
… Impulse and Virgin are providing evidence that lounges aren’t necessary in the
scheme of things so I’d argue that we don’t have market power in those areas.
(trans., p. 328)

BARA (sub. DR54) argued that the degree of discretion over lounges is not the
relevant question. Rather, the issue is whether there are substitutes that constrain
airport pricing for that service. As discussed in section 6.1, airport pricing can be
constrained by a range of potential alternatives, including the availability of off-
airport and on-airport substitutes or suppliers, and/or the intensity of demand
preference for that service (that is, the degree to which that service is required by
users).

152 PRICE REGULATION

OF AIRPORT
SERVICES
Ansett, Qantas Airways and BARA also argued that airlines could not compete
effectively without lounge space, since their customers demand it of them. BARA,
for example, noted:
… lounge facilities are … demanded by airline customers and hence it is felt necessary
to supply those services to the customers to meet that demand. It’s only viable to have
an airport lounge and to provide those services if that lounge is in fact in close
proximity to the actual gate the passengers are going through. (trans., p. 223)

Given that airports have control over access to areas near the gates, it was argued
that they have market power in the provision of this space, enabling them to extract
monopoly rents from airlines (BARA, trans., pp. 223–4).

Alternatively, the apparent strategic requirement for airlines to have lounge space
could be used by airports as a bargaining tool in trying to resolve other issues with a
particular airline. BARA argued:
… the airport may deny or artificially delay access to space in a terminal required to
provide an airport lounge in order to achieve a commercial advantage in other areas
(for example, to resolve a dispute with the airport user). (sub. 26, p. 3)

Though this is possible (perhaps even likely), it is unlikely that this type of
bargaining strategy is the preserve of airports. In the past, for instance, airlines
apparently have refused to lease space pending resolution of disagreements
elsewhere.

Further, the price of those facilities may just reflect the opportunity cost of that
space for the airport — that is, rentals the airport could earn from any other
competitive activity, such as retailing. According to SACL, ‘airlines generally
prefer windows overlooking the airfield, and locations near departure gates’
(sub. DR62, attachment A, p. 1). Given this preference, and the fact that such space
is scarce, the opportunity cost may be high. Thus, prices for lounges may reflect a
locational (scarcity) rent. Moreover, since the space has a number of uses, it is
unlikely that airports would have an incentive to leave some of it unused
deliberately, except for future operational uses. The key issue, then, would appear to
be whether overall terminal space is constrained — that is, whether expansion is
delayed so that scarcity rents can be earned. According to Hastings Funds
Management:
To date Melbourne Airport has concluded a number of successful negotiations in
relation to international lounges … with no charges of abuse of market power.
(sub. 19, p. 5)

In sum, although certain customer segments may expect business and VIP lounge
space from their airline, and this space must be located at the airport, airlines appear

MARKET POWER IN 153

PARTICULAR
AIRPORT SERVICES
to have significant discretion as to the nature of the facilities, and are therefore
likely to be responsive to changes in the price of those facilities.

There currently is no prices oversight of business or VIP lounge facilities.

6.5 Landside vehicle facilities

There has been some debate since privatisation about the composition of landside
facilities and the extent to which these are covered by regulation.

For present purposes, landside vehicle facilities can be seen as falling into two
broad categories: facilities (such as landside roads) that provide vehicle access to
the airport and its terminals; and vehicle services provided landside (such as car-
parking services and taxi waiting areas).

Vehicle access facilities

Airline passengers and other airport users (including staff) obviously require access
to the airport and its terminals. And, ultimately, only the airport operator can
provide such access.

Airline passengers choose the means by which they obtain this access, whether it be
in private vehicles, rental cars, taxis, buses, or trains. The operators of these arrival
modes, such as taxi drivers and rental car operators, have a choice about whether
they operate at the airport at all. However, the importance of airport-related
business to their overall business influences to a large extent the degree to which
airports have, and could exercise, market power in their dealings with them. In this
regard, the Australian Taxi Industry Association noted that airports are the largest
single market for the taxi industry (sub. DR61), while the Victorian Hire Car
Association (VHCA, sub. DR68) noted that airport transfers accounted for at least
80–90 per cent of the gross income of hire car operators.

SACL acknowledged that it has a degree of market power in relation to front-door

access but argued that there are costs in managing kerb-side facilities and that:
Airports around the world are increasingly realising that kerb side congestion can be a
significant problem, and efficient pricing structures can offer a means of managing this.
(sub. DR62, attachment A, p. 2)

Melbourne Airport also accepted that ‘it has a degree of market power with respect
to the kerb in front of the terminal complex’ (sub. 7, p. 72) and noted the

154 PRICE REGULATION

OF AIRPORT
SERVICES
importance of pricing to manage congestion (sub. DR66). It argued further that
charges on kerb users are justified because kerb users either:
• use other services provided by the airport, for which they should pay; or
• are in competition with other services (such as car rental and parking) for which
access is paid.

The mere fact that there are charges for competing services does not necessarily
justify charges for kerb use. Charges on competing services may reflect the exercise
of market power. Nonetheless, pricing access to ease congestion may be efficient,
assuming the airport operator does not constrain artificially the availability of kerb-
side roads.

That said, an airport operator may want to shift demand for vehicle facilities to
services provided more directly by the airport, such as car parking (as discussed
below).

Capital Airport Group (sub. DR75) denied that airports have this incentive. It
suggested that, on the contrary, it has tried to improve, not limit, the access of
transport modes to Canberra Airport. It pointed to its negotiations with the ACT
Government for the removal of Government restrictions on independent (non-
Government) bus operators and a second taxi cooperative servicing the airport.
Given this, it submitted that it:
… rejects the suggestion that it would attempt to limit access for land transport
operators under reasonable terms and conditions in order to prop up car parking
revenue. (sub. DR75, p. 6)

Nonetheless, to the extent that an increase in access charges for competing modes
shifts demand (and raises the airport’s overall revenue), there may be an incentive
for an airport operator to set excessive access charges. Thus, airports have, and may
exercise, market power in the provision of vehicle access facilities.

Car parking

Two forms of car parking are provided at airports: facilities for passengers or those
meeting passengers, and facilities for staff parking.

The ACCC (sub. 36) argued that all travellers require land access to the airport and,
although several arrival modes are available, airport market power in relation to
those alternatives gives airports market power in the provision of car parks. It also
noted that it had found evidence in its assessment of Sydney Airport’s pricing
proposal that car parking is a service ‘in which airport operators have a significant
degree of market power’ (sub. 36, p. 87).
MARKET POWER IN 155
PARTICULAR
AIRPORT SERVICES
However, other participants — such as BAC (sub. 8) and the Australian Council for
Infrastructure Development (sub. 28) — argued that airports do not have market
power in the provision of car-parking services. They cited competition from other
modes and off-airport facilities as effective constraints on any potential airport
market power. Melbourne Airport noted:
Less than 20 per cent of passengers using Melbourne Airport use car parks provided by
Melbourne Airport. In addition to the three public car parks provided by Melbourne
Airport, Qantas and Ansett provide valet parking and there are at least 6 off airport car
park operators. Thus in relation to car parks, airports have little market power …
(sub. 7, p. 72)

Examples of rates charged for on- and off-airport car parking at and near Melbourne
Airport are provided in box 6.1.

Box 6.1 Car parking at and near Melbourne Airport

There are several car parks at or near Melbourne Airport. Some off-airport operators
advertise that they provide short- and long-term parking. The following are indicative
rates at car parks on- and off-airport (obtained from the websites of the operators).
• Parking rates at Melbourne Airport’s short-term car park (mostly undercover) start at
$5 for the first hour, and $9 for 1–2 hours. The hourly rate falls as cars are parked for
longer periods — for instance, parking for 9–12 hours costs $22 and for 12–24 hours
costs $28.
• Parking rates at Melbourne Airport’s (not undercover) long-term car park are $15 for
one day, $20 for 1–2 days, $50 for one week, and $98 for two weeks. The third week
costs an additional $8. The long-term car park is within walking distance of the
terminal, though regular shuttle bus services also are provided. Melbourne Airport also
offers undercover ‘business parking’ near the Qantas terminal that costs $25 per day.
• Some off-airport operators provide undercover or outdoor parking. They provide
transport (usually shuttle buses) to the airport terminal (generally a few minutes away)
and 24-hour security. They also tend to provide other services, such as car detailing,
at additional cost. Rates and rate structures differ across providers. There are at least
six providers of off-airport car parking near Melbourne Airport. Examples of rates at
two of these providers are given here.
– One operator has 350 undercover and 450 outdoor spaces. The first day of parking
costs $10 (outdoor) or $14 (undercover). This increases to $16 (outdoor) and $21
(undercover) for two days, $48 (outdoor) and $61 (undercover) for one week, and
$84 (outdoor) and $106 (undercover) for two weeks. Additional days cost $4.50 per
day (outdoor) and $5.50 (undercover).
– Another operator provides undercover parking at $9.90 per day for the first five
days, $8.80 per day for days 6–9 and $5.50 per day thereafter. Thus, the cost of
parking would be $19.80 for two days, $67.10 for a week, and $112.20 for a
fortnight.

156 PRICE REGULATION

OF AIRPORT
SERVICES
Substitution possibilities for airport car parking appear to be most significant for
long-term parking. In some locations, a number of off-airport operators exist, and
provide competition for on-airport facilities. The number of off-airport operators
appears to be greatest for the larger core-regulated airports (particularly Sydney and
Melbourne). This competition is possible because establishing car parks does not
appear to require large and lumpy investments. The major investment appears to be
in land, which is not sunk. In addition, travellers who potentially would use long-
term parking are those most likely to find other travel modes, such as taxis, an
attractive and cost-effective alternative.

The direct supply-side substitution possibilities for short-term car parking appear to
be more limited because proximity to terminals is likely to be more important for
those using short-term than for those using long-term parking. Nonetheless, other
providers may emerge — for instance, Delta operated short-term parking near
Melbourne Airport (chapter 9), and some other off-airport operators advertise that
they provide short- and long-term parking (box 6.1). Further, rates for short-term
parking will be constrained to some degree by those set in the more directly
competitive long-term parking market. The influence of long-term rates on short-
term rates is likely to be strongest for the longer periods in the short-term car park.
Nonetheless, there also is likely to be some (if more limited) flow-through to shorter
stays.

In addition, unlike long-term parking, many of the users of short-term parking are
not airline passengers, but those meeting passengers or driving them to the airport.
They have some options about whether to use airport car parks — dropping people
off without parking, or arranging to pick them up outside terminals, for instance.
The latter has become more viable with the increasing use of mobile phones,
allowing passengers to call ‘greeters’ to pick them up as soon as planes land.
Melbourne and Perth airports, for example, note on their websites the possibility of
setting down and picking up passengers kerb-side. Airports tend to impose
conditions on this, such as limiting the time allowed, or stating that cars must be
attended at all times (for security as well as operational reasons). In addition, Perth
Airport has free five-minute parking in its car park, which would seem to be
designed to encourage the dropping off and setting down of passengers.

Nonetheless, to the extent that airports have market power in car parking, it is likely
to be higher in short-term parking. Indeed, a study of parking at San Francisco
Airport (Kanafani and Lan 1988) found that demand for short-term parking was
inelastic, with elasticities higher for long-term parking. Although the authors noted
that these results may not hold generally, the study suggests that the market power
of airports may be higher in short-term than long-term parking.

MARKET POWER IN 157

PARTICULAR
AIRPORT SERVICES
Staff may have less discretion than travellers in terms of how they reach the airport
and where they park. Taxis are unlikely to be a viable option for commuting so,
unless other public transport options are available, most staff are likely to need to
drive to (and hence park at) the airport. Thus, it would appear that airport market
power in car parking is likely to be relatively more significant in staff car parking.

Has the behaviour of airports indicated or reflected market power in car parking?

Participants cited high airport car-park prices and profits as potential indicators of
the exercise of market power — for example, the ability of Sydney Airport to levy
and sustain charges in excess of potential competitors (ACCC, sub. 36). However, it
is not clear that the Sydney Airport charges cited by the ACCC — around $13 per
day for long-term car parking, compared with off-airport charges of $10 per day —
reflect market power. It is more likely that the charges largely reflect the greater
convenience of parking closer to the airport. In other words, they may reflect
locational more than monopoly rents — that is, the value placed on the scarce
resource by consumers, not scarcity created by the airport operator (appendix C).
Further, it appears that providers of long-term car parking offer somewhat
differentiated products (in terms of rate structures and services offered (box 6.1)),
complicating comparisons of on- and off-airport car parking rates.

At a more general level, high short-term parking rates sometimes are cited as an
indication of abuse of market power. As noted above, airports are likely to have
more market power in short-term parking than in long-term parking. Thus, these
parking rates could partially reflect the exercise of market power.

However, there also are likely to be locational rents associated with the provision of
space close to terminals. In addition, the opportunity costs of having to maintain
space that may only be used for intermittent periods may be relatively high.
Although this car space may be used later in the day, there may be periods when it
is not used, and this opportunity cost would be reflected in the price. This may
explain, to an extent, the relatively high hourly rates in short-term parking, not only
at the airport but also at most CBD car parks.

Further, as noted above, short-term rates may be constrained to an extent by long-

term car parking rates. At Melbourne Airport, for example, the hourly rate falls
significantly for those parking for 9–24 hours (box 6.1). Though the cost of parking
for a day in the short-term car park is still considerably higher than the long-term
full day rate, this partially could reflect the value of proximity to terminals. People
also have a choice at Melbourne Airport of paying $28 (short term), $25 (business)
or $15 (long term) to park for a day.

158 PRICE REGULATION

OF AIRPORT
SERVICES
SACL argued that airports have incentives to price car parking so as to encourage
people to stay in the airport longer and, hence, use retail outlets:
… airports also have a strong commercial incentive to offer car park customers value
for money, as this enhances their dwell time and propensity to consume other airport
services. (sub. DR62, attachment A, p. 2)

The extent to which this effect operates in practice is unclear.

Some airports (such as Sydney and Melbourne) have indicated that their car-parking
rates are based on benchmarks, such as CBD parking rates. Some participants
suggested that using such benchmarks is indicative of the use of market power.
Professor Forsyth argued (and the ACCC (sub. 36) agreed):
If rents are true locational rents, the seller does not set the price with reference to
benchmarks (of questionable relevance); rather, the seller takes what the market offers.
(sub. 5, pp. 30–1)

However, the use of benchmarks does not, of itself, indicate an abuse of market
power. First, it is unlikely that a true monopolist would need to use a benchmark at
all. Second, in practice, the discovery of the ‘market price’ is an iterative process
that may be assisted by the use of a benchmark.

One crucial issue is whether airport operators have restricted investment in car
parking to increase prices. Since privatisation, several major airports have
undertaken and/or plan major car-park expansion. Melbourne Airport, for instance,
plans to expand car parks during 2001-02 (chapter 2). To the extent that significant
expansion of car-parking space occurs, this would not appear to be consistent with
monopoly pricing behaviour, unless demand were expanding more rapidly.2
Although expansion could coexist with ‘monopoly pricing’, this does not appear to
have been the case.

BARA (subs 26 and 41) noted that staff car-parking charges at Sydney Airport
recently increased significantly. BARA questioned whether these increases were
justified though it acknowledged that, of itself, an increase of such proportions does
not necessarily represent an abuse of market power. Prices could be expected to
reflect the opportunity cost of the land, as well as operating costs. It may be that
starting prices were inefficiently low.
• Some of the former staff car-parking area was converted to aprons, suggesting
that the opportunity cost of the space was high and, hence, that initial prices
might have been low.

2 As noted above, car parks do not appear to be natural monopolies (the costs of establishing car
parks appear to be modest). Therefore, it is unlikely that an airport operator would increase
capacity strategically to stymie competition from other potential providers.
MARKET POWER IN 159
PARTICULAR
AIRPORT SERVICES
• SACL argued that even the new prices do not allow full cost recovery (SACL,
pers. comm., 25 July 2001). If this were the case, this would suggest that the
prices do not reflect an abuse of market power. The ACCC does not appear to
have raised concerns about the price increases under its monitoring function.

Further, the market power of Sydney Airport in staff car parking may be reduced by
off-airport parking that is of comparable convenience to the SACL facility (SACL,
pers. comm., 25 July 2001). In addition, at the time of the price increase, the train
link to Sydney Airport might have provided a potential substitute for driving to, and
parking at, the airport for workers.

Some participants also suggested that airport market power in car parking could be
exercised by limiting access to the airport by companies that provide competing
services. As BARA commented:
… although an airport may lease the management of on-airport car parking facilities,
the airport nevertheless has an incentive to deny or limit access to the airport of off-
airport parking companies (such as valet parking companies). Such a strategy will
increase the demand for the on-airport parking services and increase the value of the
car park lease. (sub. 26, p. 3)

As noted above, because the airport operator has sole control over access to the
airport, there may be an issue relating to access by competing modes.

Changes in car-park charges since October 2001

On 5 October 2001, the Commonwealth Government announced changes to prices

regulation for core-regulated airports (chapter 3). Following this announcement,
some airports changed their car-parking rates.
• Darwin Airport increased parking charges from $2.40 per day to $4 per day. The
charge is invariant to time spent in the car park.
• Brisbane and Perth airports restructured their charges.
- At Brisbane, international and domestic short-term parking rates for up to
four hours were either left unchanged or fell by between 9 and 14.3 per cent.
Long-term rates increased by between 7.7 and 10.9 per cent in the domestic
terminal, and by 12.5 per cent in the international terminal.
- At Perth, short-term parking rates increased by between 3 and 5.5 per cent.
For stays over six hours, rates fell by about 3 per cent. Long-term parking
rates fell significantly, from about 3 per cent for shorter stays to over
60 per cent for stays over two weeks. Although long-term parking rates for
the first three days fell only slightly (from $16.50 to $16 per day), charges
after the third day fell from $11 to $2 per day at the domestic terminal. (A

160 PRICE REGULATION

OF AIRPORT
SERVICES
 ‘special rate’ applied at the international terminal, where the $2 daily rate was
charged after the second day.)

The changes to price regulation did not alter the regulatory treatment of car parking,
which remained monitored, for these airports. This does not mean necessarily that
the price changes did not reflect the exercise of market power, simply that changes
to price regulation did not alter the ability of airport operators to exercise it. Costs
also are unlikely to have changed immediately after the regulation was revised, so
cost changes would not appear to have driven the changes in rates.

However, the circumstances of airports changed in September 2001 (chapter 2),

which may have given rise to the revised parking charges. Ian Kew, the Chief
Executive Officer of Darwin International Airport, for example, argued that the
events of September 2001 had required a review of operations and charges
(Kew 2001).

Whether these changes reflect an abuse of market power is unclear.

• At Brisbane, charges for short-term parking fell, despite the fact that airport
operators are most likely to have market power in this area.
• Where charges have increased, ultimately (as noted in chapter 5), the long-term
profitability of the change is crucial to determining the degree of market power.
This will depend in part on the demand response. On this point, it was
commented in relation to Darwin Airport ‘that people will now just drop off and
pick up passengers and thereby avoid the exorbitant charges’ (Taxis 2001).

The significant decrease in some long-term parking rates at Perth Airport generally
would not be seen as an abuse of market power. Very low prices — if they were
significantly below the costs of providing the service — could sometimes be seen as
predatory pricing. Testing for predatory pricing is not easy. Although a number of
rules have been suggested in the literature, they can be difficult to apply and there
does not appear to be one test that is applicable in all cases (Ordover and Saloner
1989). Nonetheless, in general:
Markets in which predatory conduct is rational are characterized by imperfect and
asymmetric information, scale economies, intertemporal and intermarket cost and
revenue linkages, barriers to entry and reentry, etc. (Ordover and Saloner 1989, p. 591)

These conditions do not appear to characterise the market for airport car parking.

The nature of the restructure at Perth may suggest a form of price discrimination.
Where market power is likely to be more significant — short-term parking — rates
increased. Rates fell where competition is likely to have the most significant effect
— long-term parking and long periods in short-term parking.

MARKET POWER IN 161

PARTICULAR
AIRPORT SERVICES
What can we conclude about market power in car parking?

Overall, airports appear unlikely to have significant market power in long-term car
parking. The market power of airports in car parking is likely to be higher for short-
term and possibly staff parking, but there also are factors mitigating the extent of
market power in these facilities. For example, because many of the users of short-
term car parks tend not to be passengers, alternatives to using the car park are
available.

To the extent that airports have market power in car parking, it is likely to be
constrained as long as landside access for competing operators (of other travel
modes, such as taxis, and competing off-site parking services) is available on
reasonable terms and conditions.

Taxi facilities and charges

Airports provide a number of facilities for taxi services — such as parking areas for
waiting taxis and, in some cases, queue management systems (although taxi
companies operate these at some airports). Airports also provide areas for taxi ranks
for arriving and departing passengers.

Travellers and taxi drivers have some discretion over their use of these facilities.
Passengers can choose to travel to and from the airport using modes other than
taxis. Taxi drivers who transport passengers to the airport need to use designated
sites. However, they have a choice about whether they pick up passengers from the
airport. Anecdotal evidence suggests that many drivers who take people to an
airport then move on to other nearby areas rather than wait to collect another airline
passenger. In this case, they do not need to use queue management or taxi parking
services.

Nonetheless, to the extent that taxis and passengers use taxi services, there are no
on-site alternatives to the airport-provided (or approved) queue management
system. Viable off-site alternatives also appear limited, although it may be possible
to provide holding bays adjacent to, but off, the airport site. In addition, as already
noted, there may be issues as to whether an airport may wish to use its control over
landside access to shift demand to airport-provided car parks etc. There may,
therefore, be some access issues related to the provision and pricing of these
facilities.

Since privatisation, four core-regulated airports have levied some form of taxi
charge. Of these, only Melbourne linked the charge to a necessary new investment
proposal (submitted after the Federal Court ruled ground access fees at Canberra

162 PRICE REGULATION

OF AIRPORT
SERVICES
were charges for landside roads), although Westralia Airports Corporation (WAC)
pointed out that it also provided ‘substantial facility enhancements to cater for the
expressed needs of the taxi industry’ (sub. 21, p. 23).

Much of the debate over the charges has surrounded the issue of whether they
should be included in the price cap. This rests primarily on the question of whether
they should be considered charges for access to ‘landside roads’. Discussion of this
issue is contained in appendix E. In this section, consideration is given to the issue
of the extent to which the levies charged may reflect an abuse of market power.

Because the precise nature of the levies has differed across airports, it is difficult to
make a general judgement about whether their imposition reflects an abuse of
market power. Nonetheless, some general issues can be noted.

As discussed above, airports have control over the provision of landside access. To
the extent that the charges can be considered access charges, they are charges for
services for which airports have market power.

The Federal Court decided that the charges imposed by Capital Airport Group were
charges for access to landside roads.3 However, this decision was made in the
specific context of charges imposed by Canberra Airport, and the definitions
contained in the regulation of the time.

The issue to be addressed here is a somewhat different one. Specifically, do airports

have significant market power in the services or facilities for which taxi charges are
imposed? From this point of view, some taxi charges might not be considered
charges for landside access (in which airports do have market power).

Airports have argued that taxi charges should be seen as charges for facilities
provided, or concession charges such as those for retail space. WAC, for instance,
argued that:
… the ground facilities fee is not a levy on taxis or any other commercial vehicles for
access to the airport. The fee puts in place an equitable system of charges at the airport
for all businesses that derive revenue from the airport market, and is consistent with
charges levied on car rental operators that are in place and not regulated. Ground
facilities fees are commonplace at airports elsewhere in the world, reflecting the
principle that those who benefit from the airport’s growth, should contribute to its
revenue base. (sub. 21, p. 23)

3 The Federal Court found in a ‘line ball’ decision that the ground access charges at Canberra
related to landside roads. An appeal by the Capital Airport Group subsequently was dismissed
(appendix E).
MARKET POWER IN 163
PARTICULAR
AIRPORT SERVICES
That airports in other countries impose these fees does not, of itself, mean that their
imposition is not an abuse of market power. There are other reasons, however, that
may suggest that these charges are not entirely charges for landside access.

The charges tend not to be levied on taxis setting down passengers, so they are not
levied for all taxi access to the airport.

In addition, costs are incurred by airports in the provision of taxi facilities.

According to Melbourne Airport, for example, it incurs about $750 000 per year in
labour costs to ‘call up taxis and push them through’ (trans., p. 171). There also are
opportunity costs in the provision of substantial areas for taxi holding bays — for
example, the new facilities at Melbourne Airport will provide 790 car spaces — as
well as infrastructure costs.

Thus, charging for these services does not, of itself, reflect an exercise of market
power. Indeed, Melbourne Airport contended that it was subsidising taxi services by
not imposing charges:
We deliberately took the decision, unlike many other airports, not to impose a taxi
charge; to keep on wearing that subsidy ourselves for as long as we possibly could.
(trans., p. 170)

Given that both parties benefit in some way from the taxi management system, it
may be optimal for some sort of cost sharing to occur. What is an optimal
proportion of costs to be shared by each party may change over time.

Hence, even if there is no increase in costs (for example, due to new investment),
imposing a new charge, or increasing an old charge, does not reflect necessarily an
abuse of market power.

Overall, off-site and on-site alternatives to airport-provided taxi parking facilities

are limited but passengers and taxi drivers have some discretion in their use of these
facilities. Passengers can use other modes, and taxis do not use the parking facilities
if they only set down passengers.

The ACCC (sub. DR55) argued that, for passengers leaving the airport, access to
the kerb-side is, of itself, insufficient since taxi parking/waiting areas are required
for departing passengers. However, this is not necessarily the case at all airports. At
Canberra Airport, for example, if there are no taxis in the feeder ranks, a taxi that
has set down a passenger can pick up others directly at the rank, without incurring a
charge (Capital Airport Group, sub. DR75, appendix 5). Though not currently used
as an option, it also may be possible to have taxi waiting areas outside the airport,
with passengers able to call for a taxi when their plane arrives.

164 PRICE REGULATION

OF AIRPORT
SERVICES
On balance, the market power of airports in providing these facilities appears
moderate. The ability of airports to impose charges above efficient levels appears to
be limited if access for competing modes is provided on reasonable terms and
conditions.

Other landside access issues

The Victorian Hire Car Association (VHCA, sub. DR68) suggested that Melbourne
Airport has significant market power in the provision of access to its members. It
cited in particular its experience in relation to the imposition of the annual ‘Vehicle
Driver Authority’ fee and charges for hire car parking (and the way in which these
were implemented).

It also noted changes to the configuration of feeder ranks (which took place under
the Federal Airports Corporation (FAC) in 1992) as a demonstration of:
… Melbourne Airport’s ability to effectively control the nature of the hire car
operators’ physical access. (sub. DR68, p. 5)

The VHCA added that the airport operator reduced physical access for its members,
and that there is a:
… strong incentive for Melbourne Airport, in the absence of restraints, to increase
charges and promote more revenue positive suppliers to service this market.
(sub. DR68, p. 11)

As already noted, airports have market power in relation to front-door access.

However, it seems reasonable that the airport will want to regulate the way in which
access to its terminals occurs, for operational and safety reasons. Otherwise, it is
likely that kerb-side congestion could be a problem, even where space is not
constrained artificially by the operator.

The importance to hire car operators of revenue from the airport may make them
less able to oppose price increases. However, the fact that they are operating on
small margins and the apparent price sensitivity of their customers (VHCA,
sub. DR68) also mean the operators will be very sensitive to price increases. The
VHCA itself acknowledged that ‘hire car operators are … sensitive to minor cost
changes’ (sub. DR68, p. 8). This could diminish the market power of airports. That
the VHCA was successful in modifying Melbourne Airport’s original fee proposal
to one more palatable to its members appears to illustrate this.

MARKET POWER IN 165

PARTICULAR
AIRPORT SERVICES
6.6 Aircraft refuelling
Aircraft refuelling facilities at larger core-regulated airports are built and operated
by oil companies as a joint user hydrant installation (JUHI) on land provided by the
airport operator.4 Thus, airport operators directly provide only one element of the
refuelling service — the land. It is the lease of airport land to oil companies that
raises issues of market power.

Aircraft obviously need fuel but airlines do not necessarily need to refuel each time
they land. For larger jets on domestic short-haul flights, airlines may have some
discretion over where they refuel. The ACCC noted that ‘to minimise airlines’
running costs, aircraft generally do not carry more fuel than necessary. As such,
aircraft often refuel upon each landing’ (sub. 36, p. 82). Nonetheless, some
discretion remains.

Over longer distances, including long-haul international flights, on the other hand,
refuelling must take place at each landing. In this case, discretion over refuelling at
a location is limited to the decision of whether to fly to a particular airport at all.

Therefore, the market power an airport holds in these facilities depends on whether
there are off-airport alternatives and the degree to which refuelling can take place at
other airports can be substituted for refuelling.

For larger airports, there appear to be limited off-airport alternatives for refuelling.
• Trucking fuel from off-site to refuel aircraft at large, and often busy, airports
will be considerably more costly to airlines.
• Safety requirements may preclude trucking as a practical option.
• The alternative of aircraft taxiing to the airport boundary to refuel from an off-
site facility would not be an economic or practical option. As the ACCC noted,
‘This effectively limits the extent to which potential entrants can compete
against the airport for the provision of refuelling services’ (sub. 36, p. 82).

The viability of airport substitution for refuelling varies according to the airport and
route being considered. If, for example, an aircraft operates Sydney–Melbourne–
Canberra–Sydney, it may be possible to choose not to refuel at one of those airports.
On the other hand, for aircraft flying between Perth and the eastern seaboard, it is

4 At smaller airports, where the volume of fuel throughput is lower, arrangements may differ. At
Hobart, for example, the trucking of fuel to the aircraft is the means by which refuelling takes
place. However, the airports where this occurs are those assessed in chapter 5 as not having
market power, so this method of refuelling does not apply to the airports assessed in this
chapter.
166 PRICE REGULATION
OF AIRPORT
SERVICES
uneconomic not to refuel at each port. Likewise, for international flights, aircraft
must be refuelled at each landing. Qantas, in its submission to the ACCC review of
fuel throughput levies, noted in respect to Brisbane and Perth airports:
The only alternative to purchasing fuel from those oil companies is to bring fuel on to
the airport by land transport or refuel at other airports and carry the additional fuel to
Brisbane or Perth. Either alternative is uneconomic and would increase the cost of fuel
to Qantas by 1 to 2 cents per litre and the cost of flying to those ports even more.
(sub. to ACCC 1998b, p. 6)

The greater the proportion of long-haul flights arriving at or departing an airport,

the higher is the degree of market power that it is likely to have in refuelling. Given
this, the isolation of Perth is likely to grant it a degree of market power in refuelling
that is equal to its market power in landing. Refuelling in this case is part of the
non-discretionary bundle of airport services. In terms of Brisbane Airport, although
its most significant domestic traffic is generated from the relatively short Brisbane–
Sydney route, a high proportion of its traffic also is generated on longer routes, such
as Brisbane–Melbourne and Brisbane–Cairns (chapter 2). Therefore, it also may
have a fairly high degree of market power in the provision of refuelling services.

Sydney Airport services a significant proportion of short-haul traffic, which would

tend to lower its market power in refuelling. However, its status as a quasi-hub (for
domestic and international traffic) may offset this somewhat. On balance, it appears
that Sydney Airport has a high degree of market power in refuelling, though this
does not appear to be as high as that for Perth Airport.

The other airports may have a less significant degree of market power in the
provision of domestic refuelling facilities. For Melbourne Airport, the proportion of
its traffic generated from shorter routes, particularly Sydney–Melbourne, is far more
significant than from longer routes such as Melbourne–Perth and Melbourne–
Brisbane (chapter 2). Darwin would have a relatively high proportion of longer haul
traffic, so is likely to have market power in refuelling equal to its overall market
power. Apart from having a lower degree of market power overall (chapter 5), most
traffic through Canberra Airport is generated from short-haul routes, further
diminishing its degree of market power in refuelling. Adelaide also is likely to have
a lower degree of market power in refuelling, given its lower market power overall.

Although aircraft on international flights need to refuel in Australia, the degree of

market power of any one airport in refuelling is mitigated to an extent by the
potential for competition in attracting international services (chapter 5).

That said, airports that have market power are likely to have at least a moderate
degree of market power in the provision of aircraft refuelling services. Indeed,
Melbourne Airport noted that ‘a case can be made that there is market power with

MARKET POWER IN 167

PARTICULAR
AIRPORT SERVICES
respect to refuelling services’ (sub. 7, p. 36). The extent of market power is likely to
be highest for Perth and, to a lesser degree, Brisbane and Sydney airports.

Are fuel throughput levies evidence of abuse of market power?

At larger airports, oil companies pay licence fees and rent to the airport operator for
use of the land. A fuel throughput levy has been imposed, in addition to the land
rental, at Brisbane and Perth airports.5 The imposition of the fuel throughput levies
has generated a significant amount of debate, particularly about the extent to which
they reflect an abuse of market power and hence should be considered
‘aeronautical’. The International Air Transport Association noted that:
… airlines recognise, and accept, that charges and/or fees may be levied for the fuel
services required and used to support their operations, but believe that such activities
should be treated as ‘aeronautical’ in nature. (sub. DR56, p. 9)

The issues are outlined briefly in this section and in more detail in appendix E.

Of itself, the restructuring of the charge is not evidence of an abuse of market

power. A two-part pricing structure may be a more efficient way of pricing the
service. It also may indicate a change in attitude to risk by airport operators. Some
airport operators commented that it was reasonable that the risks be shared.
Northern Territory Airports explained:
Rather than introduce higher base lease and licence fees, airports are attempting to
share risks by making their own revenues contingent upon fuel sales volumes. (sub. to
ACCC 1998b, p. 2)

On the other hand, Shell Australia commented that:

Shell would be prepared to pay increased fuel throughput levies, if refuelling service
charges fell by a similar amount, as this would reflect usage. However, increasing both
refuelling service charges and applying fuel throughput charges is an example of abuse
of market power. (sub. DR63, p. 2)

The airports involved contended that the fact that they were simply exercising a
contractual right to impose the levy meant that they had not abused market power.
BAC, for example, argued:
BACL strongly refutes any suggestion that it has abused its market power because it
‘activated’ a fee covered by an existing contractual arrangement and determined as
reasonable by an independent expert. (sub. to ACCC 1998b, p. 5)

5 The levy at Perth Airport is imposed on the international terminal only.

168 PRICE REGULATION
OF AIRPORT
SERVICES
However, even if they were exercising a contractual right, this does not mean
necessarily that the imposition of the charges did not reflect an abuse of market
power. In addition, even if the imposition of a charge of that type would not have
been an abuse of market power, the level of the charge imposed might have been.

Neither BAC nor WAC, in their submissions to the ACCC, attempted to justify the
fuel throughput levies on the grounds of increased costs, the provision of additional
services or offsetting reductions in other charges. Thus, the ACCC concluded:
When considered in light of the lack of any cost related justification for the levies, or
offsetting reduction in charges, there is a strong case that the imposition of a fuel
throughput levy is taking advantage of market power. (ACCC 1998b, p. 7)

If the imposition of the levies had been justified on cost grounds, claims of abuse of
market power would have been more difficult to substantiate. Nonetheless, a lack of
cost justification does not, of itself, indicate an abuse of market power, either. It
may be that the previous charges were at inefficiently low levels, in which case,
there did not need to be an increase in costs to justify an increase in charges.

The critical issue, then, relates to whether the charges — of 0.4 and 0.5 cents per
litre for Brisbane and Perth respectively — were above efficient levels. On the basis
of the evidence provided to the ACCC review, it is difficult to ascertain whether the
charges exceeded an efficient level. The levies resulted in an increase in total
revenues derived from refuelling services, but the review presented no information
on expected normal rates of return from refuelling facilities. BAC contended that
the charge was a ‘reasonable market rate’ with reference to fuel throughput levies at
airports in other countries. This was supported by the determination of the
‘independent expert’, a Queen’s Counsel jointly appointed by the parties (BAC,
sub. to ACCC 1998b). Of course, the fact that fuel throughput levies are charged in
other countries may simply reflect the exercise of market power by airports in those
countries.

In its submission to the ACCC review (quoted above), Qantas implied that it would
not be economic to switch to refuelling at airports other than Brisbane and Perth
unless the levy increased to above 1–2 cents per litre. A monopolist maximising
profits may, therefore, be expected to charge a fuel throughput levy of up to 1–2
cents per litre without losing business. The fact that airport operators are charging
considerably less than this amount could indicate that there are economic, political
and/or legal constraints on their ability to maximise profits by abusing market
power, or that a more gradual approach to changing charges is preferred given the
sensitivity of the issue.

On balance, the extent to which airport operators have abused their market power is
unclear. Contractual rights are not evidence one way or the other, and the lack of
MARKET POWER IN 169
PARTICULAR
AIRPORT SERVICES
cost justification for imposing fuel throughput levies does not necessarily indicate
an abuse of market power. Moreover, the extent to which the levies are in lieu of the
rent component and/or exceed, if at all, an efficient price is not known. That said,
because these airports do have fairly high degrees of market power in the provision
of refuelling, the potential for abuse remains.

6.7 Maintenance facilities

Two types of maintenance are performed on aircraft: light or emergency
maintenance, and heavy maintenance.

Aircraft light/emergency maintenance sites

Light and emergency aircraft maintenance involves unplanned maintenance and is

essential to the safety of airline operations. Facilities to carry out these maintenance
tasks must be provided at all airports. According to the ACCC, ‘for safety and
operational reasons the major airlines would be unlikely to fly to airports that did
not have light or emergency maintenance facilities’ (sub. 36, p. 82).

In addition, it appears that off-site provision of these services generally is infeasible.

Again, the ACCC noted:
As aircraft often stop at airports for only a brief period of time, such repairs need to be
done on site. The time wasted, and costs associated with, moving the aircraft to an off-
site location for such unplanned maintenance could lead to aircraft delays and
significant additional costs to airlines. (sub. 36, p. 82)

Nonetheless, the Department of Transport and Regional Services (sub. 39) argued
that airport market power in the provision of light maintenance is low since airlines
and third parties, rather than the airport itself, provide the services on a contract or
fee for service basis.

If this is the case, any market power of airports would relate to the provision of
access to the third parties who directly provide the service. Qantas submitted that:
In order to provide these services, maintenance workers require access to land within
the airport, and aircraft parking space on which to conduct the maintenance. The airport
is a monopoly provider of these access rights and aircraft parking locations.
(sub. 48, p. 16)

SACL, on the other hand, argued that airports have an:

… incentive to provide access on reasonable terms for such facilities, to ensure the
efficient operation of the airport. (sub. DR62, attachment A, p. 2)

170 PRICE REGULATION

OF AIRPORT
SERVICES
Overall, airports with market power are likely to have some market power in the
provision of aircraft light/emergency maintenance sites, though this predominantly
is an issue of access to the site to enable third parties to provide the service. It does
not appear that access has been an issue so far.

Aircraft heavy maintenance facilities

Aircraft heavy maintenance involves periodic and planned maintenance, such as

engine overhauls or refurbishing aircraft. Heavy maintenance is essential to the
(safe) operation of airlines.

Although heavy maintenance facilities are required by airlines, the evidence

suggests that airlines have a large degree of discretion about the location of these
facilities. Indeed, heavy maintenance facilities do not need to be located at every
airport because ‘aircraft owners can simply schedule the performance of heavy
maintenance when aircraft are at an airport that has the relevant facilities’ (ACCC,
sub. 36, p. 83). Melbourne Airport (sub. 7) suggested that the market for sites can
be local, national, or even global.

A number of participants — such as Melbourne Airport, WAC and SACL (subs 7,

21 and 27) — cited recent examples of airports competing for these facilities. These
included the location of the Qantas Boeing 767 maintenance facilities in Brisbane,
rather than Melbourne or Auckland. In this case, according to Melbourne Airport ‘it
[is] important to realise that Qantas took an active decision to relocate this activity
away from Sydney’ (sub. 7, p. 69). In the case of Perth Airport, WAC submitted:
… maintenance facilities for National Jet Services established on the airport several
years ago were competitively secured in the presence of competition from other
Australian airports. (sub. 21, p. 31)

It also is feasible for these facilities to be located off the airport site or at nearby
‘secondary’ airports. Melbourne Airport, for example, noted that Ansett had
‘significant maintenance facilities located just beyond the airport boundary’
(sub. 7, p. 68). And the ACCC commented:
It may be feasible for such facilities to be located near the airport site, as long as
aircraft operators have access on reasonable terms to a road or tarmac suitable for
moving aircraft from the airport to the off-site heavy maintenance facility. For
example, Avalon Airport has heavy maintenance facilities which service Qantas
aircraft. (sub. 36, p. 82)

Ansett noted that competition between locations for heavy maintenance facilities is
driven mainly by competition between governments, not airports:
In Ansett’s Australian experience, whilst airports may offer some incentives to base
facilities such as these at their airport, generally, the greatest source of competition and
MARKET POWER IN 171
PARTICULAR
AIRPORT SERVICES
 incentive packages is between State Governments (due to the employment and state
based investment generated etc) rather than the airports in question. (sub. 42, p. 17)

If anything, however, the importance of governments — and other factors outside

the control of airports — in the bidding process, may diminish further the market
power of individual airports. According to Melbourne Airport:
… individual airports have little market power especially as location decisions seem to
be driven largely by industrial relations issues, availability of skilled labour, and
government incentives. (sub. 7, p. 69)

However, once heavy maintenance facilities are established at a particular airport,

an airline that decides to relocate its facilities may incur sunk costs. As noted by
Qantas, for instance:
These [heavy maintenance] facilities are substantial, and represent sunk investments,
with specialised labour requirements. Qantas acknowledges that in respect of future
investment in heavy maintenance facilities, it is able to exercise a degree of choice
regarding the location of such future investment (subject to a variety of labour and
locational constraints). However, Qantas has little ability to relocate existing sunk
investment. In respect of such investments, Qantas is to a degree captive to the market
power of airport operators. (sub. 48, p. 16)

This, in the short term, may give the airport at which the facilities are located some
market power. Nonetheless, when negotiating the contract for heavy maintenance
facilities, airlines also could negotiate long-term pricing and service agreements for
other airport services to cover a period that would allow them to recoup heavy
maintenance establishment costs. This would reduce the market power of the
airport. Overall, it appears that the market power of individual airports for heavy
maintenance facilities is negligible.

6.8 Flight catering facilities

Flight catering facilities do not appear to be necessary for all air services. Although
they are necessary for long-haul international flights, on shorter flights (and smaller
aircraft — ACCC, sub. 36), extensive meals are not required.

Even where they are required, there appear to be viable supply-side substitution
possibilities. As the ACCC (sub. 36) noted, the required infrastructure does not have
high sunk costs, nor do all airports need to have flight catering facilities. Further,
meal preparation can occur off-site, with food transported to the airport. Melbourne
Airport noted that:
Sites for catering facilities may also be provided by the airport but in the case of Gate
Gourmet, the largest independent caterer operating at Melbourne Airport, their kitchens
are located at a location near to but off the airport. (sub. 7, p. 68)
172 PRICE REGULATION
OF AIRPORT
SERVICES
In this case, the market power of airports appears to be negligible, as long as access
to the airport is provided.

The ACCC suggested that access for food transport has been an issue in one case
since privatisation. It submitted that:
In 1999 the Hyatt Hotel complained to the ACCC about Canberra Airport’s imposition
of an increase in their licence fee of some 800 per cent to transport food prepared in an
off-site facility to Qantas aircraft. The charges are now substantial. (sub. DR55,
attachment D, p. 4)

The Capital Airport Group (sub. DR75), however, argued that this was not an issue
of airport market power. It noted that the structure of the fees changed at this time.
Initially, the charge, as levied by the FAC, was applied on a fixed fee plus
percentage of turnover basis. In 1999, as a result of negotiation between the airport
and the Hyatt, the percentage of turnover component was removed, and the fixed
component increased by 250 per cent. However, as a result of the change to the fee
structure, the total charges incurred by the Hyatt Hotel fell. The licence fee
increased again, when ownership of the caterer changed. This second rise meant
that the total increase in the fixed licence fee between 1999 and 2001-02 was about
700 per cent (Capital Airport Group, pers. comm., 26 November 2001). The total
charges for the new caterer, however, also appear to be lower than they would have
been under the former fee structure.

Although airport operators could limit access, it does not appear that access for
flight catering facilities has been an issue to date.

6.9 Freight handling and ground service equipment

storage sites
Freight handling equipment is used by terminal operators, handling agents and
airlines, and ground service equipment is used for pushing aircraft on the airport
apron. Participants noted that the issues relating to freight equipment storage sites
and ground service equipment storage sites are similar, so they are dealt with
together here.

Ground service and freight handling equipment are used frequently, and their
storage is essential to airlines. Storage sites require some area of pavement (with
flood lighting), and could include freight or passenger aprons or hard stands.

Determining the market power of airports for these storage sites requires an
assessment of the extent to which they can be located off-airport, and the extent to

MARKET POWER IN 173

PARTICULAR
AIRPORT SERVICES
which airports are likely to constrain the availability of land for these purposes so as
to increase rentals above opportunity cost.

In assessing the Australian Cargo Terminal Operators access case (chapter 9), the
National Competition Council (NCC 1997) argued that off-airport storage of freight
handling equipment is technically possible but commercially infeasible, given the
additional costs it would impose on users. Off-airport storage also would reduce
operator efficiency and flexibility.

On the other hand, in submissions to this inquiry, airports argued that they do not
exercise market power in storage sites, and that economic barriers to entry are not
significant. Melbourne Airport, for instance, noted that:
Terminal operators, handling agents and airlines store a good proportion of this
equipment on their leased premises. The remainder is stored in common ‘GSE’ areas
on or adjoining aircraft parking aprons free of charge … some users have approached
Melbourne Airport seeking to pay for dedicated sites … There is currently no abuse of
market power here and little likelihood as long as land for leased sites remains
available. (sub. 7, p. 73)

WAC argued that, though it charges for these sites, these charges comprise a small
portion of its aeronautical revenues, and furthermore:
The adequate supply of airside land at PIA [Perth International Airport] provides
incentives for WAC to encourage take-up of these facilities. (sub. 21, p. 31)

Overall, it appears that issues in relation to freight handling and ground service
equipment storage sites are predominantly access related. The ACCC (sub. DR55,
attachment D) suggested that airports may inhibit access to these facilities. It based
this on the Australian Competition Tribunal’s finding that:
In the present matter SACL does want to deny access, or at least regulate access,
because it appears to want to control and decide itself who shall operate ramp handling
services at the airport. (ACT 2000, para. 12)

It seems reasonable, however, that airport operators would want to regulate access
to their site, particularly for safety reasons. It does not appear that they currently
have an incentive to exercise market power per se in providing access to these
facilities.

6.10 Freight facility sites and buildings

Freight facilities are used for the loading, unloading and short-term storage of
freight. Though these facilities are necessary to airlines, there appear to be supply-
side substitution possibilities.

174 PRICE REGULATION

OF AIRPORT
SERVICES
Both Melbourne Airport (sub. 7) and WAC (sub. 21) noted that there is significant
competition between airports for freight facilities, and that airports have little
incentive to constrain land for these facilities. Specifically, Melbourne Airport
noted:
Abuse of market power is unlikely at airports where apron and land is available for
expanding cargo facilities because it will always be profitable for airports to expand
activity … If land is not available, the issue won’t arise … (sub. 7, p. 73)

In its assessment of the Australian Cargo Terminal Operators case, the National
Competition Council also concluded that it was economic to duplicate a site off-
airport, as long as there was vehicle access to the airport. Indeed, the ACCC noted
that off-airport duplication ‘freely occurs’ (sub. 36, p. 84).

Therefore, it appears that airport market power in the provision of freight facility
sites and buildings is negligible.

6.11 Waste disposal facilities

Aircraft that land at airports need to dispose of waste such as sewage, aircraft oil
and food. Although the service is necessary, it appears that substitution possibilities
are significant. As noted by the ACCC, many alternative facilities are available and:
If the airport operator provides the airlines (or contractors who provide waste disposal
services) with reasonable conditions of access to the airport site to enable the disposal
of waste, then off-site waste disposal facilities appear a viable alternative.
(sub. 36, p. 86)

Thus, airport market power in waste disposal facilities appears to be negligible.

Technically, as with the provision of other services that require access to the airport
site, the airport operator could exercise market power by restricting access to the
airport. The airport operator is unlikely to do so in the case of waste disposal
facilities (although it could award contracts for waste disposal to a monopoly
provider).

6.12 Administrative office space

Office space at airports is used by airlines to perform administrative functions and
provide facilities, such as lockers and change rooms, for their staff.

Participants (airlines, airports and the ACCC) agreed that airlines need some form
of office space at airports. There was, however, disagreement as to the discretion

MARKET POWER IN 175

PARTICULAR
AIRPORT SERVICES
airlines have over the amount of space they procure for offices and, therefore, the
amount of market power that airports have in providing this space.

For example, the ACCC commented:

With the growth of telecommunications and the development of information
technology services, there is a reduced need for administration to be situated in any
specific location. Indeed many airlines’ administrative functions are not located on
airport, but in off-airport offices. In some cases airlines are even undertaking
administrative functions overseas because of labour cost differentials. (sub. 36, p. 84)

On the other hand, Qantas argued:

… there is an issue about how big the offices are, but certainly we need offices …
there’s a necessity somewhere for … staff to go, and they’re sort of a minimum size,
given the number of staff that we have at the international airports. (trans., p. 263)

BARA also commented:

For major international airlines travelling to and from airports in Australia there are no
substitutes for office space within the international terminal. International airlines need
office space to keep spare supplies including tickets, boarding passes, headsets, etc. As
these spare supplies are often required at short notice, off-airport space is not an
alternative. (sub. DR54, p. 29)

There also has been disagreement over the extent to which office rentals at airports
reflect abuse of market power. Hastings Funds Management, for instance, argued
that office rents at Melbourne Airport have not been increased since privatisation
‘despite the capacity to do so’ (sub. 19, p. 5). In contrast, Virgin Blue argued that
‘APAM’s [Melbourne Airport’s] charges for … office space rental … are
substantially above commercial market rates’ (sub. 30, p. 17).

Even if office rentals appear significant, the question is whether these reflect
locational or monopoly rents. BARA (trans., p. 224) argued that they reflect
monopoly pricing, not just locational rents. Professor Forsyth noted that:
… if airports are setting rents for terminal space with reference to benchmarks such as
rents in the CBD, this is symptomatic of use of market power. (sub. 5, p. 30)

However, as noted above in relation to car parking, to the extent that benchmarking
is used, this does not imply necessarily that market power is being abused.

Melbourne Airport (sub. 7) commented that users compete for terminal space that
has alternative operational or commercial uses. On the other hand, Qantas noted:
… offices for airlines … tend to be in areas which by and large the alternate use is not
retail. They tend to be right up the top in the gods, rather than areas sitting out on the
finger where you would be competing with retail. (trans., p. 264)

176 PRICE REGULATION

OF AIRPORT
SERVICES
As discussed in section 6.4, an airport, in allocating a given terminal space to
various activities, is unlikely to have incentives to leave space unused (except for
potential future operational requirements). Space will be allocated to the most
profitable uses, with users paying the opportunity cost of the space. The key issue is
whether the airport operator constrains the overall amount of terminal space to earn
scarcity rents.

On balance, it appears that, although airlines require some office space, discretion
over the amount procured at airports is fairly high, and airport market power is
moderate at most.

6.13 Commercial and retail activities

Airports, particularly larger ones, provide space for various retail facilities,
including food and beverage outlets, shops (such as duty-free outlets), and
advertising.

Participants commented on the significant expansion of retail space that has

occurred since privatisation. This, it was argued, indicates that airports do not
exercise market power in the provision of retail space. On the other hand, Qantas
(sub. 48) argued that airports have significant market power in commercial and
retail activities, and earn above normal returns in these activities.

Several factors appear to constrain the market power of airports in the provision of
retail and commercial space. First, apart from some last-minute purchases, in most
cases, visitors to an airport do not have to make retail purchases there.

Second, when leasing space to businesses, airports compete with other providers of
retail space, such as shopping centres. With regard to commercial space outside the
terminal, for example, a discount store announced in October 2001 that it was
relocating its premises to Brisbane Airport. One of its directors noted that other
developers wanted to provide them with the space they needed, but that BAC
offered the ‘quality product they needed for the long-term expansion of their
operations’ (Sommerfeld 2001).

Competition in product markets would put a ceiling on the rent a tenant can afford
to pay, while remaining viable. Some airports have retail pricing policies to ensure
that prices charged at airports are not higher than at equivalent locations. The retail
pricing policy at Melbourne Airport, for example, specifies that tenants must
charge: no more than the lower of the manufacturer’s recommended retail price, or
the price charged by comparable retail outlets; or no more than the off-airport store,
if the concessionaire operates an outlet off-airport (sub. 7, appendix 4).

MARKET POWER IN 177

PARTICULAR
AIRPORT SERVICES
Finally, in the case of duty-free outlets, there are not only off-airport options for
purchase, but travellers can decide to make purchases at airports overseas, or on
flights. The ACCC (sub. DR55, attachment D) argued that competition for duty free
is limited by the requirement that off-airport operators remove dockets from
purchases at the airport. Its argument appears to be that, because off-airport outlets
require some access to the airport, the airport operator could limit or price such
access in a way that would limit the competitiveness of off-airport outlets. (The
ACCC also noted that the access fee for this increased significantly at one airport.
However, the Commission understands that this increase was the first in several
years. Moreover, the previous charge had been set by the FAC on a network-wide,
and not necessarily commercial, basis.) It may be expected that charges for off-
airport operators would reflect the location and quantity of space used. Nonetheless,
this requirement for an airport presence may have the potential to reduce
competition between duty-free outlets in the city in which an airport is located.
However, the off-airport duty-free market in Australia is not insignificant. Further,
other options for duty-free purchases remain (particularly overseas).

Thus, retail rents at airports would reflect locational, but not monopoly, rents
(appendix C). Melbourne Airport argued that:
… expansions of space must — quite obviously — be carefully planned and integrated
into the logistics of the airport. When licensing to our concessionaires, no undertaking
is made not to build more retail space.
Accordingly it seems reasonable to conclude not that commercially imposed limitations
on space drive up prices at the airport, but rather that the location drives up the value of
sizes that can be hosted there. In this sense, as Kahn reminds us, prices determine the
rents that can be charged — not rents prices. (sub. 7, p. 3)

In addition, Melbourne Airport (sub. 7) offered some explanation as to why returns

to retail space appear to be higher in airports than in other retail areas. These
included the average incomes of international travellers (and thus the product ranges
offered) and the longer opening hours of the airport. It argued that, controlling for
these factors, returns to retail space at Melbourne Airport are not significantly
higher than those of other retail centres. For instance, its international terminal sales
of $1.80 per square metre per hour compared with $1.79 for Chadstone and $1.59
for Highpoint shopping centres. As the airport noted, the value of sales in the
shopping centres may be lowered somewhat by the presence of lower-priced
supermarkets and department stores.

The Commission has not received evidence to indicate that retail rents at airports
reflect monopoly pricing.

Overall, it appears that airports have little, if any, market power in the provision of
retail space. They compete with other providers of retail space, they do not appear
178 PRICE REGULATION
OF AIRPORT
SERVICES
to restrict retail space, and consumer purchases at airports are largely discretionary.
Retail rentals at airports appear to reflect locational rather than monopoly rents.

6.14 Summary
The degree of market power held by airports differs depending on the service being
considered. The results of the analysis in sections 6.2 to 6.13 are summarised in
table 6.1.

Table 6.1 Market power of airports in particular servicesa

Service Degree of market power

Aircraft movement facilities High

Passenger movement facilities Moderate/High
Lounge space (VIP and business) Low
Vehicle access facilities High
Car parking Low/Moderate
Taxi facilities Low/Moderate
Aircraft refuelling Moderate/High
Aircraft light/emergency maintenance sites Moderate
Aircraft heavy maintenance facilities Low
Flight catering facilities Low
Freight and ground equipment storage sites Low
Freight facility sites and buildings Low
Waste disposal facilities Low
Administrative office space Low/Moderate
Commercial and retail activities Low
a This is based on the airports that were assessed in chapter 5 to have market power. Where a range of
degrees of market power is indicated (eg moderate/high), this means that market power may vary by airport,
or by the particular component of the service in question.

The highest degree of market power is held in facilities that are necessary to users
and for which the supply-side substitution possibilities are limited. Thus, where
airports have market power, that power is most significant in aircraft movement
facilities, vehicle access, some forms of passenger processing facilities and aircraft
refuelling. With respect to aircraft refuelling, market power is most significant at
those airports, such as Perth, that have a high proportion of long-haul flights.

Market power is least significant in facilities or services for which users have
discretion over the quantity or quality purchased and/or there are significant supply-
side substitution possibilities. These include aircraft heavy maintenance facilities,
flight catering facilities, freight and ground equipment storage sites, freight facility
sites and buildings, waste disposal facilities, administrative office space, and
commercial and retail activities.

MARKET POWER IN 179

PARTICULAR
AIRPORT SERVICES
Where service providers potentially compete with the airport in the provision of
services, there may be an issue of access if the service must be provided on the
airport site.

FINDING 6.1

For those airports with a moderate to high degree of market power, the extent of
market power varies across the services provided. It appears to be most significant
in aircraft movement facilities, vehicle access, some forms of passenger processing
facility and aircraft refuelling. With respect to aircraft refuelling, market power
appears to be most significant at Brisbane, Perth and Sydney airports.
Where other providers potentially compete with the airport in the provision of
services, access may be an issue if these providers require access to the airport site.

180 PRICE REGULATION

OF AIRPORT
SERVICES
7 Conduct of unregulated airports

Analysis in chapters 5 and 6 concluded that major core-regulated airports possess

significant market power in key aeronautical services. In this chapter, the possible
pricing and other behaviour of those airports, were they to operate without price
regulation of any kind, is considered, along with potential efficiency and
distributional effects of such behaviour. In particular, aspects of the commercial
environment in which airports operate that are likely to influence their conduct, are
explored.

An exploration of these issues is prerequisite to an assessment of whether continued

price regulation is needed, and in what form, to prevent airports with market power
exercising that power.

Section 7.1 considers the effect of non-aeronautical revenues on aeronautical

pricing and investment, incentives for airports to cooperate with other providers in
the supply chain and incentives to discriminate in pricing. Also discussed is the
possible countervailing power of airlines in their dealings with airports. Section 7.2
draws together analysis in chapters 5 and 6 and section 7.1 to assess the likely
conduct of major airports if airport-specific price regulation were removed along
with the efficiency and distributional effects of such conduct.

7.1 Influences on unregulated airport pricing

The pricing behaviour of major unregulated airports with market power will be
influenced by a number of factors. These include airports’ reliance on non-
aeronautical revenues, demand uncertainty, and the scope for countervailing power
of major airline customers. The ability of, and incentives for, airports to
discriminate in pricing may also be important in determining the efficiency costs of
any market power.

Non-aeronautical revenues and aeronautical pricing

Revenues from duty-free shopping and other retailing, car parking and property
developments are a large and increasing part of airport revenues. Table 7.1

CONDUCT OF 181
UNREGULATED
AIRPORTS
compares average earnings per passenger movement1 (profit before abnormals,
interest, tax, depreciation and amortisation (EBITDA)) from non-aeronautical
activities and aeronautical activities at core-regulated airports for 1999-00.2 As
noted in chapter 2, some differences across airports reflect different organisational
arrangements, and non-aeronautical revenue for all core-regulated airports includes
lease payments received for domestic terminals operated by airlines.

Table 7.1 Earnings before abnormals, interest, tax, depreciation and

amortisation (EBITDA), aeronautical and non-aeronautical
services at core-regulated airports, 1999-00
Passenger EBITDA EBITDA
movements $000 per passenger $
Airport ’000 Aero Non-aero Aero Non-aero
Sydney 23 097 40 703 147 010 1.76 6.36
Melbourne 15 138 30 348 89 967 2.00 5.94
Brisbane 10 532 15 178 75 837 1.44 7.20
Perth 4 890 8 293 37 321 1.70 7.63
Adelaide 4 201 3 487 30 518 0.83 7.26
Canberra 1 969 1 122 5 119 0.57 2.60
Coolangatta 1 959 5 568 na 2.84 na
Darwin 1 080 306 4 388 0.28 4.06
Hobart 909 1 321 1 244 1.45 1.37
Alice Springs 781 183 1 671 0.23 2.14
Townsville 682 297 2 368 0.44 3.47
Launceston 542 443 1 609 0.82 2.97
Sources: ACCC (2001b–f; sub. 36, attachment A). na Not available.

For all core-regulated airports, except Hobart, non-aeronautical earnings in 1999-00

exceeded aeronautical earnings by a significant margin (non-aeronautical earnings
for Coolangatta are unavailable). Airports with substantial international passenger
movements (Sydney, Melbourne, Brisbane, Perth and Adelaide), on average
(unweighted), earned about $6.90 per passenger from non-aeronautical activities
compared with $1.55 from aeronautical services. This compared with $2.753 and

1 Passenger movements include international, domestic and regional passengers. Because of

different spending patterns and differences in the provision of commercial services for different
passenger groups (eg airports do not directly earn revenue from retail services provided in
domestic terminals leased by major airlines, though some part of the lease payments is linked to
throughput), average non-aeronautical earnings per passenger is unlikely to provide an accurate
estimate of average earnings from different passenger groups.
2 These estimates differ somewhat from operating profits presented in chapter 2 because
depreciation typically is proportionately lower for non-aeronautical services.
3 Excluding non-aeronautical revenue per passenger from Coolangatta Airport.
182 PRICE REGULATION
OF AIRPORT
SERVICES
$1.05 respectively for the smaller core-regulated airports, without significant (or
any) international traffic.

Airport pricing with demand complementarities

As discussed in appendix C, provided non-aeronautical earnings are directly linked

at the margin to passenger traffic growth and the price elasticity of demand for
aeronautical services is not zero, airports will have an incentive to encourage
additional passenger throughput (and therefore increase their profits) by reducing
aeronautical charges below the level they would be without the non-aeronautical
services. This point was acknowledged by the Board of Airline Representatives of
Australia (BARA):
The interdependencies combined with the economic rents achieved on non-core airport
services acts to limit the likely increase in the charges for core airport services in the
absence of effective regulation. (sub. 41, p. 15)

In similar vein, the UK Civil Aviation Authority (CAA) observed that:

… the airport itself has incentives to set aeronautical prices to reflect demand
complementarities. Where complementarities are important … the airport will have
strong incentives to set low aeronautical charges in any case (particularly if they are
able to price differentiate effectively). Therefore any potential loss of economic
efficiency is likely to be substantially reduced, and needs to be set against the
presumption that scope of economic regulation should not be unnecessarily wide unless
the efficiency arguments in favour are compelling. (CAA 2000f, p. 17)

Moreover, provided profits from non-aeronautical services reflect locational rents

rather than monopoly profits,4 economic efficiency will be enhanced by such
pricing — locational rents earned in non-aeronautical activities create no efficiency
distortion.5 As Starkie observed:
… for a profit-maximising airport company with market power, the effect of the
demand complementarities linked to the locational rents, is to attenuate the normal,

4 That is, provided airports do not constrain artificially the provision of space for non-
aeronautical facilities, profits earned from providing such space will reflect the inherent
scarcity of land proximate to airside activities.
5 To the extent that profits from non-aeronautical activities derive from the exercise of market
power, the efficiency loss arising from monopoly pricing in these activities must be balanced
against any efficiency gain from reducing aeronautical charges. However, as discussed in
chapter 6, evidence suggests, for most activities classified as non-aeronautical, that the degree
of market power held by the airport is likely to be low. It should also be noted that, because
economic efficiency can be enhanced by complementary pricing of aeronautical services, this
does not imply that single-till regulation (which, at the limit, involves a transfer of all locational
rents) must promote efficiency further (appendix C).
CONDUCT OF 183
UNREGULATED
AIRPORTS
 downward pressure on profits that arises when increased air traffic volumes are at the
expense of lower prices for the use of runways and terminals … In other words,
airports have good reason to limit the extent to which they exploit market power.
(Starkie 2001a)

According to Tretheway (1996), because of these complementarities, the airline

‘customer’ also becomes a ‘partner’ of the airport. Whether, as a result, average
aeronautical charges remain above their stand-alone, cost-recovering level or fall
below it (towards marginal cost) will depend on the size of marginal, non-
aeronautical net revenues, relative to the airport’s profit-maximising, stand-alone
aeronautical price.

While the data in table 7.1 indicate the overall significance of non-aeronautical
earnings to airport operators, to gauge possible implications for unregulated
aeronautical pricing, several qualifications are in order.
• The earnings per passenger reported are average, rather than marginal net
earnings. Though some non-aeronautical revenues are raised in a way that is
directly related to additional passenger throughput (for example, retail
concession fees), some are not. Hence, average non-aeronautical earnings may
overestimate earnings per additional passenger.
• The classification of earnings into aeronautical and non-aeronautical in part
reflects current regulatory arrangements. Estimates of non-aeronautical earnings,
for example, include lease payments for the space occupied by domestic
terminals operated by airlines. If these terminals were operated by the airports,
then terminal charges would be classified as aeronautical revenues. (Offsetting
this, however, the airport would then directly earn revenue from any non-
aeronautical activities conducted within the terminal.) In addition, some
aeronautical-related revenues included in these non-aeronautical estimates may
contain elements of monopoly rents though, as discussed in chapter 6, market
power in most non-aeronautical activities appears to be low.
• Some fixed costs may be included in cost estimates. This would mean that
reported average earnings will tend to underestimate marginal earnings. On the
other hand, some costs excluded from EBITDA calculations may be marginal
costs (for example, some depreciation).

Notwithstanding these qualifications, given the magnitude of these largely

passenger-related profits, it seems reasonable to conclude that the incentive for
unregulated airports to increase aeronautical charges is likely to be dampened,
though it is difficult to be precise about the magnitude. That said, not all passenger
groups are likely to be equally profitable for an airport on the non-aeronautical side.

184 PRICE REGULATION

OF AIRPORT
SERVICES
In this regard, Virgin Blue submitted that:
… its customers are likely to generate much less non-aeronautical revenue for airports
than the passengers of full-service airlines. This is because Virgin Blue passengers:
• are exclusively domestic, and do not engage in duty-free shopping;
• use terminals that are owned by the airports (and therefore counted as aeronautical
revenue, in contrast to lease payments from airline-owned terminals), are
significantly smaller and have a significantly reduced range of concessions; and
• are generally more value-conscious and seek to minimise overall expenditure at
airports. (sub. DR74, pp. 3–4)

On the other hand, Virgin Blue passengers, on average, are also likely to be far
more responsive to airport charges, as noted in chapter 5. This means that even if
Virgin Blue passengers do spend relatively small amounts at airports, airports could
still increase their profits by targeting reductions in aeronautical charges.

Earnings per passenger appear to be much higher for international passenger

movements than for domestic movements because of duty-free shopping and also
because airports do not operate domestic terminals leased to major airlines. Average
revenues from retail sales and car parking per international passenger at Sydney
Airport are reported to be $13.65 and $2.19 respectively; domestic passengers are
reported to spend around an average $1.46 per passenger on car parking
(Chanticleer 2001).

Airports will have an incentive to target aeronautical price reductions to the more
profitable passenger groups. (Tretheway (1996) noted that Vancouver Airport
linked into airline loyalty programs, awarding frequent flier points to car parkers.)
An airport also will have an incentive to offer price discounts only to marginal
flights/passengers to retain infra-marginal profits from aeronautical services.6

Where an airport faces capacity constraints, or the price sensitivity of demand for
aeronautical services is very low, it will have little incentive to reduce aeronautical
charges because this would merely serve to reduce profits. Sydney Airport (at least
until the events of September 2001) appeared to have excess demand for slots for
several hours a day. Sydney Airport is unlikely to offer incentives at peak times. It

6 Network Economics Consulting Group (ACCC, sub. DR55, attachment C) commented that, if
airports were effective price discriminators, then the scope for non-aeronautical profits to
temper aeronautical prices would be weakened. This would not appear to the case, however.
Scope for non-aeronautical profits at the margin would increase incentives for the airport to
price discriminate and thus expand throughput further. At any rate, if airports can price
discriminate effectively, the efficiency effects of monopoly pricing are reduced or eliminated
by the price discrimination.
CONDUCT OF 185
UNREGULATED
AIRPORTS
does, however, offer substantial discounts to ‘new’ off-peak services (Sydney
Airports Corporation Limited (SACL), sub. 27).

If passengers are relatively more sensitive to quality than price, then an airport may
attempt to attract passengers (and airlines) by improving quality. They may use, for
example, non-aeronautical earnings to finance capacity- and quality-enhancing
aeronautical investments (which, in turn, generate more non-aeronautical revenues).
Melbourne Airport commented that ‘high levels of customer satisfaction drive
greater turnover of discretionary expenditure in airports and this is highly
profitable’ (sub. 7, p. 9).

Apart from off-peak discounts at Sydney Airport, the Commission understands that
several core-regulated airports have offered assistance (pecuniary and non-
pecuniary) to both domestic and international carriers to increase scheduled flights.
Kuala Lumpur International Airport is reported to have waived landing and take-off
fees for flights with more than a 25 per cent load factor, in order to boost traffic
(Cheesman 2001).

Melbourne Airport (sub. 7) advised that it provided assistance for market

development for airlines because the cost of gaining market acceptance is regarded
as a barrier to entry for carriers. To encourage additional flights to Canberra post-
September 2001, that airport introduced various incentives (sub. DR75). That such
incentives have been, and continue to be, offered by these and other airports
suggests that airports consider that at least some airline demand is not highly
inelastic.

However, given price caps on aeronautical prices (under which prices have
continued to a significant extent to reflect the Federal Airport Corporation’s (FAC)
single-till pricing), scope for general aeronautical price reductions is likely to have
been limited. Airports also may be disinclined to reveal discounts because of:
• the impact on their commercial relations with other airlines;
• a concern that being seen to be pricing below the cap for some users may lead a
regulator to set a lower cap in subsequent regulatory periods; and
• commitments regarding non-discrimination between international carriers.

The effect on aeronautical prices

BARA (sub. 41) commented that even with the influence of non-aeronautical
revenues, in the absence of price regulation, aeronautical prices are likely to be
significantly higher than current levels.

186 PRICE REGULATION

OF AIRPORT
SERVICES
In similar vein, Network Economics Consulting Group (NECG), acting as
consultants to the ACCC, noted:
The PC’s conclusion that airports will have an incentive to reduce the prices of
aeronautical activities below the stand-alone profit-maximising level does not appear to
be robust. (ACCC, sub. DR55, attachment C, p. 13)

NECG supported its conclusion on the grounds that:

• potential monopoly profits from aeronautical services would far outweigh the
potential for profits from non-aeronautical services; and
• passenger numbers are likely to be unresponsive to aeronautical charges.

Taking the second point first, as noted above and in appendix C, the Commission
acknowledges that non-aeronautical profits will moderate stand-alone profit-
maximising aeronautical prices only if such profits are strongly linked to
aeronautical throughput and if such throughput is responsive to (effective) price
reductions. In the Commission’s assessment, evidence of incentives being offered
by airports to encourage additional services at the margin (with and without price
caps) suggests that most airports consider that airlines are responsive to such
incentives. This fact, combined with evidence that a substantial proportion of non-
aeronautical revenue is linked directly to passenger throughput, suggests that
airports will have an additional incentive to encourage passenger growth by offering
price reductions and other incentives.

As to the magnitude of the effect of such incentives on prices, the absolute size of
any price reduction will be a function of the absolute size of the marginal profits
from non-aeronautical services. The effect in percentage terms on aeronautical
prices will depend on the (unconstrained) market power of the airport in
aeronautical services. This, in turn, will depend on the elasticity of the demand
function for aeronautical services at that airport.

NECG suggested that non-aeronautical profits will have only a modest effect on
profit-maximising aeronautical prices because non-aeronautical profits are likely to
be small relative to potential monopoly profits in aeronautical services. In addition
to an apparent assumption that demand for aeronautical services is unresponsive to
price changes over a wide range, this assessment appears also to be premised on a
view that non-aeronautical profits will be low because non-aeronautical services are
subject to competition.

Even for those airports that the Commission has found to have significant market
power (chapter 5), demand is unlikely to be ‘extremely insensitive to price’ as
asserted by NECG (ACCC, sub. DR55, attachment C, p. 11). Though the major
market segments for these airports are likely to have relatively inelastic demands,

CONDUCT OF 187
UNREGULATED
AIRPORTS
even consumers in these market segments have substitution possibilities. Elasticities
could increase significantly as prices increase, constraining the scope for any large
changes in airport prices. Put another way, the demand curve facing airports may
not be very steep over the entire range.

The Commission agrees with NECG that most non-aeronautical services (car
parking, retailing and so on) are supplied under reasonably competitive conditions
(chapter 6). However, the airport derives rents from these activities not by
restricting their supply to raise the price of the relevant services, but because airport
land suitable for these potentially valuable activities is inherently scarce. Evidence
suggests that associated locational rents, linked to passenger volume, are substantial
in total and at the margin at the major Australian airports.

FINDING 7.1

Profits from non-aeronautical activities at most core-regulated airports appear

significant, especially when compared with current earnings from regulated
aeronautical charges. Though this earnings disparity may be expected to be
somewhat less if price regulation of aeronautical services were removed, there is an
incentive for airports to temper prices for aeronautical services (particularly for
additional services and new entrants), improve quality and/or increase aeronautical
capacity to encourage passenger growth and non-aeronautical revenue. The
magnitude of non-aeronautical earnings suggests that the effects on aeronautical
prices would be significant.

Demand uncertainty and supply-chain linkages

Demand for a particular airport’s services is derived from the demand for air travel
to that destination. The demand for air travel, in turn, is derived from the demand
for business meetings, holidays, family visits and so on to that destination/region.
Airports therefore are susceptible to changes in the relative attractiveness of the
community and region they serve. Thus, though airports’ market power, to a large
extent, relies on the relatively small effect of an increase in airport charges on ticket
prices, as small links in a very long supply chain, airports may have some incentives
to cooperate with other providers (for example, in the tourism and aviation
industries), especially when this cooperation is likely to reduce demand uncertainty.

For airports that serve very large population centres and a variety of market
segments (for example, business, domestic and international air travel), the effects
of a downturn in one market segment may be cushioned by other, more resilient,
segments. Nonetheless, as discussed above, large capital city airports appear to
derive particular profit from spending on non-aeronautical services by international

188 PRICE REGULATION

OF AIRPORT
SERVICES
tourists, and could be vulnerable to any reduction in that market segment. For some
smaller airports that are heavily reliant on inbound tourism, for example, a shift in
consumer preferences could have even more severe consequences for the airport.

By itself, an airport can do little to prevent a substantial shift in preferences. Even

zero landing charges are likely to have little effect in this instance because the shift
in preferences would dominate any response to lower airport charges. An airport
can, however, work with local tourism authorities and other input providers,
including airlines, to promote and improve the attractiveness of the city/region in
which it is located, with a view to reducing demand uncertainty.

Melbourne Airport commented that it ‘has been in active partnership with

successive Victorian Governments in developing markets both to and from
Melbourne’ (sub. 7, p. 12). Hobart Airport observed:
The privatisation of the airports has, to a significant degree, promoted the commercial
reality that there is an inter-relationship and reliance that needs to be fostered between
government infrastructure, transport and tourism agencies, tourism operators, the
airlines and AOCs [Airport Operator Companies]. (sub. 11, p. 1)

In similar vein, Australian Airports (Townsville) commented that it:

… is a small private company which focuses on providing quality services at a low cost
whilst returning value to the company and its shareholders. Its approach to competition
vis a vis other airports is directed at supporting its local business community in
attracting new development to the region. In that regard we offer a service that is priced
below the ‘competition’ and to that extent should support local community economic
goals. (sub. 14, p. 14)

To the extent that airports seek to cooperate with others in the supply chain, and
governments, to encourage increased tourism and air travel, it is unlikely they
would undermine this medium- to long-term growth strategy by setting very high
airport charges. The Motor Trades Association of Australia Superannuation Fund
(MTAA Super Fund) observed that:
A catch-phrase might be ‘airports do not compete, destinations compete’. A city such
as Brisbane, bidding to host a major conference or event is pitting its package of
airport, hotels, conference facilities and entertainment against the packages from
competing destinations, not only in Australia but around the globe. The airport, as a
facilitator of tourism and trade must remain competitive to ensure success.
(sub. 22, p. 36)

It also is interesting to note that several State Governments submitted that they had
developed cooperative relationships with capital city and regional airports in order
to develop air services to their regions.

CONDUCT OF 189
UNREGULATED
AIRPORTS
The Queensland Department of Tourism, Racing and Fair Trading (DTRFT) noted
that:
There is active realisation amongst most operators that airports are derived demand
assets and are not an end in themselves, but rather an access point to a region or
destination …
DTRFT is an active partner with a number of these airports in Queensland and jointly
develop detailed business cases to develop air services to these and other regional ports.
(sub. 6, p. 4)

The South Australian Government also noted that it has:

… developed a cooperative and effective working relationship with Adelaide Airport
Limited and its airline customers to help facilitate air services growth. (sub. 33, p. 1)

That privatised airports have sought and maintain such relationships does not sit
well with them exploiting whatever market power they may have.7 Of course, some
airports may consider factors affecting their long-term growth potential to be
outside their sphere of influence. In these circumstances, whether the outlook is
optimistic or pessimistic, they may have little incentive to cooperate with other
providers or to temper their pricing today with a view to longer-term pay-offs.

Countervailing power of airport users

Countervailing power is the ability of a buyer or buyers to constrain the prices of a

seller with market power to less than would be charged if buyers were ‘small’ and
uncoordinated. In principle, the outcome of dealings between two parties with
market power is indeterminate ex ante; but, to the extent that an airline or another
airport user is able to place pressure on an airport to price at a level more consistent
with efficient prices, economic efficiency is enhanced.

This issue has elicited diverging views from inquiry participants. On the one hand,
several airport operators argued that airlines possess significant market buying
power. On the other hand, airlines submitted that the absence of alternatives has
meant that they have little ability to influence prices of aeronautical services.
According to Adelaide Airport Limited (AAL):
Adelaide Airport Limited is not able to abuse any market power that it may have in the
domestic market as AAL’s main customers, Qantas and Ansett, hold significant

7 It is feasible that an airport with market power could cooperate with other providers and
organisations to increase demand in order to charge higher prices and earn higher monopoly
profits. But once that strategy became apparent, it is unlikely that the cooperation would
continue.
190 PRICE REGULATION
OF AIRPORT
SERVICES
 countervailing market power. Put simply, Qantas and Ansett are much more important
for Adelaide Airport than Adelaide Airport is for Qantas and Ansett. (sub. 20, p. 1)

Australian Airports (Townsville) argued that smaller airports were particularly

disadvantaged in negotiating with airlines:
In addressing this matter it is important to identify where the balance of power is. AAL
[Australian Airports (Townsville)] believes that the balance of power is with airlines by
virtue of their size relative to most Australian airports and especially small regional
airports. (sub. 14, p. 17)

Similarly, Capital Airport Group (trans., pp. 299–301) cited its difficulties in
negotiating agreements with Ansett and Qantas at Canberra Airport to illustrate the
airlines’ countervailing power.

Airlines presented a contrary view. BARA argued that ‘airlines have little or no
economic countervailing power when negotiating over airport charges’ (sub. 41,
p. 16):
The primary determinant of the relative bargaining strengths of the parties is the value
of their outside options. That is, the difference between the profits each party will
achieve with and without the transaction being completed.
International airlines make significant investments in developing international routes. A
decision by an international airline to commence services to and from Australia and to
and from a particular city in Australia involves significant investments. For example,
airlines incur costs in identifying a new market opportunity and establishing the
business case to fly there, dealing with regulatory requirements, establishing a physical
presence which involves establishing offices and recruiting and deploying staff,
establishing a sales and distribution network, and promoting the new route. Airlines
usually incur losses in the early period of the new route’s existence, in the expectation
that these costs will be recouped as the route’s popularity grows. Many of these costs
are unrecoverable if the airline decides to discontinue the service. (sub. 41, p. 16)

Qantas Airways (trans., p. 262 and sub. 48) and Ansett (sub. 42) made similar
claims, with which Professor Forsyth agreed:
Essentially, to run the countervailing power argument, the user has to have a good
alternative to the product on offer, so if Ansett wants to fly into Melbourne, to have
countervailing power it has to have a very good alternative that it can switch to fairly
readily and that’s usually not the case in Australia. It can be the case in Europe or in
parts of the US but not in most parts of Australia. (trans., p. 59)

The nature of countervailing power in the airports sector

There are several means by which large airport users may influence airport pricing
decisions. The main economic weapon for large buyers is the withdrawal or threat
of withdrawal of part or all of their demand for the seller’s product. Exercising
CONDUCT OF 191
UNREGULATED
AIRPORTS
countervailing power essentially involves game playing between the protagonists
and requires the ability to undertake or threaten behaviour that in the short term is
not profit-maximising (that is, profits are forgone by not engaging in potentially
profitable trades) in the expectation that this will deliver a more profitable, long-
term outcome (a better price or service).

In addition to any such economic power, large airlines or airline alliances or

associations in Australia and overseas have used their political influence and
recourse to the legal system to influence pricing by airports. (Toms (1994) discusses
airlines’ responses to the introduction of peak charges at Heathrow Airport.) In
regard to dealing with Qantas and Ansett, Westralia Airports Corporation (the
operator of Perth Airport) stated:
They have a far greater capacity and a preparedness to be able to take on legal action
against us if they feel that there is an incentive at the end of it for them, whereas our
shareholders have limited levels of resources and the ability to be able to counter those
significant legal actions is certainly constrained at Perth Airport. (trans., p. 325)

However, it is important to recognise that any economic, political or legal power

that may be provided to airport users through the current regulatory framework, is
not necessarily relevant in assessing the extent of underlying market and
countervailing power in Australian airport markets. The extent of countervailing
power in the absence of price regulation is the relevant starting point when
assessing the appropriate degree of airport regulation.

Criteria for assessing countervailing power

Aeronautical services, the area in which airports are most likely to have potential
market power, are also the activities where airports face the most concentrated and
financially-powerful users. Nonetheless, as noted by Professor King:
The existence of a single significant buyer does not automatically create countervailing
power … To determine if countervailing power is relevant, the analyst needs to
consider the bargaining position of buyers and sellers. In particular, it is important to
consider which parties will lose most from any failure to reach an agreement to trade
the relevant product. For countervailing power to exist in a market that otherwise is
deficient in competition, any losses from a break-down in bargaining need to be
predominantly borne by the seller. (ACCC, sub. 36, attachment C, p. 13).

There are several indicators of the relative strengths of the bargaining positions of
airlines and particular airports. They include the following.
• The options available to airlines and airports in a particular situation. If there
are other airlines ready to enter routes that become vacant, then the
countervailing power of airlines will be weak. If there are few alternative airlines

192 PRICE REGULATION

OF AIRPORT
SERVICES
 (or, more precisely, no airlines that could quickly slot into a vacated route), then
existing airline operators will be in a stronger position to exercise countervailing
power. Where airlines have choices in relation to airports, countervailing power
will also tend to be enhanced.
• The profitability of routes to airlines. If routes are known to be marginally
profitable, then threats of withdrawal may be more credible, particularly if those
flights are relatively more profitable to the airports (for example, due to related,
non-aeronautical revenues). On the other hand, where routes are known to be
profitable (for example, due to strong business demand), the airport will be in a
stronger position.
• The impact of any reductions in demand on airport profitability. Because
airports have high fixed costs in runways and terminals, their profitability will be
particularly sensitive to changes in demand. On the other hand, much of the
fixed assets of airlines can be moved to other routes or sold. (Nonetheless, both
BARA (sub. 41) and MTAA Super Fund (sub. 22) observed that airlines incur
substantial sunk costs when establishing new routes, costs that act as a barrier to
entry and exit.) In addition, fewer flights and lower passenger (and/or freight)
throughput may reduce airport profits from non-aeronautical sources. While the
airport will take non-aeronautical profits into account when setting its
aeronautical prices, it may be able to target the withdrawal of particularly
profitable services to the airport (for example, departing international flights,
cargo/maintenance facilities) as leverage to gain further price reductions.
• The relative size and financial position of airports and airlines and their
respective shareholders. The party best able to sustain short-term loss-making
behaviour will be in a more advantageous position.

One difficulty in assessing the extent of airlines’ countervailing power is that there
is no history, under private ownership, of unregulated operation of major Australian
airports and there are few lightly-regulated private, overseas airports. The
experience of regulation of airports in New Zealand — a regime that was in part
predicated on a view that airlines possess countervailing power sufficient to offset
market power of airports — provides some insights (chapter 11 and appendix G).
Notwithstanding some deficiencies in that system, major airports and airlines have
agreed on price changes and airlines ‘have demonstrated a willingness to withhold
airport payments and to consider court action’ (CC 2001b, p. 76).

Assessing countervailing power at Australian airports

In Australia, aeronautical charges since privatisation have been constrained by the

regulatory framework and remain constrained at the major four airports. In

CONDUCT OF 193
UNREGULATED
AIRPORTS
opposing increases in aeronautical prices, airlines generally have operated within
this framework rather than exercising any countervailing buying power that they
may possess. The form of commercial relationships between airports and airlines is
evolving slowly from the earlier model of a government authority dealing with a
protected domestic airline duopoly.

The Commission has received fragmentary evidence (much of it confidential) about

the exercise of countervailing power from both airlines and airports. Mostly the
submissions appear to describe tough commercial negotiations.

Some inquiry participants (for example, Forsyth, sub. 5) suggested that significant
price rises at Sydney Airport (‘not disapproved’ by the ACCC in April 2001)
confirmed the absence of airline countervailing power. However:
• these price increases were coming off the very low base of FAC single-till prices
(chapter 8); and
• the price increases were ‘not disapproved’ by the regulator after a lengthy
consultation process. This regulatory process — if not all its outcomes —
generally is supported by airlines.

It also is questionable whether either market or countervailing power is relevant at

Sydney Airport, given that demand for slots has exceeded the permitted 80
movements per hour for several hours each day. Even with the new price increases,
prices at these peak periods are likely to be at levels well short of market-clearing
levels.8

In these circumstances, it clearly would not be credible for an airline to threaten

withdrawal of a service for which it holds a valuable slot allocation. But neither is it
likely that Sydney Airport would have an incentive to increase prices such that slot
utilisation was pushed below the current maximum of 80 movements per hour.
Though the value of slots will decline somewhat as the result of approved increases
in Sydney Airport charges, airlines (rather than their passengers) will continue to be
major beneficiaries of scarcity rents.

Application of the criteria outlined above to Australian core-regulated airports

suggests that the smaller Phase 2 airports are likely to be more vulnerable to
countervailing power of airlines. As discussed in chapter 5, they tend to rely more
on the tourism market, in which scope for inter-airport competition (for example, in
Tasmania) is stronger. In these circumstances, airlines can credibly threaten to

8 Demand pressures on Sydney Airport have eased as a result of international events and the
withdrawal of many Ansett services. This reduction in demand for airport services, especially
in the domestic market, is expected to be short term, however.
194 PRICE REGULATION
OF AIRPORT
SERVICES
switch some services to another airport. Such substitution possibilities also will be
reflected in demand for an airport’s services, but an airline that is a dominant airport
customer may be able to do even better by threatening to withdraw a bloc of
services.

Smaller airports reliant on one or two major airlines also may be in a relatively
weak bargaining position if an airline threatened to withhold payment for a period.
While an airport could take legal action and, after meeting certain requirements
under the terms of their leases (chapter 3), eventually refuse access to a debtor
airline, it could experience a severe cash-flow crisis in the meantime, as well as
incur substantial legal costs. On this point, Professor King noted that:
… [countervailing] power may be increased if the airport itself cannot credibly reduce
its output; for example, if the airport is credit constrained and potentially faces cash-
flow problems. (ACCC, sub. 36, attachment C, p. 13)

The larger core-regulated airports (Sydney, Melbourne, Brisbane and Perth), on

the other hand, are particularly significant to airline networks and more airlines
offering a range of service levels want to fly to them. Services to and, more
particularly, between these destinations are likely to be more profitable to airlines.
Scope for airport substitution for domestic flights to these destinations is limited.

In the international market, the scope for airlines to shift some of their traffic
between Australian airports may provide them with a degree of bargaining power.
The growing importance of airline alliances and code-sharing agreements may also
enhance airline bargaining power. SACL suggested that large airports are
vulnerable to the group buying power of airlines:
… individually, and through representative bodies and alliances, airlines have a
significant degree of countervailing power. Airports derive a substantial degree of their
revenues from [a] comparatively small number of airline customers, who, individually
and collectively, are sophisticated and well organised. (sub. DR62, p. 4)

While these alliances are likely to increase airline countervailing power, member
airlines often compete within alliances, and alliances compete with each other,
which will tend to undermine their bargaining power.

As for the potential impact on profitability, a decline in some services at the larger
airports may be more easily offset because of the range of market segments served
and services offered. Melbourne Airport (sub. 37) presented analysis to demonstrate
that a hypothetical 10 per cent reduction in services to Melbourne Airport by Qantas
would reduce the airport’s operating profit by around 10 per cent, but the airline’s
by a mere 0.4 per cent. However, this ignores scope for other airlines to replace
some or all of the services withdrawn by Qantas, overstating Melbourne Airport’s
potential losses, at least in the aviation market that existed prior to September 2001.

CONDUCT OF 195
UNREGULATED
AIRPORTS
The larger airports rely heavily on non-aeronautical revenues, however, and airlines
feasibly could threaten to withdraw marginal flights that delivered high profits to
airports to obtain lower across-the-board charges. This may be assisted by airline
knowledge of airport revenues. Melbourne Airport commented that:
This is also a game in which there is significant information asymmetry. The airline
will have a good idea about how much it contributes to the airport in terms of direct
income and may even have some knowledge about the passenger based income derived
from retailing, car parks and so on and can be confident that marginal airport costs are
low … On the other hand, airports have little information about the economics of
individual airline routes. Indeed, it is unlikely that an airport would even know what the
next most profitable route was, let alone its profitability. (sub. 37, p. 11)

Airlines also could use the threat to locate discretionary infrastructure (such as
engineering and maintenance facilities) elsewhere as a bargaining chip to extract
lower aeronautical charges. However, as noted by BARA (sub. 41) and Qantas
(sub. 48), this strategy can be used only occasionally, and then only on an ex ante
basis. Once a facility has been built, the airline is locked in to using that facility at
the airport and relative bargaining positions can reverse. However, an airline could
lock in lower aeronautical charges via a long-term contract.

Larger airports are also likely to be better able to withstand reduced cash flows from
non-payment of fees, for example.

Developments since September 2001

The reduction in Ansett services since that airline was placed in administration,
coupled with rationalisation of several international carrier services to Australia, has
left Australian airports generally more reliant on the custom of Qantas. Over time,
Qantas’s dominant position in the domestic market is expected to diminish
somewhat as other operators build up their capacity. In the short to medium term,
however, Qantas’s bargaining position with many airports appears to have been
enhanced, at least in relation to the airline’s decisions about allocation of aircraft
capacity at the margin.

Qantas rejected this supposition, citing announced price increases at Coolangatta,

Darwin and Canberra airports, following the removal of price caps at those airports,
as evidence of its lack of bargaining power (trans., pp. 551–61). As discussed
below, it is not clear that these price increases indicate an abuse of market power. It
is correct, however, that a threat by Qantas to reduce domestic services to a
destination where there is excess demand for flights to that destination (all existing
flights being fully booked) would not be credible. As noted above, the airline’s
negotiating position is likely to be much stronger in relation to scheduling of
additional domestic and international flights.
196 PRICE REGULATION
OF AIRPORT
SERVICES
 FINDING 7.2

The countervailing power of airlines in their dealings with major capital city
airports appears limited. However, airlines may have a degree of countervailing
power even at those airports where there is scope for airport substitution (for
example, entry ports for international flights), where airlines form alliances and
bargain as a group, or where selective threats can be made to reduce services that
are highly profitable to airports.
For smaller core-regulated airports, airline countervailing power is likely to be
stronger, due to the commercial strength of major airlines relative to smaller
airports, the market segments served by those airports and greater scope for airport
competition.

Countervailing power of non-airline users

Apart from airlines, there are other users of airports who may possess some
countervailing power. Airport operators lease rental space to specialist retailers who
then provide retail services (including food and beverages) to airline passengers and
others visiting the airport. The large number of individual retailers and retail chains
that may seek these leases suggests there would be little countervailing power in
this sector. That said, it is possible that some retail chains may offer goods and
services that passengers value highly. In this case, the airport may seek out these
retailers, in which case the latter may have some negotiating power.

At any rate, as discussed in chapter 6, airports appear to possess little market power
in retail services. A broadly similar situation applies to car parking. Melbourne
Airport argued that:
… airports are unlikely to have significant market power in relation to services
provided to non-airline users and as such, a discussion of countervailing power is not
particularly relevant. (sub. 7, p. 16)

Several airports have introduced charges for taxi operators picking up passengers at
airports (appendix E). As taxi drivers are organised at the State and national level
and operate under the auspices of a few large companies, they would seem to have a
degree of economic and political power to countervail market power that airports
may have with regard to taxi charges. However, the Australian Taxi Industry
Association did not accept this proposition:
We must note that, while there are driver organisations operating within the taxi
industry, the extent of their representation of taxi drivers is relatively low and so
consequently, is their ability to develop ‘countervailing power’ in negotiations with
airports.

CONDUCT OF 197
UNREGULATED
AIRPORTS
 Similarly, while there are a limited number of large organisations that provide ‘taxi
brands’, these networks have no effective control over the individual taxi operators
providing services under that brand. (sub. DR61, p. 2)

Nonetheless, the activities of Melbourne taxi drivers in response to proposals for an

airport levy on taxis using airport holding facilities (appendix E) would suggest a
degree of countervailing power in this area.

FINDING 7.3

Non-airline users of airports do not appear to have significant countervailing

power, though representative bodies acting on behalf of these users may exert some
influence.

Airport price structures and price discrimination

As discussed in chapter 4, price discrimination by firms with decreasing average

costs allows common, fixed costs to be fully recouped without discouraging
marginal purchases (and/or purchasers). Fully-efficient price discrimination requires
that each consumer pays a price equal to marginal valuation for the last unit
purchased by that consumer.9 Though such fine discrimination is unlikely to be
feasible in practice, to the extent that broad groups of consumers with different
elasticities of demand can be identified and kept separate, those with higher
responsiveness can be offered a lower price.

If airports with market power effectively could charge different prices for different
customers, then the efficiency effects of exercising market power would be reduced
because the quantity supplied would increase and may even approach the efficient
level. Furthermore, an airport will have a strong incentive to discriminate in its
pricing because price discrimination brings higher profits.

As with countervailing power, this issue has brought forth contradictory evidence.
The ACCC concluded that:
… if airports are in a position to price discriminate between different customers, the
allocative efficiency losses resulting from monopoly pricing might be mitigated. At the
limit, if they were able to perfectly price discriminate, the deadweight welfare losses to
society could be eliminated altogether.
However, the Commission [ACCC] considers this irrelevant given practical realities.
The evidence to date suggests that airlines and airports have little capacity to price

9 Perfect price discrimination, in the sense that all consumer surplus is captured, is not required
for efficiency. Efficiency requires that marginal sales are not forgone.
198 PRICE REGULATION
OF AIRPORT
SERVICES
 discriminate in relation to aeronautical services … Furthermore even if they could, the
information requirements to enable such pricing behaviour are likely to be extremely
high. (sub. 36, p. 7)

On the other hand, Professor Forsyth argued that:

If it is not a busy airport, it will charge high price levels, but the dead-weight losses
from these high charges will be minimised since it will be able to implement a very
effective form of price discrimination, through size or weight related charging.
(sub. 5, p. 17)

Price discrimination under weight-based charging

Traditionally, airports have levied landing charges on the basis of maximum take-
off weight (MTOW) of aircraft. This means that a heavier aircraft pays more to land
than a lighter one. Aircraft weight per passenger tends to increase with plane
gauge — for example, according to the MTAA Super Fund ‘a B737 or an A320 has
around 1.5 passengers per tonne (at an 80 per cent load factor), whereas a B767 has
around 1.0 passenger per tonne at similar loads’ (sub. 22, p. 26). Thus average
airport charges per passenger tend to be higher for larger planes. To some extent
this may reflect the higher costs imposed on the airport by larger planes, though
Doganis (1992) concludes that the relationship between weight-based charges and
airport costs is indirect at best.

If larger planes fly long-distance routes, with passengers paying correspondingly

higher fares, then MTOW-based charges can roughly assume the properties of so-
called Ramsey pricing. That is, airport charges will be levied in inverse proportion
to the price elasticity of demand, assuming the price elasticity of demand for airport
services is inversely related to the amount of the fare.10

This is not to suggest that such price discrimination will be perfect; larger planes
(for example, Boeing 747s) may fly relatively low-fare, short routes. The ACCC
also commented that weight-based charges do not discriminate among passengers
on given flights or among airlines with varying capacities to pay (sub. DR55).

While the Commission agrees with the ACCC that weight-based charging will not
deliver perfect price discrimination, such pricing can discriminate effectively in
certain circumstances. It is interesting to note that the ACCC rejected a proposal by
SACL to introduce passenger-based charging for domestic services because it:
… was concerned that it may adversely impact on competition in the domestic aviation

10 All else given, the greater the airfare, the smaller will be airport charges as a proportion of that
fare, and hence the smaller the price elasticity of demand for airport services (chapter 5).
CONDUCT OF 199
UNREGULATED
AIRPORTS
 market, particularly in relation to new entrants … The concern is that the proposal
could reduce competition by disadvantaging low-cost new entrant airlines who carry
more passengers per aircraft. (ACCC 2001n)

In other words, in this case, weight-based charging appears to be acting as an

approximate form of Ramsey pricing because the carrier flies comparatively small
planes with high load factors. Virgin Blue was reported making a similar comment
in relation to the introduction of passenger-based charges at Coolangatta Airport
(Templeton and Mills 2001).

Though the ACCC is right to observe that weight-based charges cannot discriminate
among passengers on particular flights, that task is likely to be performed
reasonably efficiently by airlines. A landing charge levied per aircraft will be
treated as a fixed cost by airlines (given a decision to make the flight), to be
allocated across passengers in a way that reduces the (discouraging) effect on the
marginal fare-paying passenger. Thus, as observed by the MTAA Super Fund:
Airlines’ pricing strategies, and their pattern of recovery of fixed costs, mean airport
charges are a similar percentage for each fare class and a smaller absolute amount in
discount economy fares. It is therefore incorrect to depict changes in airport charges as
applying uniformly across all passengers, implying a greater proportional impact on
lower fares … Aeronautical charges also have more of the characteristics of fixed costs
than variable costs, and any changes will therefore likely be passed on more to higher
fare paying passengers. The proportional impact on discount airfares will be smaller as
a consequence. (sub. 22, p. 31)

Thus weight-based aeronautical charges combined with airline yield-management

can result in reasonably effective price discrimination.

Passenger-based charges

Several core-regulated airports have introduced passenger-based international and

domestic common-user terminal charges and some have introduced, or are
proposing to introduce, passenger-based landing charges (for example, Sydney and
Coolangatta) (chapter 2). This shift to passenger-based charging appears to have
accelerated since September 2001. Moreover, these changes seem to have strong, if
not unanimous, airline support (see above).

One rationale for the introduction of passenger-based charging for terminal use is
that terminal costs are more directly related to passenger numbers than aircraft
weight. Thus, passenger charges may provide better price signals to users and
providers of terminals about the need for new investment. The shift also may reflect
the increasing importance of non-aeronautical revenues to airports based on
passenger throughput. A passenger charge provides airports with data on passenger

200 PRICE REGULATION

OF AIRPORT
SERVICES
flows, by airline, and thus scope for encouraging more profitable passenger loads.
Tretheway has suggested that ‘passenger-based airport pricing also creates
incentives for airports to work in greater partnership with air carriers in stimulating
primary passenger demand’ (Tretheway 1996, p. 40), while shifting some of the
risks of lower load factors during economic downturns from airlines to airports.

Though passenger-based charging appears to reduce scope for price discrimination

on some margins (for example, aircraft size), scope for discrimination remains (for
example, between passengers travelling on different airlines or at different times of
day, or between international and domestic passengers). Capital Airport Group (the
operator of Canberra Airport) submitted:
In Capital Airport Group’s experience, marginal flights … can be encouraged through
the use of passenger-based charges coupled with a discount. Where airlines are
uncertain about passenger numbers during the start-up or development phase of a
marginal service or route, they have been comforted that aeronautical charges will be
‘linked’ to performance of the route. (sub. DR75, appendix 1, p. 4)

In short, passenger-based charging does not preclude effective price discrimination.

Start-up and other incentives

As discussed above, many airports also offer low entry prices and/or direct
assistance (in cash or kind) to new airlines and new services offered by incumbent
airlines. Such deals cannot be attributed to current regulatory arrangements and they
suggest that an airport has both the incentive and a reasonable ability to target
marginal airline services, subject to the need to prevent ‘churning’ of existing
services as new services.11

It is noteworthy that Melbourne and Perth airports, in applying the recent one-off
price increases allowed under price caps, have elected to impose the entire
allowable increase on domestic movements, with no net increase in charges for
international flights (chapter 2). This restructuring presumably reflects the nature of
the shocks that affected the two markets in September 2001. The domestic market
was hit by a reduction in the supply of aviation services that has required an
increase in average domestic fares to ration available capacity. In contrast, demand
in the international market slumped, placing downward pressure on fares and
threatening viability of several international carriers. These two airports are

11 While the ACCC correctly observed that international agreements prevent discriminatory
airport charges among international airlines, there is evidence that airports instead offer other
incentives (for example, marketing assistance). Moreover, these incentives often are available
for marginal services of all airlines.
CONDUCT OF 201
UNREGULATED
AIRPORTS
restructuring prices in a way that will favour what, in current circumstances, is the
more vulnerable, and presumably more price-sensitive market. (This does not mean
that the domestic market will always be less price sensitive than the international
market.) At the very least, the recent pricing outcomes suggest that airports are alert
to changes in price responsiveness at the margin and, moreover, consider that their
charging policies can affect demand.

FINDING 7.4

Airport operators have a strong incentive to discriminate in pricing in order to

increase their profits. Evidence suggests that they have scope to do this by applying
different pricing structures, including low entry prices and other incentives for
marginal flights and new airlines. Coupled with the ability of airlines to
discriminate in pricing among passengers, such pricing is likely to reduce any
efficiency losses arising either from the need to cover the fixed costs of providing
aeronautical services or from the exercise of airport market power.

Pricing at capacity-constrained airports

This section deals with a situation where capacity is constrained by exogenous

forces (for example, legislated aircraft movement limits at Sydney Airport). The
incentive for an airport to create scarcity by delaying investment is discussed in
section 7.2.

Weight-based, or even passenger-based, charging is unlikely to be efficient when

there is excess demand for airport capacity, where access to slots (landing and take-
off times) must be rationed. To increase profits — in this case, scarcity rents rather
than monopoly profits per se — a capacity-constrained airport that is not subject to
price regulation will tend to move away from weight-based or passenger-based
charges towards time-based charging of the scarce services (for example, a fixed fee
per slot). Thus, scarce runway capacity may be rationed by higher slot prices at
peak times.12 As noted by Professor Forsyth:
… where weight based charges are levied at busy, capacity constrained airports, an
inefficient pattern of use is encouraged, because low value users are not rationed away
in favour of high value users. This happens at Sydney and London Heathrow airports.
(sub. 5, p. 14)

Thus, when capacity is constrained, efficiency is served by allowing those with the
highest valuations to access the facility and effectively denying access to those with

12 What constitutes the peak can be a difficult question, however (SACL, sub. 27).

202 PRICE REGULATION

OF AIRPORT
SERVICES
low valuations. Whether the resulting allocation would promote distributional
objectives of the government is another matter.

A capacity-constrained airport (not subject to price regulation) would have an

incentive to introduce such pricing — the incentive being the prospect of extracting
scarcity rents. It may have an incentive to restrict capacity even further; this will
depend on the severity of the imposed constraint relative to the airport’s profit-
maximising price and volume.

Some participants (for example, ACCC, sub. 36) disputed whether price rationing
of excess demand for slots at Sydney Airport would promote efficiency, because
slots are already being allocated under a quantity rationing scheme. However, as
discussed in appendix H, leaving aside the issues of the regional ring fence (which
guarantees certain slot allocations to regional airlines — see chapter 3, box 3.4) and
possible protection of other regional users (outside the ring fence), there is no
expectation, under the current rationing scheme, that airlines with the highest slot
valuations (potentially carrying customers with the highest valuations) obtain
rationed slots.13 Indeed, such an outcome is unlikely.

Thus, even if airport prices still fall short of efficient, market-clearing levels, a
move towards those levels is likely to promote a more efficient allocation of slots
because some flights will be replaced by others with higher pay loads (for example,
through the substitution of larger for smaller planes and the withdrawal of marginal
flights). Nonetheless, this is unlikely to lead to significantly higher passenger
airfares at peak times because average fares to Sydney Airport at these times will
already largely reflect the scarcity value of slots, with the rents accruing to airlines
holding those slots. However, some passengers travelling on services that are
withdrawn or rescheduled to off-peak periods may be worse off. The transfer of
scarcity rents from airlines to airports also could have repercussions for an airline’s
operations if such scarcity rents are used to cover the fixed costs of services on
other routes.

13 For a given slot allocation, however, an airline will ensure that customers with the highest
valuations obtain seats by not offering discounted fares on those flights. While this promotes
efficiency, the airlines, rather than the airport or passengers, capture scarcity rents under a slot
allocation scheme. It also is possible that an airline will provide some flights to obtain/retain a
valuable slot (eg by flying smaller planes more frequently). Though the full current value of
slots used in this way will not be realised by the airline, such a strategy will reduce slots
available to competitors and ‘babysit’ them for higher-value use in the future.
CONDUCT OF 203
UNREGULATED
AIRPORTS
FINDING 7.5

At capacity-constrained facilities (where the shortage of capacity is not created or

perpetuated by the airport), efficiency requires that capacity is rationed such that
consumers with higher valuations of the service obtain access. Where capacity is
restricted, airports not subject to price regulation would generally have an
incentive to set aeronautical prices that promote efficient use of the airport.

7.2 Conduct of unregulated airports: efficiency and

distributional effects
As noted in chapter 4, the principal rationale for price regulation of airport services
is the potential for inefficiency through monopoly conduct by unregulated airports.
In the light of analysis in chapters 5 and 6, and the preceding sections of this
chapter, this section discusses likely pricing and related behaviour of unregulated
airports, and the efficiency and distributional consequences of such behaviour.

Likely pricing outcomes if airports were unregulated

Major Australian airports appear to have a high degree of market power in core
aeronautical services. In the absence of constraints on airport pricing discussed in
preceding sections, there will be an incentive for an airport to use this market
power. If it cannot discriminate in pricing, an airport will have an incentive to
increase prices above marginal (and average) costs, to the point that marginal
revenue and marginal cost are equal.

Any such increase in price will reduce consumption of the airport’s services,
resulting in the familiar monopoly, deadweight efficiency loss (chapter 4,
figures 4.1 and 4.2). Airlines and their passengers will pay more for use of the
airport’s facilities, allowing the airport to earn monopoly profits. (For a capacity-
constrained airport, so long as the restricted capacity is fully used, the profits will
reflect the scarcity of capacity, not monopoly profits.)

However, in the Commission’s view, for reasons discussed in section 7.1, the
‘traditional’ full monopoly pricing is unlikely to occur, even for those airports with
significant market power. These influences will temper aeronautical price increases.
Further, price discrimination, to the extent that it can be exercised, will moderate
any adverse efficiency effects.

It is likely, if prices of core-regulated airports were unregulated, that aeronautical

charges in most airports would be higher than current levels. In part, this is because
204 PRICE REGULATION
OF AIRPORT
SERVICES
there is fairly strong evidence (chapter 8) that current prices are inadequate for
efficient, replacement investment (even with any allowance that airports themselves
may make for non-aeronautical earnings). As discussed in chapter 8, current prices
at privatised core-regulated airports continue largely to reflect the single-till pricing
imposed by the FAC. Thus, though prices may rise in the absence of the current
pricing constraints, the relevant test is by how much unregulated prices would
exceed efficient prices (given a requirement that privatised airports must be self-
financing), not by how much they would be above historic or current regulated
prices.

Though care must be taken in comparing airports, the fact that increased charges of
the order of 100 per cent at Sydney Airport (to implement replacement cost, dual-
till pricing) were not disallowed by the ACCC suggests that there may be a need for
significant price increases at other airports to reach ‘efficient’ average price
levels.14 How far beyond these levels or, indeed, whether aeronautical prices at
airports with market power would increase in the absence of any airport-specific
price regulation, no-one can say with certainty. However, for reasons outlined
above, the Commission considers that any price increases would be constrained by a
range of market forces, including commercial interests of the airports themselves.

As noted at the outset, examples of major airports operating without regulatory

constraint are rare. Airports in New Zealand have operated under a ‘light-handed’
regulatory regime, while price caps were removed from most Australian core-
regulated airports in early October 2001.

The New Zealand Commerce Commission (CC) estimated that Auckland Airport
earned an average annual nominal rate of return on aeronautical assets of just under
13.5 per cent between 1989 and 2000 (CC 2001b). Using this estimated rate of
return as a benchmark for pricing by unregulated Australian airports, the ACCC
suggested that the four largest airports in Australia would increase revenues by
$1.4 billion over five years (ACCC, sub. DR55). In rough terms, this implies a
doubling of aeronautical charges and revenues. Several comments are in order.
• First, the accuracy of the CC’s rate-of-return calculation has been challenged by
Auckland Airport on numerous grounds, including that the CC’s removal of land
bought for a second runway from the allowable asset base is inappropriate

14 Sydney’s landing charges were considerably lower than those at other major airports (albeit
offset by a higher international terminal charge). Sydney’s costs and asset valuations also may
be higher (especially land valuation). Melbourne Airport observed that it ‘would not consider a
step increase of the magnitude achieved by Sydney Airport appropriate [for Melbourne
Airport]’ (sub. DR66, p. 19).
CONDUCT OF 205
UNREGULATED
AIRPORTS
 (chapter 11). Auckland Airport estimates that its true rate of return is around
8 per cent.
• Second, even on the CC’s own estimates, aeronautical prices at Auckland
Airport in 2000 were less than 3 per cent above the estimated efficient
benchmark price. Given scope for measurement error (and particularly given the
problems in valuing airport land), this gap would seem insignificant. Since then,
Auckland Airport has negotiated a 12.5 per cent price increase with airlines.
• Third, a critical issue in assessing likely price outcomes in Australia is whether
current prices at Australian airports are efficient. Efficiency (and consumers in
the medium to long term) may be best served by somewhat higher prices than at
present, to facilitate appropriate investment. It is noteworthy that the CC
suggested that an efficient target rate of return on aeronautical assets at
Auckland Airport was just under 10 per cent (somewhat higher than suggested
by the airport itself). This is well above reported rates of return on aeronautical
assets at the major privatised Australian airports, which in 1999-00 were
reported to range from zero to 3.7 per cent (chapter 2, table 2.4).

Since removal of their price caps on 5 October 2001, several Australian airports
(Adelaide, Alice Springs, Canberra, Coolangatta, Darwin, and Townsville) have
increased their charges, some by substantial amounts (chapter 2). Coolangatta, for
example, announced an increase of 170 per cent. Launceston and Hobart have not
increased their charges. Without any knowledge of the impact on airport revenues
and profits, it is difficult to say whether these announced price increases represent
an abuse of market power, in the sense of being unjustified by costs. Most airports
that have increased their charges have said they will still incur losses and that the
charges will be reviewed in early 2002. While Capital Airport Group (sub. DR75)
announced a price increase, it simultaneously introduced substantial rebates in a bid
to encourage additional services.

Though the Commission does not consider that it is possible to forecast with any
certainty possible increases in unregulated airport charges that would occur without
price caps, the ACCC’s suggestion (sub. DR55) that profit-maximising prices at
Sydney Airport are in the order of $500 per international passenger and $120 per
domestic passenger seems implausible. In particular, the calculations do not appear
to admit any scope for destination or modal substitution (box 7.1). Sydney Airport,
even acting as an unconstrained profit-maximising entity — and any political
pressure aside — is unlikely to charge prices at the levels suggested by the ACCC.
Nonetheless, it is acknowledged that efficient peak-period charges could be high at
Sydney Airport, reflecting the impact of slot constraints.

206 PRICE REGULATION

OF AIRPORT
SERVICES
 Box 7.1 Unconstrained pricing at Sydney Airport
Though full details of their derivation are not provided, the ACCC’s estimates
(sub. DR55) of unconstrained prices at Sydney Airport appear to overstate likely price
increases for several reasons.
• The price elasticity of demand typically increases with price. An estimate of point
elasticity at the current price therefore should not be used to forecast the profit-
maximising price mark-up unless it is known that the price elasticity does not
change much over the relevant range.
• In particular, the calculation appears to assume little if any scope for competition
from other airports and destinations, yet there is evidence that Melbourne and
Brisbane airports, in particular, compete with Sydney for international traffic. In
future, Canberra Airport, which is only two hours drive from Western Sydney, will
have international capacity and could even compete with Sydney for some domestic
traffic. In addition, travellers doubtless would explore alternative travel and
communication modes if airport charges rose by such an amount.

Allocative efficiency effects

As noted above, if the price at the margin exceeds the efficient level, then
consumption will be curtailed. This efficiency loss will measure the economy-wide
loss provided private costs and benefits equate with social values. If the use of
airports generated negative external effects, then a higher price that deters use could
bring net social benefits rather than losses. On the other hand, if a user industry
generated external benefits that were not captured fully in the private demand for
output of that industry, the negative effect of excessive airport charges would be
amplified. (In either case, however, the optimal policy may not be to raise or lower
airport charges but rather to target the externality directly. For example, appropriate
levies could be imposed on aircraft according to noise levels.)

Several participants seemed to imply that low airport charges are justified by the
assistance this renders to the Australian tourism industry (chapter 4). In this vein,
BARA (sub. DR54) criticised the Commission for failing to assess economy-wide
effects of airport price increases, suggesting that more than 4000 jobs could be lost
in the tourism industry if airport prices were not regulated. This figure assumes that
international airfares would rise by roughly 2 per cent on average if all core-
regulated airports approximately doubled their charges and that a 1 per cent increase

CONDUCT OF 207
UNREGULATED
AIRPORTS
in international airfares would result in the loss of roughly 2000 jobs in the tourism
industry.15

The calculations also assume that airports do not, or cannot, discriminate among
airlines and their passengers in a way that would mitigate any efficiency effects of
higher airport charges. Yet there is substantial evidence that they do discriminate in
their pricing in a manner that would reduce the effect of any average price increase
on marginal users.

Moreover, the modelled rise in airfares is based on estimates prepared by the ACCC
in relation to proposed fee increases at Sydney Airport. These increases occurred
within the current regulatory regime. In addition, given excess demand for slots at
Sydney Airport, it is unlikely that the rise in airport charges will have much effect
on fares to and from Sydney at peak times. At any rate, the estimates fail to take
into account the effect on fares of yield management by the airlines, which would
tend to reduce the impact of higher airport charges on the more price-sensitive users
(see MTAA Super Fund, sub. 22). At the very least, the interplay between airport
charges and airfares needs to be analysed before modelling an assumed across-the-
board fare ‘shock’.

Further, while any reduction in consumption of airport services caused by higher

charges will affect activity levels in user industries, such partial effects do not
measure the net effect on the economy as a whole. Indeed, increased airport charges
would promote economic efficiency — and national economic welfare — if they
encouraged efficient airport service provision and investment.

The Commission has deliberately not undertaken quantitative general equilibrium

modelling of the likely effects of unregulated airport pricing.
• For one thing, it is very difficult to determine how airfares would be affected at
the margin. In other words, what is the nature of the ‘shock’ to be modelled? If
airport price increases were structured in a way that the margin was not affected
significantly, consumption of airport services (and that of services provided by
related industries such as tourism) would be largely unaffected. As discussed in
this chapter, airports appear to be able to discriminate in their pricing on various
margins. And as mentioned above, the effect on fares and consumption of air
travel would also depend on airline yield-management strategies.
• While any price increase (or decrease) will have effects on other industries
through a variety of linkages, as noted above, these effects of themselves do not

15 The assumed relationship between higher airfares and job losses in the tourism sector appears
to be derived from the MONASH model of the Australian economy.
208 PRICE REGULATION
OF AIRPORT
SERVICES
 shed any light on the question of whether the price changes promote, or detract
from, efficiency. Even if it were considered desirable to subsidise the tourism
sector, it is difficult to conceive of a situation where holding airport charges
below efficient levels would be the most appropriate means of achieving this.

Overall, the Commission does not consider that general equilibrium modelling
would assist in answering the key questions that need to be answered in this inquiry,
at least without much more information than is currently available. Any modelling
of an assumed price change would beg the question of why that particular shock
was chosen. If the Commission modelled a scenario in which prices were increased
to an ‘efficient’ level, or if it modelled a scenario in which consumption of airport
services did not change because of price discrimination, it would be open to
criticism for contriving the result. Moreover, such an exercise would not shed any
light on what might happen to airport prices. In other words, the real debate in this
inquiry is about the likely ‘shock’, namely the extent and nature of any price change
at airports and whether such a change would be efficient or inefficient.

BARA (sub. DR54) also suggested, because core-regulated airports no longer are
owned and operated on a network basis, that additional efficiency losses would arise
due to ‘double marginalisation’. Double marginalisation typically refers to the
separation of a vertically-integrated production process, where each firm in the
chain possesses and exploits market power. Each firm in the production chain takes
the price charged by the input supplier as given (including the price mark-up) and
then itself prices according to its market power. The result of the cascading effect of
higher costs is consumer prices that are higher (and efficiency losses that are larger)
than would occur if the production process were integrated. What BARA could
have in mind here is, say, flights between Melbourne and Sydney, where both
airports separately exploit their market power.

In the case of airports, ‘horizontal’ separation of ownership has occurred, opening

up scope for both separate ‘monopoly’ behaviour and inter-airport competition (and
lower prices) where different airports can act as substitutes (for example, competing
as ports of entry for international flights). Such competition would tend to reduce
total airport charges below the level charged by a single network owner.

Where demands for different airports with market power are complementary (that
is, where a ‘pair’ of airports must be used for domestic flights), at issue is whether
separation of airport ownership is likely to lead to higher prices for (and lower
consumption of) such pairs. It is possible, in this situation, that airports compete for
profits by independently raising their charges. The net effect could be a higher price
being charged for using the pair of airports than would be charged by a single
operator, but the joint profits would be lower than the single-owner case. The
combined profits of the two airports would be maximised if their combined price
CONDUCT OF 209
UNREGULATED
AIRPORTS
(for using a pair of airports) was the same as for a single owner. For this reason,
BARA (trans., p. 535) suggested that airlines could be better off if all airports in
Australia were owned by a single monopolist than by several, though they
acknowledged that there were other factors involved.

Distributional effects

Aeronautical price increases (at airports with excess capacity) would redistribute
surplus from passengers and possibly airlines to airport owners. Any monopoly
airport profits would accrue to airport shareholders, some of whom are non-
residents. (Most privatised core-regulated airports are partially foreign-owned, the
maximum level being 49 per cent — chapter 3, box 3.3.) Airport shareholders also
comprise Australian residents, including superannuation funds investing on behalf
of Australian contributors (for example, the MTAA Super Fund investments in
Adelaide and Brisbane airports (sub. 22), AMP Henderson’s 49.9 per cent
ownership of Australia Pacific Airports Corporation (sub. 10),16 and Uni Super’s
stake in Adelaide and Coolangatta airports (sub. 20)). The ‘losers’ from higher
airport charges would be airline shareholders (comprising residents and non-
residents) and airline passengers (both resident and non-resident). Professor Forsyth
observed that:
In the airport regulation situation, it is not a matter of having some specified
distributional trade-offs between the different groups. There is little by way of
identifying who the gainers and losers are, and what their circumstances are.
(sub. 5, p. 11)

The ACCC, however, urged the Commission to undertake an empirical analysis of

potential distributional effects. It suggested that:
Prima facie there is a case to suggest that higher charges would result in such a transfer
[from Australian residents to non-residents]. The major airports are up to 49 per cent
foreign owned. By contrast only around 10 per cent of passengers using Australia’s
airports are foreign residents. (sub. DR55, attachment D, p. 1)

The estimate that 10 per cent of passengers using Australian airports are foreign
residents would appear to underestimate significantly the impact of higher airport
charges on non-residents. This figure accounts for arrivals and departures of non-
residents on international flights only. Data for non-residents who take domestic
and regional flights are unavailable. If, on average, international visitors take just
one domestic or regional flight while in Australia, then non-residents would account

16 APAC operates Melbourne and Launceston airports (through its subsidiaries Australia Pacific
Airports (Melbourne) and Australia Pacific Airports (Launceston) respectively).
210 PRICE REGULATION
OF AIRPORT
SERVICES
for 20 per cent of total passenger movements.17 Moreover, non-resident
international passenger movements at core-regulated airports assessed as having
most market power and therefore most capacity to raise charges (Sydney,
Melbourne, Brisbane and Perth), accounted for around 15 per cent of total
passenger movements at these four airports in 1999. International visitors accounted
for 58 per cent of all international passenger movements.

While broad groups can be identified as ‘winners’ or ‘losers’ from changes in

airport charges, a full distributional analysis cannot be undertaken without detailed
information about individuals who use and own airports. There is a myriad of
complications; for example, some passengers could be airport or airline
shareholders, or Australian residents may own shares in foreign companies that own
Australian airports or airlines (and some Australian companies may have foreign
shareholders). In addition, non-resident airport shareholders are required to pay
Australian tax on profits remitted abroad.

As discussed in chapter 4, the Commission has focused on likely efficiency effects

to guide its assessment of various regulatory options — an approach that implicitly
assumes that ‘a dollar is a dollar’ to whomever it accrues, provided the recipient is
an Australian resident. Because both ‘winners’ and ‘losers’ include non-residents, it
would be impossible to quantify any net leakage of income abroad due to higher
airport charges. On the face of it, there does not appear to be strong evidence that
higher airport charges would favour non-residents over Australian residents.

FINDING 7.6

Aeronautical charges at some airports could be expected to rise if unconstrained by

regulation, in part because current charges may be below the level required to
cover the costs of investment. Whether and to what extent prices rise above levels
required for efficient provision of aeronautical services, as the result of the exercise
of market power, cannot be forecast with certainty. However, there is reasonably
strong evidence that a range of market influences will moderate any increase in
prices and any negative efficiency effects of higher prices. Chief among these are
the effects of non-aeronautical earnings and the scope for discrimination in pricing
so that airport users are not discouraged from using airport facilities.
Higher airport charges could redistribute income from airlines and their
passengers to airports. The burden of such higher charges would be likely to fall on
both Australian residents and non-residents.

17 This assumes two movements (arrival and departure) per non-resident passenger on each
international and local service.
CONDUCT OF 211
UNREGULATED
AIRPORTS
Productive efficiency of airports with market power

BARA argued that airports with market power, if unregulated, are likely to operate
inefficiently in not watching costs:
It [monopoly pricing] will also weaken the imperatives for airports to be vigilant over
their costs resulting in further economic waste. (sub. 41, p. 1)

BARA continued:
Competition provides benchmarks enabling owners to more accurately gauge the
performance of management. More importantly, competition disciplines poor
performance in terms of low returns and exit from the market. In the absence of
competition there is scope for productive inefficiencies. Effective regulation, through
acting as a surrogate for some of the cost pressures on a firm facing effective
competition, can reduce these costs. (sub. 41, p. 30)

Given the nature of airport supply and demand, there is no competitive benchmark
against which to assess an airport’s performance. (Indeed, competitive supply — in
the sense of many providers — would be a highly inefficient way of providing
airport services.)

At issue is the extent to which airport managers will be able to operate inefficiently
(and thus reduce potential profits to owners). The airport manager has an incentive
to pursue such a strategy to the extent that it delivers non-pecuniary benefits,
including excessive ‘perks’ and a lower level of effort. Several factors seem
relevant.
• Given the competitive sale of core-regulated airports, owners will expect
managers to deliver at least the returns factored into the bid price. Indeed these
performance levels could have been factored into incentive-based contracts with
managers. That said, scope for additional profits not factored into the bid price
may make the manager’s task somewhat easier.
• Owners can benchmark an airport’s costs and performance against other airports
in Australia and overseas.
• Despite limited privatisation of airports internationally, specialist airport
management companies have emerged (for example, Aer Rianta, BAA plc and
Hochtief) that offer their expertise worldwide. Such companies could identify an
inefficient airport and sell their services to airport owners (several of which are
very large fund managers). In other words, there appears to be reasonable
competition in the supply of airport management services.

In addition, there is some direct competition between airports due to physical

proximity and competition for international flights and special airline facilities (for
example, maintenance facilities).
212 PRICE REGULATION
OF AIRPORT
SERVICES
Overall, therefore, privatised airports could be expected to operate in a manner that
is no less efficient than other industries in the economy. Moreover, they are likely to
operate more efficiently than airports owned and operated by the public sector.
Indeed, this was one reason for the sale of airport leases:
The leasing process should lead to further improvements in the operating efficiency and
quality of service of the airports by exposing them to private sector disciplines. (HoR
1996, p. 1310)

FINDING 7.7

Managers of privatised airports with market power are unlikely to have much scope
to allow inefficiencies in production.

The level of quality and investment provided by airports with market

power

As noted above, unregulated private monopolies that do not discriminate in pricing

have an incentive to restrict supply to raise prices. Over time, this would imply that
the monopoly will tend to delay investment in order to maintain this artificial
scarcity and to maximise profits.

If the monopolist can discriminate in pricing or differentiate investments to suit

different users, then the incentive to under-invest will be moderated. For example,
an airport might be prepared to build customised terminal capacity solely for a
budget airline, because full service airlines are unlikely to wish to use such a
facility.

There may be other pressures on an unregulated airport not to delay investment.

• While capacity constraints may generate higher aeronautical charges, they may
adversely affect non-aeronautical revenue, much of which is linked to passenger
volumes.
• If an airport allows high levels of excess demand to develop and to remain for
lengthy periods (for example, by not building an additional runway), then the
door may be opened to potential competitors at alternative sites. (A monopolist
may even bring forward investment to ward off potential competitors.) Though
emergence of a competitor airport in the same location is unlikely, capacity
constraints and the curfew at Sydney Airport appear to have worked to the
advantage of Melbourne and Brisbane airports, particularly with respect to
international traffic. Therefore, an airport that intentionally restricted capacity to
obtain scarcity rents may risk losing some business to other airports.

CONDUCT OF 213
UNREGULATED
AIRPORTS
Rather than seeking to delay investment, Qantas argued that:
Airport operators, on the other hand, have an incentive to overbuild and goldplate
airport assets, as their primary business is land and infrastructure construction and
management. It is therefore vital that regulatory oversight of capital expenditure is
maintained. (sub. 48, p. 22)

However, it is difficult to see why a private airport operator (not subject to price
regulation but subject to capital market constraints) would incur unnecessary
investment expenditure. If increased investment allows an airport with market
power to increase its prices, this must reflect increased willingness to pay of at least
some users; otherwise, prices could have been raised without incurring the extra
investment outlay.

Some participants also expressed concern that airports not subject to price
regulation would allow quality levels to deteriorate. Qantas observed:
Left unregulated, airport operators can be expected to use their market power through
monopoly pricing, diminishing service quality and the imposition of unreasonable
terms and conditions of access to the airport. (sub. 48, p. 29)

Quality deterioration might reflect a lack of investment or just cost-saving measures

(for example, employment of fewer staff). Essentially this is equivalent to arguing
that an airport with market power will allow prices to rise for a given level of
quality. The extent to which an airport with market power exploits that power by
increasing prices and/or reducing quality will depend on the relative quality and
price sensitivities of various users.

An unregulated airport with market power may be cautious about allowing service
quality to diminish below levels desired by users. At least some passengers and
airlines are prepared to pay for quality, particularly for aeronautical services with a
safety dimension.18 Provided those who value quality can be charged for its
provision, an airport with market power is likely to provide that quality.

The link between quality (for example, quick processing and attractive terminal
spaces) and non-aeronautical revenues will also play a role. As noted above,
Melbourne Airport noted a direct causation between overall airport quality and
increased non-aeronautical earnings. With respect to the likely investment and
quality performance of unregulated airports, Professor Forsyth concluded that:
Private monopolies do not necessarily achieve a perfect optimum in their choice on
quality level, but there are good reasons to expect that they will choose about the right
quality … This will be so because they can convert an increase in quality into

18 Under the terms of their leases and the Airports Act 1996, airports are required to meet certain
safety and other standards.
214 PRICE REGULATION
OF AIRPORT
SERVICES
 additional revenue, since users are prepared to pay higher prices for higher quality. The
private airport will be willing to make investments which improve the service to users;
such as an extension to a runway which lowers the costs faced by users, because these
users will be prepared to pay higher prices. (sub. 5, pp. 17–18)

FINDING 7.8

In principle, airports with market power may have some incentive to delay
investment or to allow quality to deteriorate, in order to maximise their profits.
However, in practice, to the extent that airports with market power can discriminate
in pricing or differentiate products for different users, those incentives will be
weakened. The scope to earn additional non-aeronautical profits from higher
quality or expanded aeronautical capacity and passenger throughput, will also
encourage the provision of appropriate quality and investment levels.

Access provision and airports with market power

As far as airport access is concerned, the experience of new entrant airlines

(Impulse and Virgin Blue) on domestic routes is instructive. Though relations
between these airlines and airports have not always been cordial, two airports
(Melbourne and Sydney, albeit not unregulated) built new terminals to
accommodate the new entrants. At least one airport, the Commission understands,
offered start-up assistance. This behaviour seems to support the view that a
privatised, vertically-separated airport will have an interest in expanding its
customer base and passenger throughput, a point acknowledged by the ACCC
(sub. 36).

Nonetheless, there are circumstances in which an airport deliberately may attempt

to deny access to an airline or another provider, for example, by imposing high
charges. Qantas claimed that airports had numerous incentives to deny access:
Even where the airport operator does not conduct a downstream business itself, it may
be economically integrated with the downstream business. For example, the airport
operator may decide to grant an exclusive lease of a car park to a car park operator.
Through its pricing structure, the airport operator may be able to extract all economic
profits from the conduct of the single car park. Effectively, the airport operator is
economically integrated with the operation of the car park. Furthermore, it may have no
incentive to encourage development of additional car parks. It may also decide to deny
airlines or other persons the right to conduct their own car parks (for example the valet
car parks currently conducted by airlines).
Secondly, the airport may deny access to a specific service to an airport user in order to
gain a commercial advantage in other areas of its business. The airport may use the
threat of access denial in order to resolve a dispute with the airport user, or achieve a
commercial outcome (such as increased prices) in respect of another airport service.

CONDUCT OF 215
UNREGULATED
AIRPORTS
 Thirdly, if the airport service is suffering a degree of congestion, the airport may
simply find it easier to deny access rather than establish mechanisms to deal with
congestion and scheduling problems. In other words, the airport would prefer a ‘quiet
life’. (sub. 48, p. 25)

The ability of an airport to earn excess profits in any activity at the airport depends
on the alternatives available to buyers. An airport, therefore, may have an incentive
to restrict access to its facilities to limit competition in the supply of those services
where some competition otherwise is feasible. As observed by Qantas, this could
occur whether the airport operates in the market in question directly or indirectly by
selling rights to others to operate a business on the airport. However, some of the
examples cited by Qantas do not appear to relate to denial of access as such.

In the Commission’s view, there are three broad scenarios in which an airport
operator may seek to deny access.
• Where an airport can earn more by ‘selling’ monopoly access rights to an
airline than providing access to all airline entrants at a lower cost. In effect, the
airport and the airline would collude to monopolise the aviation market (or at
least a particular route). In the absence of any scope for monopoly rents to be
earned in the aviation market (sufficient to compensate the airport for lost sales),
the airport will prefer an open aviation market.
• Where an airport can earn more by selling a monopoly right to an on-airport
service provider (such as a cargo handler) rather than allowing competition in
provision of such services. This scenario also requires the service provider to
earn monopoly rents in the relevant market, sufficient to compensate the airport.
• Where an airport provides services directly (or via a licensee), earning
monopoly profits by denying access (for example, to the airport’s ‘front door’)
to potential competitors in that market. For example, an airport may deny or
frustrate access to an off-airport car-park provider to earn monopoly profits from
its own car-parking operations.

In practice, the first scenario appears less likely to occur than the other two. It relies
on the earning of substantial monopoly profits in the aviation market but,
notwithstanding recent events, that market appears competitive — or, at least,
contestable — over the longer term. Therefore, it is unlikely that an airport would
see any benefit in an exclusive deal with one airline. Indeed, the evidence shows
that airports have actively encouraged services from new carriers (section 7.1). A
variant of the first scenario is that a large incumbent airline could threaten to
withdraw some of its business from an airport if the latter offered a discriminatory
(low) entry price to a new entrant. In this case, the airport would not be willingly
denying access. This scenario also requires the airline, rather than the airport, to

216 PRICE REGULATION

OF AIRPORT
SERVICES
have significant market power and thus significant buying power. Smaller airports
would be more vulnerable to such threats.

With regard to the second scenario, though it may be possible to create a monopoly
in the supply of cargo-handling services at a particular airport, for example, this will
be limited by the scope for airport substitution (which, in the case of dedicated
freight, may be substantial). Nonetheless, there has been an access dispute involving
access for freight handling at Sydney and Melbourne airports (chapters 9 and 11).
Other ground-handling services also require access to the airport. Apart from this
case, however, the Commission has received no evidence that airports have sought
to frustrate competition in these services. Indeed, Capital Airport Group
(sub. DR75) submitted that it actively promoted competition in such services.

The third set of circumstances perhaps describes the most likely access scenario
because the airport effectively competes with off-airport providers in certain
activities and controls access to the airport. Charging an access fee in excess of
costs could allow the airport to raise prices for these potentially competing services.
As noted in chapter 6, the conclusion that airport market power in the provision of
car parking is likely to be low is robust only if the airport does not use its control
over airport access to stifle competition from off-airport car-park providers or
providers of competing transport modes. Denial (or frustration) of access to
potential competitors may arise from imposing unacceptable terms of access rather
than a prohibitive access price (for example, an inconvenient location for the set-
down/pick-up point).

In this vein, there has been one access case in which Delta Car Rentals applied
successfully for declaration of landside roads at Melbourne Airport under s. 192 of
the Airports Act 1996. The dispute concerned the location of the ‘designated
meeting point’ for its shuttle-bus service, not the access charge (chapter 9).
Arbitration was never sought. Some participants also raised concerns about landside
access charges (for example, Victorian Hire Car Association, sub. DR68). As
discussed in chapter 6, though airports do have market power in relation to front-
door vehicle access, it also seems reasonable that the airport will want to regulate
the way access to its terminals occurs, for operational and safety reasons. Apart
from the Delta case, there has been no other application for ‘front-door’ access.

FINDING 7.9

An airport with market power has little incentive to deny or frustrate access to its
major customers, the airlines.
An airport with market power may have an incentive to restrict ‘front-door’ access
to off-airport providers of competing services such as car parking, or providers of
competing transport modes, though there is little evidence of this occurring.
CONDUCT OF 217
UNREGULATED
AIRPORTS
8 Review of airport price regulation:
price-cap and prices-notification
arrangements

Chapter 3 provides details of the post-privatisation framework for regulating prices

charged for airport services at major (core-regulated) Australian airports. This
framework commenced operating in July 1997 for Melbourne, Brisbane and Perth
airports (Phase 1) and in July 1998 for the eight other major privatised airports
(Phase 2). The original framework was due to operate for five years from
privatisation. However, it was substantially amended in October 2001 in response to
significant and unforseen falls in aeronautical traffic (Hockey 2001). This chapter
and chapter 9 examine the performance of the price regulation regime against the
criteria of economic and regulatory efficiency outlined in chapter 4, and the specific
objectives proposed for the current airports price regulation when it was first
established, reiterated in the terms of reference to this inquiry.

Price regulation applied to the 11 privatised, core-regulated airports is considered in

sections 8.2 (dealing with the limit placed on aeronautical prices by the CPI-X price
cap), 8.3 (examining price increases allowed above the price cap) and 8.4 (briefly
considering changes to the price caps under the October 2001 regulatory changes).
Price regulation of Sydney Airport is discussed in section 8.5.

8.1 Introduction
Following privatisation, the same economic regulation framework was imposed on
all 11 privatised core-regulated airports. It comprised several inter-related elements
aimed at preventing abuse of any market power available to airport operators, while
encouraging efficient levels of service quality and new investment. If an airport
possesses and may use significant market power then the incentives provided by the
regulatory framework, if appropriate, can bring forth improved economic outcomes.
However, if the regulation were poorly framed or administered it might detract from
the performance of an airport with market power or indeed impede the performance
of an airport without market power.

REVIEW OF AIRPORT 219

PRICE REGULATION
In October 2001, the Government removed price caps from all but the three Phase 1
airports (Hockey 2001). However, at three of the remaining airports, price
monitoring of aeronautical charges was introduced to replace price caps. All airport-
specific price regulation was removed from the other five privatised core-regulated
airports, though they still remain subject to the quality monitoring and access
provisions of the Airports Act 1996. These new regulatory arrangements are
outlined in chapter 3 and discussed briefly below and in chapter 9. Regulation of the
government-owned Sydney Airport remained unchanged.

Airports other than core-regulated airports have faced no special economic

regulation and are subject to the same regulatory environment as other Australian
businesses, in particular the Trade Practices Act 1974.

The various elements of the price regulation framework after privatisation need to
be considered as an integrated package. It is also important to recognise that the
initial regulatory framework explicitly was intended to be a transitional one, aimed
at facilitating the change from airport ownership by a government-owned
corporation to the more commercially-focused environment of private airport
ownership.

Chapter 4 has outlined some principles for efficient pricing and good regulation. For
price regulation of airports the Commonwealth Government has stressed the dual
objectives of protecting users from abuse of market power and the importance of
moving towards price determination by commercial negotiation between airports
and their customers with minimum regulatory involvement (chapter 3).

These objectives recognise the importance of the negotiation process in determining

an efficient structure of prices and developing alternative service and cost
combinations for customers. The Department of Transport and Regional
Development (DoTRD) Pricing Policy Paper observed:
The price cap is designed to be as simple and straightforward as possible, in the
interests of minimising the costs of regulatory oversight for both the ACCC and the
airport operators, while ensuring appropriate outcomes. (DoTRD 1996, p. 2)

The objectives of not only preventing the abuse of market power, but also
minimising regulatory costs, encouraging commercially-negotiated outcomes and
facilitating the access of new airlines have been reaffirmed in the terms of reference
for this inquiry.

220 PRICE REGULATION

OF AIRPORT
SERVICES
8.2 The price cap: CPI-X and prices notification of
aeronautical services
The CPI-X pricing formula (and the associated prices-notification requirements) for
aeronautical services was the centrepiece of the price-regulation regime for all of
the privatised core-regulated airports from the time of privatisation until October
2001 when it was removed for the eight core-regulated Phase 2 airports. The
DoTRD Pricing Policy Paper (1996) indicated that there was some scope for
improved performance and lower costs in producing aeronautical services in the
early years of private ownership. Due to concern about the potential market power
of airport operators in providing these services, the price cap required real
reductions in a basket of aeronautical charges to provide incentives to achieve the
expected productivity gains and to ensure that they were passed on to airlines.1

Box 3.1 (chapter 3) outlines the services included in the aeronautical basket.
Aeronautical-related services subject to price monitoring are discussed in chapter 9.

In January 1997, the Federal Airports Corporation (FAC) set new aeronautical
prices, to which the ACCC did not object, at its five largest airports, leaving prices
at the others unchanged. These charges became the aeronautical prices operating
when 11 of the core-regulated airports were privatised in 1997 and 1998. For the
five years following sale, the price cap was to set maximum annual weighted
average prices, determined by the underlying national CPI, less a discount
factor (X) specific for each airport (chapter 3). Low underlying CPI growth has
meant that, for most airports, the price-cap formula has resulted in falling maximum
average aeronautical charges in nominal as well as real terms.

Aeronautical prices could differ from the allowed maxima but any over-recoveries
needed to be made up by the imposition of lower than maximum allowable prices in
a later year or years.2 Within the cap, aeronautical prices could be set by the airport,
although any increases were to be notified to the ACCC. This offered some
opportunity for airports to adjust their pricing regimes, although changes under
price caps have been limited.

1 The Prices Surveillance Act 1983 does not provide for legal enforcement of price reductions
under the cap. Melbourne Airport (sub. 7) has indicated that, in any event, it considered that
compliance with the cap was a condition of sale of the airport lease which it is bound to
observe, a view supported by the Motor Trades Association of Australia Superannuation Fund
(sub. 22).
2 Some airports have elected to make rebates to users for prices in excess of the cap rather than
charge less than the cap in future years. Under-recoveries (that is, prices below the cap) may
also be made up in later years.
REVIEW OF AIRPORT 221
PRICE REGULATION
An important rationale for choosing the CPI-X approach was that, while
constraining prices, it provided airport operators with an incentive to improve
efficiency, as they are allowed to retain any productivity gains in producing
aeronautical services, above those implied in the X factors.3 In addition, it is
potentially a low cost and low intervention approach once the parameters of the
pricing formula have been established. As noted, a safety valve was incorporated by
allowing under- and over-recoveries of revenue to be rectified in later years.

Inevitably, because it was designed to provide relatively light-handed incentive

regulation, on its own, the CPI-X price cap was not attempting to track
economically efficient prices over time. Nonetheless, if the starting prices were
reasonable and the Xs based on feasible productivity improvements, then, in the
absence of significant demand or supply shocks, the outcomes might provide
reasonable approximations of efficient prices. In addition, the price cap was
scheduled to apply only for a designated time, hence offering the opportunity to
remedy inefficiencies in prices after a set period. The Government responded to the
significant declines in air traffic in late 2001 by removing the cap for the Phase 2
core-regulated airports and allowing one-off increases above the cap for Phase 1
airports. Chapter 10 discusses issues in resetting price caps.

Starting prices for the price cap

When the FAC took control of 17 federal airports in January 1988,4 it reviewed
airport charges (aeronautical and non-aeronautical) on a network-wide basis, with
the objective of covering network costs, including capital costs. This led to an
increase in landing charges (which then provided around 90 per cent of aeronautical
revenue) of about 6 per cent in July 1988, together with significant increases in
non-aeronautical commercial lease charges. Over the next three years to April 1991,
annual increases in landing charges (incorporating some restructuring of charges)
cumulated to a further 23 per cent.5 In April 1991, the FAC’s landing charges were
declared for surveillance under the Prices Surveillance Act 1983 (PS Act).

Landing charges remained constant from April 1991 until January 1997, although
there were some increases in other aeronautical charges over that period. In
January 1997, aeronautical charges were increased by an average (on a network

3 The rate of increase in CPI also embodies productivity increases across the economy.
4 The FAC purchased a further six airports from the Commonwealth Government in 1989 but
sold Cambridge Airport to a private buyer in 1993.
5 This compared to CPI growth between June 1988 and June 1991 of 19.8 per cent and Average
Weekly Earnings (ordinary time earnings of full-time adult males) increases of 20.5 per cent
over the same period.
222 PRICE REGULATION
OF AIRPORT
SERVICES
basis) of 10.8 per cent — this was not objected to by the ACCC.6 However, the
1997 price increases were confined to the five major airports — Sydney,
Melbourne, Brisbane, Adelaide and Perth — representing an average 12.1 per cent
increase for those airports.

The ACCC (1996) observed that, even with the average 13.8 per cent increase
originally sought, the FAC believed that the five airports concerned would only just
recover all aeronautical operating costs (including depreciation based on 1991-92
asset valuations but no returns to capital) from aeronautical revenues. In its draft
decision on Sydney Airports Corporation Limited’s (SACL) pricing proposal, the
ACCC observed the impact on returns to aeronautical assets of the FAC’s pricing
policy:
In the past airport prices were set by the FAC on a single till basis. The FAC adopted a
rate of return target for the airport as a whole, and set aeronautical charges at the level
required to meet the rate of return target. Since profitability on non-aeronautical
services was high, and typically well above the target rate of return for the airport as a
whole, this meant that returns on the aeronautical side of the business were low, often
negative. (ACCC 2001h, p. 75)

The differential price increases between airports were designed as a first step to
more appropriate airport-specific pricing that had been recommended by the Prices
Surveillance Authority (PSA 1993). In not objecting to the FAC’s January 1997
price increases, the ACCC (1996) also indicated its preference for further location-
and service-specific pricing adjustments to improve the efficiency of pricing.

Until the January 1997 increases, landing charges and, where they existed, terminal
charges, were uniform across the FAC network. While some airports made losses,
these were more than offset by profits from the larger airports. Of the original FAC
airports, over time, the number of loss-making units declined from ten (of fourteen)
in 1988-89 to three in 1996-97.7

The FAC network landing charges in place before the airports were leased were
essentially cost-based prices using a single till on a largely network-wide basis. A

6 These increases were in fees for use of airport terminals, but were charged on the same
maximum take-off weight basis as landing fees. The average 10.8 per cent rise compared to an
average 13.8 per cent applied for by the FAC (a 15.2 per cent average increase for the five
airports for which price increases were sought) and a CPI increase of 13 per cent between the
June quarter 1991 and June quarter 1996. There had also been some relatively minor increases
in terminal charges between 1991 and 1996.
7 Of the airports controlled by the FAC, Hobart and Cambridge were managed as a single unit, as
were Bankstown, Camden and Hoxton Park. Of the five further airport units (Alice Springs and
Tennant Creek being treated as a single unit for reporting purposes) the FAC purchased from
the Government in 1989, four were still making losses in 1996-97. Profits and losses were
before interest charges.
REVIEW OF AIRPORT 223
PRICE REGULATION
network-wide real rate of return on revalued assets of around 8 per cent had been
earned consistently over a number of years.8 As a percentage of airport revenue, the
cross subsidies that existed between airports were proportionately much greater for
the smaller loss-making airports than for the larger airports providing the subsidies.
Hence, any additional profits made by the larger airports in order to make up for
low or negative returns at other airports would have been only a small part of the
large airports’ revenue.9

The Board of Airline Representatives of Australia (BARA, sub. DR54) observed

that, due to network linkages, differential returns between airports under the FAC
might not have necessarily represented cross subsidies. However, the PSA, in
examining the FAC’s policy of uniform pricing across airports, considered that the
degree of complementarity of demand between airports in the network was small
and, in any event, would be reflected in the demands facing individual airports. It
concluded:
The most economically efficient use of the entire network of FAC airports requires
separate and independent determination of the landing charges at each airport.
(PSA 1993, p. 123)

The Industry Commission (IC 1992) also recommended that individual airports
should, over time, recover all costs and meet a real rate-of-return target.

The breaking up of the FAC network as part of airport privatisation means that
individual airports now operate on a stand-alone basis. Hence, consideration of the
likely adequacy of the inherited FAC prices as a starting point for future
aeronautical investment decisions at individual airports, needs to be assessed on this
basis.

There were also significant cross subsidies between airport services, although the
PSA (1993) indicated that the FAC’s basis for cost allocation made it difficult to
determine their exact extent. However, the general direction of subsidisation was
from non-aeronautical to aeronautical.

Westralia Airports Corporation (WAC, sub. 21) stated that the starting aeronautical
prices at Perth Airport did not cover operating costs and hence provided no

8 In its 1996 revaluation of assets, the FAC valued land at market value for alternate use (capped
at light industrial) and other assets at written down replacement cost (FAC 1996).
9 In 1996-97 the total losses of the loss-making airports represented 0.5 per cent of the revenue
of the five largest FAC airports. Earning an 8 per cent return on the assets of the loss-making
airports would have represented around 5.5 per cent of the revenue of the five largest airports.
224 PRICE REGULATION
OF AIRPORT
SERVICES
contribution to capital costs or a return on capital.10 Similarly, Capital Airport
Group (sub. 32) stated that the return on aeronautical assets at Canberra Airport was
negative. The ACCC (1998h) report on FAC price restructuring proposals for
Sydney Airport indicated that returns on aeronautical assets were estimated to be
negative in 1998-99. Alice Springs, Darwin and Townsville airports made losses on
their total operations in 1997-98, their final year of operation under the FAC.

The Department of Transport and Regional Services (DoTRS) indicated that leaving
these distortions in the starting prices was a conscious policy:
It was, however, neither practicable nor feasible to attempt to unwind these distortions
prior to privatisation — which was intended to accelerate the micro economic reform in
the aviation industry. Hence an important feature in the sales process was a requirement
that the efficiency gains potentially available at the major regular public transport
airports be shared with airport users. In effect, the Commonwealth chose to forego
value in the sale by requiring real price declines in aeronautical charges. The prices
oversight framework, and in particular the application of a CPI-X cap, was the vehicle
used to deliver these short term gains to aviation users. (sub. 39, p. 8)

The DoTRD Pricing Policy Paper (1996) indicated that these arrangements were to
be reviewed after five years. (However the significant downturn in Australian and
world aviation precipitated changes in October 2001.)

BARA (sub. DR54) noted that establishing the existence of genuine cross subsidies
at airports is complicated by the presence of common costs and demand
interdependencies between aeronautical and non-aeronautical services. With regard
to common costs, aeronautical revenues would need to be below the avoidable costs
of providing those services before an economic cross subsidy is established.

In the short term a significant portion of airport costs would remain even if
aeronautical operations ceased. However, the above evidence concerning FAC
aeronautical prices and revenues strongly suggests that aeronautical services at a
number of FAC airports would not have been covering their long-run incremental
costs. This is important for assessing whether existing prices are likely to justify
new investment.

The impact of demand interdependencies between airport services on the

assessment of cross subsidies involves similar issues to the single-till versus dual-
till debate considered in chapter 10 and appendix C. It is recognised there that,
because of demand interdependencies, airports will have an incentive to use part of

10 In the last year of FAC operation (1996-97), Perth Airport made a profit of $27 million and had
a return (before interest expenses) of 12 per cent on total assets. In view of this, the failure of
starting prices to cover aeronautical costs suggests that Perth Airport earned significant non-
aeronautical returns.
REVIEW OF AIRPORT 225
PRICE REGULATION
profits earned on non-aeronautical activities to contribute to returns on aeronautical
services, without representing a cross subsidy. However, this incentive will not
extend to a full single-till system as operated by the FAC and to this extent the FAC
pricing system provided cross subsidies from non-aeronautical to aeronautical
activities.

Importantly with regard to cross subsidies, the starting prices for the price caps were
not adjusted by the FAC or the regulator to remove economically inefficient cross
subsidies and hence are unlikely to be a good basis for efficient pricing. The PSA
inquiry into the FAC’s charges concluded:
The current approach where the pricing of aeronautical services (as defined in the Act)
are largely determined on a residual basis will contribute to inefficient use of existing
aeronautical and non-aeronautical activities and distort signals for investment.
(PSA 1993, p. 170)

Both the PSA (1993) and ACCC (1996) recognised significant inefficiencies in the
FAC’s aeronautical charges.

The Xs

As part of the airport sale process, the Government announced real weighted
average annual aeronautical price reduction factors (the Xs) which were to apply for
five years and varied between airports. The Xs were determined by the Government
using advice from the ACCC. The ACCC stated:
The ACCC’s advice was based on its analyses of the airports’ projected demand, costs,
expected productivity improvements and economic performance. The cap is based on
the prices charged by the FAC before the airports’ privatisation. (ACCC 2000a, p. 10)

In addition, the ACCC (1998a) indicated that one of the factors considered was a
modest level of capital expenditure based on expected investment requirements over
the five years of the initial regulatory framework. It also indicated that Adelaide and
Coolangatta airports were exceptions with no amounts included for new investment.

Direction No. 2411 to the ACCC from the Minister for Financial Services and
Regulation stated that the value of each airport’s X reflected productivity
improvements that the Government considered could be made in the provision of
aeronautical services at each airport. However, a weakness in the application of the
price cap has been the failure to enunciate clearly how the Xs for each airport were
determined. Transparency and certainty are both attributes of good regulation that
are absent in the application of the X values.

11 Direction No. 24, October 2001, replaced previous Directions (chapter 3).

226 PRICE REGULATION

OF AIRPORT
SERVICES
Regardless of how they were calculated, airport operators were aware of the X
values (and starting prices) before they bid for leases. They could make their own
assessments of their ability to achieve the efficiency improvements and volume
growth needed to make the resultant prices viable over the five-year period in which
the price caps were to apply.

Traffic volumes are crucial in determining whether an airport achieves expected

aeronautical productivity growth. The FAC (1998) observed that increases in traffic
volumes had been the driving force in the Corporation’s performance — passenger
numbers growing by over 80 per cent between 1988-89 and 1996-97. A number of
airport operators (for example, Brisbane Airport Corporation (BAC), sub. 8; Gold
Coast Airport (Coolangatta), sub. 12; and Northern Territory Airports (Alice
Springs), sub. 25) indicated that traffic growth up to 2000 had been lower than the
historical trend while others (for example, Melbourne Airport, sub. 7) reported
strong traffic growth.

All of the core-regulated airports had experienced slower growth in passenger

numbers since privatisation than in the five years prior to privatisation. This was
particularly so for Brisbane, Darwin, Perth and Launceston airports. The events of
September 2001 led to declines in volumes for all core-regulated airports, which are
now being reversed to varying degrees. The longer-term impact on traffic is likely
to become clearer during 2002.

While all lessees were aware of the starting aeronautical prices and Xs, in order to
estimate allowable aeronautical prices under the cap, they also had to form their
own expectations of likely CPI movements.12 Many costs of aeronautical services
are likely to move independently of the factors determining the CPI. For example, it
has been argued by some airports that the low increases in the underlying CPI since
1997 have been unfavourable to airport operators. WAC observed:
We have also seen CPI work very much against us. Now, we realise that that’s an
external fact, but the reality is that the average rate of CPI in the lead-up to the
privatisation of the airport — and we would think that was in some way taken into
account in the CPI-X determination — has certainly deteriorated from our point of
view quite significantly, and CPI is in fact the largest single-value driver of our
business. (trans., p. 321)

WAC continued that many of its costs are fixed and not related to the CPI while
some of its non-aeronautical revenue, such as property rents, were adversely
affected by low CPI growth (trans., p. 321). However, consistently low underlying
CPI growth appears to have been an important factor in the decline in interest rates

12 In addition, bidders would have needed some view about likely price movements that would be
allowed outside the cap for necessary new investment and other cost pass-throughs.
REVIEW OF AIRPORT 227
PRICE REGULATION
in recent years and hence may have indirectly provided airports with savings in the
servicing of debt and equity capital.

While the implementation of the CPI-X formula has been relatively routine, the
price outcome is likely to be crucial to future profitability and investment at
airports. The ACCC observed:
The current prices are a carry over from the prices charged by the FAC before
privatisation. These charges were determined on a network basis. They were also
determined on a single-till basis. This means that the current charges are unlikely to
closely correlate to aeronautical costs. (sub. 36, p. 109)

Knowledge of the growth and productivity assumptions underlying the Xs

designated for each airport would assist in assessing the extent of any distortions.

FINDING 8.1

The single-till basis of the starting prices (also incorporating some cross subsidies
between airports), and the real declines in aeronautical prices at most airports
under the price cap suggest that, for many airports, aeronautical prices by the end
of 2000-01 may have been below the level necessary to justify future aeronautical
investment.

Of course, at least for the first five years, buyers of privatised airports would have
factored in their expectations of the price effects of the CPI-X cap, coupled with
scope for price increases for necessary new investment (NNI). However, while the
X values were known, there was a lack of transparency regarding the types of
investments that were or were not included in the Xs, as well as (as discussed
below) a lack of initial definition of allowable investment for cost pass-through to
prices.

The possibility of obtaining increased prices to compensate for NNI (section 8.3)
has created scope for ameliorating adverse effects the CPI-X cap might have on
investment undertaken to expand or improve aeronautical capacity. However,
incentives for replacement and maintenance investment would still remain low if
the price cap has generated low aeronautical prices. Chapter 10 discusses possible
approaches to resetting these parameters if a CPI-X approach were to be continued
or reintroduced for any airports.

Implementation issues

Because CPI-X is a formula approach to price setting, there is limited scope for
regulatory discretion once the starting prices and Xs have been established. For the
life of the initial regime, these parameters were established by the Government

228 PRICE REGULATION

OF AIRPORT
SERVICES
before the airport leases were sold. The major regulatory discretion has occurred in
deciding the extent of cost pass-throughs for new investment and government-
mandated security requirements (discussed in section 8.3). However, one matter
open to interpretation is the reach of the basket of aeronautical services subject to
the price cap. Two issues have been dominant here, the facts of which are discussed
in detail in appendix E.

First, there has been dispute between the ACCC and a number of airports regarding
whether charges on taxi operators collecting passengers at an airport are part of the
definition of aeronautical services (landside roads) and, hence, part of the basket
under the cap. Second, as part of its monitoring of aeronautical-related prices, the
ACCC (1998b) recommended that fuel throughput levies, which have been
introduced at two core-regulated airports, should be transferred from price
monitoring to be included in the price-cap basket because it considered there was a
strong case that their implementation represented an abuse of market power. To
date, the Government has not responded to the ACCC recommendation and fuel
throughput levies remain subject to price monitoring.

In establishing a price-regulation regime, the important issue in deciding what

services should be subject to greater scrutiny is the extent of airport operators’
market power in those services. The ACCC contended that airports have taken
advantage of market power by introducing taxi charges and fuel throughput
levies.13 The International Air Transport Association (IATA, sub. 43) argued that
any fees on fuel suppliers should only cover the costs associated with the provision
of fuel facilities.

Whatever the degree of market power in refuelling (chapter 6), airports appear to be
able to extract charges greater than the costs of these services. Whether the
implemented charges represent an abuse of market power cannot really be assessed
fully without considering the constraint placed on aeronautical returns by the price
cap.

For core-regulated airports the issue is complicated by possible commitments given

to bidders for airport leases regarding the treatment of these charges. Regardless of
this, the ACCC (1998b) observed that directions under the PS Act required it to
monitor fuel throughput levies and the Act required it to pay particular regard to the
need to discourage firms from abusing market power. It also indicated (sub. 36) that
it had received legal advice that taxi charges would be covered by the definition of
aeronautical services used for the price cap. The Full Federal Court has confirmed

13 Melbourne Airport has sought a taxi levy through the NNI provisions, based on the costs of
building improved services for taxi operators. The ACCC did not object to a charge of $0.66
per taxi, compared to the $1.40 originally proposed by Melbourne Airport (appendix E).
REVIEW OF AIRPORT 229
PRICE REGULATION
this interpretation in a case involving Canberra Airport, at least for the particular
facts of that case (appendix E).

Such alleged contradictions between the sale process and the regulatory framework
increase the risk premium bidders will place on future sales, thereby decreasing
returns to taxpayers and adding to the costs of implementing regulation. In
undertaking sales of their assets, governments need to balance the objectives of high
sale value and the efficient subsequent regulation of the activity concerned.

The taxi issue arose because of a lack of detail in the guidelines and legislative
instruments regarding the extent of services covered by the term ‘landside roads’.
Capital Airport Group (sub. 32) and WAC (sub. 21), amongst others, claimed that
airport bidders were promised that ground transport charges could be introduced
outside the cap. The ACCC argued that:
The Commission’s concern is that the matter of taxis could easily have been addressed
in the regulatory instruments. The failure to explicitly address the issue has resulted in
unnecessary uncertainty for airport operators and airport users. It has also resulted in
substantial costs to the various parties because of the litigation process.
(sub. 36, p. 112)

These problems also highlight the difficulties of imposing an industry-specific

regulatory framework via the PS Act, which contains its own set of broad regulatory
objectives and criteria, and which may not reflect adequately the Government’s
desired regulatory framework for a particular industry.

Notification of proposed price increases for aeronautical services

For companies declared under the PS Act, the requirement to notify proposed price
increases is the usual mechanism through which the ACCC examines the firm’s
prices. However, for the privatised core-regulated airports, the main regulatory
constraint on prices for aeronautical services has been the price cap. While airports
are required to notify proposed increases in prices of such services, the ACCC is not
expected to object to notified price increases that do not result in an airport
breaching its price cap. DoTRD observed:
The ACCC will not object to price changes to aeronautical charges unless they breach
the price cap. This leaves scope for airport operators to continue to rebalance charges
within the overall price cap on aeronautical charges set for the airport.
(DoTRD 1996, p. 4)

Hence there is limited scope for the notification process to affect pricing other than
applications made under the NNI test and other cost pass-through provisions. The

230 PRICE REGULATION

OF AIRPORT
SERVICES
ACCC (sub. 36) has observed that the mandatory nature of the CPI-X Direction
(No. 20) sat uneasily with its prices notification responsibilities under the PS Act.

In addition, low CPI increases in recent years have meant that the CPI-X formula
has, for most airports, generated average nominal price declines for notified
services. Hence notifications of proposed price increases have been unnecessary for
most airports unless restructuring of prices within the cap was undertaken. Some
airports have exceeded the cap in individual years, but because this has not involved
price increases, the notification process has not been invoked. Canberra and
Townsville airports, each with X values of 1 per cent, have both been permitted
increases in average aeronautical prices under the price cap — 0.5 per cent in
1998-99, 0.7 per cent in 1999-00 and 1.8 per cent in 2000-01, while the cap
permitted increases of 0.3 per cent at Launceston Airport (X of 2.5 per cent) in
2000-01.

8.3 The price cap: allowable price increases above the

cap
Depending on the adequacy of starting prices and the method of determining the Xs,
use of a CPI-X formula could be aimed at generating appropriate aeronautical prices
without, for example, any need for further adjustments for investment (chapter 10).
However, the post-privatisation regulatory framework allowed the ACCC to
sanction aeronautical price increases above the price cap in certain situations where
the cap may inhibit expenditure necessary for efficient provision of aeronautical
services. Since October 2001, these provisions are only relevant to the three Phase 1
airports still subject to the price cap and Sydney Airport which is still subject to
price notification.

Necessary new investment

The most significant exception to the price cap related to recouping costs of NNI in
aeronautical assets.14 For the 11 privatised core-regulated airports, the NNI
provisions were formalised in directions under the PS Act (chapter 3) and
implemented by the ACCC using the PS Act legislative criteria for performing
prices oversight (section 17(3)) and the criteria in Direction No. 20 (chapter 3,
box 3.2).

14 The removal of the price cap for the eight Phase 2 core-regulated airports means that NNI
provisions now only apply to the three Phase 1 airports plus Sydney Airport.
REVIEW OF AIRPORT 231
PRICE REGULATION
The NNI provisions were one of a number of instruments designed to encourage
appropriate aeronautical investment by airport operators. These include clauses in
airport sale agreements and leases specifying minimum levels of investment over
each of the first two five-year periods following privatisation, and requirements for
ongoing maintenance of airport structures and maintenance, to at least current
standards, of those parts of the airport needed for aircraft access (chapter 3). Also,
the lessee must provide for the use of the airport site as an airport. In addition, the
quality monitoring provisions of the Airports Act should assist in encouraging
investment in facilities to maintain quality of service, although airports are likely to
have sufficient market incentives to maintain quality (chapter 9).

Since privatisation there have been 18 proposals for NNI cost pass-throughs
involving total expenditure of over $200 million. The major expansion of airport
infrastructure that occurred under the FAC has reduced somewhat the immediate
need for new aeronautical investment at a number of airports.15 The key issues in
assessing NNI are whether it is an efficient and effective means for obtaining the
prices needed to generate efficient levels of aeronautical investment, for facilitating
access to airports for new airlines, and for fostering the commercial relationships
that will engender dynamic efficiency in airports and their users.

Incentives provided by NNI

The NNI process establishes incentives for airport operators and users that are
important in determining whether efficient levels of new aeronautical investment
are achieved.

Incentives for airports

In return for undertaking new investment, current NNI provisions offer airports
aeronautical price increases either agreed to by users or determined by the ACCC.
The ACCC still has the power to object to price increases negotiated between the
parties.

The NNI provisions appear predicated on the following:

• the prices allowed under the price cap may not be adequate to deliver an
adequate return on new investment; and

15 From 1988-89 to 1996-97, the FAC invested $1.6 billion in airport infrastructure (including
non-aeronautical facilities).
232 PRICE REGULATION
OF AIRPORT
SERVICES
• given scope for cost pass-through under NNI provisions (essentially rate-of-
return regulation for NNI), there need to be appropriate checks on airport
investment proposals, including majority user support for such proposals.

On the first point, as discussed above, evidence suggests that the CPI-X framework,
applied to FAC starting prices, has resulted in aeronautical prices that may not, of
themselves, justify new aeronautical investment. Therefore the NNI test is
important in allowing efficient new investment to achieve appropriate returns.
Airport operators bid for the leases on the basis of the CPI-X and NNI procedures
operating for at least the first five years of the lease.

However, even for investments that replace or expand capacity (and hence maintain
or increase revenue), if aeronautical prices are too low, an airport is not likely to
proceed with them. Melbourne Airport observed:
The necessary new investment arrangements exist, in our view, solely because without
them airports would not invest because airports require a price increase across the cost
base as a whole to justify investment, because at the current prices you would not
invest at all. So the necessary new investment arrangements, in our view, have to be
seen very clearly as a fix for the problem of prices being below what you might call
efficient investment prices. (trans., p. 167)

The potential for the airport to earn non-aeronautical revenues from capacity-
expanding or quality-enhancing investments may ameliorate such a problem
somewhat. In addition, some new investments may be undertaken based on the cost
savings they deliver to the airport operator, rather than the revenue they generate.
The ACCC does not allow price increases for cost-saving investments
(ACCC 2000b).

On the second point, it is not clear that airports have an incentive, either with or
without the current price regulation, to undertake extravagant investment — so-
called gold plating. Only if they consider that regulators will grant prices that will
provide above normal returns on assets — prices that they believe can be realised in
the market — are airports likely to over-invest. Capital market pressures on
privatised airports — many of which are dominated by sophisticated shareholders
such as fund managers and large corporations — should discourage unproductive
expenditure (chapter 7).

However, NNI provisions could encourage inefficient substitution of new capital for
other inputs, given the scope to receive price increases for new investment. WAC
considered that the ACCC’s interpretation of the term ‘investment’ was too
restrictive:
… the narrow definition suggested provides an incentive to airport lessee companies to
select more costly solutions that unquestionably qualify as capital rather than what may

REVIEW OF AIRPORT 233

PRICE REGULATION
 well be more economic short-term investment solutions that are denied because of the
definition. (sub. 21, p. 18)

The ACCC (2000a) indicated that it considered each NNI proposal on its merits and
could allow pass-through into higher prices of incremental operating and
maintenance expenditure flowing from new investment. Appropriately applied, this
approach could avoid biasing expenditure choices between alternative methods of
undertaking a proposal which qualified as new investment. However, it would not
deal with distortions to the choice between an option that would not qualify as new
investment and an alternative approach that would.

Airport operators also may have an incentive to try to convince regulators to allow
them inefficiently high prices for new investment. While for firms in competitive
markets charging inefficiently high prices would result in lower profits, a firm with
market power could increase profits if allowed to follow such a strategy.16 Of
course, there is an incentive to submit ambit claims if the regulatory arbitration
process involves determinations that fall somewhere between the competing claims
of airports and users.

Incentives for airlines

If the airline market were competitive (that is, if airlines had negligible market
power), and if they were targeting similar parts of the market, the interests of
airlines should lead to their agreement to required investment at efficient prices at
the appropriate time. Delays in approving efficient NNI prices would not be in the
airlines’ interests. Hence, the requirement in the pricing guidelines that users with a
significant interest support the new investment and associated charges would seem a
potentially useful criterion for assessing new investment proposals. BARA argued:
The interests of BARA members are promoted by encouraging airports to invest when
it is efficient to do so. No one stands to lose more than the airlines if necessary new
investment is not delivered in a timely manner. It would make no sense for airlines to
support a regulatory system that ultimately failed to deliver an adequate stream of
investment in airport services. (sub. 41, p. 38)

However, there are several reasons why airlines, particularly incumbent airlines,
might have incentives that conflict with efficient provision and pricing of new
investment. If an investment in increasing aeronautical capacity provides benefits to
new entrants, existing airlines will have an incentive to delay that investment. Even
if the existing airlines expect to receive net benefits in a direct sense from such an

16 As noted in chapter 7 there are various other incentives (such as the potential to earn non-
aeronautical revenue) and constraints which will lessen the extent to which airports seek to
maximise aeronautical profits.
234 PRICE REGULATION
OF AIRPORT
SERVICES
investment, the total impact on them may be negative because of the facilitation of
increased competition. The significant reductions in airfares that have occurred on
routes with new entrants indicate the size of these potential costs for incumbent
operators.

The Australian Airports Association (AAA) argued:

Airlines have been utilising the procedural aspects of the ‘necessary new investment’
(NNI) provisions as their primary means by which to undertake regulatory gaming, and
thereby stifle new aeronautical investment activity at airports and consequently stifle
the opportunities available to the new entrant airlines. (sub. 15, p. 4)

BAC observed that its problems were limited to certain projects:

In some instances those projects are projects that provide benefits to new entrant
airlines either greater than the incumbent airlines or to an equal extent. So they’re using
those mechanisms, or the NNI mechanism, to stall those projects that could provide
some benefits to their competitors. (trans., p. 208)

Qantas Airways (sub. 48, p. 21) stated that airport operators’ claims that it had been
involved in gaming were ‘entirely without merit’. It argued that access to terminal
facilities was the only airport barrier to the entry of new airlines and observed that
Qantas had not participated in regulatory decisions regarding investment and
pricing for new domestic terminals. Impulse’s view (sub. 18), discussed below, was
that incumbent airlines had frustrated some airport investment in order to inhibit
new entry.

The ACCC (sub. 36, appendix D) pointed out that the shared nature of airport
terminals has encouraged strategic behaviour by potential tenants trying to minimise
their share of the terminal costs. Such behaviour has been observed in negotiations
regarding the new terminals at Melbourne and Adelaide airports. It is difficult for a
regulator (or an airport) to disentangle the genuine from the strategic aspects of such
claims. Different airlines do have different requirements for terminal and other
facilities, but they are likely to exaggerate these when negotiating airport charges —
even in the absence of a regulator.

The ACCC also noted that there can be incentives for higher-cost operators to try to
raise the costs of low-cost competitors, even if this means increasing their own
costs.17 Virgin Blue observed:
The Adelaide terminal is one which is considerably out of whack with the operating
strategy of Virgin Blue and is significantly higher cost, and that is a very real concern,

17 The higher-cost operator may not be less efficient but may simply offer a higher level of
service as part of its business strategy.
REVIEW OF AIRPORT 235
PRICE REGULATION
 where a major incumbent airline actually might have the capacity to impose a
significant increase in the operating costs of a new entrant airline. (trans., p. 354)

In addition, the incumbent airlines have had a long history of dealing with an airport
operator pricing on a single-till basis. Switching to an NNI regime predicated on
new aeronautical facilities covering their full cost through aeronautical price
increases is a significant shift in approach which might be expected to encounter
some resistance.

Melbourne Airport argued:

The current arrangements fail to provide an environment in which normal commercial
arrangements, between airport operators and airlines, will emerge. Airlines see the
opportunity to involve the ACCC in every matter, as the source of better short-term
financial outcomes than they could achieve through commercial negotiations. They
effectively seek to game the process. In a situation where they have nothing to lose
(other than the asset not being provided at all) why would they do otherwise?
(sub. 7, pp. 32–3)

SACL suggested that this problem was compounded by conflicting objectives of the
regulator.
Under the existing regulatory structure, airlines do not have an incentive to reach a
compromise position during negotiations, as the ACCC’s assessment process of price
notifications may achieve a better outcome for them because of the conflict between the
ACCC’s consumer advocacy and efficiency considerations. (sub. DR62, p. ii)

Such behaviour may have been sensible for airlines in the case of some investments
if they saw airport operators as effectively captured in their financial investment in
an airport, and constrained by current lease and safety requirements to undertake
certain capital expenditures, even if the investments are not financially viable at
current aeronautical prices. For example, WAC (sub. 21) indicated that due to
safety, security, environmental or regulatory requirements, it had undertaken capital
investments that were not approved for NNI pass-through and did not provide an
adequate return at current aeronautical prices.

However, airlines would also have an incentive to object to airports seeking

excessive prices (which could be achievable in the market and profitable for an
airport with market power) for undertaking new investment. The regulator is faced
with determining which airline claims are justified from an efficiency perspective
and which are based on other incentives facing the airlines.

The CAA noted the inevitability of strategic behaviour under regulation:

The regulator also cannot avoid the fact that the regulatory system creates incentives
for ‘gaming’ by the airports and users, in order to influence the price cap to their
respective benefits. (CAA 2001c, p. xiv)
236 PRICE REGULATION
OF AIRPORT
SERVICES
Developing NNI procedures

The DoTRD Pricing Policy Paper (1996), and the PS Act criteria (s. 17(3)) and
associated Ministerial Directions, provide limited guidance for applying the NNI
provisions. NNI was not defined and the extent of investment expenditure included
in the X factors generally has not been forthcoming. Finalised ACCC guidelines
were not published until April 2000 — nearly three years after the first leases
commenced. These guidelines included the threshold matters of defining
‘necessary’, ‘new’ and ‘investment’ and, in applying a cost-based regulation
approach, a variety of other issues, such as cost of capital and common cost
allocations between aeronautical and non-aeronautical services. Hence, while initial
uncertainty was to be expected in applying the new system, this has been
heightened by the lack of guidelines on important parameters.

This may not have been a problem if the intention underlying the DoTRD Pricing
Policy Paper (1996) — that commercial relationships between airports and their
immediate customers would generate negotiated outcomes — had been achieved.
However, in general, successful negotiations on major issues have not been the rule
and regulatory intervention has been continually required.

The ACCC (2000b) indicated that user support would be an important determinant
in its assessment of whether a particular project should be classed as new
investment. If such support were not forthcoming, the ACCC defined new
investment as a change in fixed durable inputs that does not simply seek to replace
natural degradation of capital. WAC (sub. 21) claimed that airport buyers when
bidding for leases assumed that NNI provisions included replacement investment.
However, ACCC (2000b) cited DoTRS in (April 1999) correspondence to the
industry as expecting essential and on-going investment to be funded from ongoing
revenue streams.

As well as considering the meaning of the term NNI, decisions on whether

particular investment should be covered by NNI procedures are also linked to what
was included when calculating the Xs in each airport’s CPI-X formula. If the
starting prices and/or the calculation of the X factors included an allowance for
certain types of investment, then there is a clear case for their exclusion from NNI.
Airport operators purchased leases with full knowledge of the aeronautical starting
prices and the X factors. However, the appropriate expectation for NNI issues is not
so transparent. In contesting NNI applications, airlines have argued that certain
capital expenditure was included in the Xs and hence should not attract NNI cost
pass-through. However, as discussed above, what was contained and what was
omitted from the Xs has never been specified exactly and airport operators appear to
be particularly uncertain in this regard.

REVIEW OF AIRPORT 237

PRICE REGULATION
Northern Territory Airports stated:
Clearly the Government has been remiss in not divulging how the ‘X’ factors were
derived and thus has created an expectations gap between airlines, operators and indeed
the regulator. (sub. 25, p. 15)

The AAA argued:

The ACCC has incorrectly interpreted the scope of NNI provisions to exclude
replacement capital expenditure. The impact of this incorrect definition of NNI by the
ACCC has been the rejection and deferral of investment in replacement aeronautical
infrastructure as airport owners are unable to justify capital expenditure on assets that
do not allow for the recovery of full costs and a return on their capital. (sub. 15, p. 4)

On the other hand, Qantas (sub. 48) considered that current prices already contained
a component for depreciation of existing assets that would allow for re-investment.
However, while existing aeronautical prices may provide some return on
replacement investment, the important issue for generating efficient levels of
investment is the sufficiency of that expected return in the future.18

It is not clear whether any current disincentives to investment (for example, as

evidenced by delays in the resurfacing of a runway at Perth Airport) reflect the
interpretation of NNI by the ACCC, or rather indicate that starting aeronautical
charges were insufficient to justify replacement investment on a dual-till basis.

The application of NNI procedures also has suffered from a lack of clarity in the
original pricing guidelines and subsequent legislative instruments. The pricing
guidelines have no legislative effect and, in applying the NNI regime, the ACCC
has had to operate under the PS Act and related Directions. These Directions do not
refer to certain of the objectives and policy intentions of the prices-oversight
regime, such as encouraging commercial negotiations. However, the ACCC
(sub. 36) expressed concern regarding the deterministic nature of existing
Ministerial Directions for airports and considered that this sat uneasily with the
discretion given to the ACCC under the PS Act.

In arguing for greater use of cost–benefit analysis in evaluating NNI proposals,

Professor Forsyth observed:
At present, the ACCC works to a list of general criteria, which in themselves are
reasonable enough, but which do not lead to a systematic evaluation of investments.
(sub. 5, p. 26)

18 Because it followed a network-wide single-till approach, the FAC undertook significant new
and replacement aeronautical investment even though aeronautical prices may not have
provided the revenue needed to justify them. Profits from non-aeronautical activities made up
the shortfall and enabled the FAC to achieve the accounting rate-of-return objectives imposed
by the Commonwealth Government.
238 PRICE REGULATION
OF AIRPORT
SERVICES
The problems in using the PS Act also are highlighted by the Commission in its
review of the PS Act (PC 2001c). It observes that the PS Act has many deficiencies
as a means for prices oversight and argues that there is little justification for it as a
generic price-oversight mechanism. Thus, in the few cases where any price control
might be deemed necessary, it argued that this would be best implemented via
industry-specific legislation.

Participants’ views

Participants’ views differ significantly regarding the operation of the NNI

provisions. In general, incumbent airlines have been happy with the process, new
airlines have varied in their opinions and the airport operators generally have
expressed dissatisfaction. The views of those involved in framing and implementing
the regulations also differ.

Airlines’ views

The DoTRD Pricing Policy Paper (1996) placed emphasis on user support for new
investment and charges. Among airlines, incumbents and new entrants had
conflicting judgements. BARA stated:
In BARA’s view the current regulation of the privatised airports in Australia has acted
to constrain, to some extent, the market power of the airports while encouraging
efficient investment in airport expansion. (sub. 26, p. 1)

Ansett considered that the regulatory regime provided some incentive for airlines to
oppose excessive prices that otherwise would be lacking.
In the case of the regulated Australian airports, the current economic regulatory regime
provides the Ansett/Air New Zealand Group with an incentive to scrutinise the costs
and prices of airports. This is because the regime allows the airlines to have sufficient
influence over these costs to justify the resources required to scrutinise them …
(sub. 42, p. 38)

Qantas (sub. 48) considered that there was some room for improvement in the
administrative approach to assessing NNI applications, but that airports’ complaints
of the process retarding investment, or of airlines ‘gaming’ the system, were invalid.

Impulse Airlines, then a new entrant into domestic trunk routes, was critical:
Impulse’s introduction in the Canberra marketplace also affords an interesting example
of lack of support for infrastructure development. Canberra International Airport had
tried for 2 ½ years before the arrival of Impulse to gain support from both Qantas and
Ansett to re-develop apron space which had been in its current state for 28 years. It was
not until the introduction of Impulse that CIA felt able to take their case to the ACCC.

REVIEW OF AIRPORT 239

PRICE REGULATION
 Whilst arguments over rates of return were at stake, this is not the key issue. The key
issue is that an airline competitor was able to stymie infrastructure development for 2 ½
years in order to harm a competitor and ultimately the airport. (sub. 18, pp. 2–3)

Impulse (sub. 18) also felt that the price cap had meant that infrastructure quality
had to be compromised at the new Melbourne and Sydney express terminals.

Virgin Blue (sub. 30) (a new low-fare carrier) contended that NNI pass-throughs
outside the price cap should be allowed only for investments that would not be
undertaken without the pass-through. It considered that incremental non-
aeronautical revenues should be considered when examining NNI proposals such as
Melbourne Airport’s domestic express terminal.

Such differing views of two new airlines regarding pricing of the same investment,
indicate the difficulty for the regulator in identifying facts, the targeting of different
market segments, and gaming behaviour. Impulse and Virgin Blue had different
business strategies which may have led to their divergent attitudes to the Melbourne
express terminal, but they may also have attempted to use the regulatory framework
to their advantage. Similar issues arose regarding Virgin Blue’s use of the proposed
Adelaide multi-user terminal.

Impact of NNI provisions on investment and commercial relationships

DoTRS, which coordinated development of the regulatory framework, considered

that the application of NNI provisions had not worked as expected and had hindered
new investment and the development of commercial relations.
At present the level of regulatory intervention is excessive and may in fact be
counterproductive in that it is providing an incentive for regulatory gaming by
stakeholders for commercial advantage rather than allowing market forces to operate
and encourage commercial negotiation. There is also considerable evidence growing
that the regime is creating disincentives to invest in aeronautical infrastructure and
economic efficiency is being adversely affected. (sub. 39, p.19)

It noted the strategic behaviour of airlines in negotiations for the proposed

multi-user terminal at Adelaide and regulatory intervention in the agreement
between Impulse Airlines and Melbourne Airport for use of the domestic express
terminal.

However, the ACCC argued that its involvement had not deterred investment:
The investments undertaken by airport operators to date suggest that the Commission’s
[ACCC] pricing decisions have not deterred investment in airports. In particular the
experience suggests that the Commission [ACCC] has adequately allowed for the risks
facing airport operators. (sub. 36, p. 29)

240 PRICE REGULATION

OF AIRPORT
SERVICES
That new investment has taken place, particularly in the early stages of the new
untried regime, does not constitute proof that the regulatory arrangements have not
had important disincentive effects. Inevitably, airports initially will attempt to make
use of the new regime. It is their response to this experience that will determine
longer-term implications.

In varying degrees, all airport operators have been critical of the NNI regime, in
particular the way in which it has been implemented. Northern Territory Airports
submitted:
We find the ACCC’s comments about its own supervision of new investment at
Australia’s airports at odds with the reality we have had to go through. (sub. 50, p. 17)

It indicated (sub. 50) that because of the cost, delays and uncertainty of the
application of the NNI process, it had deferred a number of investment projects
indefinitely. These included projects approved for price increases by the ACCC but
for which costs (and hence the necessary price increase) had risen during the
extended regulatory process. Because of significant reductions in traffic, these
investments have still not proceeded despite the October 2001 regulatory changes
(which removed the price cap from Darwin Airport).

Melbourne Airport argued:

The necessary new investment arrangements are a makeshift attempt to address the
critical issue of under-investment. The limitations of this approach are apparent
however. Poor administration and the scope for gaming by airlines raise risks and so
costs for airports. Any resulting under-investment is likely to be felt most keenly by
new entrants into the airline business. (sub. 7, p. ii)

Following Melbourne Airport’s exposure to the NNI process when developing a

new common-user terminal, it indicated (trans., p. 156) that, in future, it would
delay new investment until regulatory approval of proposed related price increases
had been received.

BAC (trans., p. 206) indicated that it was following a similar approach, while Gold
Coast Airport stated:
There is no certainty that commercial terms agreed with users will be endorsed by the
ACCC …
Because of this uncertainty GCAL [Gold Coast Airport Limited] has deferred initiating
any necessary new investment until the process becomes more clearly defined. This
decision will mean additional maintenance costs are incurred in the short to medium
term as current pavements are showing signs of stress due to increasing use of heavier
aircraft by both Qantas and Ansett. (sub. 16, p. 12)

REVIEW OF AIRPORT 241

PRICE REGULATION
However, BARA stated that its members found no evidence or rationale for an
investment slow-down by airports and cautioned against taking airports’ complaints
at face value:
… airports might argue that the current arrangements constrain investment as a path to
convincing the Commission that complete deregulation is a better alternative.
(sub. 41, p. 39)

Ansett (sub. 42) stated that despite some initial problems, the relationships between
airlines and airports under the current regulatory arrangements were improving and
are better than those in New Zealand under a less intrusive regime.

The Motor Trades Association of Australia Superannuation Fund identified a more

fundamental problem for investment under regulatory oversight:
Even the best intentioned and informed regulator will make mistakes and proper
assessment of investment proposals take time. Requiring ACCC approval of proposed
price increases or new charges then necessarily creates risk, uncertainty and delay, all
of which increase the costs of investment subject to assessment (and the required rate
of return). These alone would create a disincentive to new aeronautical services
investment. (sub. 22, p. 54)

Australian Airports (Townsville) (sub. 14) indicated that even if investment

eventually takes place, the NNI process was slow and cumbersome and airlines
have proven difficult to deal with. It also considered that price increases needed to
justify new investment had been quickly negotiated with airlines and the local
community at its non-core-regulated Mount Isa Airport (sub. 14).

Encouraging commercial negotiation in establishing prices for airport services was

a key objective of the current price-regulation framework. While there is little scope
for negotiation in applying the CPI-X formula, the NNI test held a clear expectation
that negotiations between airports and their users would often generate acceptable
outcomes for the regulator to approve. The AAA felt that this had not occurred:
It was a clear plank of the Government's policy objectives to encourage the
development of mature commercial relationships between airlines and airports. I think
it’s clear that that hasn’t happened. To some extent it’s understandable it hasn’t
happened because the commercial reasons for failing to reach agreement are just so
strong. (trans., p. 142)

Professor Forsyth (sub. 5) argued that negotiated agreements where users are
willing to pay the costs of investments are preferable to regulatory intervention.
Negotiated agreements use the greater knowledge of the participants rather than the
often imperfect information available to the regulator. However, evidence suggests
that, in the current regulatory framework, the relatively easy recourse to the ACCC

242 PRICE REGULATION

OF AIRPORT
SERVICES
combined with the incentives facing airlines appears to have forestalled the
negotiation process.

The Commission’s review of the National Access Regime (PC 2001a) analyses the
potential disincentives that regulation can have on investment in infrastructure. In
particular, it observes that additional risk is created by the uncertainty about
regulatory decisions and future changes in the regulatory framework, both of which
can have significant impacts on the profitability of long-lived investment.

Compliance costs

There has also been concern, particularly among smaller airports, regarding the cost
to airport operators of the NNI process. Airport operators have also complained of
the expense of even minor projects needing to go through the full NNI process.

Because of concern about setting national precedents, domestic airlines may oppose
even small NNI increases at small airports. Possible international precedents might
give incentives for international airlines (through BARA) to take a similar
approach. Melbourne Airport submitted:
It must also be added that airlines (we suspect at the urging of IATA) often see any
regulatory decision as having the potential to be a global precedent, and therefore often
argue points that they privately concede, have no merit. (sub. 7, p. 33)

The requirement for even small investments to be subject to full NNI reviews if cost
pass-throughs are sought, appears to have added to the regulatory costs and
detracted from the quality of outcomes of the regime. WAC (sub. 21) argued that all
projects were considered equally under the regulatory guidelines regardless of size.
Capital Airport Group (sub. 32) indicated that in its experience even small NNI
projects had been subject to excessively detailed scrutiny.

The potential benefit of the NNI process of preventing inappropriate price increases
while encouraging appropriate new investments to be undertaken, needs to be
balanced against the direct and other costs of the regulation process. In the case of
small investments, potential benefits are very small while the direct regulatory costs
may be relatively high. This is particularly so for smaller airports that are less likely
to be able to group a range of small projects to reduce the size of direct regulatory
costs relative to the investment concerned.

Melbourne Airport (sub. 7) stated that a conservative estimate of the combined cost
of Melbourne and Launceston airports complying with the price-regulation regime
(including quality monitoring) was $500 000 per annum.

REVIEW OF AIRPORT 243

PRICE REGULATION
From an airline perspective, Ansett (sub. 42) submitted that, with the exception of
Sydney Airport, the bulk of its costs under the current regulatory framework were
similar to what they would expect to incur in dealings with suppliers in a
commercial environment.

Evaluation of NNI provisions

For aeronautical services as a whole, the NNI regime has moved prices somewhat
above the single-till approach to pricing used by the FAC. The broad thrust of the X
factors in the CPI-X formula would appear to have been to generate aeronautical
prices that essentially maintained the extent of single-till cross subsidies between
services contained in the FAC prices existing at the start of privatisation. Increases
in non-aeronautical profits gained since then were not factored into the Xs. Price
increases obtained under NNI would have further ameliorated this situation, but
only for the expansion in the capital base involved.

To the extent that the single-till starting prices of the current regime did not allow
for a sufficient return on replacement investment, the exclusion of replacement
investment from the interpretation of new investment means that the current NNI
process could never generate aeronautical prices that covered full aeronautical costs.
If lessees bid on the basis that NNI provisions excluded replacement investment, the
bid prices should have reflected this and, at least for the five years of the initial
regulatory framework, necessary replacement investment should have been
undertaken at existing aeronautical prices. However, in the absence of clear and
transparent statements of what was included in the Xs and the lack of initial detailed
guidelines on the application of the NNI process, it is difficult to determine what
should be expected of airport lessees in regard to undertaking replacement
investment.

As the key interface between airports, airlines and the regulator, the NNI provisions
have not generated the commercial relationships and light-handed regulation
envisaged by the DoTRD Pricing Policy Paper (1996). No doubt some of these
problems reflect teething difficulties of a new regulatory system. Definitions and
procedural guidelines needed to be developed and experience had to be obtained in
working with the new system. The parties needed to adjust to the new environment
— the airports from the FAC network framework and the airlines (operating in a
rapidly changing airline industry) to dealing with fellow private companies rather
than a government enterprise.

Nonetheless, the incentives provided by the regulatory framework suggest some

endemic problems that, if not addressed, will not dissipate with time and
experience. The easy low-cost access to a regulator reduces the incentives for users

244 PRICE REGULATION

OF AIRPORT
SERVICES
to negotiate with airports, while low aeronautical charges mean that airports view
all new investment as a means of obtaining price increases.

Because of the short period in which it has been in operation, whether the NNI
process has stimulated efficient levels of new investment is not clear. Complaints
from most airlines have concerned the aeronautical prices sought to cover new
investment rather than a failure by airports to invest. The observed level of new
investment may still have been forthcoming if base aeronautical prices initially were
set high enough to justify new investment rather than relying on pass-through
provisions to increase prices selectively in return for approved new investment.
Market incentives for airport operators not to restrict output and investment
artificially are discussed in chapter 7. Certainly the NNI approach has generated
significant regulatory intervention with its associated costs and increased regulatory
risk. Given the apparently inefficiently low starting prices, a major area of concern
for investment expenditure is the combined impact of the CPI-X and NNI
provisions on returns for replacement investment.

The combination of the price cap and NNI rules would seem to have avoided any
exploitation of market power that airport operators might have attempted in the
absence of such regulations. However, in the short period in which the regulation
has been operating, the NNI process appears to have created significant regulatory
hurdles for aeronautical investments.

FINDING 8.2

The necessary new investment provisions have not promoted the commercially-
negotiated outcomes that were envisaged by the architects of the regime. This has
been partly due to the need to develop criteria and procedures for necessary new
investment after purchase and for participants to adapt to the very different
business environment following airport privatisation.
However, the observed difficulties also point to some fundamental problems. In
particular:
• the lack of transparency regarding what investment was considered to be
included in the base aeronautical prices and what was to be covered by
necessary new investment, with resultant effects on incentives to invest;
• the incentives for some participants to approach the regulator rather than
achieve commercially-negotiated solutions;
• the high costs of complying with the regime; and
• the regulatory risk due to the uncertainty and delays introduced by the need to
have every investment-related price increase vetted by the regulator.

REVIEW OF AIRPORT 245

PRICE REGULATION
Other cost pass-throughs

As part of the price-cap arrangements, the ACCC is required to allow aeronautical

prices above the cap to cover the direct costs (not including rate of return or
overhead costs) of government-mandated security requirements. The various
Ministerial Directions which have allowed the pass-through have not defined direct
cost. The ACCC has interpreted direct cost to mean costs that would have been
avoided if the security requirements had not been imposed, and to include
depreciation and rate of return on capital. The ACCC has received over 30 security
cost pass-through proposals.

In December 1998, DoTRS made changes to security regulations that placed the
responsibility for international passenger and baggage screening at international
terminals on airport operators. WAC claimed that the ACCC’s initial decision on
allowable rates of return on the capital investment required had created inefficient
results:
The ACCC initially refused to allow airports to earn a return on capital invested to
purchase security equipment of greater than the cost of debt. This created a
dis-incentive for airport operators to invest in the required capital equipment, thus
allowing the security contractors to supply and charge for the equipment at rates which,
presumably did allow for an equity return. (sub. 21, p. 25)

WAC indicated that the ACCC later allowed somewhat higher rates of return.
However, by then a number of airports had committed to external security
contractors to provide the equipment as the return on contractor’s capital embedded
in their charges was fully recoverable in higher aeronautical prices.

The price-cap regulations also allow congestion charges employed as part of an

airport demand management scheme under the Airports Act, to be passed through
the cap. There have been no such charges imposed at the privatised core-regulated
airports. In recent years, government attitudes with regard to the more congested
Sydney Airport have not favoured peak charges, due largely to concerns about their
impact on fares for regional commuters. Issues of congestion pricing and price cap
regulation are discussed further in chapter 10.

8.4 Recent changes to price regulation

As outlined in chapter 3, in response to the significant decline in traffic volumes
flowing from the suspension of Ansett’s operations and international events, the
Government announced substantial changes in price regulation of the privatised
core-regulated airports. For the Phase 1 airports, the regime has been altered nine
months before existing arrangements were due to finish and, for Phase 2 airports, 21
246 PRICE REGULATION
OF AIRPORT
SERVICES
months before. The Government has indicated that further consideration would be
given to these prices-oversight arrangements in the light of developments in airport
prices and the Commission’s final report (Hockey 2001).

The Government directed the ACCC to allow the three Phase 1 airports still
remaining under the current price cap (due to expire in June 2002) increases in
aeronautical charges above the levels currently allowed by the cap. These increases
— treated as one-off pass-throughs under the cap — were 6.7 per cent for Brisbane
Airport, 6.2 per cent for Melbourne Airport and 7.2 per cent for Perth Airport.19
Price monitoring remains for aeronautical-related services at these airports.

For all of the Phase 2 airports, the price cap on aeronautical services previously
scheduled to operate until June 2003 has been removed. For three of these,
Adelaide, Canberra and Darwin, the cap has been replaced by price monitoring of
aeronautical services by the ACCC. This is discussed further in chapter 9.

While the effects of the new regulatory environment will take some time to work
themselves out, there have been some immediate effects on prices under the price
cap which are discussed below. A proper evaluation of the new arrangements will
require a considerably longer period of operation.

Melbourne Airport has utilised the additional 6.2 per cent price cap pass-through
allowed to it under the October regulatory changes to rebalance charges between
domestic and international services. Landing charges will increase by around
11 per cent from the start of 2002. However, for international services, the terminal
charge has been reduced by an equivalent amount meaning there will be no net
increase in aeronautical charges for international flights. The difficult international
aviation climate and the opportunity to increase Melbourne’s and Australia’s share
of the international aviation market were cited (Barlow 2001) as the rationale for
this price discrimination.

Perth Airport also increased aeronautical charges by the full 7.2 per cent of base
period prices allowed by Direction No. 24. Like Melbourne Airport it made no
increase in the charges for international services but increased landing fees for
domestic regular public transport services by around 19 per cent from November
2001.

19 The ACCC has interpreted the percentage price increases allowed by Direction No. 24 to refer
to the prices applying when the airports were privatised in July 1997 and has deemed that the
resultant dollar amount increases include GST. As a result the price increases expressed as a
percentage of current prices have been less than the percentages in Direction No. 24.
REVIEW OF AIRPORT 247
PRICE REGULATION
WAC indicated that the adjustment to charges allowed by Direction No. 24 made
up only a small portion of the revenue lost due to recent declines in aeronautical
traffic:
It merely reflects the fact that there has been an extraordinary drop in airport volumes,
and provides a small pricing adjustment as a partial compensation for that.
(trans., p. 574)

Brisbane Airport instituted a uniform rise in aeronautical charges from November

2001 to achieve its allowable increase.

Melbourne Airport was critical of the manner in which the new regulatory
arrangements were introduced:
In our view in particular the Treasury should make clear the reasons and the
methodology used to arrive at this situation, because otherwise we’re just going to end
up with another situation which is exactly analogous to that debate around what’s in the
value of X … Indeed, all that’s happened really is that regulatory uncertainty has been
increased … (trans., p. 403)

The basis for the one-off price increases above the cap has not been revealed and
this presents similar concerns to those raised above for the determination of the Xs
in the price-cap formula. This may increase unnecessarily regulatory uncertainty for
both airports and their users. Also, establishing good commercial relations between
the parties may be made more difficult if there is lack of transparency in important
elements of the regulatory framework.

8.5 Price regulation of Sydney Airport services

Sydney Airport is the only core-regulated airport that has not been privatised.
Instead, the Commonwealth Government chose to transfer responsibility for its
management and operation under a long-term lease to Sydney Airports Corporation
Limited (SACL), a new government-owned public corporation, in July 1998. The
Government is currently in the process of privatising operation of the airport by
selling all of the outstanding shares in SACL. Following the suspension of Ansett
services and developments in international aviation and finance markets, the sale
process has been deferred until early 2002. Unlike the privatised core-regulated
airports, the price regulation framework at Sydney Airport was not changed by the
Government’s October 2001 announcement.

248 PRICE REGULATION

OF AIRPORT
SERVICES
Regulatory and operating environment

Since it took over the operation of Sydney Airport, SACL has been subject to prices
notification for aeronautical services and monitoring for aeronautical-related
services under the PS Act for the same groups of services as the privatised airports
(chapter 3). It has not had a price cap imposed but, like all declared companies, it
must inform the ACCC of proposed price increases for notified aeronautical
services. This regime is more extensive than the price regulation that applied to the
FAC. Additional elements include the introduction of price monitoring for some
services, the inclusion of necessary new investment as a justification for price
increases and the monitoring of service quality under the Airports Act.

Sydney Airport is now operating as an individual airport facing some degree of

competition from other major airports, rather than as part of an Australia-wide
network of FAC airports. Because of its relatively high traffic volumes, government
restrictions on airport hours and number of aircraft movements, and its relatively
small land area, Sydney Airport suffers from greater pressure on its capacity than
other Australian airports, particularly at peak times.

The ACCC pricing decision

As mentioned above, in January 1997 the FAC (with ACCC approval) increased
aeronautical charges at Sydney and three other major airports (chapter 2). These
were the last price increases before the FAC was disbanded. In the case of Sydney
Airport, the FAC made a further price notification which resulted in a large
restructuring of aeronautical charges from October 1998, but one that did not lead to
an overall increase in revenue. Landing fees were cut by around 50 per cent but
international terminal charges were increased by up to 30 per cent, leading to a net
increase in aeronautical charges for international services and significantly lower
charges for domestic flights. In addition, the peak-period landing surcharge of $250
(applicable to all aircraft) was abolished amid Government concerns about its
impact on regional airlines.

The major objective of this restructuring was for charges to international aircraft to
reflect more closely the cost of terminal facilities used by them. However, the
resultant landing charge of $2.92 per tonne (MTOW) was only a little over half of
those at the other core-regulated airports. Hence, while the price restructuring
resulted in reasonable returns on investment in the international terminal, it also
created very low returns from aeronautical revenue on other aeronautical
investment.

REVIEW OF AIRPORT 249

PRICE REGULATION
Against this background, in December 1999, SACL presented a draft aeronautical
pricing proposal, containing substantial price increases, to its customers and the
ACCC for consultation purposes. The draft subsequently was modified
(SACL 2000) and in October 2000 was submitted to the ACCC for assessment
purposes. The new proposal remained substantially the same as the original draft
and sought average price increases of around 130 per cent.20 The ACCC carried out
further extensive consultations before releasing a draft decision in February 2001
(ACCC 2001h). Following further public consultation, and the formal price
notification by SACL on 28 March, a final decision was released in May
(ACCC 2001i).

In response, the ACCC decision presented aeronautical charges to which it would

not object. These charges generated revenue increases of 87 per cent of SACL’s
proposal. This involved a 14 per cent reduction in SACL’s proposed increases for
runway and international terminal fees, but no changes to the proposals for other
less significant charges. The main factors generating lower approved price increases
were a lower valuation of land, a smaller allowable rate of return on capital, lower
depreciation and a reduction in allowable operating costs. Several of these issues are
discussed below.

The SACL proposal was based on the ACCC’s cost building-block approach to
determining allowable prices. This involves establishing the efficient projected level
of various cost components (for example, operating costs, depreciation and return
on capital) and then determining prices which, based on projected volumes, would
provide the revenue needed to cover these costs.

Hence, while for the first five years of operation aeronautical price regulation at the
other core-regulated airports is being based on incremental changes (determined by
the price cap and cost pass-through provisions) to FAC prices, Sydney Airport
regulation has now moved to a complete rebasing of prices. In particular, its
aeronautical prices are now freed from the inter-service subsidies that were part of
the FAC approach to pricing. Because non-aeronautical revenue traditionally had
provided significant profits at Sydney Airport, large aeronautical price increases
were required to generate adequate profits on aeronautical activities on a stand-
alone basis.21

20 SACL made further adjustments to various cost components of its proposal during the ACCC’s
consultation process.
21 ACCC (1998h) indicated that, in 1998-99, Sydney Airport was forecast to recover only
88.6 per cent of aeronautical costs, excluding a rate of return on capital. Higher depreciation
charges on significant quality-enhancing capital investment undertaken between 1998 and 2000
also required higher prices.
250 PRICE REGULATION
OF AIRPORT
SERVICES
In determining allowable aeronautical prices for SACL, the ACCC modelled
allowable revenue over a five-year period.22 It then generated a constant nominal
price for five years which was forecast to provide SACL with a net present value of
cash flows equivalent to this projected allowable revenue stream. The ACCC stated:
The Commission [ACCC] considers that — in the event that the current regulatory
arrangements remain in place — it would be appropriate for prices of aeronautical
services at Sydney Airport to be subject to further review at the end of five years
following the introduction of the new charges. (ACCC 2001i, p. 203)

The current regulatory framework allows for applications for NNI cost pass-
throughs into prices during this period.

While the Commission does not see the need for a detailed commentary on the
ACCC decision in this report, some aspects of the decision and related issues are of
particular relevance to future price regulation at Sydney Airport and any other core-
regulated airports for which ongoing price declarations or monitoring might be
considered appropriate. These are discussed below and in appendix F (land
valuation).

Single till versus dual till

In setting aeronautical charges, the FAC traditionally used a network-wide single

till.23 Its 1996 and 1998 pricing proposals for Sydney Airport moved towards an
airport-specific approach but continued with the single till. While the DoTRD
Pricing Policy Paper (1996) indicated that, with respect to the privatised airports,
the Government was not mandating a single-till approach to setting aeronautical
charges, this was in the context of a price cap essentially based on single-till starting
prices which only changed gradually to allow for (dual-till) NNI increases or to
accommodate CPI-X adjustments.

SACL’s aeronautical price proposal, in arguing for an average 130 per cent increase
in aeronautical charges, moved to a dual-till approach. Costs and revenues of non-
aeronautical services (including aeronautical-related services) were not considered
when justifying the proposed prices. The ACCC draft report (ACCC 2001h) largely
accepted the SACL approach but did take into account returns on certain
aeronautical-related services in determining aeronautical prices to which it would

22 In undertaking this exercise the ACCC used projections of traffic volumes and productivity
growth and made some allowance for increases in SACL’s input prices.
23 Partly reflecting its high land values, Sydney Airport tended only to achieve rates of return on
assets similar to the network average. Hence its charges were not cross subsidising other
airports. The January 1997 price increases, which applied only to the five major airports, lifted
Sydney’s rate of return by around 1 per cent.
REVIEW OF AIRPORT 251
PRICE REGULATION
not object. Following a Government Direction (No. 22, April 2001) not to take any
non-aeronautical services into account, the ACCC final report (ACCC 2001i)
adopted the conventionally accepted dual-till approach proposed by SACL.

To the extent that locational rents create above-normal returns in non-aeronautical

activities, use of a single-till approach is likely to distort investment decisions in
both aeronautical and non-aeronautical services. Because any above-normal profits
in non-aeronautical services would be totally applied to reduce aeronautical prices,
additional aeronautical investment would receive below-normal returns.24
Similarly, airport operators would have reduced incentives to achieve additional
locational rents — rents that reflect consumers’ valuation of certain
non-aeronautical services being available at an airport (appendix C).

In its draft report, the ACCC argued that SACL possessed market power in the
provision of certain non-aeronautical activities, such as car parking, and that
above-normal returns in these activities should be subtracted from allowable
revenue in calculating aeronautical prices. In response to Direction No. 22, the
ACCC reversed this approach in its final decision.

The Commission has several concerns with the ACCC’s draft report approach. First,
it is important to establish that any above-normal returns in non-aeronautical
services reflect the exercise of market power and not just locational rents (transitory
or long-term) or short-term capacity shortages. If it is the latter, then confiscating
the returns is likely to cause inefficient investment and operating decisions,
particularly where a facility, such as Sydney Airport, is capacity-constrained.

Second, even if excess profits on non-aeronautical services reflect market power, it

would be better to address the issue directly by constraining this market power
rather than requiring cross subsidisation of aeronautical charges — especially when,
because Sydney Airport is slot constrained, lower aeronautical charges largely will
benefit airlines holding slots, not their customers. While the ACCC does not have
the power to object to non-aeronautical charges, it can recommend that they become
services subject to notification or monitoring. However, this option may be
constrained in the future by any Government commitments to bidders for SACL
regarding the regulatory framework for Sydney Airport. In this regard, it is
important that the Government’s intentions regarding the regulatory treatment of
non-aeronautical revenues from services in which the operator may have market
power be made clear before the sale process is completed.

24 Under a dual-till system, to the extent that aeronautical investment has complementary impacts
on non-aeronautical profits, the airport operator would take these into account when assessing
whether to undertake this investment.
252 PRICE REGULATION
OF AIRPORT
SERVICES
The ACCC (sub. DR55) contended that its draft SACL decision was for discussion
by participants and not part of its final decision and, hence, was not relevant to the
current inquiry. However, in reaching its final decision based on giving special
weight to the Government’s policy set out in Direction No. 22, the ACCC argued:
The Commission [ACCC] considers that this policy may differ from the approach that
would be appropriate in order to give effect to the objectives set out in chapter 2 [of
ACCC (2001i)]. (ACCC 2001i, p. 94)

This suggests that the ACCC still considers that inclusion of certain non-
aeronautical revenue in determining aeronautical prices has merit.

The implementation of dual-till pricing at Sydney Airport means that the outcome
of price regulation now differs markedly from that at the privatised airports. The
dual till removes the constraints on new or replacement aeronautical investment that
may occur under a pure single till or the current legacy of the FAC’s single till.
These issues are discussed further in chapter 10 and appendix C.

Cost of capital

Particularly for an airport with high asset values like Sydney, returns on capital are
a significant component of costs. The quantum of returns to capital allowed in the
building block approach is a function of asset values and the allowable rate of
return. There are a number of complex and strongly debated theoretical and
empirical issues to be resolved in estimating appropriate rates of return. This
potentially makes for a significant range of possible rates to be applied by the
regulator.

In the Sydney Airport decision, the ACCC’s assessment of SACL’s real weighted
average cost of capital was 6.8 per cent compared to SACL’s proposal of
7.75 per cent — resulting in a reduction in allowable annual revenue (on the
ACCC’s assessed asset base) of a little over $13 million, leading to aeronautical
prices about 6 per cent lower than sought by SACL.

The range of potential uncertainty surrounding estimates of the appropriate cost of

capital illustrates the potential disputes or errors in determining cost-based prices.
Hence, prices set by the regulator may well be inefficient — either too high or too
low. Even if these highs and lows balance out in the long run (as they might if the
regulator does not bias the prices one way or the other), the added risk to the
regulated firm will now require higher rates of return if efficient outcomes are to be
achieved. If a regulated firm has market power it will be able to offset periods of
inefficiently low regulated prices by taking advantage of the periods of high prices.
However, if it does not have market power it would not be profitable for it to avail

REVIEW OF AIRPORT 253

PRICE REGULATION
itself of the periods of excessive prices. The potential costs of different directions of
error in regulatory prices are discussed in the Commission’s report on the National
Access Regime (PC 2001a).

Land valuation

Although Sydney Airport is a relatively small airport (in terms of area) by

Australian capital city standards, its close proximity to a highly-valued capital city
centre means that the value of the airport site is also very high. Markedly different
views have been put as to how airport land should be valued in setting aeronautical
prices and the resulting significant differences in valuation would have an important
impact on cost-based aeronautical prices (appendix F).

BARA (2000) argued that because the airport would continue to operate as an
airport regardless of the return on land, its existing land should be valued at an
opportunity cost of zero (although any new land acquisitions should be included at
purchase price).25 On the other hand, SACL (2000) argued that, being a
non-specialised asset, the relevant valuation of aeronautical land for cost-based
price setting was market value. Efficiency and distributional arguments have been
presented for and against these approaches (appendix F).

Clearly, a zero land value fails to provide appropriate signals to Government

regarding the potential value of relocating to an alternative airport site and does not
give the airport operator (and airport users) the correct signals and incentives for
making optimal use of existing airport land. If, due to inappropriate regulatory
valuation of land, increased returns obtained from more efficient land use result in
regulatory price reductions, there will be no incentive for airport operators to seek
out such opportunities. If the potential efficiency benefits of private operation are to
be realised in a cost-based regulation framework, an appropriate value and return on
land is required.

The ACCC (2001i) considered SACL to be highly constrained in its land use, both
in regard to sale of the airport and within airport use. It argued that this limited the
relevance of alternate use land values in setting regulated prices. In addition, it
identified a number of complications in determining the opportunity cost of land for
setting efficient aeronautical prices (appendix F).

As a result, the ACCC adopted an approach of establishing the historical cost of the
various purchases of land that went into making up Sydney Airport and then

25 A similar view regarding the appropriate valuation of land at all airports was put by IATA
(sub. DR56). It argued that land should be treated as a non-interest bearing investment with no
cost being borne by aeronautical users.
254 PRICE REGULATION
OF AIRPORT
SERVICES
indexing these costs to the present day using the CPI. In effect, the value of land
purchases dating back as far as 80 years was maintained relative to an average
consumer goods basket. Any new acquisitions and switches between uses of airport
land will be valued at current prices at the time of purchase and then indexed by the
CPI. The ACCC considered that this approach would provide appropriate signals
for efficient use of the existing site and for new aeronautical investment. However,
while this is an arithmetically deterministic approach, it essentially provides an
‘arbitrated’ solution to land values which lies somewhere between the competing
claims.

The ACCC (sub. DR55) argued that the indexed historical cost approach was based
on economic principles, in particular, that it provided appropriate incentives for new
investment. For marginal land purchases, historical cost initially will equate to
opportunity cost but, over time, that link will be lost if land prices move
differentially from the CPI. Moreover, historical cost will not reflect opportunity
cost for transfers of existing land between aeronautical and non-aeronautical uses
nor will it represent the economic value of the current site. For the existing site, the
use of indexed historical cost to value land for price regulation renders false
precision to an inherently arbitrary valuation. As noted by Dr Gannon for the
ACCC:
Without a satisfactory empirical methodology for estimating opportunity cost,
alternative bases of valuation (such as indexed historical cost) are obliged to be adopted
by default. These are poor surrogates as they are unlikely to bear a reliable relationship
to the underlying opportunity cost. (sub. DR55, attachment B, p. 1)

In the case of Sydney Airport, the paucity of data on the purchase price of the
airport site makes indexed historical-cost-based estimates particularly problematic.
The ACCC (2001h) claimed that an important advantage of historical cost was that
it was readily identifiable. However, SACL (2001) observed that the prices paid for
a significant portion of the airport site were not known. As a result, the ACCC
(2001i) had to apply price data available for a portion of the site to estimate the
historical cost of the remaining land.

The ACCC also was critical that the Commission did not provide a more detailed
examination of implementing the opportunity cost principle. Dr Gannon argued:
… there are also formidable methodological difficulties in practical empirical
application of the principle to estimation of the value of the total land at a major capital
city airport site. Unfortunately, these practical difficulties are not discussed in the PC
Draft Report. (sub. DR55, attachment B, p. 1)

The Commission considers that, realistically, the current inquiry can examine only
the broad principles of price setting and regulation rather than undertake detailed
analysis of the numerous individual elements such as land valuation. If detailed

REVIEW OF AIRPORT 255

PRICE REGULATION
price regulation is imposed, then the latter is the role of the price regulator. While
recognising the deficiencies of an historical cost approach, the ACCC has argued
(sub. DR55) that it had limited time to consider the complex issue of measuring
opportunity cost of Sydney Airport land. Nonetheless the SACL draft pricing
proposal, containing a current cost land valuation, was originally released in
December 1999.

In estimating opportunity cost of Sydney Airport land it is important to separate the

land value from other external costs and benefits of moving to another site, raised
by the ACCC (ACCC 2001i; sub. DR55). Regardless of perceived constraints that
might be placed on the use of airport land, the site’s opportunity cost for the
community remains its value in the best alternative use. Pricing based on valuations
other than this will mean that signals regarding consumers’ willingness to pay for
the efficient cost of using the Sydney Airport site for airport services will be
obscured. If it is desired to include various external costs and benefits in setting
prices at Sydney Airport, this should be done explicitly rather than by allowing
these to confuse the measurement of opportunity cost of the site itself. In regard to
this issue, the CAA considered that its price regulation role did not extend to
broader planning issues:
A full analysis of the demand and costs of airports would need to take into account the
wider externalities created by airport activities … The CAA believes that it is important
that these costs and benefits be properly taken into account in decisions to expand
capacity. However, taking direct account of these externalities is not one of the CAA’s
statutory objectives as economic regulator of airports. This is the role of the planning
system. (CAA 2001c, p. xii)

In addition, other activities in the economy are typically priced and operated on the
basis of obtaining returns on the opportunity cost of the land they use. If pricing of
aeronautical services fails to take account of such returns, then demand for those
services (and demand for other inputs used to produce them) will be inefficiently
raised relative to other activities.

For an airport with significant excess demand, the lower production-cost-based

aeronautical prices resulting from using historical cost land values rather than
opportunity cost would give particularly inappropriate price signals. Before the
suspension of Ansett operations it is possible that the aeronautical prices that would
efficiently allocate available capacity at Sydney Airport at certain times of the day
would have already significantly exceeded production cost charges based on any
feasible land valuation method.

Valuation of Sydney Airport land also raises distributional issues. As the airport is
owned by the Commonwealth Government, if land is under-valued and this is
reflected in pricing, there will be a redistribution (particularly of peak-period rents)

256 PRICE REGULATION

OF AIRPORT
SERVICES
from the taxpaying community at large to airline owners or airline passengers.26 As
noted by SACL:
In the case of Sydney Airport, which remains Commonwealth-owned, the capture of
any rents by the airport is likely to have social equity benefits as compared with their
capture by airlines. Such rents could then be used for funding health, education,
transport etc, rather than be distributed to the predominantly foreign owners of SACL’s
airline customers. (sub. 27, p. v)

Given that the Government is in the process of selling Sydney Airport, it would
seem preferable for aeronautical prices to reflect appropriate land values now and
hence be fully incorporated in the sale price received by the community, rather than
be adjusted later and possibly provide windfall profits to the private operator.

FINDING 8.3

If regulation of Sydney Airport aeronautical charges continues to involve prices set

by the regulator on a production-cost basis, aeronautical land should be valued at
its opportunity cost rather than at indexed historical cost.

Price restructuring

In its 1996 and 1998 price notifications for Sydney Airport, the FAC made some
restructuring of aeronautical charges in an effort to align more closely revenue from
particular aeronautical services with the costs of those services. These changes were
relatively simple and involved increases in charges for use of the international
terminal while (in 1998) making offsetting large reductions in landing charges. In
not objecting to these restructurings, the ACCC strongly supported the move
towards user-pays pricing in order to provide appropriate signals for efficient use of
airport services and for new investment decisions.27 BARA, Ansett and Qantas did
not oppose the 1998 changes to price structures.

In its October 2000 price proposal, SACL (2000) made further changes aimed at
aligning prices with costs and also to improve the efficiency of use of congested
facilities. The Commission considers that such developments in pricing offer
important improvements in the efficiency of use of airport facilities, particularly at
an often congested facility like Sydney Airport. More flexibility in pricing and
operations should help bring forth the greater efficiency benefits from privatisation.

26 Some of the peak period rents that accrue to airlines may have supported flights between other
cities on which yields may cover marginal, but not average, costs.
27 The ACCC objected to the quantum of the proposed 1996 FAC price increases but not the
concept of restructuring prices.
REVIEW OF AIRPORT 257
PRICE REGULATION
Specific charges for use of facilities, such as aerobridges, provide airport users with
the appropriate information for making efficient decisions about consuming
particular services. Time-based charges for use of scarce capacity, such as apron
space, dissuade excessive use of the facility and encourage more efficient methods
of operation, ration existing capacity to those who value it most and provide
appropriate investment signals to the airport operator to expand capacity where
possible.

SACL also changed the basis of the international terminal charge from a per tonne
(maximum take-off weight (MTOW)) per landing charge to a per passenger levy,
and the runway charge from per tonne (MTOW) per landing to a per tonne
(MTOW) per movement (landing and take-off). The switch to a per passenger
charge reflected the view that terminal costs were more directly related to passenger
numbers. As noted below, with airline agreement, other charges for international
airlines have also been switched to a per passenger basis. By moving to per
passenger charges, SACL effectively has transformed the terminal charge to a short-
term marginal cost to airlines when considering attracting additional passengers. In
turn, this may affect the extent of airlines’ price discrimination aimed at attracting
marginal passengers as the relevant charges now vary directly with passenger
numbers rather than being fixed imposts per aircraft of a particular size.

The ACCC did not object to any of SACL’s proposals aimed at improving the
efficiency of the price structure.

In August 2001, the ACCC did not object to a revenue neutral proposal by SACL to
institute a per passenger service charge for international passengers to replace a
series of existing MTOW-based charges covering landing and take off, terminal use
and security services. However, for domestic services, the ACCC did object to a
proposal to replace the existing aircraft weight-based charges for aircraft
movements and security services with a per passenger fee. It considered there were
concerns regarding the impact of the change on new airlines which currently had
more passengers per unit of aircraft weight than the incumbent operator. However,
the ACCC indicated that it was not able to address these concerns adequately in the
21-day statutory period available for considering pricing proposals under the PS
Act.

SACL (2000) has also indicated that it is considering the possibility of introducing
time-of-day charges and the ACCC has encouraged the development of peak-period
pricing for Sydney Airport.28 However, in the absence of peak pricing, the ACCC’s

28 Under its conditions of use of Sydney Airport, SACL has, at its discretion, introduced discounts
for airlines introducing new services during off-peak periods and for rescheduling services
from a peak to an off-peak period.
258 PRICE REGULATION
OF AIRPORT
SERVICES
discounting of SACL’s proposed prices exacerbates the problem of excess demand.
To the extent that there already is excess demand for peak-period slots, the resultant
smaller increase in charges is likely to benefit airlines holding those slots rather than
be passed on to passengers travelling at those times.

FINDING 8.4

The aeronautical price increases implemented at Sydney Airport place its pricing
on a fundamentally different and more appropriate (dual-till) basis than that at
other core-regulated airports. The significant range of possible outcomes for a
number of cost parameters (for example, land values, cost of capital etc) indicates
the imprecision of regulatory price setting and the potential for inefficient prices
being established by regulation.
If significant excess demand returns at Sydney Airport at peak times,
production cost-related or rate-of-return based regulation would be likely to
constrain the setting of efficient aeronautical charges for those times.

REVIEW OF AIRPORT 259

PRICE REGULATION
9 Review of airport regulation: price
monitoring, quality monitoring and
access regulation

Price-cap and prices-notification arrangements for aeronautical services provided by

core-regulated airports are evaluated in chapter 8. This chapter evaluates another
element of the price regulation of airport services — price monitoring of
aeronautical and aeronautical-related services. The quality of service provision post-
privatisation, and ACCC monitoring of quality (established as a complement to
price regulation), are also examined.

Airport services are subject to access regulation and general competition law.
Although access regulation does not directly regulate prices of airport services, it
does provide for the regulation of terms and conditions, including prices, of access
to airport services. This chapter also analyses the application of these provisions to
core-regulated airports. General competition law has not been applied to airports to
date (see chapter 11 for further discussion of these provisions).

The criteria for efficient regulation (outlined in chapter 4) are used to evaluate price
monitoring, monitoring of service provision quality, and access regulation.

9.1 Price monitoring of airport services

Prices of some services not subject to the price cap (designated as ‘non-
aeronautical’ services) were monitored at core-regulated airports following the sale
of airport leases. These arrangements changed for some airports in October 2001,
with monitoring of aeronautical services replacing price caps at Adelaide, Canberra
and Darwin airports, and monitoring discontinued, along with price caps, at the
other core-regulated Phase 2 airports.

Monitoring of aeronautical prices

As noted in chapter 3, the changes to prices oversight arrangements announced in

October 2001 included the introduction of monitoring of previously price-capped
aeronautical services at Adelaide, Canberra and Darwin airports. The monitoring
REVIEW OF AIRPORT 261
REGULATION
process by the ACCC will be the same as that applying to aeronautical-related
services (see section below).

Since October, all three of the above airports have increased their airport charges
(chapter 2). As noted in chapter 5 in relation to the exercise of market power, the
effects of these changes will take some time to become fully apparent.

Under current monitoring provisions of the Prices Surveillance Act 1983 (PS Act),
the ACCC could respond to price changes by, for example, investigating some price
increases. The relevant Commonwealth Minister could then direct the ACCC to
undertake an inquiry. However, as noted in chapter 3, the Commonwealth
Government has said that it will take into account price changes, together with the
Productivity Commission’s final report, when considering future airport prices
oversight arrangements.

Monitoring of aeronautical-related prices

Aeronautical-related services at core-regulated airports, including Sydney Airport,

were never subject to the price cap but were subject to price monitoring following
the granting of leases (chapter 3). Aeronautical-related services include aircraft
maintenance sites and buildings, aircraft refuelling, and check-in counters and
related facilities (for a comprehensive list see chapter 3).

Changes to price regulation at airports in October 2001 included the removal of

price caps and price monitoring of aeronautical-related services at several Phase 2
core-regulated airports (Coolangatta, Alice Springs, Hobart, Launceston and
Townsville).

In implementing prices oversight arrangements in 1998 for Phase 2 airports, the

Treasurer explained why price monitoring, pursuant to s. 27A of the PS Act, was
considered to be necessary:1
Price monitoring will allow the ACCC to collect data where the airport operator may
have scope to exercise market power but where coverage of the services under the more
formal price cap arrangements is not considered warranted. Any abuses of market
power detected through the prices monitoring arrangements will be the trigger for
consideration of stricter forms of prices oversight. (Costello 1998)

The ACCC is responsible for monitoring aeronautical-related services. It is directed

by the Minister to monitor prices, costs and profits (s. 27A, PS Act) and report to
the Minister and the public on that monitoring (s. 27B). Guidelines set out the
regulatory information requirements for airport operators (ACCC 1998f).

1 A similar comment was made regarding Phase 1 airports (Costello 1997).

262 PRICE REGULATION
OF AIRPORT
SERVICES
If monitoring indicates that further investigation is required, the Minister can direct
the ACCC to undertake a public inquiry or approve the ACCC holding an inquiry
(s. 18). Powers for the ACCC to obtain information for monitoring purposes,
including inquiries, are provided under s. 32, together with penalties for non-
compliance.

Evaluation of aeronautical-related price monitoring

The ACCC reports annually on its monitoring of aeronautical-related services at

core-regulated airports. The ACCC has not considered that it need investigate prices
of any aeronautical-related services, with the exception of fuel throughput levies
(appendix E). In other words, airport operators do not appear to have been
exercising market power in relation to most of these services. This may explain why
the present inquiry has not received many comments from participants regarding the
operation of price monitoring, other than those relating to fuel throughput levies.

The price monitoring process, including the public dissemination of information,

might have eased any community concern about the potential for operators of
privatised airports to exercise market power, and provided assurance that the market
is functioning appropriately. The ACCC commented:
Realistically from time to time there are likely to be areas of the economy where there
is considerable public concern about particular pricing outcomes. Government is likely
to want to respond to these community concerns. In this situation a price oversight
power is required that allows Government to respond. Price monitoring which requires
the firm to provide specific cost, profit and price data at regular intervals can be used in
the first instance. (sub. 10 to PC 2001c, p. 38)

Price monitoring of aeronautical-related services requires that certain financial

information on airport operations be made publicly available, thereby facilitating
the transparent reporting of airport performance. While this may have encouraged
competitive benchmark comparisons and affected behaviour of some airport
operators, this inquiry has not been provided with evidence on these matters.

As noted above, the ACCC, in its monitoring role, has identified no particular
monopoly pricing problems apart from the imposition of fuel throughput levies. The
ACCC claimed that such levies represented an abuse of market power on the part of
airport operators, although this was disputed by airport operators. It recommended
that refuelling services be subject to greater prices oversight (appendix E). Although
not raised in its regulatory reports, the ACCC (sub. 36) also expressed concern
about another monitored aeronautical-related service provided by airport operators
— car-parking charges.

REVIEW OF AIRPORT 263

REGULATION
In evidence to this inquiry, the Board of Airline Representatives of Australia
(BARA) considered that price monitoring was ineffective in controlling the
‘monopoly power’ of airports, citing increases in fuel throughput levies and staff
car-parking charges as examples of the exercise of monopoly power that price
monitoring had failed to curtail (sub. 41). (See chapter 6 for a discussion of this
issue.)

The preparation of standard information on prices, costs and profits (as set out in the
ACCC guidelines) does not appear to have been particularly costly for airport
operators, because the information requirements are not large. Nonetheless, Sydney
Airports Corporation Limited (SACL), operator of Sydney Airport, commented:
While the impact of prices monitoring is far less troublesome when compared to the
price approval process, it nevertheless does entail compliance costs for both airport
operators and the ACCC and experience to date suggests that such cost has generated
no net public benefit. (sub. 27, p. 49)

Compliance costs are likely to have been higher where airport operators have been
required to comply with additional information requests from the ACCC for its
public review of fuel throughput levies. Costs have also been incurred by the ACCC
in administering the price monitoring process, particularly when undertaking its
public review of fuel throughput levies.

One contentious issue arising from the current price monitoring process is the
potential for regulatory creep, that is, the potential for transferring aeronautical-
related services from outside the price cap to within it. Regardless of whether
additional regulation was warranted (chapter 6), uncertainty has been created for
airport operators in not knowing whether monitored services would remain outside
the price-cap basket. Moreover, any changes in Commonwealth Government policy
relating to the price-cap basket that result in direct adverse consequences for
particular parties raise issues of sovereign risk.

The issue of the imposition of fuel throughput levies by airport operators illustrates
the potential for such effects and its ramifications (appendix E). The ACCC, in
exercising its monitoring power under the PS Act, and bearing in mind the
Treasurer’s 1998 press release (Costello 1998), undertook a review of the
imposition of fuel throughput levies by two airport operators. A discussion paper
was released, submissions were sought, additional information was requested from
the relevant airport operators, a report was prepared with conclusions and
recommendations, and the report was released to the public and submitted to the
relevant Minister. Thus the review met many of the requirements for an inquiry as
set out in the PS Act, although the ACCC did not conduct a formal ‘inquiry’ under
the Act.

264 PRICE REGULATION

OF AIRPORT
SERVICES
The review recommended that refuelling services, including fuel throughput levies,
be transferred from price monitoring to inclusion within the price cap, even though
many airport operators had expectations that fuel throughput levies were to remain
outside the price cap. The Commonwealth Government has yet to respond to the
recommendation.

The ACCC approach to car parking and other price monitored aeronautical-related
services at Sydney Airport, in its draft response to SACL’s proposal to increase
aeronautical prices, also illustrates the potential for services subject to price
monitoring to become subject to stricter price regulation. Although the ACCC did
not recommend that any of these services become subject to prices notification, it
noted that, in the absence of a detailed analysis of aeronautical-related services at
Sydney Airport, ‘above normal’ revenues from aeronautical-related services should
be taken into account in determining allowable revenue from notified aeronautical
services (ACCC 2001h).2 However, the ACCC did not present evidence of abuse of
market power on the part of SACL in relation to any of these services, despite the
Treasurer having specified that abuse of market power would be the trigger for
stricter price regulation (see above). The ACCC moved away from this draft
decision in its final decision, following a Commonwealth Government Direction
that revenues from these aeronautical-related services should not be taken into
account (chapter 3).

9.2 Monitoring of quality of service provision

Quality monitoring was established as a complement to price regulation at core-
regulated airports. Under price-cap arrangements (and prices notification at Sydney
Airport), airport operators may have an incentive to reduce service quality as a
means of cutting costs. Several inquiry participants noted that, for this reason,
monitoring of service quality by a regulator is essential (for example, International
Air Transport Association, sub. 9; Forsyth, sub. 5; Qantas Airways, sub. 48).

As noted in chapter 3, changes to airport price regulation were announced in

October 2001. While all price regulation was removed at several core-regulated
airports (Coolangatta, Alice Springs, Hobart, Launceston and Townsville), quality
of service monitoring under the Airports Act 1996 continues to apply. Several other
core-regulated airports which are now subject to lighter-handed regulation
(monitoring of aeronautical services rather than price caps) also remain subject to
quality monitoring (Adelaide, Canberra and Darwin).

2 As noted in chapter 3, aeronautical-related services at Sydney Airport are monitored under

Direction No. 25 (replacing Direction No. 21), pursuant to s. 27A of the PS Act, October 2001.
REVIEW OF AIRPORT 265
REGULATION
Quality of service provision by airport operators

Monitoring by the ACCC of the quality of services provided by airport operators is

described in chapter 3. In particular, until the regulatory changes in October 2001,
quality of service monitoring did not apply to any services and facilities that were
not subject to price regulation; for example, domestic terminals that were operated
by airlines. Since October, the quality of particular services and facilities3 is being
monitored, under the existing Airports Act, at the five Phase 2 airports which are no
longer subject to price regulation.

The ACCC has reported on the quality of services provided by airport operators at
Melbourne, Brisbane and Perth airports (Phase 1) since 1997-98, and Sydney
Airport since 1998-99. Phase 2 airports were not required under the Airports
Regulations 1997 to provide quality of service information to the ACCC until the
end of 2000-01. The ACCC report is yet to be published.

According to the ACCC, in 1999-00 quality of service for the three Phase 1 airports
and Sydney Airport was ‘generally satisfactory’. Brisbane Airport appeared to rate
highest, with airport users (including airlines and passengers) ‘very satisfied’ with
most of the services and facilities provided. For example, passengers rated waiting
time at check-in facilities as ‘good’ to ‘excellent’, and airlines rated the quality of
runways, aprons and taxiways as ‘satisfactory’ to ‘excellent’. Quality remained
comparable to previous monitoring periods (ACCC 2001b).

The ACCC reported that Perth Airport users were satisfied with most of the services
and facilities provided. However, the ratings varied substantially for some services.
For example, the standard of runways was rated from ‘very poor’ to ‘good’. There
were marginal declines in the ratings from 1998-99 (ACCC 2001d).

At Melbourne Airport, users were also satisfied although there was a decline in
airline satisfaction, for example, with regard to gate allocation (ACCC 2001c). The
report for Sydney Airport was similar — users generally were satisfied although
airlines were dissatisfied with some services and facilities. Six airlines using
Sydney Airport rated gate availability as ‘very poor’ to ‘poor’, and seven rated
availability of aerobridges as ‘very poor’ to ‘poor’. The ACCC was unable to reach
a conclusion on whether service quality had declined from the previous reporting
period (ACCC 2001e).

Participants to this inquiry have also commented on the quality of services provided
by airport operators. Melbourne Airport noted that it had won a number of awards

3 The same services and facilities as were quality monitored prior to the regulatory changes of
October 2001.
266 PRICE REGULATION
OF AIRPORT
SERVICES
each year from 1997 (sub. 7). SACL stated that it ‘provides high levels of quality of
service for passengers and airline customers’ (sub. 27, p. 33) and recently had
completed major improvements, including an upgrade of the international terminal,
that had significantly increased service quality. Brisbane Airport was named 2001
‘Airport of the Year — for Aviation Excellence’ by the Australian Airports
Association.

Qantas, on the other hand, commented that there had been a reduction in the quality
of services at some airports, and cited Westralia Airports Corporation, operator of
Perth Airport, as an example (trans., p. 262). The airline was also critical of the
quality of some services at Sydney Airport, as was BARA, which noted that the
baggage-handling system at Sydney Airport (international terminal) was
considerably worse than that at airports in other countries. BARA was also critical
of the quality of particular services at Melbourne Airport (trans., pp. 228–9).
However, BARA also noted that anecdotal evidence from members suggested that
the quality of services provided by privatised airports had improved (sub. 41).

ACCC monitoring process

The ACCC’s reporting (summarised above) on the quality of service provision by

airport operators needs to be viewed in the context of the process adopted by the
ACCC to monitor quality. There are some problems relating to the process that
affect, or have the potential to affect, the robustness of the ACCC’s conclusions.
These problems are not unique to airport quality monitoring in Australia; they have
been raised in debate about this issue in other countries such as the United
Kingdom. 4

First, airport operators do not have direct control of many of the services provided
at airports. For example, check-in, customs and immigration, cargo processing and
on-time airline services are not the sole responsibility of airport operators. Other
organisations involved, such as Airservices Australia and Australian Customs
Service, are noted in chapter 3. Yet the services they provide can affect the services
provided by airport operators. For example, aircraft delays can be affected by
factors beyond the control of airport operators, and passenger perceptions of delays
can be influenced by customs processing. This issue is recognised by the ACCC in
its regulatory reports.

Second, to date ACCC reports on quality appear to have placed emphasis on survey
results rather than on the objective indicators (described in chapter 3), perhaps
because there have been few significant changes in the objective indicators for

4 See, for example, CAA (2000d) and related submissions, and Betancor and Rendeiro (1999).
REVIEW OF AIRPORT 267
REGULATION
airports. Passengers’ perceptions are subjective and, as noted above, may be
influenced by factors outside the control of airport operators. Surveys of airlines’
views also constitute a subjective form of assessment. Although the regulatory
reports indicated that there were many favourable responses by airlines about
services provided by airport operators (also noted by Qantas, sub. 48), Melbourne
Airport commented:
It seems to us that individual airlines use this [airline surveys] as a device to have a
‘free kick’ at airports and often use it as a forum for a ‘pay back’ against an airport for
taking a particular line in a commercial negotiation. (sub. 7, p. 41)

The ACCC (2001e) commented in respect of its monitoring of Sydney Airport, that
at the time of the 1999-00 survey of airlines 15 of the 23 airline respondents to the
survey were taking legal action against SACL.

Both SACL (sub. 27) and Melbourne Airport (sub. 7) considered the airline
responses to be evidence of ‘regulatory gaming’. However, BARA considered that
airlines had ‘no real incentive to act in this manner’ (sub. 41, p. 58).

Third, airport operators may be undertaking maintenance or building work that

temporarily disrupts operations and affects service quality. This disruption is not
necessarily taken into account in the ratings, although the ACCC does make a note
of such issues in its reports.

Fourth, there are some technical problems with airline and passenger surveys. For
example, as BARA noted, responses from airlines that rarely use an airport are
given equal weighting to major user airlines (sub. 41). Another problem relates to
the sample surveys of passengers at airports. Sampling biases may be associated
with passenger surveys, particularly as a result of the apparently non-systematic
way in which non-responses are handled in face-to-face interviews at airports.

Finally, there is the potential for a misleading assessment of the objective

indicators, depending on how a change in these indicators is interpreted. For
example, a reduction in the number of aerobridges at an airport may be viewed as
adverse performance if it is assumed that more is better than fewer and that all
airport users desire the same high quality levels. Virgin Blue, a low-fare airline
compared with Qantas, prefers more basic services and facilities, such as no
aerobridges, to match its customer profile.

268 PRICE REGULATION

OF AIRPORT
SERVICES
Evaluation of quality outcomes

Bearing in mind the above problems, the ACCC reports and inquiry participants’
comments suggest that the overall quality of service provided by operators at
Phase 1 airports has not deteriorated under the price cap since privatisation.

This outcome is not entirely unexpected. As noted in chapter 7, airport operators

have incentives not to reduce service quality, particularly if passengers and airlines
are highly sensitive to quality levels. Related to this are an airport’s commercial
(especially retail) operations, which are an important, and increasing, source of
airport revenue. The success of these operations depends heavily on growth in
passenger traffic and the provision of an environment conducive to retail spending
by passengers. Neither is likely to be achieved by a deterioration in aeronautical
service quality. However, BARA commented that airports have an incentive to
reduce the quality of services and cited, as an example, Sydney Airport’s
investment in additional remote gates5 that reduced costs for the airport but
increased costs for the airlines (sub. 41). In this regard, it is relevant to note that
Sydney Airport may face somewhat different incentives from those of other core-
regulated airports because it is constrained by limited airside space.

Airport operators are at the commencement of long-term leases — 50 years with an

option to renew for a further 49 years. If airport operators seek to remain
commercially viable over this period, the deliberate running-down of the quality of
existing assets in the short run is unlikely to enhance that viability. One of the key
objectives of Australian Airports (Townsville) in its Townsville Airport 2018
Master Plan is to improve service quality (sub. 14). Under the Phase 1 and Phase 2
lease agreements, any developments at an airport, such as extending a building or
constructing a road, must have regard to ‘reasonably expected’ quality standards,
and there are procedures to address non-compliance.

Several participants (for example, the Australian Airports Association (sub. 15) and
Northern Territory Airports (sub. 25)) commented that investment at airports had
been impeded by the regulatory regime in relation to necessary new investment
(NNI), and hence had restricted potential improvements in service quality. In other
words, future quality levels may be affected by application of past or current
regulatory arrangements. (Chapter 8 provides more detail on the effect of the
ACCC’s implementation of NNI provisions on investment by airport operators.)

5 Remote gates facilitate the embarking and disembarking of passengers when the gates adjacent
to a terminal are full, thereby enabling capacity to increase during peak-traffic periods. Remote
gates are distant from the terminal, usually requiring passengers to travel to and from the
terminal by bus.
REVIEW OF AIRPORT 269
REGULATION
Evaluation of ACCC monitoring

As noted above, there are some problems relating to the process that affect, or have
the potential to affect, the robustness of the ACCC’s monitoring process. It was also
noted that both BARA and Melbourne Airport were critical of various aspects of the
quality monitoring process.

BARA, more generally, commented that the process is ‘ineffectual’ and both parties
also noted that the ACCC did not appear to use much of the information collected.
BARA supported Melbourne Airport’s suggestion that:
If quality of service monitoring is to be a feature of any future regulatory system, which
we believe it should, the scope of monitoring and its objectivity needs to be seriously
examined. Any new set of arrangements must be auditable, systematic (rather than ad
hoc and anecdotal), and possess safeguards against gaming. (sub. 7, p. 41)

ACCC monitoring relates to movements in indicators of quality, rather than

standards.6 ACCC monitoring therefore makes no judgement about the absolute
level of quality, leaving this as a matter of commercial negotiation between airport
operators and users. The Commission considers this to be appropriate but, as
mentioned above, there is likely to be a judgement implicit in the monitoring
assessment process that higher quality of service provision is preferable to lower
quality.

Several airport operators (for example, Melbourne Airport, Brisbane Airport

Corporation (BAC) and SACL; subs 7, 8, 27) commented that they would, in the
absence of ACCC monitoring, undertake their own monitoring of service quality
because it is considered to be good business practice. Melbourne Airport stated that
‘quality of service monitoring is important and even if the law did not require it,
Melbourne Airport would undertake it in some form or other’ (sub. 7, p. 40).
Melbourne Airport and BAC use the information obtained from monitoring to help
identify areas for improvement in discussions with airlines.

In the United Kingdom, even though there is no regulatory obligation to monitor

service quality, BAA plc (part-owner of Australia Pacific Airport Corporation
(APAC)7), undertakes its own quality of service monitoring, including extensive
passenger surveys, to monitor trends in quality and to identify problem areas in
need of remedial action (CAA 2000d).

6 For a discussion about quality monitoring and standards, see CAA (2000d), and Betancor and
Rendeiro (1999).
7 APAC operates Melbourne and Launceston airports (through its subsidiaries Australia Pacific
Airports (Melbourne) and Australia Pacific Airports (Launceston) respectively).
270 PRICE REGULATION
OF AIRPORT
SERVICES
The ACCC monitoring process inevitably imposes compliance costs on airport
operators, including direct staff costs, costs associated with outsourcing of
passenger surveys, and the cost of overheads. These costs appear to be modest at
major core-regulated airports. Melbourne Airport noted that ‘the marginal
regulatory cost is well below the full cost’ because, as noted above, it would
undertake quality of service monitoring in any event (sub. 7, p. 42). And Capital
Airport Group considered that reporting requirements were not ‘overly burdensome’
(sub. DR75, p. 9). The annual regulatory cost is likely to be of greater significance
to the smaller Phase 2 airports, such as Alice Springs and Launceston.

The public dissemination of information on service quality by the airport operators

and the ACCC appears to be a positive feature of the monitoring process.
Transparency of reporting can encourage airport operators to improve service
quality and can form a basis for improved consultation and negotiation between
airport operators and users. From the perspective of airport operators, the public
nature of the monitoring process can provide a credible source of evidence about
service quality. Moreover, comparisons of quality ratings among airports may
facilitate competitive pressure among operators to improve service quality.

FINDING 9.1

The overall quality of service provision by airport operators at monitored airports

does not appear to have deteriorated since monitoring commenced. This outcome
reflects to a large extent the commercial incentives for airport operators not to
reduce service quality.
Although there are some problems with the quality monitoring process, it provides
transparency of reporting and facilitates some comparison between airport
operators at relatively low cost.

9.3 Access regulation

While access regulation does not directly regulate the prices of airport services, it
does provide for the regulation of the terms and conditions — including price — of
access to the services of essential facilities. Access regulation as an alternative or
supplement/complement to direct price regulation of airports is discussed in
chapter 11.

Two separate legislative instruments provide for access to airports — an airports-

specific instrument (s. 192 of the Airports Act, which applies to privatised core-
regulated airports), and a general instrument (Part IIIA of the Trade Practices
Act 1974 (TP Act), which potentially applies to all airports). These regimes are

REVIEW OF AIRPORT 271

REGULATION
described in chapter 3. In this section, the application of these access regimes to
airports is assessed. The Commission’s report on the National Access Regime
(PC 2001a) provides a detailed discussion of Part IIIA.

Although the objectives of s. 192 are not enunciated in the Airports Act, the second
reading speech for the Airports Bill suggested the primary policy intent was to
facilitate access for new passenger airlines (HoR 1996, p. 1308). That the
Commonwealth Government introduced an airports-specific access regime for
privatised airports suggests it considered the Part IIIA provisions would not
adequately facilitate access to the services these airports provide. However, as
Part IIIA was in its infancy when s. 192 was introduced, the applicability to airports
was unclear at the time. As it turned out, Part IIIA has since been used to gain
access to airport facilities (see below).

Undertakings

Section 192 provides for airport operators to lodge undertakings setting out terms
and conditions under which access to airport services will be provided with the
ACCC (chapter 3). However, lodgement of undertakings has been limited (box 9.1).

Box 9.1 Access undertakings

Melbourne and Perth airports — the only airports to submit access undertakings during
the designated period following privatisation — submitted draft undertakings to the
ACCC in early 1998. The undertakings covered a range of services within and outside
the price cap. Both undertakings committed airport operators to provide access at
prices consistent with the price cap for capped services, and at prices determined by
the operator or through negotiation between the parties for services outside the cap.
The undertakings also established mechanisms for dispute resolution.
The ACCC did not accept either draft undertaking. In the case of Perth Airport, the
ACCC expressed concerns about the enforceability of the undertaking, the adequacy
of information provided to access seekers as part of the negotiation process and the
dispute resolution provisions. In relation to the Melbourne Airport draft undertaking, the
ACCC was concerned about the enforceability of the undertaking, pricing for services
outside the price cap and the limited scope for negotiation. Neither airport pursued
lodgement of an undertaking beyond this point.
Sources: ACCC (1998c, d; 2000a, e).

APAC stated that it did not pursue lodgement of an undertaking beyond the draft
stage because:
… we had formed a view that any undertaking that would be acceptable to the ACCC
would actually leave us in a worse position than if we were to be subject to declaration
272 PRICE REGULATION
OF AIRPORT
SERVICES
 and subsequent arbitration. This was particularly the case where the ACCC was seeking
an undertaking that effectively extended into operational and commercial areas not
intended to be subject to Part IIIA. (APAC 2000b, p. 4)

The ACCC’s rejection of the draft undertakings submitted by Melbourne and Perth
airports appears to have discouraged other airport operators from submitting
undertakings. SACL stated:
The experiences of Melbourne and Perth airports in (unsuccessfully) pursuing access
undertakings resulted in no Phase 2 airports even attempting this route. The costs,
uncertainty and ultimate lack of flexibility have been factors in the deterrent. SACL has
similarly investigated the resources required to complete a successful Part IIIA
Undertaking and decided it is too onerous and uncertain to warrant pursuit.
(sub. 27, p. 46)

Airport operators’ ability to offer undertakings for some facilities was also limited
by other factors. For example, airport operators were not in a position to offer
undertakings for those domestic terminals for which long-term leases were held by
incumbent airlines (chapter 3).

The relatively short period in which undertakings could be lodged (12 and
24 months following privatisation for Phase 1 and 2 airports respectively), at a time
when airport operators were also required to develop draft airport master and
environment plans (chapter 3), and the fact that there was no scope for appeal
against an ACCC decision not to accept an undertaking, may also have been factors
contributing to the limited use of the undertakings provision.

The ACCC determines whether a given service is declared under s. 192 following a
request for a determination from an access seeker. As no undertakings were in place
at the end of the designated period, airport services at privatised core-regulated
airports were declared automatically. Rather than listing declared services, criteria
for declaration of services were specified in the legislation (chapter 3).

Access determinations

To date, application of the access provisions to airports has been limited. Delta Car
Rentals (Delta) and Virgin Blue sought determinations that services at Melbourne
Airport were covered by the automatic declaration of ‘airport services’ under s. 192.
Australian Cargo Terminal Operators (ACTO), now deregistered, sought
declaration of various services at Melbourne International Airport (MIA) and
Sydney International Airport (SIA) under Part IIIA (box 9.2).8

8 ACTO applied for voluntary deregistration in October 1999. Its status, according to the
Australian Securities and Investments Commission, is ‘deregistered’.
REVIEW OF AIRPORT 273
REGULATION
 Box 9.2 Application of access regulation to airports

Section 192

Delta Car Rentals

Delta operated an off-airport car rental service and short- and long-term car parks
close to Melbourne Airport, and provided a shuttle bus service to transport passengers
to and from the terminals. (Delta acquired National Car Rentals and signed a licensing
agreement with Europcar International in late 2000. It now operates an on-airport car
rental service and an off-airport car-park service.) When Melbourne Airport was
operated by the Federal Airports Corporation (FAC), Delta paid a fee per car rental for
access and operated from the landside drive area adjacent to the terminals. The terms
and conditions of access offered by Melbourne Airport following privatisation required
Delta’s car rental operation to pay a turnover-based access fee and to operate its car
rental service from the designated meeting point in the car park opposite the terminals.
While a reduced turnover-based access fee was agreed through commercial
negotiation between the parties, Delta did not accept the airport operator’s requirement
that it operate from the designated meeting point. On 21 October 1998, Delta
requested that the ACCC make an access determination for the service of ‘the landside
drive area to pick up and drop off passengers’.
On 17 May 1999, the ACCC determined that the service of ‘the provision of landside
roads and associated vehicle facilities for dropping off and picking up passengers at
Melbourne Airport’ is a declared ‘airport service’ under s. 192. Neither party has sought
arbitration of the terms and conditions of access to the service. The declaration will
expire on 1 July 2002, when the Minister’s determination under s. 192 that airport
services at the airport are declared for the purpose of Part IIIA expires.

Virgin Blue
On 2 March 2001, Virgin Blue lodged an application with the ACCC for a determination
that the service of the use of the new multi-user domestic terminal (MUDT) ‘for the
purposes of processing arriving and departing domestic airline passengers and their
baggage at Melbourne Airport’ was an ‘airport service’. The application followed
negotiations over the terms on which Melbourne Airport would grant Virgin Blue access
to the MUDT on a long-term basis.
The ACCC released its draft determination in October 2001, stating that, even though
the service met the criteria for definition as an ‘airport service’ under s. 192,
determining that domestic terminal facilities are ‘airport services’ was unlikely to result
in significant benefits. Thus the ACCC used its discretionary powers under s. 192 to
determine that the use of the MUDT was not an ‘airport service’ (ACCC 2001l).
(Continued next page)

274 PRICE REGULATION

OF AIRPORT
SERVICES
 Box 9.2 (continued)

Part IIIA

Australian Cargo Terminal Operators

On 6 November 1996, ACTO — a provider of cargo terminal services to international
airlines — sought access to various freight handling and related services provided by
facilities owned by the FAC at Melbourne International Airport (MIA) and Sydney
International Airport (SIA). These included the services provided by freight aprons and
hard stands needed to load and unload international aircraft, areas for moving and
handling freight, equipment storage areas, and land to construct cargo terminals.
The National Competition Council (NCC) recommended on 8 May 1997 that the
services provided by the freight aprons, hard stands and freight moving and equipment
storage areas at MIA and SIA be declared. It recommended declaration of the services
at SIA for a period of five years, and declaration of the services at MIA such that
declaration expired 12 months after leasing (to allow an undertaking to be lodged
during the designated period under s. 192). The NCC did not recommend declaration
of land for constructing cargo terminals at either airport as it considered ACTO could
construct cargo terminals off-airport. The Treasurer accepted these recommendations
and declared the services on 30 June 1997. The declaration of services at MIA expired
on 1 July 1998.
The FAC and, subsequently, SACL, applied to the Australian Competition Tribunal for
a review of the Treasurer’s declarations at SIA. It proposed the adoption of a tender
process to allocate rights to provide ground-handling services as an alternative to
declaration. The Tribunal handed down its decision affirming the Treasurer’s decision
to declare the services on 1 March 2000. The declarations at SIA are effective from
1 March 2000 to 28 February 2005. The ACTO declarations are the only declarations
made under Part IIIA in any industry to date.
Sources: ACCC (1999a, d; 2001j); ACT (2000); NCC (2001); PC (1998a).

Issues arising from access determinations

Although determinations under the access provisions to date have been limited, they
provide some insights into interpretation of the criteria by regulators and the courts,
compliance costs and declaration processes.

Delta determination

As outlined in box 9.2, Delta sought access to the landside drive area adjacent to the
terminals, while APAC argued that Delta should operate from the designated

REVIEW OF AIRPORT 275

REGULATION
meeting point:
A key obligation for an airport operator is to efficiently manage the differing
requirements of airport customers in a way that ensures the good order and
management of traffic flow in and out of the airport … In order to ensure that the
airport forecourt operates safely and without unacceptable congestion certain categories
of users (buses, taxis, limousines, and, now, off-airport car rental courtesy buses) are
required to use specific facilities outside the scope of general public roads. It is the case
that other off-airport car rental companies are currently utilising the DMP [designated
meeting point area] in a satisfactory manner. (APAC 1998, pp. 5–6)

The ACCC considered that Delta’s definition of the service as ‘the landside drive
area to pick up and drop off passengers’ was too narrow. It therefore used its
powers under ss 192(4A) and (4B) to define the service as ‘the provision of landside
roads and associated vehicle facilities for dropping off and picking up passengers at
Melbourne Airport’, where ‘landside roads and associated vehicle facilities include
landside roads, roads through carparks, kerbside parking areas, the designated
meeting point area (accessed via the carpark road) and any other vehicle access
facilities’ (Commonwealth of Australia Gazette, No. GN 25, 23 June 1999,
p. 1864).

A number of submissions to the ACCC investigation argued that this broader

definition was inappropriate, including BAC, Budget Car Rentals, Avis Australia
and Ansett. By broadening the definition of the service to include substitutes for the
landside drive such as the designated meeting point, the question becomes whether
landside access to the airport, rather than access to the landside road outside the
terminal, is necessary for operating and/or maintaining civil aviation services at the
airport. This broadening of the definition of the service in the Delta determination
may have resulted in the service meeting the s. 192 declaration criteria when it
otherwise might not have done so. Nevertheless, in April 1999 the ACCC (1999d)
determined that the service was a declared airport service. APAC (2001) suggested
in a submission to the National Access Regime inquiry that, if declaration had been
sought under Part IIIA, it is unlikely that the application would have met either the
competition test or the national significance test.

Virgin Blue determination

As noted in box 9.2, Virgin Blue requested that the ACCC determine whether the
services provided by the use of the multi-user domestic terminal (MUDT) were
‘airport services’ under s. 192 of the Airports Act. The ACCC (2001l) has released
a provisional finding that the relevant service is not an airport service for the
purposes of s. 192.

276 PRICE REGULATION

OF AIRPORT
SERVICES
In assessing Virgin Blue’s request, the ACCC defined the relevant service as:
The use of domestic terminals and related facilities for the purposes of processing
arriving and departing domestic airline passengers and their baggage at Melbourne
Airport. (ACCC 2001l, p.19)

Although a slightly broader definition of the service than proposed by Virgin Blue,
it is similar to the definition suggested by Australia Pacific Airports (Melbourne)
(the operator of Melbourne Airport), and consistent with Qantas’ proposal.

The ACCC also concluded that this service met the criteria for definition as an
‘airport service’ under s. 192(5) of the Airports Act. The ACCC found that the
relevant service was:
• necessary for civil aviation (as required under s. 192(5)(a)); and
• provided by significant facilities which cannot be economically duplicated (as
required under s. 192(5)(b)).

However, in reaching this provisional finding, the ACCC made use of its
discretionary powers granted under s. 192(4B) of the Airports Act. The ACCC
argued that services provided by the MUDT were regulated under the notification
provisions of the PS Act. Prices for the use of the MUDT had already been
approved by the ACCC under the NNI provisions of the notification. Hence, the
ACCC argued that:
In these circumstances it is not obvious that determining domestic terminal facilities are
‘airport services’ would yield significant benefits. The issues which would be
canvassed in the event that the ACCC was later called upon to determine a dispute over
access to the MUDT are likely to overlap considerably with the issues considered under
the NNI approval process. In the present circumstances it is not, therefore, obvious that
a determination along the lines sought by Virgin Blue would promote competition.
(ACCC 2001l, p. 26)

The ACCC has, in effect, decided to circumvent the arbitration process which may
have been initiated under the access provisions (Part IIIA) of the TP Act if the
service had been declared an ‘airport service’ under s. 192. The ACCC decision to
use its powers under s. 192(4B), while demonstrating the scope for discretion, may
also increase uncertainty among industry participants as to what services would be
defined as ‘airport services’ in any future determinations. The ACCC noted that the
draft determination:
… reflects the factual landscape surrounding Virgin Blue’s application. It is
emphasised that the ACCC might form a different view on the appropriateness of a
determination of similar facilities in other circumstances. (ACCC 2001l, p. 27)

REVIEW OF AIRPORT 277

REGULATION
ACTO determination

The ACTO declarations at MIA and SIA under Part IIIA (box 9.2) illustrate that the
Part IIIA declaration criteria tests are also open to interpretation. While ACTO
sought access to various freight-handling and related services, the National
Competition Council (NCC) and Australian Competition Tribunal found that the
relevant facilities providing those services were the whole of MIA and SIA. Thus, it
was these facilities that were judged against the natural monopoly and national
significance tests.

The ACTO declarations also raised a number of other issues relating to the
interpretation of the Part IIIA criteria. These are considered at length in the
Productivity Commission’s report (PC 2001a) on the review of the National Access
Regime. The Commission recommended that the first declaration criterion be
strengthened,9 and that subsequent declaration decisions be reviewed in the next
review of Part IIIA with a view to determining whether further strengthening of
criteria is required.

The SIA declarations under Part IIIA (box 9.2) illustrate that there is scope for
considerable delays in the declaration process. ACTO applied to the NCC for
declaration of a number of services at SIA in November 1996. While the Minister
declared the services approximately eight months later, the FAC applied to the
Australian Competition Tribunal for a review of the decision. The Tribunal handed
down its decision affirming the declarations on 1 March 2000 — more than three
years after ACTO’s application. The administrative and legal costs of these
declarations are likely to have been significant, though insofar as precedents have
been determined, these costs should not all be attributed to this specific case.

Summary

The limited use of s. 192 to date makes it difficult to assess its contribution to
providing access at privatised core-regulated airports. The Delta determination
supports the view that the s. 192 declaration criteria are somewhat less stringent
than the Part IIIA criteria. However, the draft determination on Virgin Blue
indicates that even when the s. 192 criteria are met, the ACCC can use its
discretionary powers to make a negative determination. Whether airports warrant
continuation of special access provisions is discussed in chapter 11.

9 Declaration would have to promote a substantial increase in competition in another market,

rather than simply promoting competition in that market.
278 PRICE REGULATION
OF AIRPORT
SERVICES
The extent to which automatic declaration of airport services and relaxation of the
declaration criteria under s. 192 might have contributed to the achievement of the
Commonwealth Government’s stated objective of facilitating access for new
domestic passenger airlines is unclear. Domestic terminal lease agreements require
incumbent domestic airlines, on application, to make some gates at their terminals
at some core-regulated airports available to new entrants (chapter 3). However, the
likelihood that new entrants will apply for access to these gates has been reduced by
the agreement by operators of a number of airports to develop new domestic
common-user terminal facilities. (Brisbane Airport already had common-user
facilities.) The potential for increased passenger throughput may have provided the
incentive to develop these facilities, coupled with scope for charging for these
facilities under NNI provisions. If airport operators had not developed common-user
terminal facilities, incumbent domestic airlines (as lessees of domestic terminals),
rather than airport operators, are more likely to have been subject to access
determinations. This suggests that s. 192 is unlikely to have been instrumental in
facilitating access for new domestic passenger airlines.

FINDING 9.2

Though privatised core-regulated airports have facilitated access for new entrant
airlines, airport-specific access provisions (s. 192 of the Airports Act) do not
appear to have been instrumental in achieving this outcome.

The successful use of Part IIIA by ACTO to obtain access declarations at MIA and
SIA also demonstrates that the general access provisions can be used to gain access
to airport services. However, these declarations have led to some concern about
interpretation of the Part IIIA criteria. The Productivity Commission in another
inquiry has recommended somewhat stronger declaration thresholds (PC 2001a).

REVIEW OF AIRPORT 279

REGULATION
10 Regulatory options: cost-based and
incentive regulation

Under the terms of reference, the Commission is required to identify and assess
alternatives to the price regulation implemented following the sale of airport leases.
In this and the next chapter, the main regulatory options are explored. This chapter
outlines cost-based and incentive regulation options. Further, issues relating to the
form and implementation of a price-cap regulatory regime are discussed. Issues
concerning the coverage of regulation — relating to single-till and dual-till
regulation — also are discussed. Chapter 11 assesses other regulatory options: price
monitoring, and access and anti-competitive conduct regulation, including price
undertakings and commercial agreements.

10.1 Cost-based and incentive regulation

Cost-based regimes have a long history in the United States, the most common
being rate-of-return regulation. Incentive-based regulation — including price caps1
and benchmark regulation — was developed partly in response to the problems of
cost-based regulation. Cost-based and incentive regulation are discussed below, as
are the relative merits and disadvantages associated with these approaches to
regulation, and the options available for stricter price regulation of airports in
Australia.

Cost-based regulation

Cost-based regulation sets prices that are based directly on the costs of the regulated
firm. Although there are other types of cost-based regulation,2 rate-of-return
regulation is the predominant form. Implementation of rate-of-return regulation

1 The term ‘price cap’ covers a multitude of regulatory regimes, including where the firm has all
prices subject to a strict price ceiling with no adjustment process. The process described here is
based on the Littlechild (1983) Retail Price Index (RPI)-X form of the price cap.
2 For example, return-on-cost allows the firm to charge a specific mark-up over total costs.
OPTIONS: COST-BASED 281
AND INCENTIVE
REGULATION
involves the regulator imposing a ‘fair’3 return on capital assets and then setting
prices to deliver this outcome. The objective of rate-of-return regulation is to ensure
that prices are set at a level that allows ongoing supply of the goods or services in
question (including investment) by the regulated firm, but are not set so high as to
allow excess profits.

Though cost-based regulation could in principle achieve efficient pricing, in

practice it has several major disadvantages.
• The first is the complex informational requirements. The regulator, using
information provided by the firm on its projected operating and capital costs,
asset base and projected sales, must determine what constitutes a ‘fair’ rate of
return (including an assessment of risk), whether the operations and proposed
investment are efficient, and then what would be appropriate unit prices over the
regulatory period.
• Given the marked information asymmetry between the firm and regulator there
is an incentive for cost-padding by the firm, including, for example, through
higher labour costs.
• The regulated entity may tend to use more capital than if it were unregulated. If
the allowed rate of return is set above the cost of capital, then the firm has an
incentive to expand the capital base to increase profitability.4 This is known as
the Averch–Johnson (A–J) effect.5 Thus, the regulated firm may have an
incentive to use an inefficient mix of inputs in production, resulting in an
inefficiently high capital/labour ratio which, in turn, results in higher production
costs for a given amount of output.
• The prices for each good or service often are set individually such that price
equals cost. This requires the allocation of any common costs.6 This essentially
arbitrary process may increase efficiency losses by moving away from more
efficient discriminatory pricing that recovers common costs more from where
demand is least responsive to price changes.

3 ‘Fair’ is commonly used to describe a desired rate-of-return for the industry and includes a
risk-adjusted return on capital.
4 There may be other reasons for the monopolist to over-invest, for example, preemptive
investment to discourage new entrants.
5 After Averch and Johnson (1962), who first analysed the problem. For this result to hold, the
‘fair’ rate of return must be above the cost of capital. Kahn (1988) notes that the A–J effect may
offset the effects of under-investment due to monopoly and encourage risk-taking and output-
expanding investment.
6 Common costs may be allocated in a variety of ways, including according to revenue and
output shares, though essentially any allocation is arbitrary. The relative merits and problems of
allocating common costs are discussed by Europe Economics (2001).
282 PRICE REGULATION
OF AIRPORT
SERVICES
• Rate-of-return regulation that ties prices to costs incurred by the firm may not be
appropriate where there is congestion and new facilities need to be built.
Efficient pricing needs to be linked to opportunity costs.
• The risk of any investment may be borne largely by customers, to the extent that
the regulated firm is able to pass through costs into higher prices.

The fundamental problem with rate-of-return regulation is that because the

regulation is effectively cost-plus and because prices are typically reset on an
annual basis, the regulated firm will not benefit from cost-reducing changes as any
resulting savings are passed through to the customer.

There have been attempts to improve the outcomes from cost-based regulation.
Benchmarking techniques have been applied to overcome information problems
inherent in rate-of-return regulation, though with limited success (discussed below).
Another approach has been profit-sharing (or sliding-scale) regulation which is a
variant of rate-of-return regulation. In this approach, after the profit results have
been calculated for a designated period, any excess profits or profit shortfalls are
shared with the customers. This can be done through either refunds or future price
adjustments. These methods are designed to narrow the gap between costs and
revenues, thereby increasing allocative efficiency, without adversely affecting the
incentives for productivity improvements. A profit-related sliding scale requires an
explicit rate-of-return mechanism, so the information requirements can be
cumbersome.

Incentive-based regulation

In view of the poor incentive for productive efficiency of cost-based regulation,

other forms of regulation have been developed. These attempt to provide regulated
firms with appropriate incentives for efficient supply and price structures, while at
the same time encouraging firms to implement efficient price levels (and earn
normal profits) over time. As summarised by Vogelsang:
… [incentive-based regulation] means that the regulator delegates certain performance-
related decisions to the firm and that the profits of the regulated firm depend on
performance measures of the regulator. Incentive regulation makes use of the firm’s
information advantage. The regulator thus controls less behavior but rather rewards
outcomes. (Vogelsang 2001, p. 1)

Price caps are the main incentive mechanism used by regulators, though there are
many variations that fall under the price-cap banner. Price caps have their
theoretical basis in the work of Vogelsang and Finsinger (1979), with Littlechild
(1983) developing a practical method for the implementation of price caps for the
UK telecommunications industry (box 10.1).
OPTIONS: COST-BASED 283
AND INCENTIVE
REGULATION
 Box 10.1 Vogelsang and Finsinger and Ramsey pricing
Economic efficiency generally requires that prices reflect marginal costs. However,
because airports have natural monopoly characteristics and substantial fixed costs,
pricing at marginal cost is not feasible in the long term. A desirable objective in this
situation is to price to cover average costs but in a way that minimises efficiency
losses. A standard economic solution in multi-product firms has been to implement
‘Ramsey pricing’, where the setting of prices reflects the inverse of the demand
elasticity for the good or service. Thus, the less price responsive that demand for the
good or service is to price changes, the higher the price.
Vogelsang and Finsinger (1979) proposed a method whereby a firm, under certain
assumptions, will set prices that satisfy Ramsey pricing conditions and earn sufficient
revenue to stay in business. This mechanism allows a firm to choose its own price
structure as long as it remains within a certain average price level for the regulated
basket of goods or services. Based on this theoretical approach, Littlechild (1983)
developed price caps that were first applied in the United Kingdom.

A price cap specifies the maximum price for a good or service over a certain period.
Typically, price increases are constrained to a level determined by an index —
commonly the rate of inflation (which serves as a proxy for exogenous rises in the
prices of inputs) — minus an X factor that (predominantly) accounts for expected
productivity improvements in the regulated firm. Any cost savings achieved beyond
the rate of inflation minus X adjustment accrue to the firm within a given regulated
period.

The major advantages of price caps are that:

• They are less susceptible to the risk of over-capitalisation by the firm and avoid
the incentives for productive inefficiency that accompany cost-plus regulation.
Given that the regulated firm has the right to retain profits earned during the
designated period, there is an incentive to produce as efficiently as possible and
to increase sales (and thus, if there are economies of scale, reduce unit costs).
Some of these expected benefits will be shared with consumers via the setting of
the X.
• They allow for the adjustment of prices of goods and services being regulated,
subject to the requirement that the weighted average increase in prices of goods
and services in the ‘basket’ is within the cap. This allows for flexibility in
pricing, including Ramsey pricing (box 10.1), and adjustment to accommodate
consumer tastes and preferences, leading to a gain in overall efficiency that can
benefit both consumers and producers.
• The risk of investment, in principle, is shared between the regulated firm and its
customers, because a rate of return is not specifically guaranteed.

284 PRICE REGULATION

OF AIRPORT
SERVICES
• Price caps are forward looking in that they encompass forecasts of expected
productivity gains as well as future demand and cost levels.

As with cost-based regulation, however, the application of price caps confronts a

range of difficulties. A major practical issue, if prices are to move towards efficient
levels over time, is that the Xs must be reset at regular intervals in line with the
costs of the regulated firm (including a reasonable rate of return on assets). In other
words, there is a tendency for price caps to converge over time to cost-based, rate-
of-return regulation, with many of its associated incentive problems. One option is
to set starting prices and/or the Xs without reference to the costs or rate-of-return of
the firm, but this approach could have serious consequences if prices were to move
too far away (lower or higher) from efficient levels (arguably the current situation
for many core-regulated airports — chapter 8).

Typically, therefore, price-cap regulation is applied with regular reference to actual

costs of the firm (and/or costs of other firms in the industry). Box 10.2 briefly
outlines the building-block approach to assessing allowable costs used by the
ACCC to determine ‘efficient’ prices.

When there is regular reference to the costs of the firm, the essential remaining
difference between cost-based and price-cap regulation is the interregnum offered to
the regulated firm, during which it can retain profits arising from cost savings
beyond those incorporated in X values. But, viewed in this way, price-cap
regulation simply amounts to cost-based regulation with longer lags: in other words
if a price-cap regime were reviewed yearly with resetting of prices or adjustment of
the X factor related purely to the costs of the firm, then the two regulatory regimes
— and their outcomes — would be indistinguishable. The major additional feature
of price caps is that they can be used to stimulate a more efficient cost base.

Thus, if the review periods for a price-cap regime are short, then the incentives for
cost reduction are limited because prices closely track actual costs. If the review
periods are longer (perhaps three to five years), then the incentive to reduce costs by
the regulated firm — particularly early in the period — is stronger, because it can
retain part of the profit resulting from any cost-saving measures. (This assumes, of
course, that the regulator does not intervene within the designated regulatory period
and that the resetting of the price caps is always forward looking.) However, as the
price-cap review is approached, there may be an incentive for the airport to inflate
costs to obtain a lower X over the next period.

OPTIONS: COST-BASED 285

AND INCENTIVE
REGULATION
 Box 10.2 Regulatory determination of ‘efficient’ prices
Determination of allowable prices is a difficult task. The ACCC has recommended an
’accrual building block’ approach, based on forecasts of the cost of service over the
regulatory period. The maximum allowable revenue (MAR) is a sum of the projections
of the return on capital, depreciation and operating and maintenance expenditure.
Having determined the level of costs, allowable prices are determined after forecasting
the level of demand for the airport’s services. While estimating operating and
maintenance expenditure is relatively straightforward, the assessment of the other
components requires difficult judgements. Some of the major decisions required to
determine the MAR are:
• Assets (other than land) valuation. Asset values are critical in determining both
allowable depreciation charges and returns received by capital. The estimated cost
of capital is applied to the depreciated asset base to determine the revenue required
to compensate for the opportunity cost of funding the assets. There is no uniformly
accepted approach to the valuation of assets. The ACCC recommends that a
replacement cost method — depreciated optimised replacement cost (DORC)
valuation — be used. However, an historical-cost approach — such as depreciated
actual cost (DAC) — may not be inappropriate for firms where capital costs are
relatively stable over time.
• Land valuation. For price regulation of core-regulated airports, the valuation of land
can be particularly important as a result of the relatively high proportion of high-
valued land in airport assets. It generally would not require a DORC valuation
because it tends not to depreciate, although the optimal land input can be
influenced by technological developments and input price changes. Opportunity cost
and historical cost are two approaches that have been used by regulators to value
land. These two approaches can produce very different land valuations and hence
allowable revenue and prices. The historical cost approach used by the ACCC in
the valuation of land for Sydney Airport resulted in a land valuation significantly
below the current replacement cost valuation proposed by Sydney Airports
Corporation Limited (SACL). Given that the next best use is not directly observable,
there are several approaches to estimating the opportunity cost of the land, all of
which require the resolution of difficult conceptual and practical issues.
Nonetheless, an estimate of opportunity cost of the land generally remains the
appropriate approach to land valuation for production-cost based price regulation.
However, cost-based regulation may not be appropriate where airport capacity is
restricted (appendixes F and H).
• Weighted average cost of capital (WACC). WACC estimates the required rate of
return to be earned by debt and equity providers. While the debt costs to a firm are
relatively straightforward to assess, the required rate of return on equity is not. The
capital asset pricing model (CAPM) is widely used. This method requires the
measurement of two contentious variables: the beta (a measure of the risk of the
firm relative to total market risk) and the market risk premium.
Sources: ACCC (1999c; 2001i); PC (2001a).

286 PRICE REGULATION

OF AIRPORT
SERVICES
Where price caps do not include allowance for investment in the starting prices or
the Xs, the cost-based reviews required to assess price increases allowable for
individual investments will tend to move price caps closer to a rate-of-return
regulation approach.

Another significant potential problem with price caps relates to the need for
maintenance and/or improvement of quality. Due to the structure of price caps,
there generally is an incentive to run down existing assets and degrade quality. With
a price cap in place, a firm may have an incentive to implement cost reductions at
the expense of service quality. Thus, price-cap regulation typically is complemented
by monitoring of service quality to ensure standards do not deteriorate.

An increase in quality usually requires an increase in labour and/or capital

expenditure. However, when the firm is subject to a price cap, it may not be able to
recoup all of the extra costs via higher prices. The regulated firm may gain
additional revenue from increased sales, but this may or may not be enough to
compensate for the extra costs incurred. Thus, price caps normally are
complemented by specific provisions to accommodate quality-enhancing
investments.

Price caps also may constrain efficient pricing when capacity constraints are being
approached. In particular, in the calculation of X values, the costs used are those
incurred by the firm, not the costs imposed on others in the form of congestion.
Efficient prices are those that would allocate the scarce services to those who value
them most highly, not prices determined by costs incurred in operating the airport or
even ‘opportunity’ costs that reflect another (non-airport) use (appendix F).

The ACCC (sub. DR55) observed that airport operators have scope to introduce
peak-period pricing by restructuring charges within a price cap. The extent to which
price restructuring is able to address excess demand problems will depend partly on
the volumes in peak and non-peak periods. If the peak is long and reflects
significant excess demand, it is likely that market-clearing prices would generate
more revenue than allowed under a production-cost-based price cap. Broadly
speaking, the minimum efficient price in off-peak periods would still have to cover
the marginal cost of providing aeronautical services at those times. Prior to
September 2001, weekday volumes at Sydney Airport reached the 80 movements
per hour limit in five peak hours, and flights at these times represented around
40 per cent of daily movements. Introducing peak-period pricing may result in
existing shoulder periods also becoming fully utilised.

Finally, price caps, when they are fixed for other than a short term, are not very
flexible. If an airport’s volumes and/or costs are changing rapidly, adherence to the

OPTIONS: COST-BASED 287

AND INCENTIVE
REGULATION
price cap may result in inefficient prices. As concluded by the Motor Trades
Association of Australia Superannuation Fund (MTAA Super Fund):
… when the firm is already operating relatively efficiently, demand conditions are
changing rapidly and unpredictably or the price cap is applied to preserve low or
negative levels of profitability, there is a strong likelihood that the controls and the X
factor will cause the firm to set prices that generate worse outcomes than might have
arisen without direct prices oversight. (sub. 22, p. 18)

Where outcomes are significantly different from expectations, governments may

respond by reviewing the regulatory framework, as occurred following the
significant decline in airport traffic volumes in September 2001.

Further, any approach involving regulators establishing prices, carries potential

costs associated with the likelihood that the regulator will not establish the efficient
price. The possibility of even random departures from efficient pricing in regulatory
decisions increases the riskiness of investment. This risk may be further increased
by the political attractiveness of lower rather than higher prices, creating a bias
towards inefficiently-low regulated prices. The dual role of the ACCC as a regulator
and consumer advocate (alluded to by Sydney Airports Corporation Limited
(SACL), sub. DR62), increases the possibility of such outcomes. In addition,
pricing rules used to implement regulation may limit efficient pricing approaches
such as price discrimination.

Forms of price cap

There is a range of options available for the implementation of a price-cap regime.

The following discussion covers some of these.

Revenue yield versus tariff basket

In the United Kingdom, regulation of airports uses a ‘revenue-yield’ approach to

price caps (box 10.3) — that is, airport charges are capped by reference to the
revenues generated per passenger. The main alternative is the current system used in
Australia, which is sometimes referred to as the ‘tariff-basket’ approach.

288 PRICE REGULATION

OF AIRPORT
SERVICES
 Box 10.3 Tariff-basket and revenue-yield price caps
In determining the weighted average index for prices, two approaches are common.
The current approach used at Australian airports is a ‘tariff basket’ that uses a
weighted sum of prices based on revenue shares in the previous period. The regulated
airport is restricted to increasing the weighted sum of prices by no more than the CPI-X
factor each period. Within the basket, the airport is able to increase or decrease
individual prices for goods or services, as long as they do not violate the weighted sum
price cap.
In the United Kingdom and Europe, some airports are subject to a regime referred to
as the ‘revenue-yield’ approach. The price cap is based on a maximum allowable
revenue yield per passenger. In the United Kingdom, where airports have been
regulated under a single-till regime, the maximum allowable revenue yield per
passenger is calculated by dividing total allowable revenue for aeronautical services by
the anticipated number of passengers. The airport can then set the prices for
aeronautical services so as not to exceed the maximum average revenue yield per
passenger. Like the tariff basket approach, prices for individual airport services can be
set freely by the airport as long as the average revenue constraint is not exceeded.

Each approach has advantages and disadvantages.

• The revenue-yield approach tends to be more flexible, because there are no
constraints on the charges that can be levied, provided the airport adheres to the
average revenue constraint. Introducing new charges is also relatively
straightforward in the revenue-yield approach, whereas in a tariff basket, prices
need to be re-weighted. Given that the weights normally are determined using
the previous year’s revenue, to implement a new charge, new revenue weights
need to be calculated.
• Under a revenue-yield approach, there needs to be a homogeneous common unit
of measurement. For UK airports, passengers are used as the unit of comparison.
This provides incentives for airports to concentrate on passenger throughput,
possibly at the expense of other services.7
• The tariff-basket approach provides better incentives to move to efficient
pricing. Under the revenue-yield regime, when there is unused capacity, the
airport may set prices below incremental cost for some services and recover
these costs on other services.
• Under both regimes, as under price caps generally, encouraging appropriate
quality is an issue, although there appear to be better incentives under the tariff-

7 If, for example, passengers are the base unit, there is little incentive to attract dedicated freight
flights, because revenue yield per passenger will increase and possibly exceed the cap,
requiring an offsetting reduction in aeronautical charges.
OPTIONS: COST-BASED 289
AND INCENTIVE
REGULATION
 basket approach, because the revenue-yield approach is focused on a
homogeneous unit (typically passengers). The tariff-basket approach has better
incentives to provide different levels of service quality, because there is an
incentive to price different dimensions of quality according to each service’s
incremental costs and relevant demands.

The UK regulator, the Civil Aviation Authority (CAA 2001c) concluded, after
consideration of the advantages and disadvantages of the two approaches, that the
tariff-basket approach should replace revenue yield at Heathrow and Gatwick
airports. According to the CAA, the tariff-basket approach provided a ‘good basis’
for the proposed default price cap (see below) because, in particular, it avoided the
dilution problems associated with the revenue-yield approach.8 However, it
proposed maintaining the revenue-yield approach at Stansted and Manchester
airports, which are less congested.

Default price cap and service quality

Under a default price cap, an airport would be required to provide a specified level
of service at a given (or maximum) price. For any services beyond this standard,
each airline would have to negotiate directly with each airport. The extra services
could be separate services or services of a higher quality. This process has the
potential to encourage commercial negotiation between individual airports and
individual airlines, and targets regulation at the core services required by airlines.
The CAA noted that:
… a mechanism [default price cap] that facilitated greater direct contracting between
users and the airports could have significant benefits in terms of better use of existing
airport facilities, and in providing better incentives to ensure that the airport invests in
new facilities that users want. (CAA 2001c, p. 185)

The CAA (2001c) proposed that a default price cap be implemented at Heathrow
and Gatwick airports where there is evidence of significant excess demand and
price caps are likely to be tightly binding. Separate service contracts to provide
service ‘top ups’ or ‘top downs’ need not be treated as airport charges, so could be
made outside the price cap. The CAA also proposed that within the proposed price
cap there should be a term reflecting airport quality of service, based on a weighted
basket of quality attributes.

A default price cap and an adjustment for quality were considered to be unnecessary
for Stansted and Manchester airports. It was considered that at these airports, which

8 If a contract is entered into at prices lower than the default cap, then unless an adjustment is
made, the airport could increase prices to those users remaining in the default price cap
(CAA 2001c).
290 PRICE REGULATION
OF AIRPORT
SERVICES
have excess capacity, the price cap would not be binding and therefore would not
act as a key driver of airport incentives to invest in additional capacity or to enter
into contracts between the airport and users.

Nonetheless, as the CAA conceded, the derivation of a quality term (or several) for
each UK airport is difficult. Numerous policy decisions are required, including in
relation to:
• the quality attributes to make up the basket;
• the ‘par value’ for each quality attribute in the basket;
• the relative weighting of each quality attribute; and
• the valuation of each quality attribute (in each period) (CAA 2001c).

In this regard, the CAA commented that:

… as there are more dangers in mis-specifying and measuring outputs and thereby
creating undesirable distortions, the proposal is that the service quality dimension of the
price cap should be relatively modest … but set within a framework that allows greater
weight to be placed on them as information systems become more robust, including
concerning costs and benefits. (CAA 2001c, p. 246)

The inclusion of a similar quality term in price caps at Australian airports would be
no less complex and has the potential to amplify the risk of regulatory error already
inherent in setting price caps. In any event, Australia faces a somewhat different set
of circumstances from those of the United Kingdom, in that domestic terminals in
Australia are mainly operated under long-term leases by the major domestic
airlines. This precludes some of the potential for quality differentiation in the
delivery of airport services by the airport; the level of the provision of services
associated with runways, and aprons, is largely non-discretionary. The use of
services, such as aerobridges, and international and domestic common-user
terminals, may be somewhat more discretionary, with airlines demanding various
levels of service.

It also is likely that there would be major disagreements between airports and
airlines as to whether a particular level of service corresponded to the level
specified under the default cap. Westralia Airports Corporation concurred, noting
that the default price cap:
… would be impracticable. The level of service requirement is difficult to specify and
may differ between users. It’s also very hard to measure, and it does depend at an
airport on the level of service provided by third parties. (trans., p. 566)

OPTIONS: COST-BASED 291

AND INCENTIVE
REGULATION
Demand-based variation

At Hamburg Airport, the price cap is designed to share increases in profits due to
exceptional passenger growth. Within a dual-till regime, the X factor is adjusted
according to a sliding scale based on passenger growth. If this growth exceeds
3 per cent then, for every extra 1 per cent growth, there is an increase in the X factor
by 0.5 per cent. (There is no provision for X to fall if passenger throughput falls
short of forecast levels.) This sharing of the extra revenue with consumers is based
on the premise that the extra profitability of the airport is highly correlated with
passenger growth, reflecting non-aeronautical revenues and/or lower unit costs.

Yardstick regulation/benchmarking

Benchmarking or yardstick regulation attempts to separate the regulated firm’s price

structure from its own reported costs. This is done through a benchmarking exercise
which may use a variety of econometric and mathematical programming techniques.

Benchmarks can be used within a price-cap regulatory regime for various tasks.
These include assessing the efficient cost structure of the firm to determine the level
of prices and assessing the value of equity in the determination of the cost of
capital. In setting or resetting the X factor, it is common to benchmark future
productivity gains against past productivity gains within an industry. As well,
benchmarking is used to assess quality performance and commonly is used within
the firm (rather than by the regulator) to assess the firm’s performance across a
range of activities.

The aim of benchmarking under a price-cap regime is to improve the incentives for
a firm by reducing the regulator’s reliance on the firm’s own costs to set the price-
cap parameters. This could provide incentives for cost efficiencies and to invest
appropriately.

Problems of benchmarking

A range of problems, both theoretical and practical, hinders the implementation of

yardstick regulation. As noted above, in practice it is very difficult to ignore
completely the cost base of the individual firm in setting the price parameters. The
costs of setting incorrect prices can be very high. If prices are set too low, then there
is the threat of bankruptcy or at least under-provision and under-investment and, if
prices are set too high, the firm will reap windfall profits.

292 PRICE REGULATION

OF AIRPORT
SERVICES
As Professor Forsyth stated:
The alternative to cost-based regulation is yardstick or benchmark regulation. You
often find references to this and how it’s desirable but regulators always avoid it. They
avoid it simply because it’s risky from the firm’s perspective and therefore from their
perspective. Regulators can be punished if an industry is too profitable — and one can
quote various examples of that. So the safe thing for a regulator to do is to rely heavily
on cost and not too much on benchmarks. So there’s a real problem of getting
benchmark regulation accepted if that’s what’s wanted. (trans., pp. 60–1)

The unreliability of benchmarking is due to the difficulty in achieving comparability

across firms. The conditions under which firms operate vary and each firm will face
different demand structures. Thus each firm may provide a different mix of services
and different levels of quality. Yardstick regulation requires that firms are
homogeneous or that differences are easily identifiable. If this requirement is not
met, a series of judgements needs to be made by the regulator. Australian Airports
(Townsville) did not consider that:
… it is possible to determine an appropriate benchmark due to wide disparity in
demand for, and quality of, services amongst airports. (sub. 14, p. 31)

Application of benchmarking methods, such as data envelopment analysis (DEA), is

becoming more sophisticated. However, even if there were an ideal measurement
method, it would remain difficult to differentiate between efficiency and other
sources of cost differences. Thus, it is difficult to ignore the firm’s own cost
structure in setting prices — a point confirmed in a decision in the Victorian
Supreme Court (May 2001) regarding the resetting of price caps in the electricity
distribution industry (Supreme Court of Victoria 2001; NERA 2001a).

Because there is no objective method for choosing the explanatory variables or for
allocating unexplained cost differences, the benchmarking process requires arbitrary
judgements at crucial points. This leads to an increase in the level of uncertainty and
increases the risk to the firm, which in turn may also raise its cost of capital.

10.2 Future regulation: price caps and Australian

airports
Airport participants in this inquiry generally expressed a preference for no airport-
specific regulation or at least less intrusive regulation — with price caps being their
preferred alternative if that were not accepted. Melbourne Airport recommended
limited regulation but, if explicit regulation were to proceed, suggested a ‘tariff-

OPTIONS: COST-BASED 293

AND INCENTIVE
REGULATION
basket style price cap’ (sub. 7, p. 58). However, the first preference of Australian
Airports (Townsville) was:9
A broad CPI based price cap set for some airport services prices with scope for pricing
reviews for new investment or other special circumstances and no other prices
surveillance or monitoring. (sub. 14, p. 3)

The airlines (and their representatives) have proposed a range of regulatory options,
though generally advocating stricter price regulation than supported by the airports.
Virgin Blue proposed that ‘the CPI-X price cap should be applied to all aeronautical
services’ (sub. 30, p. 25). The Board of Airline Representatives of Australia
(BARA) (sub. 41 and trans., pp. 532–3) suggested rate-of-return regulation as one
sensible regulatory option for airports:
In BARA’s view, an effective, simple and relatively efficient approach to regulating the
price of airport services is rate of return regulation over each airport as a whole.
(sub. 41, p. 53)

One of the main features of this option would be that:

… the airport would be able to set prices across all airport services with complete
freedom, so long as the overall allowable rate of return is not breached. (sub. 41, p. 53)

In assessing future regulatory options, the Commission is required to take into

account the regulatory principles referred to in the terms of reference, cited at the
beginning of this report and in chapter 4. For example, regulation should promote
efficiency and encourage commercially-negotiated outcomes, and target those
aeronautical services and airports where the airport operators have the most
potential to abuse market power.

BARA’s option for consideration — rate-of-return regulation across the whole

range of an airport’s services — ignores the widely-acknowledged problems of this
form of regulation discussed earlier. In particular, if implemented, their proposal
would regulate returns in areas where airports held little, if any, market power.
(This is discussed further in section 10.3 in the context of the single/dual-till issue.)

As discussed above, there are several possible price-cap models as well as a number
of issues and options that need to be canvassed. These include the inter-related
issues of the starting prices, the resetting of X factors and the treatment of new
investment. In addition, the coverage of the price cap needs to be examined, along
with whether further modifications to the price cap are required to allow for
congestion pricing. The efficiency, incentive and compliance effects of each of
these options are also discussed.

9 Australian Airports (Townsville) considers that the balance of market power (at least for
regional airports) lies with the airlines and considers that regulation is required for this reason.
294 PRICE REGULATION
OF AIRPORT
SERVICES
There seems no reason to prefer the use of a revenue-yield approach over the
existing tariff-basket approach to a price cap, particularly if incentives to invest are
considered to be appropriate. It also is clear that yardstick regulation cannot be
relied upon fully to supplant direct analysis of the costs of the firm. Benchmarking
could be one component of price regulation but ultimately firm-specific costs also
will need to be taken into account. Benchmarking has its place in comparing the
performance of airports to both airports within Australia and overseas. It could also
be used when establishing maximum prices between the resettings of the cap. As
well, the terms of reference for this inquiry state that price regulation should
facilitate benchmarking comparisons of airports.

The default price cap raises the issue of the appropriate specification of regulated
services. Although the default price cap has attractive theoretical underpinnings, the
derivation of a quality term that does not distort incentives is likely to be difficult
and complex. The users of aeronautical services will have different base
requirements, the minimum requirement will differ across airports, and disputes are
likely as to whether minimum specified standards are being met. In practice,
effective implementation of a default price cap may not be feasible.

Starting prices, X values and treatment of investment

As discussed above, price-cap regulation generally requires regular reference to

actual costs of providing the regulated services. If this is not done, then prices could
move away from efficient levels (either too high or too low) with associated adverse
effects. Therefore a fundamental issue in designing price caps is to attempt to
ensure, at least over time, that prices roughly converge to their efficient levels. In
particular, to promote dynamic efficiency, price caps should be designed to provide
incentives for efficient investment.

In a competitive market, any new investment will be subject to a straightforward

cost–benefit analysis to determine whether the investment is viable at expected
prices, demand and costs. With regulated firms, the process is reversed. The need
for investment is determined by reference to expected future market conditions, and
the price that is needed to justify the project is then determined.

While an airport with market power may have some incentives to under-invest
(chapter 7), price caps introduce a different set of incentives and potential
distortions. As observed by Professor Forsyth:
Price-caps are a simple form of control, but they open up many subsidiary problems.
Further, quite detailed regulation is needed to address quality degradation problems and
investment matters. (sub. 5, p. 35)

OPTIONS: COST-BASED 295

AND INCENTIVE
REGULATION
There are four broad types of investment that need to be considered: cost saving;
replacement/maintenance; capacity enhancing; and quality enhancing. In practice,
more than one of these attributes is likely to be associated with a particular
investment project.

Cost-saving investment

Cost-saving investment typically reduces the cost of operating and/or maintaining

current assets. Under price-cap regulation, airports have an incentive to embrace
cost-saving investment because they can retain profits, at least for the term of the
regulatory period. Thus, there should be no need for additional incentives for the
airports to undertake this type of investment, provided the airport can retain at least
some of the savings.

Replacement/maintenance investment

This is investment that is aimed at preserving the service potential of an existing

asset. It is also the area that has caused considerable contention in applying the
necessary new investment (NNI) arrangements (chapter 8). If regulated prices are
broadly adequate for long-term supply of the airport service, then:
To the extent that this investment represents the most efficient means of maintaining
without enhancement existing service potential, it is most appropriate to treat such
investments as though they were a form of maintenance expenditure.
Given the neutral impact on both costs and revenues over time from planned like-for-
like cyclic renewal of assets, investments with a maintenance purpose should be
included within the price cap. These would provide no basis for modifying the price
cap parameters. (Network Economics Consulting Group (NECG), quoted in ACCC,
sub. 36, p. 101)

In other words, if prices are set at an efficient level, then replacement of existing
capital can be expected to proceed, because it would be commercially viable. As
stated by SACL:
The reason that an NNI arrangement is required is that existing prices are below
incremental cost. If prices approximated incremental cost, new investment could be
funded from expected increased revenues from the volume growth, rather than unit
price increases. (sub. 27, p. 12)

Capacity-enhancing investment

Capacity-enhancing investment typically allows for traffic growth over time. If the
increase in capacity increases revenue, the less the requirement for additional cost

296 PRICE REGULATION

OF AIRPORT
SERVICES
pass-through under a price cap to provide appropriate incentives. As pointed out by
Professor Forsyth:
When extra capacity enables extra traffic to be handled, revenues will increase. There is
no general reason why the price of using the airport should increase as its traffic
expands … (sub. 5, p. 26)

And the MTAA Super Fund noted that:

Moreover, although a price capped firm is unable to raise prices as demand expanded,
it would still have an incentive efficiently to invest to expand capacity as long as
regulated prices gave a sufficient margin over operating costs. (sub. 22, p. 53)

However, if regulated prices are such that an airport currently is making a loss on
the provision of a service, then there is no incentive for the airport to increase
capacity — it will just lose more.10

Quality-enhancing investment

Quality-enhancing investment increases the level of quality for a given service (for
example, an upgrade of terminal facilities). Under a price cap, an airport is unlikely
to be able to capture the extra benefits from a quality-enhancing investment, unless
the quality improvement encourages spending on services outside the cap or
increases demand for the airport’s services. It is likely that extra incentives,
probably an increase in prices, will be required to encourage this type of investment
by airport operators.

Where the benefits of such investment are likely to accrue to existing airlines, their
involvement in the approval process may be appropriate. However, airlines can
have divergent quality requirements. Even if existing airlines benefit, they may
oppose a quality improvement that would facilitate new entry.

As noted above, in practice it may be very difficult to differentiate between types of

investment, including capacity- and quality-enhancing investments. Because of the
lumpiness of most airport investment, capacity-enhancing investment typically will
result in higher quality, at least in the short term. For example, duplicating a runway
will (among other outcomes) reduce the time delays for existing flights, which
translates into a better service for the passenger.

10 Unless: (a) the airport expects a very large increase in demand such that unit costs fall; (b) costs
fall over all output ranges due to technological improvements; or (c) extra capacity encourages
additional spending on services outside the price cap.
OPTIONS: COST-BASED 297
AND INCENTIVE
REGULATION
Alternative price-cap treatment of investment

Apart from cost-saving investment, which the price-capped firm always will have
an incentive to undertake (because it can retain the benefits within a period), the
incentives to undertake the remaining types of investment depend crucially on the
detailed design of the price cap. The treatment of investment within a price cap is
inter-related with the starting prices for the regulated services and how the Xs are
set. Thus, for example, if there were no cost pass-through provisions for any
investment, then the starting prices and Xs would have to be set to accommodate all
future investment.

Figure 10.1 sets out three possible approaches to implementing price caps — each
of which, in principle, would provide for efficient investments and prices, at least
over time. These three approaches are not exhaustive; for example, the situation in
which starting prices are above levels required to provide for all investment is not
shown. In such a case there would be no need for provisions for the pass-through of
investment, while the X values could be made sufficiently large to drive prices to
their efficient levels over time.

The three interdependent decision variables listed down the left-hand side of
figure 10.1 are:
• starting prices;
• the setting of X values; and
• the scope of investment cost pass-through provisions.

The fourth (bottom) row represents other possible pass-through provisions for costs
outside the airport’s control — for example, noise levies, and mandated safety and
security provisions.

Three approaches to price caps are outlined, differentiated by starting prices.

Approach A assumes starting prices are at a level below that required to justify any
efficient investment (except cost-saving investment). The analysis in chapter 8
suggests that this was the situation at a number of core-regulated airports when
privatisation occurred. Under approaches B and C, starting prices are set at a level
that is considered adequate for encouraging efficient replacement/maintenance and
capacity-enhancing investment over the regulatory period. Approaches B and C
differ in their treatment of quality-enhancing investment.

298 PRICE REGULATION

OF AIRPORT
SERVICES
Figure 10.1 Price-cap approaches

A B C
Starting Starting prices that do not Starting prices that allow Starting prices that allow
prices allow for any investment for expected replacement/ for all expected
maintenance and capacity investment
enhancing investment

1 2 3 4 5 6

Setting Productivity Productivity Productivity Productivity Productivity Productivity

X only Rep/main I Rep/main I only Quality I only
values Capacity I Quality I
Capacity I

Scope of Rep/main I Quality I Exceptional I Quality I Exceptional I Exceptional I

pass-through Quality I
provisions for Capacity I
investment

Scope of other
pass-through Noise, safety, congestion etc Noise, safety, congestion etc Noise, safety, congestion etc
provisions

Rep Replacement. Main Maintenance. I Investment.

The solid arrows within each of the three approaches represent six feasible paths,
given the starting price decision, that could, in principle, deliver efficient pricing
and investment outcomes. (The path illustrated by the dashed line is discussed
later.)

If starting prices are below the levels required to cover any expected investment,
then there is likely to be a need to allow price adjustments for capacity-enhancing
and replacement/maintenance investment as well as for quality-enhancing
investment. All paths (1, 2 and 3) in approach A allow for anticipated productivity
growth to be incorporated in the X factor. However, the calculation of the X values
differs across the three paths with respect to the types of investment incorporated. In
path 1, the X values incorporate expected productivity improvements only, whereas
path 2 X values incorporate an allowance for replacement/maintenance and
capacity-enhancing investment as well as an allowance for productivity growth. In
path 3, the X values incorporate all of the investment and expected productivity
gains incorporated in path 2, as well as an allowance for quality-enhancing
investment.

OPTIONS: COST-BASED 299

AND INCENTIVE
REGULATION
The Xs in paths 4, 5 and 6 incorporate an adjustment for anticipated productivity
gain. However, starting prices are at levels adequate for replacement/maintenance
and capacity-enhancing investment. Hence, paths 4, 5 and 6 differ only in their
treatment of quality-enhancing investment. In path 4, the X values incorporate
expected productivity improvements only. In path 5, quality-enhancing investment
is incorporated into the X values, whereas in path 6 it is incorporated into the
starting prices.

Given decisions about setting starting prices and Xs, each of the paths requires
varying scope for the pass-through of investment costs. Path 1 allows for all
investment costs (except cost-saving investment) to be to passed through, whereas
under paths 2 and 4, only the costs of quality-enhancing investment can be passed
through. Under paths 3, 5 and 6 there is no need for ongoing investment assessment
(as investment is incorporated in the starting prices or Xs), but there is a need, in
exceptional circumstances, to allow for variations in cost pass-throughs provided for
investment.11 This will cover situations unforeseeable at the time the Xs are set.
(Where there is provision for assessment of investment on a continuing basis
(paths 1, 2 and 4), there is no need for such an ‘escape’ clause.)

While, in principle, each of approaches A, B and C can provide efficient outcomes,

they:
• require different levels of regulatory involvement (at different times);
• provide a range of incentives for strategic behaviour and commercial negotiation
by the parties;
• have different effects on the time paths of prices; and
• have different effects on investment incentives.

Advantages and disadvantages of different approaches and paths

Approach A has the immediate appeal of administrative simplicity in that there is no

need to engage in the lengthy and complicated process of estimating appropriate
prices (based on an airport-wide cost assessment) that would cover all expected
investment (box 10.2) or, alternatively, to use a benchmark to set prices. Price caps
were developed to reduce the disincentives of tightly linking regulation of prices to
the assessment of a firm’s costs. However, as noted above, prices eventually need to

11 Exceptional circumstances for investment may include situations requiring unexpected

investment, or the delay or cancellation of investment that has been allowed for in the cap.
These could be due to a significant change in demand — such as experienced by Brisbane
Airport following the Asian economic crisis or by all airports after the events of September
2001 — or a radical improvement in technology that affects costs significantly.
300 PRICE REGULATION
OF AIRPORT
SERVICES
converge towards efficient long-run levels. Thus, if starting prices are not at
efficient levels, then other adjustments eventually will be needed.

Path 1 is the easiest path to implement at the start of the regulatory period. Prices
are carried forward and only expected productivity improvements need be
calculated. However, this ostensible simplicity is counteracted by the need for an
ongoing examination of every proposed investment (excluding cost-saving
investment). This assessment is likely to be complicated by the uncertainty
surrounding the existing price levels. To determine the amount of adjustment to the
prices for aeronautical services, the appropriateness of each investment will need to
be determined, as well as a dollar value for all the investments. Then there will need
to be an assessment of whether current prices will sustain these investments; if not,
an upward adjustment of prices will be required.

Some other aspects of path 1 are appealing. As discussed in chapter 8, the process
of cost pass-through can provide an opportunity for consultation between users and
the airport, possibly with a view to commercially-negotiated outcomes rather than
regulated ones (box 10.4). Also, the need for assessment of investment proposals by
the regulator occurs only when the investment proposal is submitted, therefore
avoiding the information problems and risks associated with planning investment
several years ahead.

A significant disadvantage of adjusting prices as new investment comes on line is

that it can be administratively complex and costly for all parties involved. Under the
current regime, this complexity has been exacerbated by the need to apply for
approval of very small investment projects if a price increase is sought (chapter 8).
As the ACCC stated:
So far the Commission has received dozens of applications for new investment
proposals, including for small items such as drainage works, baggage room fans,
installation of doors to passenger terminals and revegetation works. The process
imposes administrative costs on the airport operator, airport users and the regulator and
in some cases may delay the investment works. (sub. 36, pp. 105–6)

On this point Qantas Airways agreed with the ACCC that:

… continuous new investment applications to the ACCC are administratively
inefficient. (sub. 48, p. 22)

In addition, incumbent airlines may have an incentive to frustrate new entrants by

delaying capacity-enhancing investments and could use new investment approval
procedures to achieve this delay (chapter 8). As stated by Impulse:
It is our belief that the current regime is not working as appropriate levels of
infrastructure development since the entrance of new competitors such as Impulse has
not occurred. (sub. 18, p. 3)

OPTIONS: COST-BASED 301

AND INCENTIVE
REGULATION
 Box 10.4 Project control groups
User support and the encouragement of commercial negotiation is an important aspect
of a regulatory regime. Given that there are a relatively small number of large users
(the airlines), negotiation with individual airports is feasible. Brisbane Airport has
established project control groups, comprising airport and airline representatives, to
assess a range of investment proposals. This process has been established
voluntarily, although user support is one of the criteria that the ACCC uses to guide its
assessment of investment proposals. The process generally appears to have been
successful. However, with Brisbane’s 2000 investment proposal, there was
disagreement between the airport and the airlines on whether three proposed projects
were appropriately considered ‘necessary new investment’ (NNI).
NECG, as consultants to the ACCC, (sub. 36, attachment D) raised some issues
concerning formalisation of this process within a price-cap regime, including the
behaviour of the users and the potential for strategic behaviour within the control
group. This behaviour includes:
• cheap riding — where one of the parties to the agreement understates the value (to
it) of the project being considered;
• forced riding — where larger airlines may force through an investment that causes
disproportionate costs on a smaller airline; and
• barriers to airline entry — where large incumbent airlines may vote against entry-
facilitating investment.
Source: ACCC (sub. 36, attachment D).

Paths 2 and 3 (approach A) again have low starting prices but require some
assessment of investment at the beginning of the regulatory period for incorporation
into the X values. Path 2 will require a relatively high level of regulatory
involvement because investment needs must be assessed in setting the X factors as
well as on a continuing basis. Also, there will need to be a distinction made between
replacement/maintenance, capacity-enhancing and quality-enhancing investment. In
practice, this could be a very difficult and, at times, arbitrary determination.

Under path 3, all foreseeable investment is included in the X factors. There is no

need to differentiate between types of investment. However, the investment
requirements for the regulatory period will have to be forecast by the airport (and
users) and assessed by the regulator. These process will require detailed information
on expected future costs and demand growth during the regulatory period. Except
for unforeseeable circumstances, new investment will have been set for the
regulatory period.

Path 3 shares characteristics with paths 5 and 6 in that all investment must be
assessed at the beginning of the regulatory period. Approaches B and C are
302 PRICE REGULATION
OF AIRPORT
SERVICES
characterised by the setting of the initial prices at levels adequate for the specified
investment. Compared with paths 1, 2 and 4, the ongoing information requirements
for paths 3, 5 and 6 are reduced — though there remains the substantial task of
determining the initial efficient price levels and Xs for aeronautical services. Major
elements of such an exercise are outlined in box 10.2.

Under the system in the United Kingdom, prices are reset every five years based on
a review of expected costs and demand. New investment over the regulatory period
is estimated and incorporated into the anticipated costs after a consultation process
with the industry. This system is similar to paths 3, 5 and 6, as all investment is
incorporated in the starting prices or the X values at the start of the regulatory
period. Although the UK approach has been to accept airport investment proposals
with minimal regulatory assessment, the potential always remains for intense
regulatory involvement in the investment decision-making process during price-cap
reviews.

The longer the price-cap period, the more difficult it is to predict the level and cost
of required investment. In addition, airports may have an incentive to overstate
investment requirements, to procure excessive up-front price increases, while
airlines may be tempted to understate their forward investment requirements to
achieve lower charges.12 Once an investment program is approved, there is an
incentive for airport operators to delay the investment process, because pricing
parameters have been fixed for the regulatory period. Hence, the regulator must
ensure the integrity of the investment plan by monitoring during the regulatory
period. These tendencies, and other failures of the process, have led the CAA
(2000e) to request feedback from airports and airlines on ways to ‘incentivise’
investment.

The potential inflexibility in the forward-looking approach would need to be

addressed. If, for example, after privatisation charges at Australian core-regulated
airports had been established based on a forward investment plan, then some
unanticipated new investment — such as the domestic common-user terminal at
Melbourne Airport — might have been frustrated. Conversely, unexpected falls in
demand may require the deferral of planned investments, as has occurred with the
Adelaide Airport multi-user terminal. An escape clause for both the regulator and
the airport would be advisable to accommodate unforeseen events that have major
ramifications for airport investment.

12 Such an approach would only be beneficial to airlines if they are still able to achieve their
desired airport investment profile in spite of airport charges being set too low.
OPTIONS: COST-BASED 303
AND INCENTIVE
REGULATION
Alternative paths?

The ACCC suggested several price-cap approaches (sub. 36). Essentially, all would
carry forward price levels from the current regulatory regime (approach A).13
However, the proposed approaches are unclear with respect to
replacement/maintenance and capacity-enhancing investment. The ACCC (sub. 36)
referred to its preferred option as a ‘hybrid’ approach. This would allow only pass-
through (upward adjustment of prices within the regulatory period) of the costs of
investment for major projects. Again, starting prices would be kept at the level
applying at the end of the current price cap; alternatively, an initial minor
adjustment could be made for expected small new investments.

If starting prices were not adjusted for smaller investment projects these would be
funded through adjustments to the Xs. The ACCC recommended that the distinction
between small and large investment be based on an (unspecified) dollar value. It
recognised that ‘the ‘X’ values should be consistent with the approach adopted in
relation to new investment’ (sub. 36, p. 110).

Another categorisation for the distinction between investment in the ACCC’s hybrid
approach is ‘developments that could not be anticipated at the time the ‘X’ values
were set’ (sub. 36, p. 107). This suggestion is somewhat similar to path 6, if prices
were set at the appropriate level at the start of the regulatory review period. If the
prices are not set in this way, then the process would proceed on a similar basis to
paths 3 or 5, in that all foreseeable investment would be assessed at the beginning of
the regulatory period and prices (including Xs) would be adjusted to accommodate
the expected investment program.

If there were no provision made for replacement/maintenance and/or capacity-

enhancing investment in the X values (and starting prices were not adequate for
such investments), then efficient pricing would not be achievable. For example,
even in principle, the path represented by the dashed line in approach A
(figure 10.1) cannot promote an efficient outcome (in the sense of aeronautical
revenue covering stand-alone aeronautical costs). With starting prices too low to
cover investment and without specific provision for replacement/maintenance and
capacity-enhancing investment, airports’ ability to invest in
replacement/maintenance or capacity investment could be impaired (unless the
investment were funded from non-aeronautical revenue).

13 The ACCC considered that an allowance for smaller new investments could be factored in
either through adjustment to the X factors or starting point prices.
304 PRICE REGULATION
OF AIRPORT
SERVICES
The ACCC indicated that its hybrid approach would allow higher prices to recover
the costs of projects that would qualify as major development projects under the
Airports Act 1996.
The pass through provisions should not distinguish between ‘replacement’ and ‘new’
investments. In other words the pass through provisions should be available for
investments irrespective of whether they replace existing assets or add to capacity or
quality. (sub. DR55, p. 17)

With the aim of lessening regulatory costs, the ACCC hybrid approach would allow
for smaller new investment (but presumably not replacement investment) being
included in the X values or in the starting point prices. The difference between the
two alternatives is a matter of the timing of the effect on allowable prices of such
investment over the regulatory period. If small new investments are included in the
starting prices, there will be a relatively small initial price increase, whereas
including them in the Xs will lead to each annual price change being marginally
higher than otherwise.

Over time, as more aeronautical infrastructure is replaced, prices under the hybrid
approach will tend to increase towards those necessary to cover aeronautical costs
(including appropriate returns on assets).14 The ACCC’s argument for the extension
of cost pass-through to major replacement investment, which is excluded under the
current framework, would seem to indicate a recognition that current aeronautical
prices under the price cap may not be sufficient to justify replacement investment.
The ACCC (sub. 36) observed that current prices were unlikely to reflect
aeronautical costs. Its apparent reluctance for starting prices to be reset to the levels
needed to cover such investment (at least at the beginning of a new regulatory
period) appears to reflect a concern that a rise in prices would deliver windfall gains
to airport operators. The ACCC explained that:
Introducing new CPI-X price caps for regulated airports raises the question of how to
set starting point prices and the ‘X’ values. The Commission [ACCC] proposes that
prices from the current regulatory framework should be carried over to form the
starting point for a new price cap. The alternatives, such as setting starting point prices
to reflect costs, are likely to result in significant increases or decreases in charges.
Given that these starting point prices relate to existing, mostly sunk assets, there is little
if any reason to make such a change from an economic efficiency perspective. Instead
the main effect of such a change would be a distributional one, either a transfer from
airlines and their passengers to airport operators or vice versa. (sub. 36, p. 10)

14 If the starting prices already include some returns to capital, then allowing price increases for
replacement investment could eventually lead to prices overshooting efficient levels. The price
increase allowed would need to be tempered by the depreciation and returns on existing assets
already contained in the starting prices.
OPTIONS: COST-BASED 305
AND INCENTIVE
REGULATION
The scope for windfall gains (or losses) would, however, depend on the
expectations of bidders regarding aeronautical price regulation (including NNI pass-
through provisions) at the time of bidding. As discussed in chapter 8, bidders might
have had some grounds for expecting regulated outcomes different from those that
transpired. Moreover, the price-cap regime was locked in for five years only and
described as a transitional regime. Bidders also were informed that single-till
pricing would not be mandated (DoTRD 1996) and yet this approach remains
incorporated to a substantial degree in current prices courtesy of the Federal
Airports Corporation charges used as the starting prices for the post-privatisation
regime. The Government’s commitment to a dual-till approach to establishing
maximum regulated prices was confirmed in its direction to the ACCC regarding
treatment of non-aeronautical revenue in the assessment of SACL’s pricing
proposal (Direction No. 22).

Though a price-cap approach that perpetuates existing prices, with appropriate

adjustments for investment as the need arises, has some attractive features
(including that prices adjust slowly and that there is scope for consultation between
the parties about investment proposals), such a model, for reasons outlined above,
may encourage strategic behaviour, particularly by airport users, that frustrates
appropriate investment. It also may be administratively costly and protracted.

The ACCC is correct in observing that failure to provide sufficient returns on sunk
airport assets is likely to have limited efficiency effects, at least in the short term.
This is so for any industry with excess capacity. However, regulators have
recognised that longer-term considerations make short-run marginal-cost pricing —
the logical conclusion of the ACCC’s sunk costs argument — inappropriate on
efficiency grounds. In its decision on Sydney Airport, the ACCC (2001i) took a
longer-term view and allowed prices to rise to a level that provided returns on sunk
assets.

The Commission considers that setting starting prices for any new price cap to
reflect efficient long-run aeronautical costs offers a superior approach that would
reduce ongoing regulatory involvement with its associated costs and risk of error.
The CAA has made a preliminary proposal to rebase price caps for a number of UK
airports and observed that:
While the regulatory system would allow the incremental returns from additional
investments to cover their incremental costs eventually, it provides poor signals for the
airports and users as to the real costs and benefits of delivering additional outputs.
(CAA 2001c, p. xvii)

Where rebasing price caps would involve significant increases in aeronautical

prices, adopting a gradual path to the appropriate level is an option. Indeed some
airports suggested that this is the approach they would adopt.
306 PRICE REGULATION
OF AIRPORT
SERVICES
Whichever price-cap approach is selected, it needs to be implemented in a way that
clearly and transparently sets out how appropriate investment will be
accommodated, and it should set prices that will encourage appropriate investment.

Implementation

A price cap could be implemented under the:

• Prices Surveillance Act 1983 (PS Act);
• Trade Practices Act 1974 (TP Act); or
• Airports Act 1996.

The current price-cap regime is administered under the PS Act (chapter 3). In its
review of that Act, the Productivity Commission (PC 2001c) has proposed that the
Act be repealed. A CPI-X price cap could be administered within the TP Act either
via generic provisions (under Part IIIA as an industry code, for example) or an
industry-specific provision. Industry-specific provisions also could be placed in the
Airports Act, with the ACCC appointed to administer the cap.

Given that price-cap regulation concerns the competitive conduct of airports, its
inclusion within the TP Act rather than the Airports Act would seem desirable if the
PS Act were repealed. The Commission also considers that any price-cap regime for
airports should be implemented as an industry-specific amendment to the TP Act
(as for telecommunications (PC 2001d)).

Summary

In theory, price caps can be devised that could, at least over time, deliver efficient
prices and investment. The important questions are the likelihood of achieving such
a result in practice, and the direct and indirect costs involved in implementing the
regime. The various approaches outlined require different levels of regulatory
intervention and provide for different opportunities for commercial negotiation and
strategic behaviour by the parties. Major trade-offs relate to the level of ongoing
assessment of investment proposals and the time over which any adjustment needed
to bring prices to efficient levels are made.

The scope for game-playing is likely to be reduced if it is made clear to all parties
which price-cap approach has been chosen and, therefore, how different types of
investment will be accommodated. Nonetheless, the risk of regulatory error
remains, as does the almost inevitable convergence of price caps towards cost-based

OPTIONS: COST-BASED 307

AND INCENTIVE
REGULATION
regulation and the substantial informational requirement and poor incentives for
efficiency inherent in that form of regulation.

In particular, because firms always possess more knowledge about the firm and its
markets than possessed by the regulator (information asymmetry), the firm
generally is in the best position to determine future cost structures, demand
forecasts and many other important facets of its business. It is difficult for a
regulator without this detailed knowledge to determine appropriate prices and
investment. For some airports, the importance of, and inherent problems in, valuing
airport land make regulatory determination of efficient airport prices particularly
difficult (appendix F).

Given this limited information and the resultant likelihood of regulatory error,
pricing rules applied by regulators can have an adverse impact on investment in
essential infrastructure. While firms will make investments based on an expected
probability distribution of returns, the common objective that regulated prices
should provide a firm with the market average rate of return, on average, will result
in below average returns. While the regulated price aims to preclude excessive
returns, if market conditions generate below normal returns, the firm is not
adequately compensated.

As noted in the Commission’s review of the National Access Regime (PC 2001a),
the result of setting prices too low is likely to be a failure to invest. This is
particularly so where there is lumpy investment with decreasing costs in which case
the investment may not be undertaken if the regulated price is set too low. Hence
the resultant costs to the community of inefficiently low regulated prices are
expected to be significantly greater than the costs that would flow from regulated
prices being set a little too high. The CAA (2001c) also noted the significant
distortions of imposing a regulated price that is too low and consequently argues for
a price cap slightly higher than the regulator’s best guess of efficient prices.

FINDING 10.1

Incentive regulation has some advantages over cost-based regulation. However,

given the tendency of price caps to converge towards cost-based regulation (with
associated high levels of regulatory involvement and risks of regulatory error, and
the consequences of these for investment and long-term provision of services),
price-cap regulation should be implemented only when there is clear evidence that,
without such regulation, economic efficiency would be seriously impaired.
Where price caps are implemented, the approach adopted for investment should be
spelled out clearly and transparently to all relevant parties from the outset, in order
to reduce the risk of inefficient outcomes and excessive gaming.

308 PRICE REGULATION

OF AIRPORT
SERVICES
10.3 Coverage of price caps and cost-based regulation
A crucial issue in implementing rate-of-return or incentive regulation for airports is
determining which services should be regulated. There are three broad options:
• regulate the prices of all services provided by (or at) the airport;
• regulate prices of only those services in which the airport has market power; or
• regulate prices of only those aeronautical services in which the airport has
market power.

The terms of reference for this inquiry set out several principles to guide future
price regulation, including that ‘future prices regulation should be applied to those
aeronautical services and those airports where airport operators have most potential
to abuse market power’ (in essence, the last option above). Moreover, core-
regulated airports were privatised on the basis that a single till would not be
mandated (DoTRD 1996) and the Government issued a Direction (No. 22) to this
effect in relation to Sydney Airport in April 2001. Nonetheless, several participants
argued in favour of broader application of price regulation on efficiency grounds.
These arguments are assessed here and in more detail in appendix C.

Rate-of-return regulation across all airport activities

As one option for regulation of airports, BARA (sub. 41) suggested imposing rate-
of-return regulation across the entire range of services provided by airports (but not
for individual services). The rationale for regulating all airport services is that, for a
decreasing cost, multi-product monopoly to cover its fixed costs, while minimising
efficiency losses of pricing above marginal cost, prices should be set such that
markets with relatively inelastic demand bear a greater share of fixed and common
costs than do markets with more elastic demand (Ramsey pricing). In principle,
rate-of-return regulation applied across the entire range of airport services would
encourage such pricing. It also has the practical advantage of not requiring cost
allocation across various services.

While this model may be appropriate if firms have market power in all the services
they provide, if profits arise from locational advantages rather than market power,
then rate-of-return regulation could have undesirable consequences. As discussed
below, if an airport operator were regulated such that additional rents earned could
not be retained, the incentive to earn such rents would be removed. This could mean
that airport land was not used in the most efficient way. Therefore, rate-of-return
regulation may not bring about lower aeronautical prices. Rate-of-return regulation
also is likely to introduce incentives for the airport to allow costs to increase across
the whole range of airport activities (section 10.1).
OPTIONS: COST-BASED 309
AND INCENTIVE
REGULATION
Single-till price-cap regulation

Single-till price-cap regulation, whereby a price cap is imposed only on aeronautical

activities, but where the setting of the price cap takes into account some or all
profits earned in non-aeronautical activities, is a variation on the theme of rate-of-
return regulation of all airport services. The major difference is that the single-till
price-cap model effectively forces cross subsidisation of aeronautical activities from
any above-normal non-aeronautical profits, and provides some incentives for cost
minimisation in the provision of non-aeronautical and aeronautical activities, at
least during the period of the price cap.

One argument put for a single till is that profits from non-aeronautical activities
may reflect market power rather than locational rents. If this were the case,
however, then it may be more appropriate to reduce prices directly to promote
efficient consumption and protection of consumers of the service in question. The
single till would not prevent excessive pricing of non-aeronautical services by
airports; it merely would require that any monopoly profits were transferred from
consumers of those services to the owners of airlines and/or their passengers. As
Kahn observed:
It is no more consistent with economic efficiency or fairness if prices for restaurant
meals, duty-free sales, car parking or other commercial services at airports are set at
excessive levels, than if airlines were subjected to excessive charges for aviation
services. Moreover, the inefficiency resulting from the former monopolistic pricing
would not be mitigated, but compounded if the excess revenues were used to hold other
airport charges below the level of marginal cost. (Kahn 1991, p. 20)

To the extent that a single till discouraged an airport operator from earning any
monopoly profits in non-aeronautical activities, this would be a desirable, if
possibly unintended, consequence of such an approach to regulating aeronautical
prices.

BARA, Ansett and Qantas (subs 41, 42, and 48) suggested that efficient airport
pricing requires that some or all locational rents earned at the airport be applied to
reduce aeronautical charges. BARA provided some illustrations effecting this result.
However, even in highly-competitive industries such as petrol retailing (cited by
BARA, sub. DR54), price-taking producers (or land-owners) in relatively high-yield
locations (whether because of convenience or natural phenomena) will earn and
retain locational rents.

Ansett argued in a similar vein that:

In the world of aviation, airline operators can be considered the anchor tenants of an
airport. Hence, the value of an airport is largely driven by airline operations, even when

310 PRICE REGULATION

OF AIRPORT
SERVICES
 the land is on ‘prime’ real estate. If the market for airport services was competitive,
airlines would not be expected to contribute to ‘locational rents’. (sub. 42, p. 22)

What an anchor tenant pays, however, will depend on the relative attractiveness of
the location. At very attractive (and profitable) locations, the owner of a shopping
complex is likely to have a choice of anchor tenant in which case higher rentals can
be charged. In other cases, there will be a ‘scarcity’ of tenants and they will be
rewarded accordingly.

As discussed in chapter 7 and appendix C, airports are likely to ‘reward’ airlines for
additional passenger throughput in recognition of any demand complementarities
between aeronautical and non-aeronautical services. The airline, in this case,
provides a scarce and profitable resource (passengers) to the airport. But, there does
not appear to be any reason that airlines (and their passengers) should receive all
locational rents earned at airports. In this regard, the CAA recently concluded that it
‘does not accept that airlines have any intrinsic rights over net commercial revenues
generated at the airports’ (2001c, p. 13). Importantly, others contribute to the
earning of rents (including the airport lessee who develops and allocates appropriate
spaces within the airport, thus promoting the creation of such rent, the airport land
owner and concessionaires).

Another argument for a single till, addressed above in relation to BARA’s

suggestion for rate-of-return regulation, is that a multi-product monopoly (with a
requirement to be (just) self-financing) should be encouraged to set prices in each
market to minimise efficiency costs — that is, to implement Ramsey pricing (Crew
and Kleindorfer 2001).

This argument appears to rest on two assumptions — namely, that there are
significant common costs in supplying aeronautical and non-aeronautical services,
and that demand for non-aeronautical services is less elastic than demand for
aeronautical services. If these assumptions hold, then common costs could be
covered more efficiently by increasing the prices of non-aeronautical services
relative to aeronautical services. However, the extent of common costs in the
provision of aeronautical and non-aeronautical services appears limited. There may
be some shared terminal costs but, even in this instance, costs are likely to be
separable. In addition, as the ACCC (sub. 36) pointed out, for those non-
aeronautical services sold in competitive markets, prices cannot be raised above
marginal cost. Hence there would appear to be limited scope for prices to exceed
marginal cost in order to make a contribution to any common, fixed costs.

A case for a single till can be made on the grounds that non-aeronautical earnings,
to the extent that they are locational rents, could be ‘taxed’ in a non-distorting
manner (that is, without affecting economic behaviour) and used to reduce
OPTIONS: COST-BASED 311
AND INCENTIVE
REGULATION
aeronautical charges to ensure short-run marginal-cost pricing of those services. In
other words, locational rents could be used to pay for the airport’s aeronautical
assets. As discussed in appendix C, there may be an in-principle case for such cross
subsidisation, at least at uncongested airports where the marginal costs of providing
aeronautical facilities are likely to fall below average costs and where pricing of
aeronautical services would otherwise result in some marginal sales being lost.15

However, the disincentive effects of this for airport development of both

aeronautical and non-aeronautical services may cause substantial efficiency
losses.16 In part, this is because an airport operator that knows that any additional
locational rents earned will be used to reduce aeronautical charges (by more than
the amount an airport operator may reduce such charges voluntarily — see chapter 7
and appendix C) will have a reduced incentive to earn such rents in the first place.17
This may result in:
• the allocation of airport space by the operator to lower-valued uses, in which
case some or all rents are forfeited (representing an efficiency loss);
• the ‘giving away’ of rents to existing concessionaires (such as retailers) by way
of low rentals, in which case the rents continue to be earned but there is no
transfer to aeronautical service users;
• the ‘giving away’ of rents to concessionaires but where competition for the rents
amongst potential concessionaires means that rents are dissipated to some degree
in rent-seeking activities; or
• over-capitalisation of, and cost-padding in, non-aeronautical activities by the
operator to reduce measured rents (thus also generating an efficiency loss).

Moreover, aeronautical charges could be pushed below their marginal cost.

Particularly at capacity-constrained facilities, a single till would push prices away
from market-clearing levels and exacerbate excess demand. In this situation, airport
profits from non-aeronautical activities would be transferred mainly to airlines (if
the latter were allocated slots under a demand management scheme).18

15 If an airport discriminates in its pricing reasonably efficiently, the marginal distortion will be
small. In this case, a single till largely would redistribute income without promoting efficiency.
16 To the extent that an airport can retain additional profits during the price-cap period, there will
be some incentive to make short-term investments that result in higher non-aeronautical profits.
17 Though it may be feasible to design an arrangement that would provide incentives for the
airport operator to earn additional locational rents (such as an up-front agreement under which
an airport operator pays expected infra-marginal rents), single-till arrangements in practice tend
to appropriate ‘excess’ (marginal) rents.
18 If demand were not rationed, then queuing, for example, could dissipate surplus.
312 PRICE REGULATION
OF AIRPORT
SERVICES
In addition, as noted above, owners of leases for core-regulated airports bought the
leases on the basis that they could retain profits from services that were not subject
to the price cap. The (discounted) expected value of this profit stream largely would
have been factored into bid prices, benefiting the Commonwealth Government (and
taxpayers). Thus, the Government effectively has imposed an up-front ‘tax’ on
expected locational rents.

Price regulation of services where an airport has some market power

A dual-till approach refers to regulation where only those services in which the
airport is considered to have market power are subject to a price cap. Typically the
regulated ‘till’ includes only those aeronautical services in which the airport has
market power. A dual till may introduce some distortions in favour of non-
aeronautical investments, particularly if the price cap, over time, effectively limits
the rate of return allowed on aeronautical services. On the other hand, compared
with a single till, because a dual till allows aeronautical cost recovery, airports may
have a greater incentive to invest in aeronautical (as well as non-aeronautical)
activities and increased capacity.

A major practical issue in dual-till regulation is the need to distinguish services in

which the airport has market power from those in which it has little market power,
and then to separate the costs of providing these classes of service. For Australian
core-regulated airports, this task largely has been done under current price-cap
arrangements, though there is some dispute about airport market power in some
designated aeronautical-related services (for example, refuelling, and taxi and car-
parking facilities).

In chapter 6 it was found that major airports are likely to have a relatively high
degree of market power in refuelling services. If continued price regulation of
airports were considered appropriate, then because refuelling is closely related to
aircraft movements, its inclusion in a price cap for aeronautical services at those
airports with market power would seem appropriate. This could conflict with
undertakings given to airports prior to sale, however.

Car-parking and taxi facilities involve somewhat different circumstances — first,

because in the Commission’s view, airport market power in these facilities is likely
to be moderate (at most) and, second, because these services are not part of the
bundle of core-aeronautical services provided by airports and required by airlines.
Thus, even if it were considered that these services should be subject to some form
of price regulation, it may be more appropriate to apply regulation separately.

OPTIONS: COST-BASED 313

AND INCENTIVE
REGULATION
FINDING 10.2

The Commission considers that if price-cap regulation were to be continued for any
airports, then the cap should apply only to those aeronautical services in which the
airport has substantial market power. Profits earned in non-aeronautical activities
should not be taken into account in setting the cap.
If an airport exercises significant market power (as opposed to earning locational
rent) in any non-aeronautical activity, separate price monitoring or other
regulation may be appropriate.

314 PRICE REGULATION

OF AIRPORT
SERVICES
11 Regulatory options: price monitoring,
access provisions and general
competition law

This chapter continues the discussion of alternative regulatory options from the
previous chapter. Price monitoring, access provisions and general competition law
are discussed as alternatives or supplements to other forms of price regulation.

11.1 Price monitoring

Price monitoring can be described as a lighter-handed alternative to price caps or
cost-based regulation. Though firms may be required to provide information on
prices, costs and profits, there is no direct regulatory control over prices charged or
revenues or profits earned. The impact of monitoring on firms’ pricing decisions is
through moral suasion, publicity, and the explicit or implicit threat of stricter forms
of price regulation. Monitoring may also influence bilateral negotiations between
airlines and airports because airlines will have access to the monitored data, while
airports may not have as much information on the airlines. Thus, monitoring can be
a less explicit or intrusive method for influencing prices than price caps or cost-
based regulation, though it may have similar effects on pricing and costs.

Price monitoring generally has been used in areas where scope for monopoly
pricing is limited but there are still some concerns about pricing. As noted in the
second reading speech of the Competition Policy Reform Bill, which introduced
price-monitoring provisions into the Prices Surveillance Act 1983 (PS Act):
Price monitoring may be appropriate where there is concern about the effectiveness of
competition, a history of price problems or community concern about price levels or
movements, or where industries have been recently reformed or deregulated. (HoR
1995, p. 2800)

Price monitoring at Australian airports

Since 5 October 2001, prices, costs and profits of aeronautical services at Adelaide,
Canberra and Darwin airports have been subject to price monitoring, replacing the
price caps that previously applied to these services. The prices, costs and profits of
OPTIONS: PRICE 315
MONITORING AND
ACCESS
aeronautical-related services at all price-capped airports were monitored by the
ACCC until 5 October 2001; since then, aeronautical-related services have been
monitored only at those airports where either price caps continue to apply or where
aeronautical services are subject to price monitoring.

The ACCC also monitors the quality of selected aeronautical and aeronautical-
related services at all core-regulated airports under Part 8 of the Airports Act 1996
(chapter 3). Operators of core-regulated airports are also required under Part 7 of
the Airports Act to provide the ACCC with specified audited financial statements
and reports to supplement the information that they are required to produce under
corporations law (chapter 3). These monitoring arrangements are discussed in
chapter 9.

Some inquiry participants advocated price monitoring as an alternative to price

caps. Although the ACCC (sub. 36) supported continued price caps for aeronautical
services at major airports, it supported replacement of price caps by price
monitoring of aeronautical services,1 as a transitional measure, at some others. The
South Australian Government suggested that price monitoring should replace the
price cap at Adelaide Airport:
… aviation support activities (as defined in Declaration 84) and aeronautical charges
should be subject to prices monitoring by an appropriate regulatory body. Monitoring
should be less costly for all parties, but provide an incentive for airports to be
responsible in setting of prices within a transparent framework that clearly allows for
tighter control if the Commonwealth Government considered an abuse of market power
was occurring. (sub. 33, p. 2)

The Department of Transport and Regional Services (sub. 39) supported a

monitoring model that encouraged commercial negotiation, with monitoring of
prices and service quality backed up by scope for regulatory intervention only if any
market power were abused.

However, some participants were of the view that light-handed forms of price
regulation, such as monitoring, would not constrain the potential for abuse of
market power by airport operators effectively — for example, the Board of Airline
Representatives of Australia (BARA), Qantas, Ansett and Virgin Blue
(subs 41, 48, 42, and DR74). Virgin Blue commented that:
… the price monitoring framework proposed by the [Productivity Commission’s] Draft
Report is a weak and inadequate solution to the market power enjoyed by airport
operators. The proposal would create significant uncertainty in the industry and would
promote a wide range of regulatory gaming behaviour. For example, airport operators

1 The ACCC proposed that the range of aeronautical services be extended to include some
services previously monitored as aeronautical-related services.
316 PRICE REGULATION
OF AIRPORT
SERVICES
 could reap very high monopoly rents in the five years of the monitoring if they thought
either that increased regulation would not be implemented or that the increased
regulation would still be ‘worth it’. Alternatively, they could use their market power to
raise prices to a lesser but still significant and inefficient degree until the end of the
probationary period, then raise prices even higher after that period had ended.
(sub. DR74, pp. 9–10)

Price monitoring of airports in New Zealand and the United Kingdom

Airports in New Zealand and some airports in the United Kingdom (the revenues of
which exceed a specified threshold) are currently subject to light-handed regulatory
regimes that involve price monitoring.

Case 1: Light-handed regulation of New Zealand airports

Since 1988, there has been no direct price regulation of aeronautical services
provided by New Zealand’s three largest airports — Auckland, Christchurch and
Wellington. Instead, these airports have been subject to a ‘light-handed’ regulatory
regime where disclosure and consultation requirements, the countervailing power of
airlines, and the threat of the introduction of direct price control are intended to act
as checks on the abuse of any market power (see box 11.1 and appendix G for a
more detailed discussion of the NZ regulatory system).

Box 11.1 Light-handed regulation  NZ style

Economic regulation at the major airports (Auckland, Christchurch and Wellington)
involves:
• the disclosure of prices, terms and conditions for contractual arrangements, costs,
performance measures, and financial performance indicators;
• the use of the Commerce Act 1986 to control anti-competitive behaviour; and
• the prospect of stricter regulation, such as price control, if market dominance is
abused.
The regulatory regime aims to encourage negotiated outcomes between parties,
without the need for direct intervention by the Government. To this end, airports are
required to consult with airlines about charges. Where proper consultation processes
are not carried out, airport users can initiate legal proceedings.
Airport companies are subject to general competition law (Commerce Act 1986). Under
Part IV, the Governor-General, on the recommendation of the Minister of Commerce,
can impose price controls where competition has been restricted.
Sources: CC (2001a); PSA (1995); appendix G.

OPTIONS: PRICE 317

MONITORING AND
ACCESS
Regulatory costs of the New Zealand system
It appears that the NZ regulatory system has resulted in a high incidence of
litigation between airport users, airport operators and, in some instances, the
Government (appendix G). In its recent draft report on airport regulation, the NZ
Commerce Commission (CC) noted:
The major airlines have demonstrated a willingness to withhold airport payments and
consider court action. Litigation occurred twice in connection with [Wellington
Airport’s] setting of charges … Such litigation imposes substantial costs on the airport
… Nonetheless, litigation has been repeated for subsequent consultation outcomes at
other airports. (CC 2001b, p. 76)

The design of the regulatory system seems to have contributed to the willingness of
airport users to resort to litigation. For example, the meaning of the legislative
requirement that airports ‘consult’ with ‘substantive’ airline customers about price
changes has been frequently tested in the courts (appendix G). The willingness of
airlines to engage in litigation has been exacerbated by the lack of any dispute
resolution options other than the legal system.

In addition, since 1989, there have been three regulatory reviews of the system. The
current review by the CC, examining whether price controls should be introduced
for ‘airfield activities’,2 commenced in 1998, around the same time as the
privatisation of Auckland and Wellington airports. This might have encouraged
strategic behaviour by both airports and airport users to attempt to influence the
outcome of the review.

Have New Zealand airports abused market power?

Auckland Airport was privatised in 1998. In September 2000, Auckland Airport

increased landing charges for large commercial aircraft by 8.5 per cent. In
September 2001, landing charges were increased by a further 5 per cent, with
another 5 per cent increase scheduled for September 2002. Prior to these increases,
landing charges had not been increased since March 1992.3

2 Airfield activities, as defined in the Airport Authorities Amendment Act 1997, cover facilities
and services that enable the landing and take-off of aircraft, including: the provision of
airfields, runways, taxiways, and parking aprons; facilities and services for air traffic and
parking control; airfield lighting; and services to maintain and repair airfields, runways,
taxiways, and parking aprons for aircraft. The terms of reference do not require the CC to
examine the provision of ‘passenger terminal activities’ and ‘aircraft and freight activities’
(which are also defined under the Act) (CC 2001a).
3 International landing charges for aircraft with a maximum take-off weight (MTOW) of more
than 40 tonnes were increased by 3 per cent in 1992 to contribute to the costs of the
development of a new international terminal building. Upon completion of the building the
318 PRICE REGULATION
OF AIRPORT
SERVICES
This phased increase in charges of 18.5 per cent was agreed to by all of Auckland
Airport’s ‘substantive’ customers with the exception of its largest airline customer,
Air New Zealand. In October 2000, Air New Zealand instigated legal proceedings
against Auckland Airport on the grounds of an alleged failure to consult, and
withheld payment of the increase in the landing charges. However, in November
2001, the two parties agreed to a 12.5 per cent increase in airport charges (which
will apply to all airlines).

Christchurch Airport introduced new landing charges in January 2001 — the first
increase in charges for 10 years. The charges were accepted following a
consultation process with the airlines. Charges were set for three years and involve
price restructuring. Landing charges were previously levied on the basis of the type
of aircraft; they are now calculated based on maximum take-off weight (MTOW)
(CC 2001b).

Wellington Airport commenced consultation with airlines in July 2001 with a view
to setting new charges from July 2002 (CC 2001b). In the 12 years since this airport
was corporatised, charges in nominal terms have increased by around 50 per cent.
However, most of this increase came from a one-off increase in 1992. Since then
charges in nominal terms have increased by 9 per cent.

As noted above, the CC currently is assessing whether prices of airfield activities at

these airports should be controlled. According to the CC’s draft report released in
July 2001:
• … airfield activities … are supplied or acquired in a market in which competition is
limited or is likely to be lessened. [And]
• … there is evidence that it is necessary or desirable for the prices of the airfield
activities supplied by [Auckland and Christchurch airports] to be controlled in
accordance with the Commerce Act in the interests of the acquirers of airfield
activities. (CC 2001b, p. xliv)

However, based on an assessment of the net efficiency benefits of introducing price

control, the CC found that only airfield activities supplied by Auckland Airport
should be controlled. The final report is expected to be released in the second
quarter of 2002.4

charges were reduced by 3 per cent. In 1997, international and domestic landing charges
increased for small aircraft (less than 6 tonnes) (CC 2001b).
4 The CC was to release its final report in November 2001, but this has been delayed because:
‘The [CC] is working through a range of complex issues and needs to consider all of the
information received. The report is expected to be finalised in the second quarter of 2002’
(CC 2001c).
OPTIONS: PRICE 319
MONITORING AND
ACCESS
For the period 1988-89 to 1999-00, the CC found that average annual rates of return
exceeded target ‘competitive’ returns at Auckland and Christchurch airports by
3.71 and 2.01 per cent respectively. In contrast, Wellington Airport, over the same
period, did not achieve this benchmark rate of return. The CC also estimated a
benchmark ‘competitive’ price for 1999-00 and compared this with actual airport
charges.5 The CC found that prices exceeded the competitive benchmark by less
than 3 per cent at Auckland Airport, while Christchurch and Wellington airports
were charging below ‘competitive’ benchmark prices (Christchurch by 10 per cent
and Wellington by 34 per cent) (CC 2001b).6 The CC forecast that, based on the
announced price increases at Auckland Airport, the differential between actual
charges and the competitive benchmark would increase to 15.5 per cent at
Auckland, and to 28 per cent at Christchurch Airport by 2002-03.

The CC’s conclusions supporting the introduction of price controls at Auckland

Airport appear, in many instances, to be based on relatively small estimated margins
and to be highly sensitive to the assumptions made, including those regarding land
valuation. Crucial to the CC’s estimation of excessive returns at Auckland Airport is
the exclusion from the asset base of land held for the development of a second
runway. The CC’s stated rationale for this was that ‘today’s acquirers of airfield
activities should only bear today’s costs’ (CC 2001b, p.xxvii).

However, as argued by Kahn, it may be efficient to include (at least) part of the
costs of future capacity, before it becomes operational, in today’s prices:
The asserted principle that ‘today’s users should only bear today’s costs’ is
meaningless in an industry with lumpy, long-lived assets and fails to recognise the
causal effect of today’s demand on the investment requirements of tomorrow, and
accords neither with fairness nor economic efficiency … In situations in which it
conflicts with efficient pricing — specifically with an efficient behaviour of prices over
time — the principle explicitly adopted by the [Commerce] Commission itself dictates
that it be ignored. (Kahn 2001b, p. 14)

In the case of Auckland Airport, where airport users have agreed to the need for the
development of a second runway by 2007 (AIAL 2001), it would seem efficient for
Auckland Airport to incorporate (at least some of the costs) of developing increased
capacity into current landing charges.

Other concerns have been raised about the assumptions used in estimating future
returns to airports, and the costs and benefits of imposing price control (box 11.2).

5 Airport charges, for this purpose, were derived by dividing total landing charge revenue by
tonnes landed. The ‘competitive’ price was derived from the asset base and weighted average
cost of capital (WACC) determined by the CC.
6 If these results are correct, they imply that these airports were earning their highest rates of
return prior to privatisation.
320 PRICE REGULATION
OF AIRPORT
SERVICES
For example, the CC assumed that both Auckland and Christchurch airports would
fully implement their pricing proposals. As discussed above, since the release of the
CC’s draft report, Auckland Airport has agreed to reduce the proposed increase
from 18.5 per cent to 12.5 per cent.

Box 11.2 Participants’ comments on the Commerce Commission’s price

control study
In responding to the Commerce Commission’s (CC’s) draft report (into the need for
price control of aeronautical services in New Zealand) a number of participants raised
concerns about the methodology used by the CC to determine ‘benchmark’ returns and
prices against which the performance of Auckland, Wellington and Christchurch
airports were measured. (Consideration of these substantive issues has required the
CC to take additional time in making its final decision.) The issues raised include:
• Asset (other than land) valuation — the CC adopted the historical cost approach to
asset valuation. A number of participants argued that a more appropriate valuation
method would be to use a replacement cost approach, such as depreciated optimal
replacement cost valuation, to provide a more realistic assessment of the value of
airfield assets (AIAL 2001; CIAL 2001; WIAL 2001; NERA 2001b).
• The exclusion of certain assets on the basis that only ‘used and useful’ assets
should be incorporated in the asset base. This has been most contentious with
regard to land held by Auckland Airport for the development of a second runway. A
number of key participants considered that it is reasonable to recover the costs
associated with increasing capacity that had been agreed to by major airline
customers (AIAL 2001; NERA 2001b; Kahn 2001b).
• The modelling of future returns to airports was considered to be simplistic. The CC
assumed demand, costs and the capital base would be constant over the modelling
period. It was also assumed that price increases announced by Auckland and
Christchurch airports would be achieved in full yet Air New Zealand and Auckland
Airport concluded an agreement in November 2001 which reduced announced
charges (NERA 2001b; AIAL 2001).
• The dynamic efficiency gains to be achieved from the introduction of price control
were over-estimated. The CC assumed that Auckland Airport presently incurs
NZ$6.7 million per annum in dynamic efficiency losses and that this figure could be
halved through price controls. These gains were based on assumptions that
Auckland Airport’s investment in the land held for an additional runway was
inefficient. The CC measured the costs of holding this land by comparing the
airport’s current returns on the land to the returns available in its next best use —
land for residential housing. Airports disagreed that investment in land was
inefficient (NERA 2001b; AIAL 2001).

In summary, while the ACCC (sub. DR55) described the NZ approach as a failed
model, a closer examination demonstrates that, notwithstanding some deficiencies,

OPTIONS: PRICE 321

MONITORING AND
ACCESS
it has not been a failure. For example, several major airports have agreed on new
aeronautical prices with their major airline customers and these pricing outcomes do
not appear excessive. The CC’s own analysis indicates Auckland Airport’s landing
charges in 2000 were less than 3 per cent above the estimated competitive
benchmark. Charges at Wellington and Christchurch airports were estimated to be
below the competitive benchmark.

Case 2: Light-handed regulation of UK airports

The approach to regulation of UK airports is outlined in box 11.3 and appendix G.

The performance of light-handed regulation of non-designated airports holding
Civil Aviation Authority (CAA) permission is less well-documented than the
performance of price caps. The CAA, as part of its current quinquennial review of
price caps at designated airports (due to take effect in 2003), is undertaking a
fundamental review of its approach to the economic regulation of those airports.
However, the CAA is not examining the effectiveness of a light-handed approach to
regulation as part of its current review.

Box 11.3 Light-handed regulation  UK style

A system of light-handed economic regulation applies to UK airports that meet a pre-
determined revenue threshold (other than the four largest airports that are subject to
price caps (Heathrow, Gatwick, Stansted and Manchester)). A key component of this
regime is public disclosure of airport charges and accounts. Airports must provide the
Civil Aviation Authority (CAA) with their annual statutory accounts and schedules of
airport charges. Airports do not need to seek CAA approval before they revise charges,
but they must notify the CAA of new charges before they take effect.
Although airports subject to light-handed regulation are not subject to price caps, the
option of imposing further restrictions is available to the regulator at any time if the
airport is deemed to have abused its market power. On receipt of a complaint, the CAA
may investigate and recommend actions or impose conditions on the airport if
necessary. Additional restrictions can be applied in cases where it is determined an
airport has: unreasonably discriminated against any class of users (or particular user)
of the airport; unfairly exploited its bargaining position relative to users generally; or
levied charges that are unduly low and cause (or are designed to cause) damage to
another airport.
Airports subject to light-handed regulation also can be designated by the Minister of
State under s. 40 of the UK Airports Act and become subject to a price cap.
Sources: CAA (2000b); appendix G.

However, there is some evidence that the threat of regulation under the Airports Act
has acted as a deterrent to the potential abuse of any market power (Kunz 1999;
322 PRICE REGULATION
OF AIRPORT
SERVICES
Starkie 2001b). Two Scottish airports operated by BAA plc, Glasgow and
Edinburgh, were subject to a designation application (that is, to impose a price cap),
which the UK Government rejected because there was no evidence of abuse of a
monopoly position or of inefficiency. When announcing its decision, the
Government added that it believed the threat of designation provided a strong
incentive for BAA plc to control its charges. Possibly reflecting this threat, BAA plc
capped its prices at both airports voluntarily — initially with the formula Retail
Price Index (RPI)-3 (Starkie 2001b).

Approaches to price monitoring

The major potential advantages of price monitoring over stricter price controls are:
• the scope it provides for commercial relationships to develop;
• lower levels of regulatory intervention in price-setting and hence reduced
opportunity for regulatory error and consequent distortions in production and
investment; and
• lower compliance costs.

On the other hand, to ensure that market power is not abused, lower levels of
regulatory intervention in price setting must be balanced by a credible threat that
abuse of market power can and will be identified and appropriate action taken.

Therefore, in order to achieve the appropriate balance, key issues in the design of a
price monitoring regime are specification of:
• the information to be disclosed (to assist in determining whether market power
has been abused and also to promote informed commercial relationships); and
• the nature of the ‘threat’ or over-arching constraint on abuse of market power,
including in what circumstances action can be taken and the form that action will
take.

If these elements are not clearly spelt out, there is a risk that light-handed regulation
could become ineffective or, indeed, as intrusive as stricter price controls. Professor
Forsyth noted, in the context of regulation of airports in New Zealand, that this
could result in inefficiencies similar to those that occur when stricter forms of price
regulation apply:
The threat to regulate is not the same as actual regulation, but its impacts on the firm
may well be much the same. The regulated firm does not know what behaviour on its
part will induce the regulator to impose formal regulation. One possible trigger might
be its profitability; if the firm earns high or supernormal profits, the regulator may
intervene. If this is what the airports believe, this shadow regulation would have the

OPTIONS: PRICE 323

MONITORING AND
ACCESS
 same effects as cost plus regulation. The incentive to keep costs at a minimum would
be weak, since lower costs and higher profits will result in regulation being imposed.
The inefficiencies created by cost-plus regulation will also be present with this shadow
regulation. (sub. 5, p. 19)

Triggering stricter regulation

The potential for inefficiencies may be alleviated to some extent by defining the
behaviour on the part of the regulated firm that would trigger stricter forms of price
regulation (or, indeed, ‘good’ behaviour that would not trigger stricter regulation).
Nonetheless, clearly defining such behaviour may be difficult — high prices may be
a signal that new investment is required rather than an indication that monopoly
prices are being charged; high profits may reflect entrepreneurial skills rather than
market power, and increases in prices may simply reflect changes in costs or that
prices previously were too low. In a capacity-constrained airport, higher prices may
be a means to allocate the available capacity efficiently. This suggests that a broad
set of principles is likely to be preferable for guiding efficient behaviour to specific
criteria that if applied in isolation may not be consistent with efficient outcomes.

Specific criteria for triggering regulatory intervention could also encourage strategic
behaviour to this end. There is a need for a credible threat of stronger regulation, for
reasons discussed above. If it were made clear, however, that any such regulation
would not be reintroduced within a predetermined period, there would be less
potential for the undermining of bona fide commercial negotiations. A review at the
end of that period could then assess whether stricter forms of price regulation,
further monitoring, or any other action were warranted at individual airports. The
monitoring period would need to be long enough to encourage commercial
negotiation, but not so long that the threat of reintroduction of stricter forms of price
regulation was not an effective deterrent against abuse of market power.

The review at the conclusion of the monitoring period could take into account
changes in the competitive environment in which an airport operates and the
behaviour of the various parties during the monitoring period. Information collected
through monitoring could form part of that assessment.

While the regulator is likely to have expertise in analysing monitoring information,

as a matter of principle it would be desirable to separate policy and regulatory roles
by having the review conducted by an independent body, as noted by the
Commission in its review of the PS Act (PC 2001c).

324 PRICE REGULATION

OF AIRPORT
SERVICES
Information disclosure

The effectiveness of a monitoring regime also would depend on whether disclosure

of information was mandatory or voluntary. Although the compliance costs of
voluntary disclosure may be lower, a legislative requirement for disclosure could
help ensure compliance and allow monitoring of information that is not publicly
available. Penalties for non-compliance may also encourage monitored firms to
comply with disclosure requirements.

Comprehensive requirements for disclosure that are spelt out clearly at the start of
the monitoring period, and not altered, may help to minimise regulatory (in this
case, monitoring) creep. Although there may be occasions when it is desirable to
vary disclosure requirements or seek further information, doing so may increase
compliance costs and regulatory uncertainty. Additional information could be
sought and provided on a voluntary basis, although there would be no guarantee that
the monitored firm would provide the requested information.

Information that could be disclosed under a price-monitoring regime includes

charges for aeronautical and aeronautical-related services, revenues, the level and
incidence of operating and capital costs, various measures of profitability, service
quality and productivity. Data for non-aeronautical activities might also be
appropriate, for example, in relation to cost and revenue allocation. The disclosure
of such information may:
• help in assessing whether any market power is being abused;
• facilitate comparisons of monitored variables between airports (both within
Australia, and between Australian and overseas airports); and
• allow airport users to negotiate better with airport operators.

The information that airport operators in New Zealand are required to disclose is
detailed in box 11.4.

Compliance costs of monitoring

BARA (sub. 41) stated that the compliance costs of monitoring are not necessarily
lower than the compliance costs associated with the current regulatory
arrangements. Sydney Airports Corporation Limited (SACL) (sub. 27) also noted
that price monitoring incurs compliance costs. However, the compliance costs of
regulation will be reduced where the information required for monitoring is less
detailed and its analysis by the regulator less complex than for price caps. Airport
operators already may be required to collect and disclose some of the required

OPTIONS: PRICE 325

MONITORING AND
ACCESS
information under other legislative provisions such as corporations law, although
perhaps in less detail.

Box 11.4 Information disclosure requirements in New Zealand

The Airport Authorities (Airport Companies Information Disclosure) Regulations 1999,
which apply to financial statements prepared on or after 1 January 2000, require
specified airport companies to disclose information, which can be summarised as:
• segmented financial accounts for identified airport activities;
• passenger charges and charges for identified airport activities, and the methodology
used to determine these charges;
• the basis for allocating assets to identified airport activities;
• details of asset revaluations and any reports on which revaluations were based;
• operating costs of identified airport activities;
• weighted average cost of capital (WACC) and the methodology and calculations
used to determine WACC;
• numbers of passenger and aircraft movements;
• interruptions to services; and
• the number of people employed in identified airport activities.
Source: CC (2001a).

Indeed, even in the absence of price monitoring, airport operators may collect at
least some of the required information for their own purposes, and may make it
publicly available of their own accord, as has been suggested by Melbourne Airport
(PC 2001b). It was noted in chapter 9 that airport operators have incurred costs
complying with current monitoring arrangements, though costs have not been large
because the information requirements have not been onerous.

Confining monitoring only to those airports and services where there is clearly
scope for abuse of market power could help to minimise any compliance costs.
Ensuring that information is not required too frequently could also help. However,
this should be balanced against the need for regulators and others (such as airport
users) to gain an accurate picture of changes in monitored variables over time.

FINDING 11.1

Where airport market power is not substantial, or where there are commercial
constraints on the misuse of market power, price monitoring has significant
advantages over stricter forms of price regulation. Provided there is no easy
recourse to regulatory intervention, a price-monitoring regime can promote

326 PRICE REGULATION

OF AIRPORT
SERVICES
efficient outcomes while reducing the risk of regulatory failure. Price monitoring
also has the potential to reduce compliance costs and promote commercial
negotiations.

Implementation of price monitoring

Price monitoring could be implemented under the PS Act, the Trade Practices Act
1974 (TP Act) or industry-specific legislation. Price monitoring of aeronautical and
aeronautical-related services at some core-regulated airports currently is conducted
under s. 27A of the PS Act. However, in its inquiry into the PS Act, the
Commission concluded that the PS Act is an unsatisfactory vehicle for conducting
inquiries and monitoring (PC 2001c). It recommended that the PS Act be repealed,
and a new section incorporated in the TP Act to provide for monitoring and
inquiries (see below).

The TP Act contains a number of existing provisions under which price monitoring
could be conducted:
• s. 28 provides for the functions of the ACCC in relation to dissemination of
information, law reform and research;
• s. 29 requires the ACCC to comply with directions of the Minister and
requirements of Parliament;
• Part IVB provides for voluntary or mandatory codes to be declared by the
ACCC; and
• s. 29B details the functions and powers of the National Competition Council
(NCC) (PC 2001c).

While these provisions do not provide specifically for price monitoring, they permit
the Minister to direct the ACCC or NCC to undertake research and analysis, and
could allow the Minister to direct either organisation to conduct price monitoring.
However, as noted in the review of the PS Act, these provisions lack a clear
framework for defining the objectives of monitoring and ensuring its appropriate
application.

The TP Act could also be amended to provide for price monitoring. This could be
done through either generic provisions as proposed in the Commission’s review of
the PS Act or industry-specific provisions. The Commission argued that the
inclusion of a new part in the TP Act:
… would consolidate the various legislative elements of Australia’s competition policy
into one legislative package. It would complement Part IIIA and Part IV of the TP Act
as an element of Australia’s competition policy. (PC 2001c, p. 84)

OPTIONS: PRICE 327

MONITORING AND
ACCESS
The Commission’s recommendation to repeal the PS Act and incorporate a new part
in the TP Act, providing powers to undertake public inquiries and to implement
monitoring programs, can be summarised as follows:
• firms may only be subject to monitoring following a recommendation from a
public inquiry;
• a public inquiry could only be initiated where the market is of national
significance and there is evidence of monopolistic pricing;
• the inquiry would be conducted by a body other than the regulator;
• the information to be disclosed would be determined during the inquiry process;
• the ACCC would be responsible for collating and auditing monitoring
information and its public dissemination, but could not make any determination
on the appropriateness of prices or make recommendations to government under
this monitoring provision; and
• monitoring would be for a limited period, and the monitoring declaration
automatically revoked at the end of that period.

Industry-specific amendments to the TP Act could be made to provide for price

monitoring of airport services. The Postal Services Amendment Bill 2000 tabled in
Commonwealth Parliament proposed, among other things, to amend the TP Act by
introducing industry-specific provisions (section XID) to regulate the prices charged
by private firms for standard letter services. While the Bill was not enacted, it
would have been the amended TP Act, rather than the PS Act, that provided for
price regulation of postal services. Alternatively, industry-specific price-monitoring
provisions could be incorporated into the Airports Act.

The Commission is of the view that any criteria for monitoring incorporated into
industry-specific legislation should reflect generic monitoring criteria and be
consistent with the monitoring principles discussed in the Commission’s review of
the PS Act.

Another issue relevant to the implementation of a price-monitoring regime is by

whom it would be administered. A number of organisations have expertise in
monitoring prices, for example:
• the ACCC conducts price monitoring in a number of sectors, including airports,
under various legislative instruments;
• the Bureau of Transport Economics (BTE) monitors prices, costs and other
variables in a range of transport-related industries; and

328 PRICE REGULATION

OF AIRPORT
SERVICES
• the Productivity Commission monitors the performance of government trading
enterprises and conducts national and international benchmarking of the
performance of key Australian industries.

As noted above, under its recommendation to repeal the PS Act and insert a new
section in the TP Act, the Commission proposed that the ACCC be responsible for
collating and auditing information and ensuring compliance. This would ensure the
ACCC’s expertise in monitoring prices is applied in the new regime. However,
Melbourne Airport, although not opposed to the ACCC taking such a role,
suggested that other agencies such as the BTE may also be able to fulfil that role,
particularly if the aim of monitoring is to inform policy makers rather than to trigger
intervention in pricing decisions.

11.2 Access provisions and general competition law

Reliance on access provisions and general competition law and the implicit or
explicit threat of re-regulation to constrain potential abuse of market power would
allow airport operators to set prices without regulatory intervention or oversight
provided they did not breach the provisions of Part IV and were not subject to
arbitration under Part IIIA of the TP Act.

Regulating access to and anti-competitive conduct of airports

Part IV

Though Part IV of the TP Act proscribes certain practices that can create market
power through restricting competition or through mergers, and makes illegal the use
of market power to harm or eliminate competitors where market power is already
established, Part IV relies on ex post remedies for abuse of market power, and
monopoly pricing per se is not proscribed. Although s. 46 can address pricing issues
such as price fixing or predatory pricing, it does not directly address situations
where market power results in monopoly prices. There would be no contravention
of s. 46 unless a pricing decision were taken for one of the purposes it proscribes.
The legal costs associated with prosecution for contravention of the provisions of
Part IV may also be significant, and it may be difficult to prove an offence has
occurred.

Though it would be possible to amend Part IV to address pricing issues relating to

airports, amending generic provisions for an industry-specific purpose is unlikely to
be an appropriate approach due to its effects on other sectors of the economy.

OPTIONS: PRICE 329

MONITORING AND
ACCESS
Consideration of changes to Part IV is likely to be part of the Government’s
announced review of the TP Act, but it is outside the terms of reference of this
inquiry. Thus, the discussion below is focused on access regulation. Part IV would
continue to apply regardless of other legislative arrangements.7

Access arrangements

Airport operators currently are subject to s. 192 of the Airports Act and Part IIIA
(the National Access Regime) of the TP Act. The five-year, automatic declarations
of privatised, core-regulated airports cannot be renewed under s. 192. However,
new airport-specific access arrangements or Part IIIA could continue to complement
any future price regulation of airport services or, indeed, provide an alternative to an
airport-specific price-regulation regime. Prices of airport services could be
regulated indirectly through regulation of the terms and conditions of access to an
airport service, if a service were declared.

There has been limited application of access provisions and general competition law
to airports to date (chapter 9). However, reliance on these provisions as an
alternative to direct regulation of prices could result in greater use of these
provisions by airport users.

Access regulation could have some advantages compared with price regulation. It is
activated only when a user complains and provides some encouragement of
commercial negotiation and agreements (PC 2001a). Though compliance costs can
be high, once precedents are set, the need for arbitration may decrease and the costs
of any arbitration may decrease as precedents are set and participants become more
familiar with the process.

Some participants argued that the general access provisions available under Part
IIIA (and the provisions of Part IV) provide sufficient regulation to prevent airports
abusing any market power. For example, the Australian Airports Association argued
that:
Due to the significant countervailing power of airlines, and the active competition
between airports for a range of important services, a strong argument can be made that
both the Prices Surveillance Act … and the automatic declaration of airport services for
the purposes of Part IIIA of the Trade Practices Act 1974 under s. 192 of the Airports
Act 1996 should be removed. This would leave airports subject to declaration under
Part IIIA on a case by case basis in the same way as any other infrastructure is liable to
Part IIIA, and subject to the usual application of Part IV of the [TP Act]. (sub. 15, p. 5)

7 Part VII of the TP Act confers on the ACCC power to authorise practices otherwise proscribed
under Part IV.
330 PRICE REGULATION
OF AIRPORT
SERVICES
Nonetheless, application of Part IIIA to airports to date has raised a number of
concerns, including the degree to which the declaration criteria are open to
interpretation, that the national significance test is linked to the facility rather than
the service provided by the facility, and that there is considerable scope for delays
in the declaration process (see chapter 9 and PC (2001a) for a more detailed
discussion).

In the final report for its review of the National Access Regime, the Commission
proposed a number of amendments to the current Part IIIA regime to address these
and other concerns. Those relevant to airports include:
• modification of the declaration criteria, including a requirement that declaration
would promote a substantial increase in competition;
• inclusion of pricing principles to improve certainty for access seekers and
providers;
• provision for the ACCC to conduct multilateral arbitrations following
consultation with the parties to the dispute;
• provision to lodge undertakings after a service has been declared;
• appeal rights for decisions on undertaking applications; and
• exemption from Parts IV and VII of the TP Act of the terms and conditions of:
arbitrated determinations for declared services; agreements reached under
certified regimes or negotiated under accepted undertakings; and private
agreements for declared services covered by registered private contracts.

If the Commission’s proposals were implemented, several concerns in relation to

the application of the current Part IIIA regime to airports would be addressed.

An airport-specific access regime?

As it stands, s. 192 of the Airports Act effectively lapses once initial access
declarations expire. (As noted in chapter 3, the Minister is required to specify the
expiry date of the determination, and no power is conferred on the Minister under
s. 192 to renew the declaration once it has expired (ACCC, pers. comm., 5 July
2001).) These declarations expire on 30 June 2002 for Phase 1 airports and on
30 June 2003 for Phase 2 airports.

Though s. 192 could be amended to continue to operate in its current form, at issue
is whether there is a case for continuation of an airport-specific access regime and,
if so, what form that regime should take.

OPTIONS: PRICE 331

MONITORING AND
ACCESS
The stated rationale for s. 192 was the Government’s desire to encourage access to
airports by new airline entrants. However, for reasons discussed in chapter 9, there
are strong commercial incentives for airports to promote access by new airlines and
s. 192 does not appear to have been instrumental in influencing this outcome.
Importantly, two airports have built new terminal facilities for entrants, making gate
allocation at existing domestic terminals unnecessary. This is not to say that access
issues with respect to new entrants will not arise in future.

Moreover, that s. 192 has not been invoked extensively need not mean that it has
not influenced airport behaviour in other circumstances. As discussed in chapters 6
and 7, airports may have some incentive to frustrate access to other elements of the
airport where, by doing so, competition in a particular market (such as car parking)
can be constrained and monopoly profits earned by the airport operator. However,
airport operators do not seem to have strong reasons to deny or frustrate airline
access to the airport.

Another possible rationale for airport-specific access arrangements is the range of

airport services and the large number of potential access seekers, raising the
transactions costs of access regulation. This could warrant more streamlined access
provisions than provided in Part IIIA. The ACCC argued that:
The difficulties experienced with the existing declaration mechanisms under the
Part IIIA suggest that a declaration procedure whereby access seekers must first seek
declaration of services before being entitled to access imposes a significant barrier to
entry. This would seem to be unnecessary in the context of an industry where there is
clearly a need for an effective access regime and a great deal is known about the nature
of services provided at airports. (sub. 38, p. 12)

Qantas made a similar argument:

While the application of the declaration criteria is relatively straightforward to a large
range of airport services, the procedural requirements under Part IIIA to demonstrate
the criteria in each access application is burdensome and time consuming …
(sub. 48, p. 27)

Though Qantas stated it has ‘experienced substantial difficulty in negotiating access

to a range of services at airports’, it has not sought declaration of any airport
services at any airport under either s. 192 or Part IIIA (sub. 48, p. 24).

The only application of Part IIIA to airports to date has been the Australian Cargo
Terminal Operators (ACTO)8 declarations at Sydney and Melbourne International
Airports (chapter 9). Indeed, these are the only declarations in any industry to date
under Part IIIA. As noted in chapter 9, there were significant delays involved in

8 As noted in chapter 9, ACTO has been deregistered.

332 PRICE REGULATION
OF AIRPORT
SERVICES
these declarations and subsequent appeals, and the administrative and legal costs are
likely to have been significant. However, insofar as precedents have been
determined, these costs should not all be attributed to this specific case. It is also
likely that clarification of the Part IIIA declaration criteria through the Australian
Competition Tribunal review of the ACTO declarations at Sydney International
Airport would result in fewer delays in future declaration decisions.

In broad terms, an airport-specific access regime may be warranted where there is a

justification either for weaker declaration criteria and/or different processes and
procedures.

Currently, privatised core-regulated airports are declared automatically under

s. 192, though access seekers must apply to have an airport service declared in order
for arbitration of terms and conditions for that service to occur. Box 11.5 sets out
the declaration criteria under s. 192 and Part IIIA respectively.

The Part IIIA declaration criteria are designed to ensure that the access framework
is applied only where access would promote competition in a related market, it is
uneconomic to develop another facility to provide the service, and the facility is of
national significance. The criteria for declaration as an airport service under s. 192
generally are weaker than the Part IIIA criteria.9 There is no requirement under
s. 192 that access to a service promote competition in another market, and the
facility need not be of national significance. Some participants noted (for example,
Hastings Funds Management (sub. 19) and SACL (sub. 27) that this could result in
broader coverage of services under s. 192 than would occur under Part IIIA.

The ACCC (sub. 38) suggested that the requirement under s. 192 that ‘facilities
cannot be economically duplicated’ should be amended to reflect the (possibly less
stringent) Part IIIA requirement that ‘it would be uneconomical for anyone to
develop another facility to provide the service’. It also advocated ‘specification of
certain up-front declared services, coupled with a mechanism for declaration of
other services (or variation of the initial list) in the future’ (sub. 38, p. 15).

Qantas (sub. 48) recommended continuation of an airport-specific access regime

because:
… it is both unnecessary and administratively inefficient to demonstrate the
applicability of the declaration criteria each time an access application is made in
respect of an airport. It is for that reason that the Federal Government enacted s. 192 of
the Airports Act. As the Productivity Commission is aware, s. 192 deems each airport

9 A possible exception is that criterion (b) under Part IIIA, which requires that it be
uneconomical for anyone to develop another facility to provide the service, is potentially a
weaker test than the s. 192 test that the facility providing the service cannot be economically
duplicated (PC 1998a).
OPTIONS: PRICE 333
MONITORING AND
ACCESS
 service to be a declared service for the purposes of Part IIIA. Section 192 does not
completely remove inefficiency from the administrative process, but it does improve it
greatly. (sub. 48, p. 27)

Box 11.5 Declaration criteria in existing legislation

S. 192
S. 192(5) of the Airports Act states that an airport service means a service provided at
a core-regulated airport, where the service:
(a) is necessary for the purposes of operating and/or maintaining civil aviation services
at the airport; and
(b) is provided by means of significant facilities at the airport, being facilities that
cannot be economically duplicated;
and includes the use of those facilities for those purposes.

Part IIIA
S. 44G(2) of the Trade Practices Act states:
The [National Competition] Council cannot recommend that a service be declared
unless it is satisfied of all of the following matters:
(a) that access (or increased access) to the service would promote competition in at
least one market (whether or not in Australia), other than the market for the service;
(b) that it would be uneconomical for anyone to develop another facility to provide the
service;
(c) that the facility is of national significance, having regard to:
(i) the size of the facility; or
(ii) the importance of the facility to constitutional trade or commerce; or
(iii) the importance of the facility to the national economy;
(d) that access to the service can be provided without undue risk to human health or
safety;
(e) that access to the service is not already the subject of an effective access regime;
(f) that access (or increased access) to the service would not be contrary to the public
interest.
Sources: Airports Act 1996; Trade Practices Act 1974.

Further, Qantas advocated that s. 192 declaration criteria be amended so that a

service should only be required to be used for the purposes of operating and/or
maintaining civil aviation services at the airport, rather than be necessary for the
purposes of operating and/or maintaining civil aviation services. It also considered
that s. 192 should apply to all core-regulated airports (Sydney Airport in particular),
334 PRICE REGULATION
OF AIRPORT
SERVICES
rather than just to privatised core-regulated airports, and that provision should be
made for airport operators to lodge access undertakings following declaration under
s. 192.

In its review of the National Access Regime, the Productivity Commission found:
The current approach of a national access regime operating in tandem with industry-
specific regimes has significant advantages. In effect, it draws on the strengths of both
the generic and specific approaches, while avoiding some of the pitfalls of a one-
dimensional solution.
Some changes to both Part IIIA and Clause 6 of the Competition Principles Agreement
are nonetheless required to strengthen the access framework and to discourage
unwarranted divergence across industry-specific regimes. (PC 2001a, p. 122)

In other words, while industry-specific regimes may be warranted, those regimes

should as far as possible reflect the principles in the general access framework.

The Commission’s report also suggested amendments to Part IIIA to strengthen

declaration criteria such that facilities and services would be declared only where
there is a likelihood of efficiency benefits. The amendments include a
recommendation to require access (or increased access) to promote a substantial
increase in competition (PC 2001a).

In contrast, the ACCC’s proposal for up-front declaration of certain airport services
would seem to by-pass altogether the need for assessment of likely efficiency
outcomes of declaration, while the Qantas proposal would weaken further s. 192
criteria relative to current Part IIIA declaration criteria. Given that an airport’s
incentives to deny access are limited, the case for considerably weaker declaration
criteria than those applied to other industries does not seem to be strong. Despite its
proposal for listing declared airport services, the ACCC also observed:
When a service provider is vertically separated it will usually have little incentive to
deny access. While the service provider may exploit its market power by setting high
prices it is unlikely to manipulate other terms and conditions to limit access.
Nevertheless the negotiate-arbitrate provisions allow an access seeker to seek
arbitration over non-price terms and conditions. This could result in unnecessarily
intrusive arbitration over detailed operational matters. (sub. 36, p. 94)

This suggests that application of the negotiate–arbitrate access model should be

confined to those situations where airports clearly have an incentive to deny access
and where declaration of the service would promote efficiency benefits. In the
Commission’s view, application of Part IIIA criteria to airports would promote this
outcome. If changes proposed to the Part IIIA declaration criteria in the
Commission’s report on the National Access Regime were implemented, this
conclusion would be strengthened.

OPTIONS: PRICE 335

MONITORING AND
ACCESS
To some extent the ACCC’s proposal is designed to address procedural problems
created by the large number of services provided by airports and multiple users of
airport services. Qantas (trans., p. 275) also was of the view that a negotiate–
arbitrate model was not suited to industries such as airports where there are many
users and multiple transactions.

In addition, Melbourne Airport stated:

Part IIIA is predicated on there being a single access provider in dispute with a single
customer. It seems to us that if Part IIIA were to be relied upon, airports could face the
possibility of being constantly in arbitration with airlines and the ACCC. (sub. 7, p. 56)

However, this could be addressed to some extent within Part IIIA by allowing
‘multilateral’ arbitration of terms and conditions of access to a declared service. As
noted, the Commission has recommended such an amendment to Part IIIA in its
review of the National Access Regime. If such a change were not implemented
within the National Access Regime, an industry-specific regime that allowed for
multilateral arbitrations (but which was consistent with Part IIIA in other respects),
may warrant consideration.

FINDING 11.2

There do not appear to be any grounds for airport-specific access arrangements

that set lower declaration thresholds than Part IIIA of the TP Act. An airport-
specific regime that allowed for multilateral arbitrations might be warranted if such
arbitrations were not provided for in the National Access Regime, as recommended
separately by the Commission in its review of that regime.

Undertakings

Some participants supported a model under which airport operators would develop
‘undertakings’, which set out the terms and conditions under which airport services
would be provided.

Part IIIA undertakings

Part IIIA currently provides for airport operators to lodge access undertakings with
the ACCC, provided the services in question have not been declared for access
purposes. Airport services at privatised core-regulated airports were declared
automatically for the purposes of Part IIIA under s. 192, though they were given a
period of grace during which they could submit undertakings. Currently there is no
provision for undertakings covering declared airport services to be lodged
(chapter 3). However, declarations under s. 192 will expire on 30 June 2002 and

336 PRICE REGULATION

OF AIRPORT
SERVICES
30 June 2003 for Phase 1 and 2 airports respectively, after which airport operators
will be able to submit undertakings under Part IIIA.

As discussed in chapter 9, shortly after their privatisation, two airports submitted

draft undertakings but these were not acceptable to the ACCC. An apparent
reluctance to pursue the undertakings further might have been due to a perception
on the part of airport operators that lodgement of an undertaking acceptable to the
ACCC was less appealing than the risk of declaration and subsequent arbitration. If
airport operators considered that there was a greater likelihood of application of
access regulation in the absence of direct price regulation, they may be more likely
to submit undertakings. Further, changes to Part IIIA proposed in the access inquiry,
such as provision for appeal of decisions not to accept undertakings, may increase
the likelihood that undertakings will be lodged successfully. The proposal to allow
lodgement of undertakings after a service has been declared may also provide some
encouragement.

Qantas (trans., pp. 275–6) suggested that an access model based on an industry code
such as the National Electricity Code (NEC) (box 11.6) may be a desirable
approach to regulating prices for airport services. In particular, Qantas argued for
regulation of minimum service levels:
Qantas believes that the most effective and practical means to ensure minimum levels
of service quality within the airport industry is to include within the regulatory
framework service level commitments. These are included in other regulated industries
such as electricity. (sub. 48, p. 5)

In the final report for its review of the National Access Regime, the Commission
noted that the access arrangements in the NEC ‘result in a more prescriptive
approach to the provision of access than regimes in some other sectors that rely
more on commercial negotiations between the parties’ (PC 2001a, p. 450). The
Code effectively regulates prices and other terms and conditions of access. If such a
regime were imposed on airports it would impose considerably stronger regulation
than currently applies.

Airport-specific undertakings

Some participants in this inquiry supported less prescriptive, industry-specific

undertakings. Though SACL (sub. 27) preferred complete deregulation of prices, it
outlined a proposal for a ‘prices and quality undertaking’ in the event that the
Commission was of the view that some form of regulation was warranted. SACL
suggested that s. 192 of the Airports Act could be amended to allow airport
operators to develop prices undertakings setting out the services covered, the prices

OPTIONS: PRICE 337

MONITORING AND
ACCESS
to be charged for services, arrangements for varying prices, and the service levels to
be provided.

Box 11.6 National Electricity Code

The national electricity market is governed by the NEC, which is jointly administered by
the ACCC, the National Electricity Code Administrator (NECA) and State regulators,
and has three separate but related elements:
• an access code — rules governing the way in which participants gain access to, and
connect with, transmission grids and distribution networks, including the principles
for regulating access prices;
• market rules — rules governing how generators and customers trade through the
common pool, and how the physical system is controlled; and
• administrative arrangements — rules governing how disputes are settled, how the
code is enforced and changed, and establishing transitional arrangements for each
participating jurisdiction.
Each transmission and distribution business owner or operator is required to submit an
access undertaking consistent with the code. To streamline the process, NECA
submitted an access code on behalf of service providers pursuant to s. 44ZZAA of the
Trade Practices Act 1974. Service providers also have the option of submitting
individual undertakings.
The ACCC, in the case of transmission services, and the relevant State regulator, in
the case of distribution services, is required to determine an annual revenue
requirement for the regulated assets under the principles set out in the code, and
approve prices for the use of networks by third parties within that revenue requirement.
The code also requires regulators to incorporate CPI-X or other incentive-based
variants in the revenue caps to encourage providers to improve productivity.
Sources: ACCC (2000e); PC (2001a).

On the other hand, Melbourne Airport proposed:

… that s. 192 be repealed and replaced by an arrangement whereby those airport
services that are considered to have market power sufficient to warrant price regulation
may be regulated under a ‘pricing undertaking’. Where airports were considered not to
require regulation, or services provided at larger airports were considered contestable,
regulation would be undertaken under the standard provisions of the TP Act.
(sub. 7, p. 57)

In a subsequent submission, Melbourne Airport supported the SACL proposal for

an approach allowing airport operators to develop prices and quality undertakings,
though it continued to advocate replacement of s. 192 by an industry-specific
regime. Melbourne Airport suggested that development of undertakings by airport
operators should be optional, and that the likelihood of declaration under Part IIIA

338 PRICE REGULATION

OF AIRPORT
SERVICES
would provide an incentive for the operators of large airports to develop
undertakings. In its view, industry-specific legislation should specify:
• Items that a PQU [prices and quality undertaking] must contain (eg duration,
definition of services etc) similar to section 91 of the Airports Act 1996 in relation
to Major Development Plans.
• Requirement to consult and inform the Minister of the outcome similar to
sections 92 and 93.
• Matters the Minister must have regard to in approving a PQU similar to Section 94.
These might include the impact on efficiency and investment, impact on
competition … in other markets, the public interest (and in particular the interests of
the travelling public) and ensuring that the PQU does not lead to any abuse of
market power.
• Provisions for variations similar to section 95.
• Relationship between the PQU and the National Access Regime more generally.
• Enforceability of charges levied under the PQU. (sub. 37, p. 14)

The ACCC suggested that, although airports could be required to submit for
approval a set of price and non-price terms and conditions applicable to the
provision of the declared services at the airport and a dispute resolution mechanism:
… experience with airport regulation in Australia to date suggests that there are a
number of difficulties associated with putting in place ex ante terms and conditions for
access to core airport services. These arise primarily from the fact that airports provide
a range of interlinked but different (non-homogeneous) services, and that the dynamics
of the provision of those services is constantly changing in response to changing market
conditions. As a result, it is difficult to establish generic terms and conditions that are
sufficiently precise as to ensure that disputes will not arise in relation to their
application in a particular case. (sub. 38, pp. 14–15)

Yet the ACCC suggested an alternative model that involved:

… the imposition by law (whether in the enabling legislation or in subordinate
instruments) of certain ‘standard access obligations’. These obligations could provide
that airport operators provide access to the declared services on the basis of non-
discriminatory, objective and transparent terms and conditions. The obligations could
also include an obligation to ensure that prices reflect costs.
Appropriate mechanisms would be required to enable the enforcement of those
obligations, whether by way of court action initiated by the Commission [ACCC]
and/or the arbitration of disputes as to whether those obligations have been fulfilled.
… legislative obligations could be enforceable in the courts. Alternatively, or in
addition, access seekers could be empowered to notify disputes for arbitration; in that
event, the arbitration should be limited to determining whether or not the legislative
obligations have been complied with. (sub. 38, p. 15)

OPTIONS: PRICE 339

MONITORING AND
ACCESS
The mandatory nature of this model, combined with requirements for non-
discriminatory terms and conditions and mechanisms for enforcement of
obligations, could result in significant regulatory intervention in firms’ pricing
decisions and inefficient outcomes (for example, if discrimination were prevented).

Just as the Commission does not see a strong case for setting weaker declaration
criteria for airports, it does not consider that airports should be compelled to submit
or meet prescribed access obligations. At the same time, the Commission does not
consider that airports should be allowed exemption from declaration for access
purposes on the basis of undertakings that otherwise would not satisfy Part IIIA
criteria. In other words, insofar as third party access is concerned, airports should be
treated neither more strictly nor more leniently than other industries.

FINDING 11.3

There are insufficient grounds for an airport-specific access undertaking regime.

Commercial agreements

Several participants suggested models that promoted commercial agreements,

backed up by either general access provisions of the TP Act or by special regulatory
arrangements. Westralia Airports Corporation proposed a model:
… whereby airports and airlines enter into commercially negotiated aeronautical
services agreements, incorporating the scope and pricing of airport services, the
conditions of use by airlines of the airport, the process for consultation and exchange of
information and quality of service. Such agreements when accompanied by the access
provisions of the Trade Practices Act and the significant countervailing market power
of airlines would constitute an effective body of counter-measures to alleviate concerns
over abuse of market power by airports. (sub. 21, p. 1)

It suggested these agreements may require legislative support such as a Government

policy directive, which should provide guidance for abuse of market power and
define key parameters of the airport–airline consultation environment that would
apply in the absence of agreement between the airport and airlines. The Department
of Transport and Regional Services (sub. 39) supported a model that combined
monitoring of prices and service quality with provision for development of
commercial agreements between airport operators and users, backed up by scope for
regulatory intervention if any market power were abused (section 11.1).

The model advocated by Ansett, which in its view would provide flexibility for
airports and airlines to reach commercial agreement, while providing a framework
to moderate the market power of airports, is a variation of the industry-specific

340 PRICE REGULATION

OF AIRPORT
SERVICES
approach to the development of commercial agreements. Though the model was not
outlined in detail, its key features were:
• broad specification of the elements to be contained in the contractual agreements
between airlines and airports;
• recourse to arbitration only if agreement on … individual elements cannot be
reached, but with clear mechanisms to discourage unnecessary recourse to the
arbitrator; and
• [specification of the] powers and procedures of the arbitrator. (sub. 42, p. 47)

While a shift towards commercial agreements is attractive, in the Commission’s

view, agreements (or their contents) should not be made mandatory. This would
undermine the fundamental objective of promoting commercial relationships.
Development of agreements on a voluntary basis, on the other hand, may promote
commercial relationships and efficient outcomes. For example, if provision were
made for a review of regulatory arrangements, and development of agreements was
to be taken into account by the review, airport operators could be encouraged to
develop agreements. The possibility of declaration under Part IIIA could also
provide such an incentive.

Development of guidelines for commercial agreements, including consultation and

dispute-resolution mechanisms, might be of some assistance to airport operators and
users, at least during the transition from regulated prices. Some participants
supported regulatory approval of such agreements. However, this would require a
prescriptive approach to their content. On the other hand, provision for lodgement
of agreements voluntarily entered into might promote transparency and user
acceptance. Agreements could be lodged if they specified, for example:
• agreed terms and conditions (including prices) of airport use, including the
services covered, service quality and information to be provided by the parties;
• dispute-resolution mechanisms (for example, scope for independent arbitration);
• the period covered by the agreement; and
• consultation and other procedures to be followed to vary the agreement.

Importantly, however, acceptance of lodgement would not involve an assessment by

the regulator of the prices, terms or conditions agreed between the parties.

That such guidance from government may be helpful for commercial entities in a
commercial environment is perhaps surprising. The reason for it is the historical
legacy of decades of government ownership of this industry and the pervasive
effects of regulation.

OPTIONS: PRICE 341

MONITORING AND
ACCESS
FINDING 11.4

Facilitation and encouragement of agreements between airport operators and users

has the potential to promote commercial relationships and efficient outcomes.

342 PRICE REGULATION

OF AIRPORT
SERVICES
12 Appropriate future regulation: the
Commission’s assessment

In this chapter, the Commission outlines its assessment of the need for price
regulation of particular airports and airport services, and its preferred option for
such regulation.

12.1 The Commission’s task

It is worth repeating that the purpose of this inquiry, as set out in the terms of
reference, is:
… to examine whether new regulatory arrangements, targeted at those charges for
airport services or products where the airport operator has been identified as having
most potential to abuse market power, are needed to ensure that the exercise of any
such power may be appropriately counteracted.

In making its assessment and recommendations, the Commission also is asked to

take into account several ‘principles’:
(a) the [current] CPI-X price cap applied to aeronautical charges during the first five
year period of private ownership will no longer operate;
(b) future prices regulation should be applied to those aeronautical services and those
airports where airport operators have most potential to abuse market power;
(c) airport operators may propose to the inquiry alternative approaches to prices
regulation which would provide equal or better protection to users;
(d) prices regulation should minimise compliance costs on airport operators and the
Government;
(e) prices regulation should promote the efficient operation of airports;
(f) prices regulation should facilitate benchmarking comparisons between airports,
competition in the provision of services within airports (especially protecting
against discrimination in relation to small users and new entrants), and
commercially negotiated outcomes in airport operations; and
(g) the Commission may recommend more effective forms of prices regulation than are
currently in place, where this may be necessary.

APPROPRIATE 343
FUTURE
REGULATION
As noted in chapter 4, in essence these principles (many of which mirror the
Competition Principles Agreement) suggest that any future regulation should be the
least required to target the source of any problem and promote efficient outcomes,
while being applied in a way that fosters market outcomes where feasible, imposes
minimal compliance costs on all parties, and promotes transparency and
competition. Importantly, the terms of reference guide the Commission to
recommend regulation only where it is necessary to promote efficient outcomes.

Developments since the draft report

As noted in chapter 1, on 5 October 2001, the Commonwealth Government

announced major changes to airport price regulation in response to the effects on
airports of both the terrorist attacks in the United States and the cessation of
services by Australia’s second-largest carrier, Ansett. The Minister for Financial
Services and Regulation stated that the changes were ‘consistent with the findings
in the Productivity Commission’s draft report’ (Hockey 2001). He added that ‘the
Government will give further consideration to the prices oversight arrangements at
airports in the light of developments in airport prices and the Productivity
Commission’s final report’.

12.2 The Commission’s assessment of the need for

regulation
The terms of reference require that regulation should be targeted only at those
airports and those aeronautical services provided by airports where there is most
potential to abuse market power.

The Commission considers that the four largest Australian airports — Sydney,
Melbourne, Brisbane and Perth — have a high degree of market power in core
aeronautical services and therefore warrant some form of airport-specific price
regulation, additional to general third-party access and competition regulation. As
discussed in chapter 11, these general provisions of the Trade Practices Act
1974 (TP Act), while potentially regulating some terms and conditions (of declared
services) and proscribing certain practices of firms with market power, would not
regulate prices of aeronautical services per se.

The Commission is not convinced that Adelaide, Canberra or Darwin have

substantial market power. In this respect they may be comparable to Cairns Airport
(the sixth-largest in Australia, albeit State owned), which is not subject to any price

344 PRICE REGULATION

OF AIRPORT
SERVICES
regulation. However, some form of monitoring may be appropriate for these
airports, at least as a transitional measure.

Remaining core-regulated airports, because they appear to have much less market
power, or because, given their size, the costs of regulation would far outweigh any
potential benefits, should not be subject to any airport-specific economic regulation,
including price or quality monitoring. (They would continue to be subject to the
TP Act (as would all airports) and relevant provisions of the Airports Act 1996.)

For those airports with moderate to high market power, the Commission considers
that, based on its assessment of various regulatory options in chapters 10 and 11,
two types of regulation, incentive-based price caps and price monitoring, merit
consideration. The choice essentially comes down to forms of regulation that best
promote incentives for efficient outcomes. Of the stricter forms of price regulation,
CPI-X price caps provide the best incentives but inevitably they bring significant
risks of regulatory error. Given that airports face several commercial constraints and
incentives that will moderate their behaviour, the Commission sees significant
advantages in a more light-handed approach involving price monitoring. Therefore,
two regulatory options are outlined, differentiated by their treatment of Sydney,
Melbourne, Brisbane and Perth airports.

12.3 Option A: dual-till price caps

The stricter option essentially would continue the regulatory structure introduced on
5 October 2001. However, it proposes a different approach to the setting of price
caps at the Phase 1 airports (Melbourne, Brisbane and Perth), designed to improve
incentives for aeronautical investment and to reduce the need for ongoing regulatory
intervention in investment decisions. Price monitoring arrangements would also be
modified to reflect monitoring principles outlined in the Commission’s concurrent
Review of the Prices Surveillance Act 1983 (PC 2001c). Under this option, price
regulation at Sydney Airport would continue largely unchanged.

Option A: Melbourne, Brisbane and Perth airports

For Melbourne, Brisbane and Perth airports, an annual CPI-X price cap would be
set for five years, complemented by price and quality monitoring.

The price cap would apply only to those aeronautical services in which the three
airports have been assessed by the Commission as having a high degree of market
power. Broadly speaking, this means all services designated as aeronautical services
for the purposes of current price caps plus aircraft refuelling services (box 12.1).

APPROPRIATE 345
FUTURE
REGULATION
However, some assets (such as landside roads) that are used for both aeronautical
and non-aeronautical purposes will have to be apportioned between these activities.

Average prices allowed under price caps should broadly reflect the anticipated cost
of efficiently providing regulated aeronautical services on a dual-till basis
(box 12.2). In other words, unlike current price caps where inherited single-till
prices are adjusted incrementally as new investment is put in place, the cap should
correspond to anticipated efficient dual-till prices.

Box 12.1 Which airport services should be regulated?

The terms of reference stipulate that future prices regulation should be applied only to
those aeronautical services (and those airports) where airport operators have most
potential to abuse market power. Having assessed the degree of market power in the
various services provided by airports, the Commission considers that if a system of
price caps were retained then the caps should incorporate:
• Aircraft movement services and facilities — airside grounds, runways, taxiways,
aprons; airfield lighting, airside roads, airside lighting; airside safety; nose-in
guidance; aircraft parking; visual navigation aids; aircraft refuelling; and
• Passenger processing services and facilities — forward airline support areas;
aerobridges, airside buses; departure lounges and holding lounges (not VIP and
business lounges); immigration and customs services areas; public address
systems, closed circuit surveillance systems, security systems; baggage make-up,
baggage handling, baggage reclaim; public areas in terminals, public amenities,
public lifts, escalators and moving walkways; flight information display systems; and
landside roads.
Thus, apart from the inclusion of aircraft refuelling in the bundle ‘aircraft movement
services and facilities’, there would be virtually no change from current regulatory
coverage. However, some assets (for example, landside roads) would have to be
apportioned between regulated and unregulated activities of the airport. The provision
(and pricing) of taxi holding facilities that should relieve landside road congestion and
the need for additional roadway, should not be subject to the price cap.
Some services (apart from aircraft refuelling) that currently are designated
‘aeronautical related’ — staff car parking, check-in counters and related facilities, and
aircraft light and emergency maintenance sites — should continue to be subject to
price monitoring but not a price cap. However, some other services currently
designated as aeronautical related should not be subject to further monitoring: public
car parks; freight equipment storage sites; freight facility sites and buildings; ground
support equipment sites; and heavy maintenance sites and buildings.

Price caps would continue to be complemented by monitoring of the quality of

regulated services (see details under Option B, section 12.4). Quality monitoring
should identify any failure to invest appropriately on the part of airports. However,

346 PRICE REGULATION

OF AIRPORT
SERVICES
planned investment incorporated in the price cap may also need to be monitored
directly to ensure that projects are not being intentionally delayed by the airport.
Airlines would be expected to play a role in developing the forward investment
program to be incorporated within the price cap (ideally, in negotiation with
airports) and their input should be sought in monitoring implementation of that
program.

Box 12.2 Dual-till, replacement-cost price caps

Any price caps should be developed on a current-cost basis. They should incorporate
the cost of required (optimised) aeronautical assets (including anticipated efficient
capital expenditure) and anticipated efficient operating costs (with reference to
productivity benchmarks where feasible) over the five-year regulatory period. In
addition, they should be calculated on a dual-till basis. That is, assets used and profits
earned in non-aeronautical activities should not be taken into account in determining
allowable aeronautical prices.
Implementation of this current cost, dual-till approach would require both an
assessment and revaluation of the aeronautical asset base. While historical costs may
make for a simpler assessment, they are less likely to encourage efficient production
and investment in this industry, especially where the current opportunity cost of
significant assets is substantially above their historic cost (eg aeronautical land at
Sydney Airport).
Capped prices at these airports would then be calculated on a broadly similar basis to
current aeronautical prices at Sydney Airport (but with land appropriately valued).
Moreover, whereas at Sydney prices were increased by around 100 per cent (from
previous Federal Airports Corporation prices) in one step, any significant price changes
could be phased in over the regulatory period via the X factors (appropriately adjusted
for anticipated productivity and demand growth), to give airports and airport users time
to adjust to the new basis for price setting.
With price caps calculated in this way, prices would not require adjustment for
aeronautical investment foreseen and agreed to at the commencement of the
regulatory period. However, there still may need to be provision for appropriate cost
pass-through of large, unanticipated investments (and possible downward adjustment
if planned investments do not proceed).
Price caps should continue to be calculated on a tariff-basket basis and hence allow
price restructuring within the cap.
The Commission does not consider that setting price caps on a dual-till basis
necessarily would deliver windfall gains (or losses) to airport operators. The outcome
would depend on the expectations about future prices and regulation that were
factored into operators’ bids for the purchase of the leases.

APPROPRIATE 347
FUTURE
REGULATION
Price monitoring of some services currently designated as ‘aeronautical related’, in
which market power has been assessed as moderate (for example, check-in
counters, staff car parking) would continue at these airports (box 12.1).

Option A: Sydney Airport

For reasons discussed in box 12.3 and appendixes F and H, in the absence of
defined slot rights and trading mechanisms, regulation preferably should allow
aeronautical prices at a capacity-constrained Sydney Airport to reflect more closely
the opportunity costs incurred by airlines and their passengers in using the facility
(that is, the value of slots to those who miss out at peak times), rather than simply
reflecting the efficient costs of production.

At the very least, provided capacity remains constrained, aeronautical prices should
not be required to decline in real terms and should be adequate to encourage
efficient, feasible expansion of aeronautical capacity at that facility.
• This could be implemented either by requiring notification of aeronautical price
increases above the CPI, or by imposing a CPI-X cap with the X set at zero.
• In either case, price increases should be allowed to reflect peak-period demand
(as provided for in the current regulatory framework for Phase 1 airports), and to
accommodate necessary investment.

Box 12.3 The benefits of price rationing at Sydney Airport

Prior to September 2001, Sydney Airport experienced excess demand for aircraft
movement slots for several hours a day. This situation is likely to resume in the not-too-
distant future. The limited supply of slots is currently due to a movement cap and
curfew, and the regional ‘ring fence’. It is likely that the market-clearing price at peak
times exceeds airport production costs.
Appropriate peak/off-peak charges at Sydney Airport would promote more efficient use
of the airport and, given the massive investment required to establish a second airport,
also would provide important signals to the Government about the need for a new
facility and the costs of the various constraints imposed on the use of Sydney Airport.
Though prices approaching market-clearing levels at peak times would be likely to
promote use of the airport by those who value it most — even if ring-fenced regional
flights were quarantined from such price increases, in line with announced Government
policy — the airport could earn large profits, reflecting the scarcity value of slots.
(Efficient peak/off-peak charging would balance lower off-peak charges against higher
peak charges — this would tend to moderate the extent of profit increases.)
(Continued next page)

348 PRICE REGULATION

OF AIRPORT
SERVICES
 Box 12.3 (continued)
It is important to note that any rise in peak prices is unlikely to lead to higher passenger
fares. There is strong evidence that the scarcity value of the slots is factored into fares
already; it is very difficult to obtain discounted fares to Sydney at peak times. In other
words, it is the airlines that are currently benefiting from the excess demand for landing
slots at peak times.
If bidders were informed prior to the sale of Sydney Airport that efficient peak/off-peak
charges would be permitted, then any anticipated scarcity rents would accrue to the
Commonwealth Government (and taxpayers generally) in the bid price, rather than to
the new airport operator.
Though there are strong efficiency arguments for pricing the use of slots to reflect the
opportunity costs of using the facility, any decision also will be guided by potential
effects of higher charges on airlines (to the extent that they forfeit scarcity rents), and
on passengers unwilling to pay higher charges and who forfeit peak-period access.
Services on some other routes also may be affected if average costs of these flights
are being covered by higher airline yields on Sydney routes.
An alternative or complementary means of promoting efficient use of Sydney Airport
would be to allow trading (and possibly auctioning) of slot rights — that is, rights to
future airport access. The development of slot markets would require definition of the
property rights attached to slots, including the scope for the airport to alter its charges.
Whatever the decision about appropriate future policy at Sydney Airport, for that policy
to be factored adequately into the sale price (for the ultimate benefit of taxpayers),
bidders must be given clear guidance about the regulatory framework that will apply to
the airport. The need for such clarity is underscored by the uncertainty and disputation
that have been associated with the lack of transparency and specificity of the current
price-cap regime applying to the already privatised core-regulated airports.

Option A: other airports

Adelaide, Canberra and Darwin airports would be subject to modified price-

monitoring arrangements (as outlined below in Option B, section 12.4) rather than
price caps. Quality monitoring (also as outlined below) would continue at these
airports.

For all other airports, there would be no airport-specific price regulation or quality
monitoring. This would require amendment of the Airports Act so that Hobart,
Launceston, Alice Springs, Coolangatta and Townsville airports no longer were
designated as ‘core-regulated’ airports.

APPROPRIATE 349
FUTURE
REGULATION
Option A: review and access provisions

An independent, public review would need to be conducted towards the end of the
five-year period to ascertain, on the basis of that experience, whether there should
be any future price regulation of price-capped and price-monitored airports. Other
airports should be included in this review only where there is prima facie evidence
of persistent misuse of market power. Proposed review guidelines are set out under
Option B in section 12.4.

All airports, whether they are subject to price regulation or not, should be subject to
the generic provisions of the National Access Regime (Part IIIA of the TP Act). An
airport-specific access regime for price-regulated airports should be considered only
if procedural improvements, such as scope for multilateral arbitrations
recommended in the Commission’s Review of the National Access Regime
(PC 2001a), were not made to that regime (chapter 11).

12.4 Option B: price monitoring

Option B differs from Option A in that it would extend lighter-handed regulation to
Sydney, Melbourne, Brisbane and Perth airports. Under Option B, prices at these
airports would be monitored by the regulator, albeit with somewhat more detailed
reporting requirements than for price-monitored Phase 2 airports.

Option B: Sydney, Melbourne, Brisbane, Perth, Adelaide, Canberra

and Darwin airports

Price monitoring

For Sydney, Melbourne, Brisbane, Perth, Adelaide, Canberra and Darwin airports,
there would be mandatory price monitoring by the ACCC. The monitoring regime
would continue for five years.
• During this probationary period, the regulator would not have the power to alter
unilaterally the monitoring regime or impose stricter price regulation.
• Information requirements would be specified at the commencement of the period
and could not be amended without agreement of the parties. As outlined in
box 12.4, the Commission envisages somewhat more detailed monitoring of
Sydney, Melbourne, Brisbane and Perth airports (specifically, requiring
reporting of the weighted average cost of capital (WACC) for aeronautical

350 PRICE REGULATION

OF AIRPORT
SERVICES
 assets). Otherwise, proposed information requirements for the seven airports
broadly would be consistent with existing disclosure and reporting requirements.
• As under current monitoring and reporting arrangements, information would be
formally audited. There also should be provision to treat some data as
commercial-in-confidence where disclosure could undermine commercially-
negotiated agreements. An annual report presenting monitored information (in a
way that protects commercial-in-confidence material) would be made publicly
available, with commentary (but not a pre-emptive assessment of the monitoring
regime) by the ACCC (and auditors) where considered warranted.

Box 12.4 Option B: disclosure requirements for price-monitored

airports
For Melbourne, Brisbane, Perth and Sydney airports, price monitoring of aeronautical
services (classified in box 12.1 as aircraft movement services and facilities and
passenger processing services and facilities) and selected aeronautical-related
services (also outlined in box 12.1) would require:
• for each of these services, detailed information on:
– the allocation of costs to the provision of the service; and
– revenue, including the number of units supplied, the charge, the basis for
charging, and total revenue derived.
• at an aggregate level, information on revenue, costs and profitability, comprising:
– profit and loss account;
– balance sheet;
– cash flow statement;
– expense allocations;
– schedule of assets;
– an estimate of the weighted average cost of capital (WACC); and
– operational statistics, including information on passenger numbers and aircraft
movements.
For other facilities and services provided by the airport operator (broadly speaking,
non-aeronautical services), only aggregate information would be reported (the same as
that for aeronautical services outlined above, except that WACC information would not
be required). Detailed information for each service would not be required for services in
this category.
For Adelaide, Canberra and Darwin airports, information requirements would be the
same as for Sydney, Melbourne, Brisbane and Perth airports, except that WACC
estimates would not be required for any monitored service.

APPROPRIATE 351
FUTURE
REGULATION
Quality monitoring

Modified quality monitoring arrangements would continue at all price-monitored

airports. The list of quality-monitored services should be reviewed to match those
services subject to detailed price monitoring. Quality indicators also should be
reviewed to ensure that the quality of monitored services is within the control of the
airport operator (chapter 9). Public reporting by the ACCC of quality of service
outcomes at airports would continue as part of the broader requirements for price
monitoring.

Commercial agreements

Voluntary commercial agreements between airports and users (including non-airline

users) would be encouraged by providing some guidance as to the form such
agreements could take (chapter 11). For example, agreements should incorporate
agreed price and service levels and/or quality of service provision undertakings
(possibly with penalties for non-compliance); a commitment to the exchange of
information; consultation procedures; and dispute-settlement mechanisms (for
example, provision for independent arbitration).

Review criteria
An independent public review would need to be conducted towards the end of the
five-year monitoring period to ascertain whether there should be any future price
regulation of those airports. Other airports could be included in the review only
where there is prima facie evidence of persistent misuse of market power.
• Factors to be taken into account by the review in assessing whether airports have
abused market power should be specified at the start of the regulatory period.
Suggested criteria are summarised in box 12.5. Efficient pricing principles for
airports are discussed in chapter 4 and also have been developed more generally
by the Commission in its reviews of the National Access Regime and
Telecommunications Competition Regulation (PC 2001a and 2001d). They are
predicated on the need to avoid excessive prices, without the regulator
attempting to set prices too precisely, with the resultant risk that prices may be
pushed too low and efficient investment discouraged.
• In line with these principles, at least at airports without significant peak periods,
efficient prices broadly should generate revenue that is not significantly above
the long-run costs of efficiently providing aeronautical services (on a stand-alone
basis). At airports with significant capacity constraints, as discussed in box 12.3
and appendix H, efficient peak/off-peak prices may generate revenue that
exceeds the production costs incurred by the airport. Price discrimination also
352 PRICE REGULATION
OF AIRPORT
SERVICES
 can promote efficiency — this may mean that some users pay a price above the
long-run average costs of providing aeronautical services, while more price-
sensitive users pay a price below it.
• Additional factors to be taken into account by the review would include whether
quality has deteriorated or failed to meet the requirements of users; the extent to
which commercial agreements have been negotiated; the level of consultation
and exchange of information; and the level (and success) of applications for
access and the number (and validity) of user complaints.

Box 12.5 The Commission’s proposed review criteria

The proposed review of price regulation of airports to ascertain the need for continued
regulation (if any) in five years time, should be guided by the following criteria.

Pricing principles
• At airports without significant capacity constraints, efficient prices broadly should
generate expected revenue that is not significantly above the long-run costs of
efficiently providing aeronautical services (on a dual-till basis). Prices should allow a
return on (appropriately defined and valued) assets (including land) commensurate
with the regulatory and commercial risks involved.
• At airports with significant capacity constraints, efficient peak/off-peak prices may
generate revenues that exceed the production costs incurred by the airport.
• Price discrimination and multi-part pricing that promote efficient use of the airport
should be encouraged. This may mean that some users pay a price above the long-
run average costs of providing aeronautical services, whereas more price-sensitive
users pay a price close to marginal cost.

Other criteria
• Whether quality-of-service outcomes have deteriorated and/or failed to meet the
requirements of users.
• The extent to which commercial agreements on prices and quality of service have
been negotiated.
• The degree of consultation with airport users and the extent of the exchange of
information.
• The number and outcome of applications for third-party access and the extent and
validity of user complaints.

Option B: Other provisions

For all other airports, there would be no airport-specific price regulation or quality-
of-service monitoring. This would require amendment of the Airports Act so that
APPROPRIATE 353
FUTURE
REGULATION
Hobart, Launceston, Alice Springs, Coolangatta and Townsville Airports no longer
were designated as ‘core-regulated’ airports.

For all airports, access provisions (and provisions for access undertakings) should
not differ from the generic provisions of the National Access Regime (Part IIIA of
the TP Act). An airport-specific access regime should be considered only if
procedural improvements, such as scope for multilateral arbitrations (as
recommended by the Commission in its Review of the National Access Regime
(PC 2001a)), are not made to that regime. The Commission sees no need to exempt
from access regulation airports that enter into commercial agreements with users.

12.5 Weighing the options

Option A is superior to the existing regime

Option A would continue to apply price-cap regulation to those airports considered

to have a high degree of market power. In addition, Option A would set price caps
that broadly reflect costs of supplying aeronautical services on a stand-alone, rather
than single-till, basis. For these reasons, Option A would, in the Commission’s
view, perform better than the regulatory arrangements in place before 5 October
2001 and those that have applied since then. In particular:
• by making allowance for anticipated investment in the price-cap parameters, it is
likely that regulatory involvement in and, therefore, scope for distortion of,
investment decisions will be reduced. In addition, airports and their users would
receive much better signals from efficient dual-till prices regarding the real costs
and benefits of supplying aeronautical services;
• airports and their customers would have greater incentives to negotiate price and
quality agreements within the overall limits of the designated caps (at least
where there is unused airport capacity), reflecting incentives to increase
aeronautical and non-aeronautical profits by expanding use of the airport at the
margin;
• the use of benchmarking, where feasible, could help to sustain the positive
incentive effects of price caps and avoid them converging to rate-of-return
regulation; and
• maximum average aeronautical prices at Melbourne, Brisbane and Perth airports
would be set on a dual-till basis, consistent with prices allowed by the ACCC for
Sydney Airport.

354 PRICE REGULATION

OF AIRPORT
SERVICES
Price caps are not warranted

The Commission has not been persuaded, however, that there is a strong case for
the continuation of price caps at any privatised core-regulated airports. This is for
two principal reasons.

The first is the ever-present risk of regulatory failure, given the severe information
problems confronting any regulator. Setting price caps inevitably entails detailed
regulatory assessment of, and involvement in, airport operations and investment
decisions. It should therefore be used only where the potential efficiency costs of
abuse of market power are significant. Even then there is a risk that regulation will
cause its own distortions to production and investment decisions. While the
Commission agrees that some transitional problems with current price-cap
arrangements may have been settled, and that the price caps proposed under
Option A should reduce regulatory involvement in investment decisions, the risk of
regulatory failure remains high. This risk has been amplified by the uncertainty that
currently pervades global aviation markets.

The second and related reason is that the ‘problem’ to be addressed does not
warrant such a heavy-handed regulatory regime. Though the four largest airports
have considerable market power, the prospect of them using that power in a way
that would generate significant costs to the economy or community is supported
neither by the evidence nor the analysis. There are strong commercial incentives
pulling in the other direction, including scope for increased profits in non-
aeronautical activities from increasing passenger volumes, and incentives to
discriminate and differentiate in pricing.

Option B provides more appropriate regulation

On balance, therefore, the Commission considers that while the undoubted market
power of the four major airports warrants some form of regulatory constraint, the
continuation of price caps is not the best approach. Option B offers a much better
regulatory mechanism for promoting the principles for regulation identified in the
terms of reference.

The ultimate objective — to ensure efficient long-term provision of airport services

— is common to both options. In the Commission’s view, Option B provides a
greater chance that this objective will be achieved, by providing a better balance
between regulatory constraint and promotion of commercial relationships.

While some participants have questioned whether commercial negotiations are

feasible in this industry, the Commission notes that airlines deal directly with

APPROPRIATE 355
FUTURE
REGULATION
numerous unregulated smaller airports, both privately- and publicly-owned, and
some unregulated larger ones (for example, Cairns). Relations may not always be
smooth, but there seems to be considerable scope for reasonable commercial
interaction between airports and users, provided there is some ultimate constraint on
abuse of any market power (which Option B provides, in addition to the TP Act).

By providing a credible threat that price controls could be reintroduced, enunciating

principles of efficient pricing to guide airport behaviour, and imposing disclosure
requirements, any tendency for airports to increase their charges beyond efficient
levels should be constrained.

The Commission also draws attention to the continued relevance of third-party

access provisions (Part IIIA) and Part IV of the TP Act to airports. The Commission
does not consider that airports have strong incentives to deny or frustrate access in
order to impede competition in a way that would be likely to lead to declaration of
aeronautical facilities (at least under the modified declaration criteria recommended
in the Review of the National Access Regime (PC 2001a)). Nonetheless, the
potential activation of this mechanism does provide users with regulation of last
resort, providing additional encouragement for airports to enter into reasonable
agreements regarding prices and conditions of airport use.

Some participants expressed concern about the need for additional public scrutiny
of any principles designed to guide airport behaviour. However, what the
Commission is proposing (box 12.5) is not detailed prescription, but rather broad
principles of efficient pricing that have been discussed extensively during this
inquiry and parallel inquiries into infrastructure regulation (including reviews of the
National Access Regime, the PS Act and Telecommunications Competition
Regulation).

Some participants also cited their interpretation of the recent New Zealand
experience with light-handed regulation in an attempt to rebut the Commission’s
price-monitoring option. But, closer examination reveals that the New Zealand
system, notwithstanding some deficiencies, has not been a failure (chapter 11). For
example, Auckland and Christchurch airports have agreed on new aeronautical
prices with their major airline customers, and those price outcomes do not appear
excessive.

While light-handed regulation currently is not the dominant model worldwide, there
are signs that the direction is changing. For example, in the United Kingdom, the
Civil Aviation Authority recently has proposed modifications to regulatory
arrangements that are in part designed to foster commercial negotiation between
airports and airlines at major airports.

356 PRICE REGULATION

OF AIRPORT
SERVICES
The Commission recognises that its preferred option would involve a considerable
shift from current arrangements at major airports — though one largely envisaged
by the architects of the regulatory arrangements at the time of airport privatisations.
The Commission also recognises that some parties may find such a change difficult,
particularly given the long history of government provision of airport services at
major airports, and pricing structures that have effectively subsidised aeronautical
charges. Nonetheless, even if price caps were to be maintained (in any form),
aeronautical charges would still need to rise to encourage efficient long-run service
provision.

The Commission considers that the full benefits of privatisation of airports are
unlikely to be realised if commercial relationships between airports and airlines
continue to be heavily conditioned by intrusive price regulation. The ongoing need
for substantial investments at major airports requires a more commercial and
cooperative approach. The potential for regulation unduly to constrain prices poses
a real risk and one that could impose significant costs on consumers in the future.

The Commission emphasises that it is not advocating deregulation of major airports.

It is proposing a probationary regulatory package designed to facilitate the transition
to a more commercial environment, while providing credible constraints on the use
of market power by these airports.

Whether or not the Commission’s preferred option is adopted, it is essential that,

prior to implementation of the chosen regulatory approach, parties are consulted on
the practicalities of proposed regulation and made aware of the various
requirements, in order to reduce uncertainty and the potential for disputation. In
particular, bidders for Sydney Airport should have a clear picture of the regulatory
framework for that facility, so that expected future airport charges can be factored
adequately into the sale price.

Recommendations 1–8 should be read in conjunction with Options A and B which

are set out in this chapter.

RECOMMENDATION 12.1

For Sydney, Melbourne, Brisbane and Perth airports, price caps and prices
notification arrangements should be replaced by mandatory price monitoring
arrangements for a probationary five-year period, as outlined in Option B.
• Airport-specific price-monitoring arrangements could be incorporated either
in the Airports Act 1996 (Airports Act) or the Trade Practices Act 1974
(TP Act), but should be consistent with any generic price-monitoring
provisions that may be introduced into the TP Act following the Commission’s
separate review of the Prices Surveillance Act 1983 (PS Act).

APPROPRIATE 357
FUTURE
REGULATION
In the event that the Government opted for a stricter form of price regulation at
these four airports, Option A should apply such that:
• annual price caps of the form CPI-X continue for five years at Melbourne,
Brisbane and Perth airports. Price caps should be set to reflect the efficient
costs of providing aeronautical services in the long run, on a dual-till basis.
Price caps should be complemented by price monitoring of some
‘aeronautical-related’ services; and
• for a capacity-constrained Sydney Airport, prices should not be required to fall
in real terms. Regulation should comprise either prices notification or a price
cap of the form CPI-X, with X set at zero. Price increases should be allowed to
reflect peak-period demand and to accommodate necessary investment.

RECOMMENDATION 12.2

For Adelaide, Canberra and Darwin airports, mandatory price monitoring of

aeronautical services and some ‘aeronautical-related’ services (as outlined under
Option B) should continue for five years. (Airport-specific price-monitoring
arrangements could be incorporated either in the Airports Act or the TP Act, but
should be consistent with any generic price-monitoring provisions that may be
introduced into the TP Act following the Commission’s separate review of the
PS Act.)

RECOMMENDATION 12.3

Quality monitoring of regulated services (as outlined under Option B) should

continue at all airports subject to price regulation; that is, at Sydney, Melbourne,
Brisbane, Perth, Adelaide, Canberra and Darwin airports.

RECOMMENDATION 12.4

Neither price monitoring nor price caps should be reintroduced for Alice Springs,
Coolangatta, Hobart, Launceston and Townsville airports. The Airports Act
should be amended so that these airports are no longer designated as ‘core-
regulated’ airports.

RECOMMENDATION 12.5

Commercial agreements should be encouraged and assisted (for example, by

providing guidelines regarding coverage) under price-monitoring arrangements,
or price caps, if they were retained at some airports.

RECOMMENDATION 12.6

Price regulation of airports should be reviewed towards the end of the five-year
regulatory period. The review should be independent and public. Its objective
358 PRICE REGULATION
OF AIRPORT
SERVICES
should be to ascertain the need for any future price regulation of airports
(including price monitoring or more stringent price regulation). In making its
assessment, the review should be guided by principles of efficient pricing plus
several other criteria set out under Option B. Agreed review criteria should be
spelt out at the beginning of the regulatory period.
Other airports should be included in the review only where there was prima facie
evidence of persistent misuse of market power (namely, evidence of inefficient
prices, poor quality etc).

RECOMMENDATION 12.7

All airports should be subject to the generic provisions of the National Access
Regime in Part IIIA of the TP Act. An airport-specific access regime should be
considered only if procedural improvements, such as scope for multilateral
arbitrations, are not made to the National Access Regime.

RECOMMENDATION 12.8

Prior to implementation of the chosen regulatory approach, airports and airlines

should be consulted on the practicalities of the proposed regulation and made
aware of its various requirements, in order to reduce uncertainty and the
potential for disputation. In particular, bidders for Sydney Airport should have a
clear picture of the regulatory framework for that facility so that expected future
airport charges can be factored adequately into the sale price.

APPROPRIATE 359
FUTURE
REGULATION
APPENDIXES
A Terms of reference: correspondence

TERMS OF 363
REFERENCE:
CORRESPONDENCE
364 PRICE REGULATION
OF AIRPORT
SERVICES
B Public consultation

B.1 List of submissions

Participant Submission no.

Adelaide Airport Limited 20, DR57, DR77
Aero-Tropics Air Services 3
Airports Council International DR65
AMP Henderson 10
Ansett Australia/Air New Zealand 42
Australian Airports (Townsville) Pty Ltd 14
Australian Airports Association 15, DR51, DR69
Australian Competition and Consumer Commission 36, 38, 49, DR55
Australian Council for Infrastructure Development 28
Australian Taxi Industry Association 4, DR61
Board of Airline Representatives of Australia Inc 26, 41, DR53, DR54, DR72
Brisbane Airport Corporation Limited 8, 44, DR73
Capital Airport Group 32, 45, 46, 47, DR52, DR60, DR75, DR76
CEO Collegiate Pty Ltd DR59
Department of Industry, Science and Resources 40
Department of Infrastructure, Planning and Environment (NT) DR78
Department of State and Regional Development (Vic) 23, DR70
Department of Tourism, Racing and Fair Trading (Qld) 6
Department of Transport and Regional Services 39
Department of Transport and Works (NT) 24
Forsyth, Professor Peter 5
Gold Coast Airport Limited 16, DR58
Hastings Funds Management 19
Hobart International Airport Ltd 11
Impulse Airlines 18
International Air Transport Association 9, 43, DR56
Launceston Airport 35
Lewis, Peter 1

PUBLIC 365
CONSULTATION
Major, Hugh DR79
Melbourne Airport 7, 37, DR66
MTAA Superannuation Fund Pty Ltd 22, 34, DR67
Northern Territory Airports Pty Ltd 25, 50, DR71
Perth Airports Municipalities Group 12
Qantas Airways Limited 48
Queensland Transport 31
Regional Aviation Association of Australia DR64
Scott-Bloxam, W.H. 2
Shell Australia DR63
South Australian Government 33
Stott, Donald 29
Sydney Airports Corporation Limited 27, DR62
Tourism Tasmania 13
Victorian Hire Car Association DR68
Virgin Blue 30, DR74
Western Australian Government 17
Westralia Airports Corporation Pty Ltd 21

B.2 Visits
Adelaide Airport Limited
Airport Co-ordination Australia
Airservices Australia
AMP Henderson
Ansett Australia
Australian Airports Association
Australian Competition and Consumer Commission
Australian Federation of International Forwarders
Australia Pacific Airports Corporation
Board of Airline Representatives of Australia Inc
Brisbane Airport Corporation Limited
Bureau of Transport Economics
Cairns Port Authority
Capital Airport Group
Commerce Commission (New Zealand)
Department of the Treasury (Cwlth)

366 PRICE REGULATION

OF AIRPORT
SERVICES
Department of Transport (NSW)
Department of Transport and Regional Services (Cwlth)
Impulse Airlines
New South Wales Treasury
OECD (Secretariat Competition Policy)
Office of Asset Sales and IT Outsourcing
Qantas Airways Limited
Queensland Transport
Regional Airlines Association of Australia
Sydney Airports Corporation Limited
The Cabinet Office (NSW)
Virgin Blue

B.3 Public hearing participants

Initial public hearings

Sydney, 26 March 2001

Australian Airports (Townsville) Pty Ltd

Australian Taxi Industry Association
Impulse Airlines

Melbourne, 2–4 April 2001

Adelaide Airport Limited

Ansett Australia
Australian Airports Association
Australian Council for Infrastructure Development
Australia Pacific Airports Corporation
Board of Airline Representatives of Australia Inc
Brisbane Airport Corporation Limited
Capital Airport Group
Gold Coast Airport
Hobart International Airport Ltd
MTAA Superannuation Fund Pty Ltd

PUBLIC 367
CONSULTATION
Northern Territory Airports Pty Ltd
Professor Peter Forsyth
Qantas Airways Limited
Sydney Airports Corporation Limited
Virgin Blue
Western Australian Government
Westralia Airports Corporation Pty Ltd

Draft report public hearings

Melbourne, 17–18 October 2001

Adelaide Airport Limited

Australian Airports Association
Australia Pacific Airports Corporation
Australian Taxi Industry Association
Board of Airline Representatives of Australia Inc
Brisbane Airport Corporation Limited
Capital Airport Group
Gold Coast Airport
MTAA Superannuation Fund Pty Ltd
Northern Territory Airports Pty Ltd
Qantas Airways Limited
Sydney Airports Corporation Limited
Westralia Airports Corporation

368 PRICE REGULATION

OF AIRPORT
SERVICES
C Aeronautical and non-aeronautical
services

This appendix explores some conceptual issues arising from the complementary
provision of aeronautical and non-aeronautical services at most major airports,
specifically:
• implications for pricing of aeronautical services;
• the distinction between locational rents and monopoly profits; and
• arguments for and against ‘single-till’ regulation.

The implications of non-aeronautical earnings for aeronautical prices at Australian

airports are discussed in detail in chapter 7. The Commission’s assessment of
single- and dual-till issues is presented in chapter 10.

C.1 The effect of non-aeronautical profits on

aeronautical charges
As noted in chapter 2, in 1999-00, all core-regulated airports in Australia, except
Coolangatta, earned more revenue from so-called non-aeronautical activities
(essentially property and retail activities from which airports receive rental and/or
royalty payments) than from the provision of aeronautical services. While these
commercial services typically complement the core aeronautical services provided
by the airport,1 they do not have to be consumed as part of the core ‘bundle’ of
aeronautical services. For example, as discussed in chapter 6, international
passengers can choose where they buy duty-free goods (and whether they buy duty
free at all); airlines can choose where to locate maintenance facilities and so on.

Despite this competition, it would appear, for most core-regulated airports, that the
profitability of non-aeronautical services exceeds that of aeronautical activities. The
report prepared for the ACCC by KPMG (sub. 36, attachment A) provides some
evidence that this result is not attributable to arbitrary allocation of costs across the
two categories.

1 Not all services are complementary, however. As noted in chapter 2, some airports are
developing business parks that appear as dependent on local business as on the airport.
AERONAUTICAL AND 369
NON-AERONAUTICAL
SERVICES
The current imbalance in profitability may reflect regulation of aeronautical
charges. In the absence of price-cap regulation, airports may raise aeronautical
charges and their profits from this sector.2

Whatever the ‘stand-alone’ profit-maximising level of aeronautical charges, if

airports are able to earn additional profits from non-aeronautical goods and services
provided at the airport, they will have an incentive to reduce the (unregulated)
prices of aeronautical services below the level that would apply without the non-
aeronautical profit. This result holds provided the demand for aeronautical services
responds to changes in their price.

The other key requirement is that the demand for non-aeronautical services (and
hence profits from these activities) flows directly from consumption of aeronautical
services, not the other way around. This condition appears to be met at airports. It is
unlikely that people primarily visit an airport to shop and decide to take a flight
while they are there. As Kahn notes:
The critical condition for such a cross-subsidization (pricing one of the complementary
products below its separate incremental cost) is that the cross-elasticity of demand for
the complementary service be high enough to compensate for the out-of-pocket losses
on the sales of the first of these, considered in isolation. (Kahn 2001a, p. 17)

Figure C.1 illustrates the possibility. The diagram is a modified version of Starkie’s
(2001c) exposition. The figure shows a stylised price of, and demand for,
aeronautical services (eg for landings or use of terminal facilities) at an airport. As
drawn, the demand curve is downward-sloping and therefore assumes that the
airport has some market power. For simplicity, marginal (and average) costs of
providing these facilities are assumed to be equal and constant.

The efficient, economic welfare-maximising price for aeronautical services is Pc.

An unregulated, profit-maximising airport, in the absence of non-aeronautical
profits, would set the price for aeronautical services at Pm (assuming price
discrimination is not feasible), at which price marginal revenue equals marginal
cost.

2 This could result from airports using their market power to push prices above efficient levels or
it may just be that current charges are below levels consistent with cost-recovery and therefore
would rise even in the absence of market power (chapter 8).
370 PRICE REGULATION
OF AIRPORT
SERVICES
Figure C.1 Aeronautical pricing with non-aeronautical profits

Paero
MRaero+MPna

Pm

Pna

MC=ACaero
Pc
Daero

MRaero Qaero

The constructed line MRaero+MPna combines marginal revenue from aeronautical

services and marginal profits (marginal revenue minus marginal costs) from the
provision of non-aeronautical services. Again for simplicity, marginal profits from
non-aeronautical activities are assumed constant per aircraft movement (and hence
MRaero+MPna is drawn parallel to the MRaero curve).

The rationale for this is that a profit-maximising airport, in setting aeronautical

prices, will take into account all additional profits that flow to it from attracting an
additional user of its aeronautical facilities. The price of aeronautical services now
will be set where MRaero+MPna intersects the MCaero curve — that is, at Pna. As
drawn, this price still exceeds the efficient price, Pc, though it is feasible, if non-
aeronautical marginal profits are large, and demand for aeronautical services is not
highly inelastic, that unconstrained airport pricing could result in aeronautical
charges below their average cost. (Indeed, they could be negative.) While the airport
forfeits profits from aeronautical activities due to the reduction in their price, this
loss is more than compensated for by additional profits from non-aeronautical
activities. This expansion of passenger throughput will continue until the marginal
profit forgone from aeronautical activities equals the marginal profit earned from
non-aeronautical activities.

In other words, where users of aeronautical services have complementary demands

for non-aeronautical services, a profit-maximising airport will have an incentive to
attract more customers to the airport by reducing its aeronautical charges. How
large the (absolute) reduction in the price of aeronautical services will be is an
empirical matter, depending on the marginal profitability of non-aeronautical
activities.

AERONAUTICAL AND 371

NON-AERONAUTICAL
SERVICES
For capacity-constrained airports where increased aircraft throughput is not feasible,
or for those airports where the demand for aeronautical services is insensitive to
price, or where non-aeronautical activities are not profitable at the margin, or are
profitable but are not linked to increased aeronautical traffic, prices for aeronautical
services will be set on a ‘stand-alone’ basis.

It is possible, even where demand and profit increases are feasible, that prices will
not be reduced for all users of aeronautical facilities. Airports (with excess capacity)
have an incentive to target price discounts at marginal, or more price sensitive,
airlines and passengers, especially those who contribute more to non-aeronautical
profits (eg international passengers).

In addition, an airport may seek to encourage traffic growth by expanding capacity

(at a given price) or encourage passenger spending by improving airport quality
rather than by explicitly reducing aeronautical prices. The mix of incentives
selected will depend on the marginal consumer being targeted. Additional flights or
a new airline may be encouraged by rebates or discounts on aeronautical charges,
marketing assistance or the provision of additional facilities. Passenger spending
may be encouraged by the provision of attractive terminals and quick processing.

Additional airport profits from related non-aeronautical activities may also augment
airlines’ bargaining power: the potential loss of airport profits will be greater for
any threatened reduction in services by airlines.

C.2 Non-aeronautical profits: locational rents and/or

market power?
The above analysis rests on the assumption that non-aeronautical activities generate
pure profits to the airport — that is, profits in excess of rates of return necessary to
maintain supply of the good or service. The existence of such profits may suggest
that any reduction in prices of aeronautical services (and a consequent reduction in
the monopoly deadweight loss caused by pricing above marginal cost) largely
would be offset by increased prices (and associated efficiency losses) in the
provision of non-aeronautical services.

However, non-aeronautical profits may reflect economic rent accruing to location

rather than market power per se. Economic rent is the payment made to a factor
above that needed to keep it in its current use. If a factor has some alternative uses,
then at the margin, it may earn no rent (with payment to the factor just reflecting its
value in an alternative use), though it may earn infra-marginal economic rent.

372 PRICE REGULATION

OF AIRPORT
SERVICES
Airports typically act as landlords to providers of retail services, maintenance
facilities, etc. Airport revenue from these activities accrues as rentals and/or
royalties to the scarce factor — in this case, designated airport land — which core-
regulated airports, in turn, lease from the owner, the Commonwealth Government.3

Provided airports do not constrain artificially the provision of space for non-
aeronautical facilities, profits earned from providing such space will reflect
locational rents. These rents in turn reflect the inherent scarcity of land proximate
and convenient to airside activities at an airport. An airport’s ability to raise prices
through artificial scarcity will be constrained by the ability of concessionaires to
locate, and consumers to purchase, elsewhere. For example, any monopoly profits
(that is, in excess of locational rents) earned from car parking are likely to
encourage off-airport providers and substitution by airport users to other travel
modes.4 Monopoly retail rentals would drive concessionaires to relocate because
uncompetitive rentals cannot be passed on to consumers (who, in turn, have a range
of choices of where to buy these goods and services).

Importantly, if profits from non-aeronautical services reflect only locational rents,

then they generate no efficiency loss. At the margin, the rental paid for land used in
a particular non-aeronautical service will reflect the opportunity cost of not using
that land to provide the highest-valued, alternative, competitive, non-aeronautical
activity.

Moreover, intervention that discourages such rents being earned unambiguously

will reduce welfare because scarce resources will not be allocated to their highest-
valued use. Economic surplus will be forfeited. This result contrasts with the
potential for intervention aimed at reducing monopoly profits to enhance welfare.
This is because monopoly profits are transfers from consumers to the monopolist
that may be brought about by an inefficient level of supply to the market.

C.3 Arguments for and against a single till

Privatised core-regulated airports in Australia have been regulated on a so-called
dual-till basis. Under a dual till, for regulatory purposes, airport services are divided
into two tills covering aeronautical and non-aeronautical services. Only those
aeronautical services in which the airport is considered likely to have most market

3 Some airport operators also may just be very good at developing profitable ventures at airports.
In this case it may be expected that skilled personnel would retain a reasonable share of any
quasi-rents in the form of higher remuneration.
4 However, the airport’s control over ‘front-door’ access may allow it to influence competition
from off-airport car-park providers and other travel modes (chapters 6 and 7).
AERONAUTICAL AND 373
NON-AERONAUTICAL
SERVICES
power are then regulated under price caps. Airports are free to set rentals and prices
for retail and other non-aeronautical activities. Car-park prices have been monitored
by the ACCC (chapter 3) but car-parking revenue has not been taken into account
when setting the price caps.

This contrasts with the single-till regulatory arrangements for major airports in the
United Kingdom, where revenues from non-aeronautical services are subtracted
from expected airport-wide costs to determine allowable revenue from aeronautical
services (appendix G). A similar system was used to set aeronautical charges at
Federal Airports Corporation-operated airports prior to privatisation — aeronautical
charges covered residual capital and operating costs (plus a real rate of return on
capital) after taking into account projected revenues from non-aeronautical
activities.

The ACCC’s draft decision on proposed price increases at Sydney Airport

(ACCC 2001h) suggested a limited single till (though described as a ‘modified dual
till’) that would have transferred around $20 million of annual car-parking and other
‘aeronautical-related’ revenue to reduce ‘stand-alone’, cost-based aeronautical
charges.

Privatised core-regulated airports appear to have been sold on the basis that a single
till would ‘not be mandated’, and the Commonwealth Government, in April 2001,
directed the ACCC to implement a dual till at Sydney Airport (Direction No. 22).
But, as outlined in chapter 10, several participants (Board of Airline Representatives
of Australia (BARA), Ansett and Qantas; subs 41, 42 and 48) argued in favour of
airport price regulation on a full or partial single-till basis on the grounds that
efficient pricing requires it.

Arguments for a single till

Market power in non-aeronautical services?

One argument put for a single till is that profits earned in non-aeronautical activities
reflect market power rather than economic (locational) rent. If this were the case,
then the single till may not prevent excessive pricing of non-aeronautical services
by airports. It merely would require that monopoly profits earned in non-
aeronautical activities were transferred to airlines and their passengers. However,
the single till could remove the incentive to earn these profits. This would be a
desirable, if unintended, consequence of a single till. If prices of non-aeronautical
services reflected market power, it may be more appropriate to reduce them directly
to promote efficient consumption and to protect consumers of the service.

374 PRICE REGULATION

OF AIRPORT
SERVICES
Producers in competitive markets do not retain rents

BARA, Ansett and Qantas (subs 41, 42 and 48) argued that efficient pricing of
airports requires that any locational rents earned from non-aeronautical activities by
airports should be applied to reduce aeronautical charges. BARA demonstrated that
if there were two identical airports (including, it must be assumed, with respect to
the relative attractiveness of their location), then competition for customers would
ensure that both airports earned normal profits. BARA is correct that if there were
no scarcity in the supply of any factor used in the production of airport services (and
no natural monopoly characteristics), long-run supply would be perfectly elastic.
Consequently, there would be no producer surplus (rent) earned on infra-marginal
units.

BARA (sub. DR54) cited the example of petrol stations that combine petrol
retailing with convenience outlets. In this industry, competition will drive down the
price of petrol such that any rents earned in the related retailing activity are
competed away. If there is no scarcity of any factor, then rents will not be earned in
the first place — that is, the price of petrol will decline only if the
petrol/convenience store combination reduces the cost of selling petrol. However, if
some petrol stations are more conveniently located than others and attract more
customers to their convenience stores, rents will be earned at these stores that will
not be bid away (unless another petrol station can open in the same location).
Therefore, even in highly-competitive industries, price-taking producers in
relatively high-yield locations (whether because of convenience or natural
phenomena) will earn and retain locational rents.5

In the case of airports, natural monopoly characteristics mean that there is only one
major regular public transport airport in each Australian capital city. This, in turn,
means that airport land that can be used to provide a range of non-aeronautical
activities (albeit complementary with aeronautical services), is scarce. (It may be
the case, if airport services could be supplied by large numbers of providers in any
location, that airport land would not be scarce. The reality is different, however.
Airport services are not supplied (efficiently) by a highly-competitive market.)
Given this scarcity, locational rents can be earned. As in any market where there is a
scarce factor, there does not appear to be any reason on efficiency grounds that
scarcity rents should accrue to consumers. Indeed, as discussed below, preventing
returns to owner-operators that reflect scarcity may discourage optimal use of the
scarce factor of production.

5 Generally, the rents will accrue to the original owner/exploiter of the site’s value. Those who
subsequently buy the site will pay the discounted, capitalised value of anticipated rents in the
sale price.
AERONAUTICAL AND 375
NON-AERONAUTICAL
SERVICES
As the UK Civil Aviation Authority noted:
Given the cost structure of airports and the diverse services which they provide … it is
unlikely that the provision of airport services would be perfectly
competitive … Moreover, given that capacity constraints are the key issue for the
industry, this characterisation assumes that scarcity rents which follow from planning
restrictions should necessarily be clawed back by the regulatory regime. This has not
occurred in the case of supermarkets … or in other markets where scarcity rents
exist … (CAA 2000f, pp. 11–12)

Demand complementarities

Demand complementarities between aeronautical and non-aeronautical services

provide a related rationale for a single till. The argument is that airlines deliver
passengers who generate non-aeronautical profits to the airport and these additional
profits should be factored into (lower) aeronautical prices. In other words, without
airlines, airports would not earn locational rents. However, who ‘deserves’ the rents
generally will be addressed adequately by the market itself. As discussed in
section C.1, an airport will have an incentive to encourage additional services and
passengers to increase its profits to the extent such complementarities exist. It does
not appear necessary to regulate to ensure such behaviour.

Common costs

Another argument for a single till is that a multi-product monopoly (with large fixed
common costs and a requirement to be (just) self-financing) should be encouraged
to set prices to minimise efficiency losses — that is, to implement Ramsey pricing
(Crew and Kleindorfer 2001). Generally, this means that markets/customers with
relatively inelastic demand should bear a greater share of common, fixed costs than
do those with more elastic demand.

In this vein, the Motor Trades Association of Australia Superannuation Fund

(sub. 22), while arguing against the need for price regulation of airports, suggested
that if price caps were imposed, a single till would promote greater pricing
efficiency compared with a dual till. BARA made a similar argument in the context
of its option for rate-of-return regulation across all airport services (sub. DR54).

This argument appears to rest on two assumptions — namely, that there are
significant common costs in supplying aeronautical and non-aeronautical services,
and that demand for non-aeronautical services is less elastic than demand for
aeronautical services. If these assumptions hold, common costs could be covered
more efficiently by increasing the prices of non-aeronautical services relative to
those of aeronautical services. However, the extent of common costs in the

376 PRICE REGULATION

OF AIRPORT
SERVICES
provision of aeronautical and non-aeronautical services appears limited. There may
be some shared terminal costs but even in this instance, costs are likely to be
separable. In addition, for those non-aeronautical services sold in competitive
markets (chapter 6), prices cannot be raised above market prices (which, in
competitive markets, will reflect the marginal cost of supply).

The ACCC compared opposing arguments of Kahn, and Crew and Kleindorfer, on
this point:
Professor Kahn argues that the apparently high returns accruing to airport operators
from retail services is related to the locational advantage of the airport. Implicit in this
argument is the point that SACL’s monopoly market power does not extend beyond
those services defined as aeronautical. By contrast, Professors Crew and Kleindorfer
explicitly state that SACL’s proposal ‘develops a multi-till approach to price
regulation, but this proposed approach leads to a number of problems, not least of
which is that it allows non-aeronautical services to be priced at monopoly levels’.
For the reasons outlined above, the Commission [ACCC] agrees with Professor Kahn
that the dual till is in general a superior approach to aeronautical pricing, provided the
services defined as aeronautical include all those in which the airport operator has
significant market power. (sub. 36, p. 74)

Using infra-marginal locational rents to promote marginal-cost pricing of

aeronautical services

It is possible that, in principle, infra-marginal locational rents earned in non-

aeronautical activities could be extracted without affecting the efficient provision of
those services. This raises the question of whether appropriation of locational rents
earned on airport land to push down prices for aeronautical activities is likely to
improve economic efficiency. It also raises the question of whether single-till
regulation, in practice, is likely to effect such a transfer in a non-distorting way.

Several issues are relevant.

• If provision of aeronautical services involves economies of scale such that
average costs exceed marginal costs, the break-even price will exceed the first-
best level for efficient use of that airport (where price equals marginal cost). An
appropriate subsidy could achieve efficient pricing but the deadweight costs of
raising a tax to pay for the subsidy (essentially the disincentive effects of
taxation) must be weighed against the efficiency gain in aeronautical services.
• In theory, economic or locational rent can be ‘taxed’ without distorting
behaviour (a non-distorting, ‘lump-sum’ tax). In practice, single-till
arrangements typically are applied in a way that distorts incentives to earn
locational rents. If airport operators know ex ante that they will not retain
additional locational rents earned in non-aeronautical activities (eg because rents
AERONAUTICAL AND 377
NON-AERONAUTICAL
SERVICES
 earned above an allowed rate of return are transferred to reduce aeronautical
prices), they are unlikely to expend a great deal of effort allocating available
airport land to its highest-valued non-aeronautical use. They may charge low
rentals to existing concessionaires. These concessionaires would then capture the
locational rents. Low rentals offered by the airport also could encourage rent-
seeking by prospective tenants which would dissipate at least some of the rents.
Alternatively, airport operators may allow their reported costs of supplying non-
aeronautical services to rise to disguise profits.
• It could be feasible to devise a system whereby rents were shared with
aeronautical users and which did not distort the airport operator’s incentive to
earn additional rents. However, such an arrangement would need to be agreed
upon and not altered unilaterally by the regulator if, for example, rents increased.
Under the UK single-till model, airport operators have some incentives to earn
additional locational rents within the five-year price-cap period, but they also
know that any additional profits in the current regulatory period may affect
revenue forecasts in the setting of price caps in future periods. Importantly, as
discussed below, buyers of airport leases at core-regulated airports in Australia
paid up-front for ‘commercial opportunities’. In one sense they have paid a
once-off tax reflecting anticipated locational rents. Alternatively, prospective
buyers could have made bids on the basis of paying annual lease payments or
royalties to the Commonwealth.
• A single till appropriates rents earned from non-aeronautical activities located on
airport land. This could create an incentive for airport operators effectively to
‘shrink’ the airport boundaries, capitalising non-aeronautical operations, locating
commercial activities off-airport or engaging in activities that yielded locational
rents but that were only indirectly related to airport activities (eg business parks).
As noted above, to the extent that any locational rents are forgone (that is, not
earned), this would represent an efficiency loss. A similar distortion may have
been created if the draft ACCC (2001h) proposal to offset roughly $20 million of
(mainly) annual car-parking revenue at Sydney Airport against aeronautical
charges had been implemented. In this instance, the locational rent accruing to
the airport from one activity (car parking) would have fallen, possibly creating
an incentive for the airport to develop other activities on land that otherwise may
have been used, more efficiently, for car parking.6

6 This assumes that the marginal locational rent flowing from car parking is not inflated by any
market power in that activity. That is, the rent received does not exceed the marginal rent
forgone from not using the land in the next-best, alternative (competitive) use.
378 PRICE REGULATION
OF AIRPORT
SERVICES
• Some rents may reflect an airport owner/operator’s entrepreneurial rather than
locational advantage. Appropriation of rents from this source could encourage
these operators to move to activities where rents were not taxed as heavily.7
• Reducing aeronautical charges can promote efficiency only if stand-alone
aeronautical prices exceed their marginal cost. As discussed above, it is possible
that an airport, internalising complementary demands for non-aeronautical
services, voluntarily sets aeronautical prices for marginal customers close (or at
least closer) to marginal cost. Airports may also charge in such a way that
marginal consumers are not deterred from using the facility (via some form of
discriminatory pricing). In both cases, the efficiency argument for single-till
pricing is weakened.
• Alternatively, marginal costs may exceed average costs, for example, as a result
of capacity constraints or decreasing returns to scale at airports. If an airport is
already experiencing excess demand (that is, charges are below market-clearing
levels), then as noted by Kahn (2001a), lower aeronautical prices largely will
redistribute profits from airports to airlines — fares will not fall (for there will be
no incentive for airlines to reduce fares below market-clearing levels), so airline
profits will rise.8 With prices below market-clearing levels, limited capacity will
have to be rationed by a non-price mechanism which, almost inevitably, will
incur efficiency costs.9
• In effect, single-till regulation can impose incentives similar to those of cost-
based regulation on non-aeronautical activities. One possible result — as for
most cost-based regulation — is that airports will have an incentive to inflate the
costs of non-aeronautical activities to reduce measured profits, with consequent
losses of productive efficiency. It also could lead to a perverse situation where
aeronautical activities effectively subsidised inefficient expansion of non-
aeronautical activities (eg if airports undertook inefficient investment in retail
facilities to increase costs and reduce non-aeronautical returns on assets).
(Starkie and Yarrow 2000)

7 In the economy as a whole, rents are taxed via capital gains and income taxes. The single till, in
the limit, imposes a 100 per cent tax on rents.
8 The ACCC (sub. 36) appeared to reject this argument, at least in relation to Sydney Airport, on
the basis that there is scope for inter-airline competition. While this potential exists, at peak
times when slots are fully allocated, airlines are unlikely to initiate a discount war for peak
traffic. This is unlikely because the scope for a more efficient airline to increase market share is
limited by the slot allocation scheme — a more efficient airline, officially, cannot buy extra
slots. In other words, airlines are unlikely to compete on price among each other for additional
customers who, without an additional slot, they cannot carry (appendix H).
9 Even in the (unlikely) event that a non-price rationing scheme allocates capacity to those who
value it most highly, administration of the scheme will not be costless and wasteful rent-
seeking is likely to be encouraged (appendix H).
AERONAUTICAL AND 379
NON-AERONAUTICAL
SERVICES
• In the case of privatised core-regulated Australian airports, an additional
complication would arise from introduction of a single till. These airports appear
to have paid for the rights to exploit commercial opportunities on airport land (as
the result of a competitive bidding process) and the land-owner (the
Commonwealth Government) largely has captured expected (infra-marginal)
locational rents already.10 In this sense, some or all of the locational rents earned
by airports are not pure profits. If these rents were redistributed to airlines and/or
their customers, then airports would be likely to suffer large capital losses.
Changing the rules could also raise concerns about sovereign risk for future
investors in airports and possibly other regulated activities.

In sum, where the airport has little or no market power in the provision of non-
aeronautical services, there may be a theoretical case to appropriate airport
locational rents to push the price of aeronautical activities closer to marginal cost.
Gains are possible only when the marginal cost of providing aeronautical services
lies below their average cost and the airport operator cannot or does not offer
inducements to marginal airport users.

However, in practice, it is highly unlikely that the single-till ‘taxation’ of airport

locational rents will be non-distorting. The efficiency losses involved could be
much larger than any efficiency gains from lower aeronautical prices (which will
depend on the elasticity of demand for aeronautical services). This risk arises
because of the incentives created for the airport not to earn locational rents (and
therefore not to allocate airport land appropriately), to dissipate such rents in higher
costs, or to encourage dissipation of rents by prospective tenants competing for
them.

Issues with dual-till regulation

A dual till, which regulates only those aeronautical services in which an airport is
most likely to have market power, is not without problems, particularly in its
implementation. A dual till requires the regulatory basket to be clearly and
comprehensively defined (see chapters 8 and 9 for a discussion of some issues
arising from the definition of current price caps). Further, if prices under a price cap
reflect airport costs, then any common costs must be allocated between regulated
and unregulated activities. Under a dual till, an airport also has a strong incentive to

10 This raises the question of whether it may be more appropriate for the Commonwealth
Government to use airport lease sale proceeds to subsidise aeronautical activities, if such
subsidisation were considered efficient. Indeed, it could be argued that the Government did this
to some extent by selling core-regulated airports subject to five-year price caps that largely
maintained single-till aeronautical prices.
380 PRICE REGULATION
OF AIRPORT
SERVICES
restrict competition from off-airport providers (eg by restricting access to the airport
or by extending the airport’s boundaries by regulation).11

A dual till also has the potential to discourage investment in aeronautical services in
favour of non-aeronautical services if allowable returns in the former are
constrained relative to returns in the latter. At the same time, however, an airport
operator may have an incentive to expand aeronautical capacity to increase
passenger throughput and, consequently, rents from commercial activities. This may
offset any tendency for under-investment in aeronautical facilities.

11 The airport also could buy up surrounding land, but presumably it would have to compete with
other buyers and pay a competitive price.
AERONAUTICAL AND 381
NON-AERONAUTICAL
SERVICES
D Characteristics of demand and
competition at core-regulated airports

This appendix summarises the main demand and competition characteristics of the
core-regulated airports. These are used in the analysis of the extent of market power
held by particular airports in providing domestic passenger services (chapter 5).

In the tables that follow, the categories ‘annual passenger movements’ and
‘proportion of passenger movements that is international’ are for 2000-01 and are
taken from the Department of Transport and Regional Services (unpublished data).
The ‘proportion of revenue that is non-aeronautical’ is taken from ACCC (2001b, c,
d, f, m), for 2000-01 for Sydney Airport and 1999-00 for all other airports. As noted
in chapter 2, non-aeronautical revenue covers a range of services, not all of which
are provided at all core-regulated airports (eg business parks). This may contribute
to some of the variation in the proportion of non-aeronautical revenue across
airports. ‘Main market segments’ generally refers to interstate overnight visitors to
the city, region or State (not the airport) (where it refers to domestic — interstate
and intrastate — visitors, this is stated). Unless otherwise stated, these data, as well
as data on modal substitution possibilities, are from BTR (2000b) for 1999.

Table D.1 Demand and competition characteristics of Adelaide Airport

Annual passenger movements : 4 464 998
Proportion of passenger movements that : 6%
is international
Proportion of revenue that is : 80%
non-aeronautical
Main market segments (percentage of : For South Australia as a whole, business (36%) and
interstate overnight visitors in 2000) visiting friends and relatives (VFR — 25%)
(SATC 2001).
Destination substitution possibilities : Relatively low, given the dominance of business and
VFR travellers.
Modal substitution possibilities : Low for business travellers. For VFR and holiday
(SATC 2001) travellers, some modal substitutes appear viable,
particularly for visitors from Victoria and parts of New
South Wales (these States accounted for 75% of
visitors to South Australia in 2000). More than half of
all interstate overnight arrivals to South Australia
arrived by private vehicle in 2000. Train services also
are available.
Airport substitution possibilities : Low. There are no proximate regular public transport
(RPT) airports.

CHARACTERISTICS 383
OF CORE-REGULATED
AIRPORTS
Table D.2 Demand and competition characteristics of Alice Springs
Airport
Annual passenger movements : 717 722
Proportion of passenger movements : 0%
that is international
Proportion of revenue that is : 61%
non-aeronautical
Main market segments (percentage of : Holiday/leisure (55% to the Northern Territory as a
interstate overnight visitors in 1999) whole). Of all travellers (including interstate and
international) to the ‘Centre’ region, 77% went for
holiday/leisure purposes in 1999-00 (NTTC 2000).
Destination substitution possibilities : High, because Alice Springs competes with other
tourist destinations.
Modal substitution possibilities : Moderate. More than half the interstate visitors to
the Northern Territory arrive by air, but a significant
proportion (over 40% in 1999-00 (NTTC 2000))
arrives by other modes.
Airport substitution possibilities : High. Yulara (Ayers Rock) Airport provides a viable
alternative to Alice Springs, especially for tourists
who travel to destinations such as Uluru.

Table D.3 Demand and competition characteristics of Brisbane Airport

Annual passenger movements : 12 467 537
Proportion of passenger movements : 20%
that is international
Proportion of revenue that is : 72%
non-aeronautical
Main market segments (percentage of : Business (32%) and VFR (32%)
domestic overnight visitors in 1999) (Tourism Queensland 2000a).
Destination substitution possibilities : Relatively low, given the dominance of business and
VFR travellers.
Modal substitution possibilities : Low for business traffic. However, for VFR and
holiday travellers, modal substitutes — particularly
private vehicle — may be viable (about 45% of
interstate visitors to Queensland arrived by modes
other than air in 1999).
Airport substitution possibilities : Low. Although there are relatively proximate airports
in Coolangatta and Maroochydore, they do not
appear to provide significant competition due to the
scale of Brisbane, the extent of its business traffic,
and its much stronger ability to service international
traffic.

384 PRICES REGULATION

OF AIRPORT
SERVICES
Table D.4 Demand and competition characteristics of Canberra Airport
Annual passenger movements : 2 107 219
Proportion of passenger movements : 0%
that is international
Proportion of revenue that is : 59%
non-aeronautical
Main market segments (percentage of : Business (30%) and VFR (34%).
interstate overnight visitors in 1999)
Destination substitution possibilities : Relatively low, given the dominance of business and
VFR travellers.
Modal substitution possibilities : High. Despite the high proportion of business and
VFR visitors, a significant proportion of visitors (71%
in 1999) arrives in Canberra by car. This may reflect
the fact that total travel times by air and car are
similar on the Sydney–Canberra route.
Airport substitution possibilities : Low. There are no proximate RPT airports.

Table D.5 Demand and competition characteristics of Coolangatta Airport

Annual passenger movements : 1 888 008
Proportion of passenger movements : 2%
that is international
Proportion of revenue that is : 45%
non-aeronautical
Main market segments (percentage of : Holiday/leisure (62%) (Tourism Queensland 2000b).
domestic overnight visitors in 1999)
Destination substitution possibilities : High. The Gold Coast competes with other tourist
destinations.
Modal substitution possibilities : Moderate. Many domestic visitors to the Gold Coast
arrive in private vehicles (69% in 1999 (Tourism
Queensland 2000b) although this figure includes
intrastate visitors so it would tend to overstate the
proportion of interstate visitors that drives).
Airport substitution possibilities : High. Brisbane Airport appears to be a viable
alternative to Coolangatta for visitors to the Gold
Coast.

CHARACTERISTICS 385
OF CORE-REGULATED
AIRPORTS
Table D.6 Demand and competition characteristics of Darwin Airport
Annual passenger movements : 1 078 481
Proportion of passenger movements : 16%
that is international
Proportion of revenue that is : 54%
non-aeronautical
Main market segments (percentage of : Holiday/leisure (55% of interstate overnight visitors
interstate overnight visitors in 1999) to the Northern Territory as a whole; and 55% of all
visitors (including interstate and international) to the
‘Top End’ region of the Northern Territory in 1999-00
(NTTC 2000)).
Destination substitution possibilities : High. Although the proportion of business travellers
to the ‘Top End’ is higher than for the rest of the
Territory, most travellers go for a holiday. Therefore,
Darwin and surrounding areas compete with other
tourist destinations, including other areas in the
Territory.
Modal substitution possibilities : Low for business travellers and other travellers
visiting Darwin only, given the relative isolation of
Darwin. For holiday travellers who visit several
regions in the Territory, the potential for modal
substitution appears to be more significant (over
40% of interstate visitors to the Territory arrived by
modes other than air in 1999-00 (NTTC 2000)).
Airport substitution possibilities : Vary by market segment. Low for those visiting only
the ‘Top End’. May be higher for those visiting
several areas within the Territory, but still likely to be
low overall.

Table D.7 Demand and competition characteristics of Hobart Airport

Annual passenger movements : 973 922
Proportion of passenger movements : 0%
that is international
Proportion of revenue that is : 67%
non-aeronautical
Main market segments (percentage of : Holiday/leisure (56% to Tasmania as a whole
interstate overnight visitors in 2000) (Tourism Tasmania 2001)).
Destination substitution possibilities : High. Hobart (and Tasmania generally) compete
with other tourist destinations.
Modal substitution possibilities : Moderate. Sea, used by 22% of arrivals to
Tasmania in 2000 (Tourism Tasmania 2001), is a
viable alternative for holiday (and VFR) visitors.
Airport substitution possibilities : High. Launceston Airport in particular appears to
provide (potential and actual) competition. The high
proportion of tourists that visits more than one area
in the State increases the potential competition
between airports.

386 PRICES REGULATION

OF AIRPORT
SERVICES
Table D.8 Demand and competition characteristics of Launceston Airport
Annual passenger movements : 522 100
Proportion of passenger movements : 0%
that is international
Proportion of revenue that is : 66%
non-aeronautical
Main market segments (percentage of : Holiday/leisure (56% to Tasmania as a whole
interstate overnight visitors in 2000) (Tourism Tasmania 2001)).
Destination substitution possibilities : High. Launceston (and Tasmania) compete with
other tourist destinations.
Modal substitution possibilities : Moderate. Sea (used by 22% of arrivals to
Tasmania in 2000 (Tourism Tasmania 2001)), is a
viable alternative for holiday (and VFR) visitors.
Airport substitution possibilities : High. There are a number of airports in relatively
close proximity to Launceston. Hobart Airport, in
particular, appears to provide (potential and actual)
competition. The high proportion of tourists that
visits more than one area in the State increases the
potential competition between airports.

Table D.9 Demand and competition characteristics of Melbourne Airport

Annual passenger movements : 16 908 656
Proportion of passenger movements : 19%
that is international
Proportion of revenue that is : 67%
non-aeronautical
Main market segments (percentage of : For Victoria as a whole, business (34%) and VFR
interstate overnight visitors in 1999) (30%).
Destination substitution possibilities : Relatively low, given the dominance of business and
VFR travellers.
Modal substitution possibilities : Low for business traffic. However, for VFR and
holiday travellers, modal substitutes — particularly
private vehicle — appear to be particularly viable for
visitors from some areas, including South Australia
and New South Wales. About 49% of interstate
overnight travellers arrived in Victoria by air in 1999.
Airport substitution possibilities : Low. There are no proximate RPT airports. Although
Essendon Airport, for example, is physically capable
of handling B737 traffic, there are administrative
problems in doing so.

CHARACTERISTICS 387
OF CORE-REGULATED
AIRPORTS
Table D.10 Demand and competition characteristics of Perth Airport
Annual passenger movements : 5 162 980
Proportion of passenger movements : 31%
that is international
Proportion of revenue that is : 71%
non-aeronautical
Main market segments (percentage of : For Western Australia as a whole, business (40%)
interstate overnight visitors in 1999) and VFR (26%).
Destination substitution possibilities : Relatively low, given the dominance of business and
VFR travellers.
Modal substitution possibilities : Low, given the isolation of Perth. Of interstate
overnight arrivals to Western Australia in 1999, 82%
arrived by plane.
Airport substitution possibilities : Low. There are no proximate RPT airports.

Table D.11 Demand and competition characteristics of Sydney Airport

Annual passenger movements : 25 779 242
Proportion of passenger movements : 33%
that is international
Proportion of revenue that is : 60%
non-aeronautical
Main market segments (percentage of : Business (33%) and VFR (34%) (Tourism New
domestic overnight visitors in 1999-00) South Wales 2001).
Destination substitution possibilities : Relatively low, given the dominance of business and
VFR travellers.
Modal substitution possibilities : Relatively low for business traffic (except for the
Sydney–Canberra route). However, for VFR and
holiday travellers, modal substitutes — particularly
private vehicle — appear to be particularly viable for
visitors from some areas, including the ACT,
Queensland and Victoria.
Airport substitution possibilities : Low. There are no proximate RPT airports.

388 PRICES REGULATION

OF AIRPORT
SERVICES
Table D.12 Demand and competition characteristics of Townsville Airport
Annual passenger movements : 731 960
Proportion of passenger movements : 0%
that is international
Proportion of revenue that is : 62%
non-aeronautical
Main market segments (percentage of : Holiday/leisure (53% of visitors to Tropical North
domestic overnight visitors in 1999) Queensland (Tourism Queensland 2000d)).
Destination substitution possibilities : High. Townsville competes with other tourist
destinations, both in Tropical North Queensland and
elsewhere.
Modal substitution possibilities : Relatively low for business traffic. For VFR and
holiday travellers, the potential to use other modes
exists. However, this option does not appear to be
as attractive as for other areas in Queensland (43%
of domestic visitors to Tropical North Queensland
arrived by air in 1999 (Tourism Queensland 2000d)).
Airport substitution possibilities : Low. There are no proximate RPT airports.

CHARACTERISTICS 389
OF CORE-REGULATED
AIRPORTS
E Fuel throughput levies and taxi
charges

The introduction of fuel throughput levies and taxi charges by some airport
operators has generated considerable debate. This appendix describes the
background to these charges, discusses their treatment under price regulation
following the sale of leases, and highlights issues for future regulation.

E.1 Fuel throughput levy

Aircraft refuelling services provided by airport operators at core-regulated airports
were subject to price monitoring by the ACCC under the Prices Surveillance
Act 1983 (PS Act) following the sale of leases (chapter 3). In October 2001,
changes to prices-oversight arrangements at core-regulated airports included the
removal of all price regulation at Coolangatta, Alice Springs, Hobart, Launceston
and Townsville airports. This meant that refuelling services at these airports were
no longer subject to price monitoring. However, monitoring of refuelling services
continues to apply at other core-regulated airports.

At no time have aircraft refuelling services been subject to price-cap arrangements,

and neither were they subject to prices surveillance (notification) prior to the sale of
leases.

Under price monitoring arrangements implemented following the sale of leases,

Brisbane Airport Corporation (BAC) and Westralia Airports Corporation (WAC)
introduced a new charge on oil companies for aircraft refuelling services — a fuel
throughput levy. Objections to the imposition of this levy from oil companies and
airlines led to a review by the ACCC, but a decision on the issue has yet to be
announced by the Commonwealth Government.

Background

Refuelling facilities at Brisbane and Perth (international) airports are known as joint
user hydrant installations (JUHIs). They consist of a fuel storage facility, a hydrant

FUEL THROUGHPUT 391

LEVIES AND TAXI
CHARGES
refuelling site and connecting pipelines. These were established, and are owned, by
joint ventures comprising several oil companies.

Airport operators charge the oil companies for use of the land by way of licence
fees and rents payable pursuant to the leases under which the oil companies occupy
the refuelling sites. This charge for aircraft refuelling services, when imposed by the
Federal Airports Corporation (FAC), was not subject to prices surveillance. Since
the sale of leases for Phase 1 and 2 airports, and the lease of Sydney Airport to
Sydney Airports Corporation Limited, this charge has been subject to price
monitoring by the ACCC.

According to BP (sub. to ACCC 1998b), in 1995 the FAC mooted the concept of a
fuel throughput levy during negotiations with the oil companies to establish a
standard licence for JUHIs. The oil companies strongly resisted the inclusion of a
fuel throughput levy in a standard licence. Negotiations stalled until 1997 when the
oil companies signed the licences (to continue for 15 years), with the inclusion of a
provision for a fuel throughput levy. The FAC, however, did not activate the
provision, although it had the power to do so. Oil companies continued to pay, on a
dollar per square metre basis, for lease of the land.

On privatisation, airport operators succeeded to the FAC’s rights and obligations

under the licence (Airports (Transitional) Act 1996), acquiring the right to impose
the fuel throughput levy.

In 1997, BAC informed Shell of its intention to impose a new charge for aircraft
refuelling services — a fuel throughput levy of 0.4 cents per litre on all fuels
supplied, distributed or transferred through the combined main/short pipeline at
Brisbane Airport. Shell objected and the dispute was referred to an ‘independent
expert’, in accordance with licence provisions, who subsequently confirmed that
‘BAC had a legal right to charge the levy and the level of 0.4 cents per litre was
reasonable in terms of BAC’s contractual rights’ (ACCC 1998b, pp. 16–17). In
July 1998, BAC introduced the levy.

WAC introduced a fuel throughput levy of 0.5 cents per litre on refuelling
operations at the Perth Airport international terminal from June 1999.

The revenue raised by the BAC and WAC fuel throughput levies, in 1999-00, was
$2.5 million and $727 000 respectively (ACCC 2001b, d). The revenue is in
addition to revenue obtained from the existing lease charges.

Many airports in other countries charge for refuelling services on a volume of fuel
throughput basis, including Amsterdam, Kuala Lumpur, Los Angeles, Manchester
and Aberdeen, and Auckland and Wellington airports in New Zealand. Wellington

392 PRICE REGULATION

OF AIRPORT
SERVICES
Airport moved in a different direction from BAC and WAC, re-negotiating with oil
companies to introduce a fixed rent charge and to reduce the existing fuel
throughput charge. In Australia, since the late 1980s, Cairns Airport (not subject to
price regulation), has charged for refuelling services in two parts: a lease charge,
and a licence agreement based on a sliding scale volume of fuel uplifted.

The ACCC review

In undertaking its monitoring role, and in response to concerns by oil companies

and airlines, the ACCC (1998b) undertook a public review of fuel throughput
levies. In undertaking its review, the ACCC paid particular regard to s. 17(3)(b) of
the PS Act:
the need to discourage a person who is in a position substantially to influence a market
for goods or services from taking advantage of that power in setting prices.

The ACCC also took account of the Treasurer’s media release on price monitoring,
which stated:
Price monitoring will allow the ACCC to collect data where the airport operator may
have scope to exercise market power but where coverage of the services under the more
formal price cap arrangements is not considered warranted. Any abuses of market
power detected through the prices monitoring arrangements will be the trigger for
consideration of stricter forms of prices oversight. (Costello 1998)

The ACCC summarised its approach as follows:

For the purposes of this report ‘market power’ refers to circumstances in which a
person is in a position to substantially influence a market for goods and services. An
‘abuse of market power’ occurs where the person takes advantage of that power in
setting prices. (ACCC 1998b, p. 14)

Based on this approach, the ACCC (1998b) addressed four issues:

• the extent of increases in the price of refuelling services and airport profits as a
result of increases in refuelling charges;
• whether fuel throughput levies could be justified through increases in costs
and/or offsetting reductions in other charges;
• whether the fuel throughput levy is an abuse of market power; and
• whether the fuel throughput levy should trigger a stricter form of prices
oversight, and the form that should take.

On consideration of information provided by BAC and WAC, and submissions

from interested parties, the ACCC formulated its conclusions and recommended
that refuelling services, and therefore both lease payments and fuel throughput

FUEL THROUGHPUT 393

LEVIES AND TAXI
CHARGES
levies, be included under the price-cap arrangements (box E.1). This means that if
revenue were to be pressing up against the CPI-X limit, revenue from aircraft
refuelling would need to be balanced by reductions in other revenue.

Box E.1 ACCC report on fuel throughput levies

The ACCC (1998b) report on the introduction of fuel throughput levies by Brisbane
Airport Corporation (BAC) and Westralia Airports Corporation (WAC) concluded and
recommended the following:
Introduction of fuel throughput levies will significantly increase the price of refuelling services
at airports where they are introduced. (p. 5)
The validity of the contractual arrangements is separate from the issue of the relationship of
fuel throughput levies to costs or reductions in charges elsewhere, and as such is not
directly relevant to the Commission’s monitoring role … (p. 30)
The introduction of a fuel throughput levy at Brisbane and Perth airports is likely to result in
some or all of that levy being passed on to the airlines refuelling at those ports. (p. 5)
A logical outcome … is higher charges to passengers on at least some airline services to
that destination. (p. 25)
… the fuel throughput levies introduced by BAC and proposed by WAC are not justified in
terms of increases in costs or through offsetting reductions in other charges. (p. 6)
There is a strong case that large airports have market power in the market for refuelling
services. (p. 7)
There is a strong case that by introducing fuel throughput levies airport operators have taken
advantage of market power that they have in the provision of aircraft refuelling services.
(p. 7)
The Commission recommends that stricter forms of prices oversight should be considered in
relation to aircraft refuelling services. (p. 8)
The Commission recommends that refuelling services are included within a CPI-X price cap.
(p. 11)

Source: ACCC (1998b).

The ACCC reported to the Treasurer and publicly released its report in
December 1998. It reconfirmed its approach this year, suggesting that refuelling
services be subject to a price cap (sub. 36). Moreover, in its draft decision on
aeronautical pricing at Sydney Airport, the ACCC (2001h) decided that ‘above-
normal’ returns from aeronautical-related services, such as fuel throughput levies
and car parking, should be taken into account in pricing aeronautical services.
Specifically, these returns were to be subtracted from aeronautical revenue to
generate a total allowable revenue from aeronautical services for the airport
(ACCC 2001h). Following a Direction (No. 22) in April 2001 from the Minister for
Financial Services and Regulation, the ACCC (2001i) moved away from this
position in the final decision.

394 PRICE REGULATION

OF AIRPORT
SERVICES
To date, the Commonwealth Government has not announced a decision on this
issue. Both BAC and WAC continue to impose a fuel throughput levy, which
remains outside the price cap.

Divergent views

Submissions to the ACCC and, more recently, to this inquiry, have highlighted the
contentious nature of the fuel throughput levy. Airport operators contended, in
essence, that they should be allowed to introduce a fuel throughput levy, and that it
should not be included under price-cap arrangements.

On the other hand, oil companies (including Shell, sub. DR63) argued that the fuel
throughput levy should not have been introduced by BAC and WAC, that its
imposition represents an abuse of market power and cannot be justified, and
therefore that the levy should be subject to stricter prices oversight. The airlines and
the International Air Transport Association (sub. DR56) in general supported this
view.

Airport investors (Motor Trades Association of Australia Superannuation Fund,

Hastings Funds Management, Australian Council for Infrastructure Development;
subs 22, 19, 28) did not make submissions to the ACCC review of fuel throughput
levies. However, they expressed their disagreement with the ACCC’s conclusions
and recommendations to this inquiry.

E.2 Taxi charges

Price-cap arrangements apply to the provision of aeronautical services, being
limited to aircraft movement facilities and activities, and passenger processing
facilities and activities. The latter includes ‘landside roads, landside lighting and
covered walkways’ (chapter 3).

Several airport operators have introduced airport vehicle access charges for taxis. In
1998, BAC introduced a fee for taxis of $1 per passenger pick-up at Brisbane
Airport. WAC also introduced a fee for taxis in 1998 — $1 per pick-up at Perth
Airport (unbooked) and $2 per pick-up (booked). The following year Capital
Airport Group introduced a taxi charge of $2 at Canberra Airport.1 Taxi charges
were recently introduced at Melbourne Airport — a 66 cent charge per taxi for

1 Northern Territory Airports introduced taxi charges at Darwin and Alice Springs airports in
1998, but removed them the following year.
FUEL THROUGHPUT 395
LEVIES AND TAXI
CHARGES
passenger pick-up (passengers are charged $1, 66 cents of which is revenue to the
airport operator, and 34 cents is a handling commission to the taxi).2

The introduction of these charges has been particularly contentious. The airport
operators argued that taxi charges did not relate to the use of ‘landside roads’ and
were not within the definition of ‘aeronautical services’, and therefore that revenue
derived from taxi charges had not (and should not be) included within the price cap.
Hastings Funds Management (part-owner of several airports) noted:
We have consistently maintained that charges for ground transport are in fact not
charges to access the land-side roads (as would be the case, for example if a toll was
placed on the road), but a concession fee for the right to conduct business by taxi and
bus operators. (sub. 19, p. 14)

The ACCC, on the other hand, considered that revenue derived from taxi charges
was covered under the definition of ‘aeronautical services’ (being included in
‘landside roads’) and therefore was included in the price cap (ACCC 1999e, f;
2000d). The ACCC treated these as ‘new’ charges under Direction No. 20, pursuant
to the PS Act, which stated that ‘charges on new or varied services are to be
factored into the price-cap arrangements if the services are declared services’.

The ACCC therefore requested that BAC, WAC and Capital Airport Group
formally ‘notify’ the ACCC of their proposals to introduce a taxi charge, which they
did, albeit on a ‘without prejudice’ basis. This inclusion has contributed to BAC and
WAC not complying with the price cap; revenue has been ‘over-recovered’ and will
have to be passed back to users within a specified time period.

Canberra Airport instigated an action in the Federal Court against the ACCC in an
attempt to have the ACCC’s decision on taxi charges overturned. In March 2001,
the Federal Court ruled, in a ‘line-ball’ decision, in favour of the ACCC — finding
that the charge was within the price cap because it related to the use of landside
roads. The Court stated:
… it is submitted for the applicant [Canberra Airport] that, on the facts, the area in
question is properly described as a car park or marshalling yard and is simply not an
access way for through traffic.
The primary submission on behalf of the respondent [ACCC], in summary, is that the
area is physically an extension of the main road …
The issue is finely balanced, with each argument capable of acceptance. My mind has
fluctuated as to the correct result. In the end I am persuaded that the area is best seen as
part of the overall road system of the Airport in a way which, for example, car parks are
not. (Federal Court of Australia 2001a, paras 21–3)

2 A $2 per vehicle pre-booked taxi fee applies to passengers departing in a taxi from the
Premium Parking area.
396 PRICE REGULATION
OF AIRPORT
SERVICES
The Full Federal Court has since dismissed an appeal by Canberra Airport,
concluding that:
In our view, the area is, in its physical aspect, sufficiently connected to the main
thoroughfare, and, in its functional aspect, properly seen as incidental to the purpose
served by that thoroughfare, to warrant itself the description of ‘road’. (Federal Court
of Australia 2001b, para. 16)

Since then, the changes to price regulation in October 2001 (chapter 3) removed
aeronautical services at Canberra Airport from the price cap. Both aeronautical and
aeronautical-related services are now subject to the same price-monitoring
arrangements. Thus, for the time being, this taxi charge issue is not relevant to
Canberra Airport in the context of price regulation.

Melbourne Airport is the only airport operator to seek ACCC consideration of the
taxi charge under necessary new investment (NNI) provisions of the price-cap
arrangements (chapters 3 and 8). However, in so doing, Melbourne Airport (2001)
noted that it considered that the charge should be subject to price monitoring, not
declaration under the price cap, and that because of different circumstances, the
Federal Court decision regarding Canberra Airport would shed little light on the
Melbourne Airport situation.

Melbourne Airport sought approval under NNI provisions to introduce a $1.40 taxi
charge to recover costs associated with development of a taxi facility, including
unrecovered road costs, taxi car-parking development costs and taxi rank labour
costs. The Victorian Taxi Association, representing the taxi industry, disagreed with
the imposition of the charge, commenting that taxi passengers would not receive
any service improvement, that the charge was an unfair tax, and that the timing of
its introduction was poor. The Association also noted that it was concerned about
the imposition of taxi charges in principle — that taxi operators may not be allowed
to pass on charges to passengers without delay, and that charges may affect the
demand for taxi services (sub. to ACCC 2001a). The Transport Workers Union,
representing a smaller but increasing number of taxi drivers, noted that taxi drivers
(content with existing facilities) saw no need to introduce new taxi facilities and
therefore objected to the charge (sub. to ACCC 2001a).

The ACCC considered the proposed taxi charge according to criteria for NNI set out
in the Department of Transport and Regional Development Pricing Policy Paper
(DoTRD 1996) and replicated in Direction No. 20 pursuant to the PS Act
(chapter 3). It found that some of the costs met the criteria and therefore could be
‘passed through’ the price cap — specifically, that 66 cents (GST inclusive) of the
proposed $1.40 charge would not be subject to the price cap (ACCC 2001a).
Following this decision, Melbourne Airport introduced a 66 cent taxi charge, as
noted above.
FUEL THROUGHPUT 397
LEVIES AND TAXI
CHARGES
E.3 Issues arising from price regulation
The debates regarding fuel throughput levies and taxi charges highlight a problem
intrinsic to a dual-till system — that given there must be two baskets of services
(those subject to a price cap, and those not subject) there is likely to be contention,
at least for some services, regarding:
• which services are within the price cap and which are not (illustrated by the taxi
charges debate); and
• which services should be within the price cap and which should not (illustrated
by the fuel throughput levy and taxi charges debates). The fuel throughput levy
debate highlights, at least for some airport operators and some services, that
whether the potential for abuse of market power exists (and is abused) can be a
contentious issue.

Similarly, there may be disagreement regarding whether services not within the
price cap are, or should be, subject to price monitoring or to no price regulation.

The taxi charges debate highlights the importance of clarity and precision in
legislation defining the basket of services to be subject to the price cap. The scope
for different interpretations of particular legislative clauses cannot be eliminated,
but it can be minimised in the drafting process. Moreover, shortcomings in the
legislation can be addressed by appropriate amendment as they arise.

The lack of clarity about what services relate to ‘landside roads’ has created
uncertainty among airport operators and users. As with fuel throughput levies, the
uncertainty has been exacerbated by the length of time that these issues remain
unresolved and the apparent difference between the ACCC interpretation and the
Commonwealth Government’s intent on these issues.

In relation to fuel throughput levies, airport operators claim to have assessed the
risks and made commercial decisions (bidding for airport leases) based, in part, on
their perception of Commonwealth Government policy at the time of sale of leases
— that refuelling services were outside the price cap.

Several airport operators commented on what they perceived to be Commonwealth

Government policy prior to, and after the sale of leases. Melbourne Airport, BAC,
WAC and Australian Airports (Townsville) put the view that they had a contractual
right to impose a fuel throughput levy because of a pre-existing FAC (a
Commonwealth Government agency) contract with the oil companies (subs 7, 8, 21;
trans., p. 16).

398 PRICE REGULATION

OF AIRPORT
SERVICES
BAC noted other Commonwealth Government representations:
… the Commonwealth Government’s Office of Asset Sales confirmed to airport
bidders during the privatisation process they would be contractually entitled to
introduce a fuel throughput fee … (sub. 8, p. 38)

Airport operators were of the view that when they exercised this contractual right,
charges (including fuel throughput levies) for aircraft refuelling services would not
be subject to a price cap. As noted above, when the FAC operated airports, charges
for the provision of refuelling services were not subject to price regulation.
Moreover, airport operators (Melbourne Airport, BAC, WAC; subs 7, 8, 21) added
that representations by the Commonwealth Government to prospective bidders for
the leases of Phase 1 and 2 airports did nothing to dispel their view that fuel
throughput levies could be implemented outside the price cap. Melbourne Airport
commented:
In relation to fuel throughput levies, it is our view that in the information memorandum
to bidders the Commonwealth did represent that that was a source of revenues available
to them that would not be within the cap. (trans., p. 169)

As a consequence, bidders factored projected revenue from fuel throughput levies

into their bid prices for the leases. WAC noted:
The bidders for PIA [Perth International Airport] factored new revenue from the fuel
levy into their bid, on the basis that pre-existing contracts allowing for the fuel levy had
been entered into by the FAC and agreed with fuel companies prior to the airport sale.
Moreover, the opportunity to implement the fuel levy was specifically referenced by
the Commonwealth Government in the Information Memorandum.
Accordingly, airports which acquired the benefits of these contracts, have merely
exercised their rights to charge the fee. (sub. 21, p. 21)

In its submission to the ACCC review of fuel throughput levies, WAC commented:
If the Government did not want WAC, or other bidding groups, to impose such a fee, it
should have made [it] perfectly clear prior to January 1997 … (sub. to ACCC 1998b,
p. 3)

The Productivity Commission is not aware of any public statements by the

Commonwealth Government (information memoranda to bidders were confidential)
specifically identifying services to be included or excluded from the price cap prior
to successful bidders for Phase 1 airports being announced in May 1997. In June of
that year, the Commonwealth Government publicly made its intentions clear
regarding refuelling services — that they would not be subject to the price cap. The
Treasurer stated that refuelling services would be subject to price monitoring
(Costello 1997). This public information was therefore available to bidders for
Phase 2 airports, although the policy was not formalised until May 1998 by

FUEL THROUGHPUT 399

LEVIES AND TAXI
CHARGES
Declarations (Nos 83 and 84) pursuant to the PS Act — after the successful Phase 2
airport bidders had been announced.

Phase 1 airport bidders may not have been clear about the implications of the formal
monitoring process. Although the Department of Transport and Regional
Development Pricing Policy Paper (DoTRD 1996), distributed to prospective
bidders for Phase 1 airports, noted that there would be monitoring of selected
aeronautical-related services, it did not specify the services and did not make
mention of the power of the ACCC to recommend that particular aeronautical-
related services be transferred to within the price cap. It was not until June 1997,
after finalisation of the bids, that the Treasurer publicly announced that there was
the potential for stricter prices oversight (Costello 1997).

In relation to taxi charges, airport operators (eg Australian Airports (Townsville),

Melbourne Airport, BAC, and WAC: trans., p. 17; subs 7, 8, 21) and Hastings
Funds Management (sub. 19) stated that Commonwealth Government advice during
the bidding stage for airports, and since (eg information memoranda), was that
revenue from vehicle access charges was to be collected outside the price cap.

In summary, with respect to these two charges, considerable uncertainty has ensued.
This seems to stem from differences in interpretation, a lack of clarity, precision and
transparency regarding the policies at, and since, the sale of leases, and time delays
in formal Commonwealth Government statements. Moreover, changes in
Government policy that may conflict with Government commitments at the time of
sale could raise issues of sovereign risk for parties that perceive themselves as
adversely affected.

400 PRICE REGULATION

OF AIRPORT
SERVICES
F Valuation of aeronautical land and
efficient land use

This appendix considers some issues that arise in the valuation of airport land for
the purposes of regulation of aeronautical prices. It begins with an in-principle
discussion of land valuation and the importance of the concept of opportunity cost.
Some practical issues associated with land valuation are then considered, drawing
on the issues raised in the recent ACCC (2001i) review of aeronautical prices at
Sydney Airport. The intention is not to recommend or suggest a detailed valuation
methodology, which would be beyond the scope of this inquiry, but rather to
explore the broad principles that can guide the valuation of aeronautical land.

Aeronautical services, particularly for large aircraft in large cities, are land
intensive. Runways need to be several kilometres in length and substantial buffer
zones between the airport and surrounding activities (particularly residential areas)
are desirable. In addition, most airline passengers value proximity to city centres,
and airports often are situated on land that would be valued highly for other uses. At
June 1997, the Federal Airports Corporation (FAC 1997) valued total land in its
airport network at around 30 per cent of its total (including non-aeronautical) fixed
assets.1

Because of its significance in the aeronautical asset base, the valuation of land can
have a major effect on the aeronautical prices that a regulator may consider justified
on the basis of recovering airport costs. In general, this has not been an issue to date
for the privatised core-regulated airports because their (average) aeronautical
charges have been determined by adjustments to FAC starting prices under price
caps, including related new investment cost pass-throughs.2 However, Sydney
Airport is not subject to a price cap; and the ACCC examined Sydney Airports

1 This was the last year in which the FAC operated all of the capital city airports. The FAC
valued land on the basis of market value for alternative use (capped at light industrial) at 30
June 1996. Other fixed assets were valued at written-down replacement cost at 30 June 1996
plus asset additions and less asset sales and depreciation during 1996-97.
2 Where new aeronautical investment has involved use of land previously used for non-
aeronautical purposes, regulatory assessment of the allowable price increase has included a
return on that land.
VALUATION OF 401
AERONAUTICAL
LAND
Corporation Limited’s (SACL) September 2000 proposal for aeronautical price
increases using a cost building block approach.

Although Sydney Airport is smaller in land area than most other major Australian
airports, it is situated close to the Sydney CBD and hence the market value of its
land is very high. SACL’s (2000) valuation of aeronautical land averaged $115 per
square metre and represents nearly 42 per cent of the aeronautical asset base.3
Participants in the ACCC’s examination of SACL’s price increase proposal
presented a variety of options for valuing land for the purpose of price regulation.
These provided a very wide range of land values and, by implication, significant
variations in cost-based aeronautical prices.

In considering Sydney Airport land values, the ACCC (2001i) drew attention to its
earlier views (ACCC 1999c) on asset valuation in general:
In determining an appropriate asset valuation methodology economic principles and
analysis do not provide an unambiguous decision rule for the valuation of sunk assets.
Rather economic principles provide lower and upper bounds — scrap value and
replacement cost. Within these bounds there is opportunity for regulatory judgement.
(ACCC 2001i, p. 132)

The scope for such ‘regulatory judgement’ appears particularly pronounced with
regard to Sydney Airport land. The ACCC reduced SACL’s proposed value for
aeronautical land of $705 million by over one-third to $452 million. In itself, this
lower land value leads to average aeronautical prices being about 8 per cent lower
than otherwise.4 Using the zero land value (for existing land) suggested by the
Board of Airline Representatives of Australia (BARA 2000), and applying the
ACCC’s assessment of SACL’s cost of capital, would have led to prices about
20 per cent lower than SACL’s request.

This wide variation in the valuation of land contrasts with other important elements
of Sydney Airport’s cost structure about which there was more agreement and in
which the ACCC, in reaching its pricing decision, varied much less from SACL’s
proposal. The ACCC accepted, for example, SACL’s valuations of other assets

3 SACL estimated the opportunity cost of land by considering the price that a developer would
pay for a similar land parcel in a similar location (after allowing for development and holding
costs). It then added the land-related costs (for example, holding costs) of developing the site
into an airport over five years. This is a ‘new entry cost’ concept for valuing land.
4 In conjunction with its proposal to include land at market values, SACL also included a
deduction of $14.7 million from allowable aeronautical revenue to allow for the forecast annual
real capital gain on land. However, because the ACCC preferred to use indexed historical cost
land values it did not take account of SACL’s proposed deduction when determining allowable
prices.
402 PRICE REGULATION
OF AIRPORT
SERVICES
(although it reduced allowable depreciation by nearly 25 per cent) and it reduced
allowable operating and maintenance expenditure by 10 per cent.

The opportunity cost concept is the appropriate approach to land valuation from an
economic efficiency perspective. It assists the government (and private investors in
airports) in making decisions on the efficient location of the airport by signalling to
users the land use costs of their air travel decisions — their willingness to pay
indicating that they value the current airport site more than its next best use. Any
move away from opportunity cost pricing will obscure these signals. Opportunity
cost also provides the correct incentives to the airport operator regarding use of
airport land.

The New Zealand Commerce Commission (CC), in its draft report on price control
for major NZ airports, argued in favour of opportunity cost as the appropriate value
of land for price regulation purposes, although it excluded land held for future
airport development:
Valuing airfield land at opportunity cost provides appropriate signals to either continue
operating the land in its existing use (as an airfield) or to put the land to alternative use
and relocate the airport. It also provides the appropriate incentives for new investment.
Opportunity cost should be determined based on the highest alternative use value of
airfield land. (CC 2001b, p. xxvi)

The ACCC, in its response to the Productivity Commission’s draft report, noted that
the Productivity Commission:
… argues that land at Sydney Airport should be valued at opportunity cost. The
Commission [ACCC] agrees with this principle. (sub. DR55, p. 18)

In a consultancy report for the ACCC, Dr Gannon argued that:

The opportunity cost principle for land valuation is widely accepted, for example, by
the ACCC, the NZCC, SACL, and BARA, although views differ substantially on its
interpretation.
But, there are also formidable methodological difficulties in practical empirical
application of the principle to estimation of the value of the total land at a major capital
city airport site. (sub. DR55, attachment B, p. 1)

The important questions are: what is the appropriate concept of opportunity cost?
And how can it be estimated in practice? Section F.1 provides an in-principle
discussion of aeronautical land valuation issues, focusing on the importance of land
valuation for:
• determining the appropriate location for an airport; and
• deciding how to use land within the airport precinct.

VALUATION OF 403
AERONAUTICAL
LAND
Section F.2 then considers the third role of land valuation  the setting of regulated
prices. This issue is explored by drawing on the recent ACCC decision in relation to
Sydney Airport.

F.1 Efficient land use

Two aspects of the appropriate valuation of airport land are considered. One is the
valuation of the site that could enter into calculations of whether an airport should
be left where it is or moved to a new site. The other is the valuation that would
guide efficient use of the existing site.

Valuation of existing airport land

Figure F.1 presents a very simplified picture of the demand for airport services and
the costs of providing them. For simplicity all airport services are bundled together
(including leases for retailing and so on) and the supply and demand for the package
is related to aircraft movements. Demand is aggregated so there are no differences
in demand at different times of day or year.

Figure F.1 Market for airport services: different valuations of airport land

AC1
Pc

AC3 AC2 D
P1

P2

P3

O A
Aircraft movements

The average cost curves (AC1, AC2 and AC3) are drawn for different valuations of
land. For AC1, it is assumed that the average costs incorporate the highest

404 PRICE REGULATION

OF AIRPORT
SERVICES
opportunity cost of the land, which could be redeveloping the land for residential
purposes. Using existing values of neighbouring land (depressed by airport noise)
gives average costs of AC2.5 Using BARA’s proposed opportunity cost of airport
land of zero could give AC3.

Assume that price is regulated on a cost ‘building-block’ approach — that is, by

adding up the production costs of the airport together with an assumed value of the
land. In the absence of a capacity constraint, the price would be determined by the
intersection of the AC1 and the D curves.

Now assume that use of the airport is constrained to OA by regulation. (Assume

also that capacity cannot be increased by buying more land and that the airport
charges a single price and has to cover its costs.) The highest level that prices can
reach under the building-block approach is P1. However, the efficient price of
airport services is Pc, at which price, demand is constrained to the available supply
and those who value the service at Pc or more will purchase the services. If the
airport were to be sold as an airport, then the revenue would be the capitalised value
of OA.OPc less running costs. (The valuation of consumers of the airport would be
larger than this, being the area under the demand curve up to the quantity OA. This
is the amount of revenue that a perfectly discriminating monopolist would be able to
raise from consumers of the services.)

The difference between Pc and P1, P2, and P3 respectively, multiplied by the quantity
OA, shows the ‘economic rent’ that could accrue to someone other than the airport
if prices were constrained to P1, P2 or P3 (some of the rent may be dissipated by
queuing). It is only the price Pc that shows the value of the airport to the operator or
potential purchaser, in its present site; it is this value that provides an appropriate
signal as to whether to sell and move elsewhere, and to a government in evaluating
the effects of regulation.6

If a building-block approach to price regulation is taken, AC1 provides a better

guide than AC2 or AC3, and also imposes a smaller burden on the non-price
rationing of the demand to the constrained supply: excess demand is smaller with

5 The position of the average cost curves is for illustrative purposes. For example, it is also
possible that AC2 may be higher than AC1 if proximity to the airport increases the value of
neighbouring land relative to its value in an alternative use.
6 It should be noted that environmental considerations are not taken into account here — they are
a necessary factor to be addressed in making any decision regarding moving etc in practice.
VALUATION OF 405
AERONAUTICAL
LAND
price P1 than P2 or P3. Also, the rents accruing to those other than the airport
operator (or government) are smaller with P1 than P2 or P3.7

The analysis is made a little more complicated if the airport were capacity
constrained only at certain times of the day. A more detailed discussion of capacity-
constrained pricing issues is contained in appendix H.

Efficient use of airport land

Now consider the appropriate land valuations to guide the use of the fixed amount
of airport land by the airport operator. Initially, the assumption is made that land (at
the margin) can be allocated between different uses. Allowed pricing freedom, the
operator would allocate land among various activities so they yielded the same
(marginal) return on each. This would be equal to the price Pc if the output is
constrained.8 For those services that could be provided on or off the airport, only
those that yield a benefit of Pc or more would be supplied on the airport. If the price
is constrained below Pc then the operator would no longer have to be so selective in
choosing what was located on-site. As noted earlier, constraining prices to below Pc
may mean that some users who value airport services below Pc may obtain the
services while those who value them higher than Pc may be excluded. So it may be
that some services that are valued (at the margin) below Pc are provided on-site,
instead of those that are valued at Pc or more. This potentially inefficient use of the
airport site would be aggravated the lower is the regulated price.

If the regulated price were to be such that a single till ‘taxed’ away from the
operator all profits from some airport activities to reduce the price of others, the
operator would have no incentive to allocate airport land efficiently and would have
an incentive to move off-site all activities on which a higher return was possible. (If
the constrained return were below P2, then there would be an incentive to sell some
airport land if that were possible.)

If a dual till were to be applied, regulating prices on some activities but not others,
then an efficient allocation of land between the two types of activity could be
obtained only if the costs allowed in determining the regulated prices incorporated
the full opportunity cost of the land. That opportunity cost would be the return
available on land from producing unregulated services.

7 Note that this is not implying that land values should be inflated or manipulated in order to
achieve prices sufficient to ration demand to the constrained supply. Land should be valued at
its opportunity cost and this value used for regulatory purposes.
8 Assume that the constraint is due to the availability of land at the airport.
406 PRICE REGULATION
OF AIRPORT
SERVICES
Now assume that there are constraints on allocating land at the margin between
uses. In particular, assume that there are two types of land — airside and landside
— and that the constraint on aircraft movements means there is no point in moving
landside land into airside use. (It is also assumed that there is no incentive to
convert airside land into landside land.) The relevant opportunity cost of land for
landside activities as a whole would be the actual cost of land off the airport site. If
the services provided on landside land are free to locate either within the airport
boundary or outside, then over time the price of landside land within the airport and
surrounding land would be expected to converge. In the long run, the value of
surrounding land would therefore indicate the lower bound on the valuation of
landside land within the airport.

However, if there were locational advantages from being on the airport site (or on
parts of it) then the locational rent should be added to the off-site value in
considering the relevant opportunity cost for allocating any particular land between
alternative on-site uses.

F.2 Sydney Airport and the ACCC decision

While there was wide support for valuing aeronautical land at its opportunity cost,
participants who commented on the Sydney pricing proposal raised a number of
issues relating to the empirical measurement of opportunity cost.

Issues relating to the valuation of land at Sydney Airport

A key issue was whether Sydney Airport would continue to operate as an airport
and the impact of this on the opportunity cost of land. The Government has
announced that it considers that Sydney Airport will be able to handle Sydney’s air
traffic growth until 2010, with a review of Sydney’s airport needs to be conducted
in 2005 (Anderson 2000). Even when a second airport is built, it is likely that
Kingsford Smith will be retained because of its locational advantage.

BARA (2000) argued, in view of this, that the appropriate opportunity cost
valuation of Sydney Airport land is zero because there were no other uses to which
it would be put. BARA considered that any purchases of new land should be valued
at its purchase price to provide appropriate incentives for SACL in making marginal
additions to land. Network Economics Consulting Group (NECG 2000b), in a report
to the ACCC, also contended that if there is no likelihood of Sydney Airport being
closed then there is no opportunity cost associated with SACL land.

VALUATION OF 407
AERONAUTICAL
LAND
Pitchford (2000) argued that if the use of the current site as an airport is considered
mandated by the Government, then the land’s sale value is not relevant for assessing
required regulatory return. He considered that for existing land, a (non-zero) value
derived from users’ valuations of airport services is appropriate.

It should be recognised that while such regulatory considerations may limit the
airport owner’s options in relation to sale of the land, they do not alter the
opportunity cost to the economy of the land itself. The opportunity cost remains the
value of the land to society in its next best available use (adjusted for any costs
needed to prepare it for that use, such as the cost of demolition of existing
facilities).

Another issue that has been raised relates to the costs and benefits of moving
Sydney Airport and how these affect the valuation of land for regulatory purposes.
The ACCC (2001i; sub. DR55) argued that, in deciding to move Sydney Airport,
the Commonwealth Government would need to take into account:
… a wide range of issues. These could include the costs of relocating the airport, the
costs of providing infrastructure to service the new airport as well as less tangible
considerations such as the costs and benefits of relocation on airport noise and travel
times. Since a complete assessment of such matters is complex, and the magnitude of
the components of the calculation is likely to change over time, the merits of using a
full opportunity cost valuation should be weighed carefully against alternative
approaches and is considered beyond the scope of this decision. (ACCC 2001i, p. 138)

However, in estimating the opportunity cost of land at Sydney Airport it is

important to separate the narrower question of land value for regulatory purposes
from the broader question of factors that will influence a decision to relocate the
airport. The two are not synonymous and should not be confused.

The Civil Aviation Authority (CAA) considered a similar issue in its review of price
caps at major UK airports. It concluded that:
A full analysis of the demand and costs of airports would need to take into account the
wider externalities created by airport activities. The most direct of these are
environmental externalities, particularly noise, road congestion and emissions … The
CAA believes that it is important that these costs and benefits be properly taken into
account in decisions to expand capacity. However, taking direct account of these
externalities is not one of the CAA’s statutory objectives as an economic regulator of
airports. This is the role of the planning system. (CAA 2001c, p. xiii)

It would be appropriate that the Government take account of all costs and benefits
(both private and social) if it were considering relocating Sydney Airport. However,
the opportunity cost of land at Sydney Airport remains its value in its next best use.
The opportunity cost valuation should be used in regulatory pricing decisions
because prices based on any another valuation will mean that signals regarding
408 PRICE REGULATION
OF AIRPORT
SERVICES
consumers’ willingness to cover the efficient cost of using the airport will be
distorted.

The ACCC decision

The ACCC (2001i) considered that because the Mascot site was likely to continue
as an airport or would need to be replaced by another airport, it was not appropriate
to adopt SACL’s proposal to use an estimate of its market value as an opportunity
cost measure of aeronautical land value in developing cost-based prices. It also
argued that prices based on market value of the whole aeronautical section of the
site were not necessary to ensure efficient use of airport land by SACL. Instead, the
ACCC valued aeronautical land by indexing the original cost of each of the land
purchases that had gone into making up the current aeronautical land by CPI
increases since their purchase. Hence the real value (at least in terms of the various
CPI baskets over the years) of the land was maintained. New land purchases were to
be included in the asset base at their purchase price, and then indexed by the CPI.

The ACCC also argued that there were complexities in estimating the opportunity
cost of Sydney Airport’s aeronautical land. These included whether purchase or sale
of the land (with associated development or clean-up costs respectively) was
relevant and the consideration of externalities (costs and benefits) of relocating to a
new site (discussed above). Hence it concluded:
… the Commission [ACCC] is not persuaded that SACL’s proposal can be considered
a measure of the true opportunity cost of land at Sydney Airport. It also concludes that
such an assessment would be highly complex and well beyond the scope of this
decision. (ACCC 2001i, p. 141)

Dr Gannon argued that there are practical difficulties with measuring the
opportunity cost of land and therefore alternative valuation methodologies, such as
indexed historical cost, may have to be used. However, he noted:
These are poor surrogates as they are unlikely to bear a reliable relationship to the
underlying opportunity cost. For example, an indexed historical cost basis (as resorted
to by the ACCC …) may have some properties that allow it to serve as a reasonable
substitute. But it could result in a valuation that is higher or lower than the real
opportunity cost. (sub. DR55, attachment B, pp. 1–2)

The ACCC did not justify using indexed historical cost as an opportunity cost
concept, but argued instead that the investment and land use signals that such
historical cost pricing provided to SACL would not be inappropriate for efficient
decision making. However, it recognised that valuing aeronautical land at current
market value would meet this criterion.

VALUATION OF 409
AERONAUTICAL
LAND
The ACCC argued that the relative simplicity of the indexed historic cost approach
was a factor in its adoption, because of:
… the difficulty of tying down an opportunity cost valuation. Given the limited time
available to it in making its decision, the Commission [ACCC] considered the
alternative of a historic cost based measure. One advantage of this approach is data
availability. (sub. DR55, pp. 18–19)

However, assumptions must be made regarding the data used in the calculation —
most notably, regarding the price index used. The ACCC used the CPI to inflate the
value of past land purchases. But changes in the general level of prices need bear no
relation to changes in the opportunity cost of land at Sydney (or any other) airport.

Prices based on the ACCC historical cost valuation do not provide signals to the
Government regarding the value that aeronautical users place on the facility
compared with its alternative use. The willingness of consumers to pay prices set on
the basis of the opportunity cost of using Sydney Airport land (measured by the full
market value) would assist the Government in deciding whether an airport is the
best use of the land.

F.3 Distributional issues

The value of land chosen in determining a regulated price has major distributional,
as well as efficiency, implications. It is argued in section F.1 that the efficient prices
for airport services at a congested airport are those that equate demand with the
(constrained) supply. A valuation of the airport that uses ‘building-block’ prices,
even those that use the full opportunity cost of the land for purposes other than an
airport, are likely to undervalue the airport, in terms of the value that users place
on it.

Increases in airport charges to reflect the value placed on using the airport would
increase the sale value of Sydney Airport — a benefit that would accrue to the
Government and the community at large. Lower charges, and a sale price based on
these, imply that rents accrue to parties other than the owner or operator of the
airport. Some of these rents currently support the reservation of slots (and low
charges) for regional airlines. Some accrue to major airlines (and their shareholders)
and are reflected in the high yields obtainable on peak-hour flights into and out of
Sydney. In part, these high yields may support flights between other destinations on
which yields may cover marginal, but not average, costs.

410 PRICE REGULATION

OF AIRPORT
SERVICES
G Economic regulation of privatised
airports: international experiences

This appendix focuses on economic regulation of privatised airports in the United

Kingdom, New Zealand, Germany and Denmark. Airports in these countries are
(generally) similar in ownership and structure to those in Australia, as they tend to
be privately owned and are run as commercial enterprises. Economic regulation at
major airports that are wholly-government-owned, or run as non-profit
organisations, such as those in the United States and Canada, is not examined.

G.1 United Kingdom

The sale of the British Airports Authority (BAA) by public float in 1987 —
becoming BAA plc in the process — was one of the first (and largest) sales of
government-owned airport infrastructure in the OECD.1

BAA was established initially in 1966, as an independent commercial enterprise

under government ownership, to manage four airports — the three major London
airports (Heathrow, Gatwick and Stansted) and one in Scotland (Prestwick).
Between 1966 and 1987, BAA acquired a further three Scottish Airports —
Edinburgh in 1971, and Aberdeen and Glasgow in 1975. Following privatisation,
BAA plc purchased Southampton Airport and sold Prestwick Airport.

A number of UK airports were corporatised around the same time as the

privatisation of BAA. Many of these corporatised airports have since been
privatised — currently around two-thirds of airports regulated by the Civil Aviation
Authority (CAA) (see below) have been privatised to some extent (MTAA Super
Fund, sub. 22). Local government ownership of airports is also common, even for
large airports such as Manchester.

1 BAA was privatised under the Airports Act 1986 and 500 million shares were sold on the
London Stock Exchange in July 1987.
INTERNATIONAL 411
EXPERIENCES
Economic regulation of airports

There are two levels of economic regulation of airports in the United Kingdom. For
a small number of ‘designated’ airports that are considered to have significant
market power, economic regulation consists of a Retail Price Index (RPI)-X price
cap. The second level comprises a system of light-handed regulation for airports
that achieve a pre-determined revenue threshold.

Economic regulation of airports in the United Kingdom is detailed in the Airports

Act 1986 and the Civil Aviation Authority (Economic Regulation of Airports)
Regulations 1986 (CAA 2000b). Regulation is administered by the CAA. As the
airport regulator, the CAA is given a number of objectives to pursue under the
legislation. These include:
• to further the reasonable interests of users of airports;
• to promote the efficient, economic and profitable operation of such airports; and
• to encourage investment in new facilities at airports in time to satisfy anticipated
demands by users of such airports.2

Light-handed regulation

A system of light-handed regulation applies to airports at which annual turnover has

exceeded £1 million pounds in two of the previous three financial years.3 Airports
that meet the revenue threshold must apply to the CAA for permission to levy
airport charges (once granted, the permission remains in force until revoked). As of
March 2000, 47 airports in the United Kingdom held CAA permission to levy
charges, including the four designated airports and all of BAA plc’s airports.

A key component of the system is public disclosure of airport charges and accounts.
All airports holding CAA permission must provide the CAA with their annual
statutory accounts, schedules of airport charges and changes (if any) to the
information provided in their original application. Airports do not need to seek

2 However, unlike other UK regulators, the CAA has been given no specific duty to promote, or
even to facilitate, competition (Cotterill 1999).
3 Airports currently excluded from regulation under the Airports Act (or the Airports (Northern
Ireland) Order) are those in the Isle of Man and Channel Islands, those owned or managed by
the CAA or a CAA subsidiary and those managed by the Government. An airport becomes
subject to economic regulation by the CAA nine months from the end of the financial year in
which it first meets the turnover qualification. Should the annual turnover at a regulated airport
fall below £1 million for two years, the Secretary of State may determine that the airport shall
cease to be regulated.
412 PRICE REGULATION
OF AIRPORT
SERVICES
CAA approval before they revise their airport charges but must notify the CAA of
the charges before they take effect.

There also is provision for placing further restrictions if an airport is considered to

have abused market power.

First, restrictions can result from complaints brought against an airport operator by
airport users and others (including other airports) for pursuing a course of conduct
specified in the Airports Act. In broad terms, complaints may be brought against an
airport that:
• unreasonably discriminates against any class of users (or a particular user) of the
airport; or
• unfairly exploits its bargaining position relative to users generally; or
• levies charges that are both unduly low and cause damage, or are designed to
cause damage, to another airport (CAA 2000b).

If the CAA considers that there is a case to answer, it will investigate and, where
necessary, recommend actions or impose ‘conditions’ on the airport to rectify the
situation.

Second, there is the threat that an airport may become designated by the Minister of
State under s. 40 of the Airports Act and, consequently, be subject to price-cap
regulation. While there are no formal criteria for designation in the Airports Act, the
Government has stated that it considers the relevant criteria are:
• the market position of the airport, including the degree of competition from other
airports and other modes;
• prima facie evidence of excessive profitability or abuse of a monopoly position;
• the scale and timing of investment, and the implications for profitability; and
• the efficiency and quality of service (CAA 2000a).

Economic regulation at designated airports

As noted above, under the Airports Act, there are provisions for airports to be
designated and become subject to more restrictive regulation than those airports
holding CAA permission. Current designated airports in the United Kingdom are
Heathrow, Gatwick and Stansted in London, and the major international airport in
the north of England, Manchester.

Designated airports must adhere to a number of conditions in addition to those

applying to airports holding CAA permission. An accounts condition, for example,

INTERNATIONAL 413
EXPERIENCES
requires designated airports to reveal, among other things, the revenue and costs
from three classifications of their activities: airport activities, other airport-related
activities and non-airport activities (CAA 2000b).

The principal feature of economic regulation of designated airports is the price cap.
In practice, economic regulation of designated airports is a hybrid model of price-
cap and rate-of-return regulation. The price cap, which is similar to the price caps
commonly applied to other regulated industries in the United Kingdom, comprises a
RPI-X price cap applied in the form of an average revenue yield per passenger
(Cotterill 1999). Airport charges subject to the price cap are those associated with
the landing, take-off, and parking of aircraft, and with the processing of passengers
through the terminals. As individual charges are not subject to the price cap, the
airport operator has a degree of discretion with respect to the level of each
individual charge and the relationship between them (Starkie 2001c).4

Application of the RPI-X price cap at designated airports differs from other UK-
regulated industries subject to a price cap in that, when resetting the price cap, the
CAA adopts a single-till approach. Under the single till, future revenues and costs
are assessed on an airport-wide basis to determine allowable average revenue yield
per passenger. That is, the regulator takes into account not only the revenue
generated from aeronautical services, but also the revenue generated by activities
such as retailing, and the provision of rental property and other services to tenants
and licensees (box G.1).

The price-cap approach to the regulation of designated airports has been in place
since 1987. The cap is reset every five years by the CAA, in conjunction with the
Competition Commission,5 after an extensive review process. Every quinquennial
review begins with a reference to the Competition Council, which is given six
months to report its analysis and recommendations concerning the price cap, and
any conclusions as to whether the airport company has acted against the public
interest. The Competition Council’s recommendations are then made available to
the public by the CAA for consultation and comment. Finally, the CAA announces
its findings on the price cap, and also indicates how it intends to implement any
public interest findings.

4 At BAA plc’s designated airports, a single revenue yield price cap applies to Heathrow and
Gatwick airports and a separate cap applies to Stansted (subject to the constraint that the charge
differential between Heathrow and Gatwick is required to increase by at least one percentage
point per year) (CAA 2001b). The allowable rate of return for BAA plc implicit in the price-cap
calculations for the period 1997–2002 was 7.5 per cent.
5 Formerly the Monopolies and Mergers Commission.
414 PRICE REGULATION
OF AIRPORT
SERVICES
 Box G.1 UK approach to single-till price-cap regulation
The Civil Aviation Authority and the Competition Council undertake a review process to
reset the Retail Price Index-X price cap at designated airports every five years
(although this process is currently under review). The price-cap model used is similar
to price caps applied in other UK regulated industries, except that a single-till approach
is used to calculate the revenues the airport may earn from the provision of
aeronautical services.
The procedure for calculating the price cap at designated airports under the single till
comprises four main steps.
1. An agreed program of capital expenditure for the airport is determined. In addition,
based on consultation between the regulators, the airports, airport users and other
interested parties, the regulator determines, for the coming five-year period,
estimates of:
– traffic and passenger numbers;
– operational expenditure; and
– commercial revenues (revenues from non-aeronautical services).
2. The value of the airport’s asset base is derived, and an allowable risk-adjusted rate
of return is determined by the regulator.
3. The overall revenue required by the airport to achieve the approved rate of return is
then estimated. The estimated commercial revenues are then subtracted from the
overall revenue requirement to obtain the residual revenue requirement.
4. The residual revenue requirement is what airports may earn from charges from the
provision of aeronautical services under the price cap. The Xs, which are set by the
regulator, are estimated to allow the airport to recover the residual revenue
requirement.
Source: Kunz (2001).

There is no explicit access regime for airports in the United Kingdom. There are
general provisions for determinations to be made on access to ‘essential’ facilities
on a case-by-case basis under UK and European law (ACCC, sub. 36; PC 2001a).

The UK experience with economic regulation of airports

As noted previously, price regulation at designated airports has differed from other
UK industries subject to a price cap in the application of a single till. The single-till
approach to airport regulation in the United Kingdom stemmed from international

INTERNATIONAL 415
EXPERIENCES
treaty obligations6 and concerns over equity and distribution. Cotterill noted that
distributional concerns have continued to be used as an argument to support the use
of the single till:
… [the Monopolies and Mergers Commission in the 1997 quinquennial reviews]
pointed for example to the large windfall profits which removal of the single till would
mean for the airport companies at the expense of the airlines and the universally
adverse reaction of the airlines to such a change. (Cotterill 1999, p. 4)

However, as reflected in the CAA findings in the current review (see below), it
appears that the objectives of the regulation (outlined earlier) are not being met. As
discussed below, significant concerns have been raised that the regulatory system is
proving to be an impediment to the efficient pricing of current airport facilities, and
that it is not providing the appropriate incentives to encourage investment in major
airport infrastructure, such as runways and terminals. Other issues raised in relation
to the regulatory system include the potential for gaming to occur, and the high
compliance costs of the system.

Particular issues include the following.

• The regulatory system may cause some reluctance by airport operators to invest
in large infrastructure projects (such as new runways and terminals). Apart from
regulatory uncertainty associated with such long-term investment, there is the
possibility that the existing system of slot allocation (where airlines have de-
facto property rights to the scarcity value of the slots) combined with the single
till, does not allow the airport operator to realise the full value of the capacity
expansion (Cotterill 1999).
• The single till has been criticised for reducing aeronautical charges at congested
airports below market-clearing levels and sometimes below the resource costs of
providing the airside services (Starkie 2001b).
• Other incentive problems relate to the denominator of the price cap — passenger
numbers. This may encourage airports to focus on passenger volumes at the
expense of other outputs (eg aircraft movements that do not generate any
additional passengers). There is little incentive to attract such traffic if the
resulting revenues must be offset by reductions in revenue elsewhere
(CAA 2001b).
• There is potential for gaming because of the information asymmetry that exists
between the regulator and the airport operator. The regulator does not have good
information on the airport operator’s ability to achieve cost efficiencies or to

6 Until 1991, the UK Government was committed to applying the single-till principle under the
terms of the Bermuda II Air Agreement with the United States, although this is no longer the
case (Starkie and Yarrow 2000).
416 PRICE REGULATION
OF AIRPORT
SERVICES
 invest efficiently. Airport operators potentially can game the system, for
example, by inflating recorded and projected costs in order to obtain a ‘looser’
price cap, or by putting forward investment proposals that are ‘gold plated’ or
are not demanded by users (CAA 2000g; Kunz 2001).7 The CAA also noted the
possibility of gaming by airport users (CAA 2001c).

The experience of the UK regulatory system with light-handed regulation of non-

designated airports holding a CAA permission are less well-documented. The CAA,
for instance, is not examining the effectiveness of this approach to regulation as part
of its current review. However, there is some evidence that the threat of designation
under the Airports Act acts as a deterrent to the potential abuse of any market power
(Kunz 1999; Starkie 2001b). BAA plc’s Scottish airports were subject to a
designation application, which the Government rejected because there was no
evidence of abuse of a monopoly position or inefficiency. When announcing its
decision, the Government added that it believed the threat of designation provided a
strong incentive for BAA plc to control its charges. Possibly reflecting this,
BAA plc capped its prices at both Glasgow and Edinburgh airports on a voluntary
basis, initially with the formula RPI-3 (Starkie 2001b).

Preliminary findings of the UK regulatory review

The CAA, as part of its current quinquennial review of price caps at designated
airports, is undertaking a fundamental review of its approach to the economic
regulation of airports within the existing framework of the Airports Act (the review
is to determine how the price caps should be set for the period 2003–08 and
beyond). The review is broad ranging in scope, and has arisen out of concerns that
the current regulatory framework is not best suited to address a number of issues
that have emerged or intensified in recent years. These include:
• that demand for access to Heathrow and Gatwick airports exceeds available
capacity;
• in the absence of an efficient market for take-off and landing slots, utilisation of
existing capacity may not be optimal;
• whether incentives for the promotion of appropriate investment in capacity are
efficient;
• the importance of unregulated revenue in setting charges; and

7 The CAA, as part of its current review of the price-cap regulation, is examining a number of
options that might result in a lessening of the potential to game. These include the use of
benchmarking (CAA 2000g) and a ‘default’ price cap (CAA 2001a).
INTERNATIONAL 417
EXPERIENCES
• the importance of service quality for customers and consumers, and the wide
variation in quality that different users may require (CAA 2000c).

The CAA (2001c) released its preliminary conclusions in November 2001. After
examining several options for setting price caps, it recommended substantial
changes to the way the caps are applied. In particular, it recommended a move to
dual-till price caps for all four designated airports. The single-till approach was seen
as being particularly unsuitable at the capacity-constrained Heathrow and Gatwick
airports because:
• under conditions of excess demand, the single till was an inefficient way of
allocating scarce capacity; and
• it did not provide airports with good incentives to invest optimally.

The CAA noted that the argument against the single till was not as strong at
Stansted and Manchester airports (which currently are not congested). For a number
of reasons, however, the CAA recommended that the single till should not be
applied at either airport.
• Stansted currently faces competition from Luton Airport, so it is not clear that it
has monopoly power. Thus:
… the case for retaining the single till at Stansted is a trade-off between a minimal or
non-existent gain in restraining prices with the costs and distortions that persisting with
a wider regulatory framework, that was only marginally binding, would generate.
(CAA 2001c, p. xvii)
• On the other hand, the CAA was not convinced that Manchester Airport
operated in a competitive environment. It also concluded, however, that although
Manchester has a significant local monopoly it faces more competition than
Heathrow and Gatwick. That Manchester ‘had not sought to expand passenger
numbers through low cost carriers’ (CAA 2001c, p. 122) was cited as one
indicator of the lack of competition it faces. Given this:
… the loss in efficiency from moving to a RRCB [Revised Regulatory Cost Base] price
cap may be higher [than at Stansted], but a demonstrated ability to price differentiate
and its local government ownership should ensure that any loss in output from a move
to a RRCB would still be low. The CAA believes that the distortions caused by
regulating the commercial business by retaining the single till would outweigh this loss.
(CAA 2001c, p. xvii)
- The CAA also was concerned about the potential impact of applying the
single till at Manchester on inter-airport competition:
… Manchester is in a situation where it is subject to increasing, albeit fringe,
competition and so tight price cap regulation runs the risk of the most undesirable
distortion of all: that low prices set by the dominant firm under regulation may
hinder the growth and development of a more competitive airport market.

418 PRICE REGULATION

OF AIRPORT
SERVICES
 Continuing with the single till, which results in airport charges that are below the
costs of providing airport services, may have this effect. (CAA 2001c, p. 123)

Moreover, the CAA did not accept that airlines had any intrinsic rights over the
commercial revenues generated at airports.

On balance, the CAA found that the existing single-till framework ‘is unlikely to be
the best basis for meeting its statutory objectives given the challenges now faced
over the next five years and beyond’ (CAA 2001c, p. xviii).

The CAA recommended as the appropriate basis of the price cap the Revised
Regulatory Cost Base (RRCB) at all four designated airports. The RRCB includes
only those aeronautical activities in which the airport operator has clear monopoly
power. The CAA identified three main reasons for modification of the price-cap
framework to the RRCB.
• It would reduce the scope of regulation to core aeronautical activities ‘where the
airport has a clear monopoly in relation to its users’ (CAA 2001c, p. xviii).
• It would generally result in prices that better reflect market conditions at
Heathrow and Gatwick airports, though prices would remain ‘well below’
market clearing levels.
• Profitability of investment in increasing capacity would be increased, ‘resulting
in a greater likelihood of more investment in aeronautical infrastructure than
under a single till’ (CAA 2001c, p. xviii).

Other options for setting the cap that were examined by the CAA included:
incremental costs, value-based incentives, and contracting between the airport and
users.
• The CAA considered that there would be significant benefits — in terms of
incentives for both investing in appropriate capacity and cost-effectiveness — if
a long-term price path could be set in terms of forward-looking incremental
costs.
- It noted, however, that setting caps on this basis is not straightforward, with
risks of both under- and over-estimating costs.
- At capacity-congested airports (Heathrow and Gatwick), prices would be well
above RRCB levels, but would result in improved resource use in the
medium term.
• Prices incorporating a premium over costs (value-based incentives) may be
appropriate where there is excess demand and ‘the value of additional outputs to
users may … be much higher than the costs to the airport of delivering those
outputs’ (CAA 2001c, p. xxi).

INTERNATIONAL 419
EXPERIENCES
• Given that real prices could rise with the new price cap, there would be
opportunities for greater contracting within the cap set. The CAA noted that it:
… would expect the airports to use this greater latitude pro-actively and responsibly to
deliver material benefits in terms of improved investment in service quality and
capacity, to engender a greater consensus with its users on the strategic direction being
taken and to pursue the objectives desired. (CAA 2001c, p. xxi)

Based on these options, additional modifications were recommended for Heathrow

Airport’s price cap to create further incentives for investment at the capacity-
constrained airport (discussed below).

As well as examining options for the setting of the price cap, the CAA identified a
number of other areas where the regulatory framework required improvement,
including in relation to service quality, the default price cap, information disclosure
and pricing structures.
• The CAA noted that users had expressed concerns about the treatment of service
quality. It added, however, that ‘specifying and fixing quality standards carries
major risks of perverse outcomes, would result in a very detailed oversight of
airport operations by the regulator, and would be likely to reduce the
accountability of the airports’ (CAA 2001c, p. xxii).
• It noted that the default price cap would give protection to users while allowing
and encouraging contracting between the airport and users. This would allow
services and prices to be tailored to user needs. Where the price cap would be
binding, it argued that:
… explicit provision should be made in the price cap condition for direct contracts for
different levels of service quality, and for specific facilities that would not be counted
as airport user charge revenues for the purpose of meeting the price cap limits. The
CAA recognises that its provisions will require some degree of regulatory oversight, in
order to ensure that direct contracts do not have a serious detrimental impact on those
users that remain on the default price cap. (CAA 2001c, p. xxiii)
• The CAA found that none of the designated airports provides sufficient
information to help users assess project proposals. It argued that airports should
provide further information to users, including demand, capacity and cost
projections, service quality, investment plans and resourcing implications. It
commented, however, that the CAA ‘does not have, and could never have, the
information to take ownership of the airports’ investment plans, and to attempt
to do so would compromise the management and accountability of the airport
and its users in delivering services to passengers’ (CAA 2001c, p. xxiv).
• The CAA argued that there was considerable scope for improving the use of
airport capacity through price differentiation. This could be done through the
default price cap, but also through a move to a tariff basket approach. It also

420 PRICE REGULATION

OF AIRPORT
SERVICES
 expressed a ‘strong view that airports should move to greater peak pricing’
(CAA 2001c, p. xxvii). It decided, however, not to impose peak pricing:
The fact that this [peak pricing] has not been forthcoming … suggests that there are
significant costs to the airport in attempting to introduce peak pricing, perhaps in the
form of pressure from some groups of users. (CAA 2001c, p. xxvi)

Summary of CAA recommendations

The major thrust of the CAA’s preliminary recommendations is to create the right
incentives for future investment so as to alleviate the capacity constraints at
Heathrow and Gatwick airports, as well as ensuring that similar situations do not
arise at the other two designated airports. The CAA also considered that the
package contained initiatives to place stronger pressure on the airports to improve
their service quality, to move to greater direct contracting, and to improve
consultation on airport development.

The price cap recommended to apply at Heathrow Airport is to be based on three

components:
• prices for current outputs (up to 70 million passenger movements per annum)
based on RRCB;
• higher set prices for additional passenger movements over 70 million passengers
per annum, based on the incremental costs of new capacity; and
• a set premium of (indicatively) £300 for each additional aircraft traffic
movement above 480 000 per annum.

Gatwick also would move to an RRCB pricing framework, but with no separate
returns for additional output. A ‘simple’ RRCB would apply at Stansted and
Manchester.

Recommendations regarding other areas of price regulation included:

• moving to a tariff basket as the basis for the price cap at Heathrow and Gatwick,
while continuing with a revenue yield cap at Stansted and Manchester airports;
• strengthening information disclosure and consultation requirements for airports;
• facilitating direct contracting outside the price cap at Heathrow and Gatwick
airports, by way of the default price cap (contracting is likely to occur at
Stansted and Manchester without the default price cap);
• introducing a service quality term into the price cap at Heathrow and Gatwick
airports (and at Stansted and Manchester ‘if a strong case is made for it’
(CAA 2001c, p. xxviii));

INTERNATIONAL 421
EXPERIENCES
• moving regulation of Stansted Airport to a stand-alone basis, and the ending of
any cross-subsidy of Stansted’s asset base from charges at Heathrow and
Gatwick airports; and
• setting separate price caps at Heathrow and Gatwick airports.

The CAA envisaged that the new prices would be phased in over the five years,
rather than being introduced at the start of the period:
This will give users in particular time to adjust to the new system, and encourage
airports to facilitate a move to greater accommodation … within it. (CAA 2001c,
p. xxix)

It acknowledged that adopting the RRCB approach would be likely to result in price
caps that generally are higher than single-till prices, and a transfer of economic
rents to airport operators. With regard to Heathrow and Gatwick, however, it noted
that:
These higher prices [for aeronautical services] are justified on the basis of increasing
the prospect that available capacity is utilised as well as possible given the serious
levels of unsatisfied demand at these airports at current prices. (CAA 2001c, p. xxix)

The CAA’s final report and recommendations are due to be released by the end of
February 2002, with the final decision on the price caps to be delivered by end-
November 2002.

G.2 New Zealand

Private ownership of major airport infrastructure is a relatively new phenomenon in
New Zealand. While eight of New Zealand’s major airports were corporatised in the
1980s, including New Zealand’s three international (and also largest) airports —
Auckland, Wellington and Christchurch — to date, only two have been privatised.
Auckland and Wellington moved to majority private ownership when the national
government sold its shares in 1998 (51.6 per cent in Auckland by public float, and
66 per cent in Wellington by sale — local governments still own a minority share).8
Most other airports of commercial significance in New Zealand remain
government-owned, usually in the form of joint ownership between the central and
local governments.

8 Christchurch Airport was corporatised in 1988 with ownership transferred to Christchurch

International Airport Limited, a company jointly owned by the Christchurch City Council
(75 per cent) and the New Zealand Government (25 per cent).
422 PRICE REGULATION
OF AIRPORT
SERVICES
Economic regulation of airports

New Zealand has adopted what has been referred to as a ‘light-handed’ approach,
with major airports not subject to any direct economic regulation. Although there
are examples internationally of government-owned major airports not subject to
direct regulation, indirect economic regulation of privatised airports is more
unusual.

The NZ approach to economic regulation of airports is in line with the approach the
Government adopted with former state-owned industries in the 1990s. In general,
this approach focused on:
• requiring the disclosure of prices, terms and conditions for contractual
arrangements, costs, performance measures, and financial performance
indicators;
• the use of the Commerce Act 1986 to control anti-competitive behaviour; and
• threats of further regulation, such as price control if market dominance is abused,
but perhaps in a different form than was previously employed in New Zealand
(PSA 1995).

The aim of this type of light-handed regulation was to facilitate negotiated

outcomes between parties, without the need for direct intervention by the
Government.

The NZ approach to economic regulation is implemented for airports through the

Airport Authorities Act 1966 and the Commerce Act 1986. The Airport Authorities
Act requires NZ airport companies to, among other things:
• consult with airlines over charges. Where proper consultation processes are not
carried out, airport users can initiate legal proceedings (s. 4.2(a)); and
• disclosure of prices, terms and conditions for contractual arrangements, costs,
performance measures and financial performance indicators (s. 3BA).

Airport companies, like other companies in New Zealand, are subject to the general
competition law provisions under the Commerce Act. Several sections of the
Commerce Act are relevant to the regulation of airports, in particular Part IV, which
enables the Governor-General, on the recommendation of the Minister of
Commerce, to impose price controls in circumstances of restricted competition.9

9 Other sections of relevance include: Part II, which covers restrictive trade practices, and also
contains provisions relating to resale price maintenance and market dominance; and Part III,
which prevents acquisitions that would result in a dominant position or a strengthening of such
a position.
INTERNATIONAL 423
EXPERIENCES
Privatised airports and light-handed regulation

In 1989, the NZ Government commissioned a review into the implications of a

light-handed approach to regulation for the ‘actual and potential abuse of monopoly
power’ by airports. The review formed part of the Government’s deliberations into
an appropriate regulatory framework for the future privatisation of airports. The
review considered that a light-handed approach to regulation was appropriate
because:
• the major airport users, the airlines, had sufficient countervailing power to
counteract the monopoly power of the airport companies which would not be
significantly diminished by the sale of the airports to private interests;
• the transparency required by the Airports Authorities Act, through the consultation
process, contributed to the exercise of the airlines’ countervailing power and that
the current level of transparency could be maintained post-privatisation, with some
small changes to the legislation;
• there were unlikely to be net efficiency gains from privatising the airports if this
was accompanied by increased regulatory intervention; [and]
• the provisions of the Commerce Act, which provide for the application of price
controls, act as a constraint on airport pricing, while the prices of services to
passengers and visitors to the airport are constrained by actual or potential
competition and by other provisions of the Commerce Act … (PSA 1995, p. 15)

The NZ experience with economic regulation of airports

As discussed in chapter 11, there has been considerable debate about the
effectiveness of the NZ system. The ability of the system to limit the abuse of any
market power held by airports has been questioned and the costs of using the
system, both to airports and airport users, allegedly have been high. The regulatory
system appears to have been characterised by a high degree of uncertainty, both in
terms of its implementation and how it may evolve over time. It also appears that
this uncertainty might have contributed to the high costs of using the system.

Since 1989, there have been three reviews of the regulatory system. As the current
review commenced at around the same time as the privatisation of Auckland and
Wellington airports in 1998, there has been potential for strategic behaviour by both
airports and airport users alike in an attempt to influence the outcome of the review.

The regulatory system has resulted in a high incidence of litigation between airlines
and airport operators and (in some instances) the Government, in part over
interpretation of key elements of the legislative framework. The most well-
documented of these was a case instigated by Air New Zealand and a number of

424 PRICE REGULATION

OF AIRPORT
SERVICES
other parties10 against Wellington International Airport Limited (WIAL) in 1992
over the meaning of the term ‘consultation’ in the context of setting airport charges.
The action was brought about after a WIAL decision to increase landing charges,
which the airlines claimed adversely affected them. In essence, the airlines claimed
that WIAL had failed to allow them to put their case because it ended the
consultation process too quickly. After the High Court of New Zealand initially
found in favour of the airlines, an appeal heard by the Court of Appeal of New
Zealand reversed the decision in favour of WIAL (PSA 1995).

The Prices Surveillance Authority (PSA) noted:

The costs of litigation, including the uncertainty of outcomes, associated with the New
Zealand approach have been considerable. These costs appear to have been
underestimated in the development of the regulatory framework. Recourse to the Court
system has not been shown to be the most appropriate or cost effective way to solve the
consultation and pricing disagreements which have arisen. (PSA 1995, p. iii)

Litigation between airports and airlines has also been cited as evidence of the lack
of countervailing power of airlines. Under the regulatory system, it has been argued,
airlines faced with increased airport charges have had to resort to litigation
(PSA 1995). However, the regulatory regime may have promoted the use of
litigation. It is also possible that airlines have been exercising countervailing power
by initiating or threatening expensive litigation.

The Commerce Commission (CC) noted in general that ‘neither Auckland,

Wellington or Christchurch international airport[s] are likely to be significantly
constrained by the countervailing power of airlines’ (CC 2001b, p. 77).
Countervailing power of airlines at these airports was said to be constrained as,
unlike smaller regional airports, the airports are unlikely to be dependent on a small
number of airlines for their business.

A key component of the NZ regulatory framework was threat of explicit price

regulation acting as a deterrent to airports exploiting (any) market power in their
dealings with airlines. The uncertainty surrounding the regulatory framework
through the 1990s has, at various stages, raised questions about the credibility of
this threat. Some argued that during the mid-1990s there was very little likelihood
of price controls being introduced. In this regard, Kunz noted:
The NZ Commerce Act is primarily directed towards preventing restrictions on
competition not monopoly pricing and, therefore, provides no forceful provisions. It
indeed allows for price regulation to be invoked, but unless an expert regulation body

10 These were Qantas, Bilmans Management Limited (Ansett) and the Board of Airlines
Representatives of New Zealand.
INTERNATIONAL 425
EXPERIENCES
 with applicable instruments is at hand, the threat of regulation is not credible at all.
(Kunz 1999, p. 39)

The PSA also noted that the disinclination of the NZ Government to introduce price
regulation generally reduced the threat of any such regulation being applied to the
airports sector:
… both the Government and the Commerce Commission [in 1995] have made clear
their reluctance to implement or recommend price control, thereby undermining the
threat of its use. (PSA 1995, p. 11)

However, the CC’s current review of the need for price regulation at New Zealand’s
international airports might have made the threat of price regulation more credible.
Forsyth noted the threat of price controls, combined with the current review, ‘would
be a deterrent to [airport operators] using market power and setting high charges’
(2001, p. 12).

Preliminary findings of the NZ regulatory review

The light-handed approach to the economic regulation of some airports in New

Zealand currently is under review. The review commenced in 1998 when the NZ
Government directed the CC to examine whether price controls should be
introduced for airfield activities11 at Auckland, Wellington and Christchurch
airports. Broadly, the terms of reference require the CC to report on four issues:
• whether there is limited competition for airfield activities;
• whether control is necessary or desirable in respect of the charges for airfield
activities;
• whether market conditions are such that the Minister should make a
recommendation resulting in the imposition of control over such charges; [and]
• the nature and extent of any price control measures that would be introduced (where
applicable). (CC 2001a, para. 31)

Although the scope of the review is limited to airfield activities, the terms of
reference require the CC to consider the impact of other airport activities.
‘However, in reporting to the Minister, the [CC] … intends to confine its
recommendations to airfield activities’ (CC 2001a, para. 33).

11 The Airport Authorities Amendment Act 1997 defines airfield activities as one of three
identified airport activities, the others being: aircraft and freight activities; and specified
passenger terminal activities. Airfield activities cover facilities and services that enable the
landing and take-off of aircraft, including: the provision of airfields, runways, taxiways, and
parking aprons; facilities and services for air traffic and parking control; airfield lighting; and
services to maintain and repair airfields, runways, taxiways, and parking aprons for aircraft
(CC 2001a).
426 PRICE REGULATION
OF AIRPORT
SERVICES
The CC (2001b) released its draft report in July 2001. The preliminary
recommendation is for price controls to be introduced at Auckland Airport only. In
addressing the terms of reference for the review, the CC found there is evidence
that:
• airfield activities … are supplied or acquired in a market in which competition is
limited or is likely to be lessened; [and]
• there is evidence that it is necessary or desirable for the prices of the airfield
activities supplied by … [Auckland and Christchurch airports] to be controlled in
accordance with the Commerce Act in the interests of the acquirers of airfield
activities. (CC 2001b, p. xliv)

However, based on an assessment of the net efficiency benefits of introducing price

controls, the CC found that only airfield activities supplied by Auckland Airport
should be controlled.

Several participants raised concerns about the methodology used in the CC’s draft
report. These concerns related to the definition of the aeronautical asset base and its
valuation, the modelling of future returns and the dynamic efficiency gains that it
estimated would be achieved through price controls (chapter 11).

To consider further the substantive issues raised in response to the draft report, the
CC has delayed the release of its final report by six months (until the second quarter
of 2002). If price control is recommended in the final report, then a decision on the
type of model will be made after a consultation process as required by the
Commerce Act.

G.3 Germany
While most major German airports remain subject to some variant of cost-based
regulation based on a single till, the recent privatisation of airport infrastructure has
been associated with a move to incentive-based regulation. Nonetheless, Hamburg
is the only airport where incentive-based regulation has been introduced.12

Hamburg Airport was partially privatised in 2000.13 A new regulatory system was
adopted for the privatised airport, specified in a public contract between Hamburg
Airport and the Ministry of Economic Affairs of Hamburg. The contract is in place
for five years, from the beginning of 2000 to the end of 2004.

12 A price cap is planned to be introduced at Berlin Brandenburg International Airport when the
current privatisation process is completed.
13 Hochtief AirPort GmbH and Aer Rianta acquired a 36 per cent share, with an option for a
further 13 per cent. The City of Hamburg remains Hamburg Airport’s major shareholder.
INTERNATIONAL 427
EXPERIENCES
The basis of the regulation at Hamburg Airport is a dual-till CPI-X price cap, based
on revenue yield per passenger. The price cap applies to landing fees, passenger
handling fees, noise level charges and aircraft parking fees. The airport must also
attain quality of service targets under the regulatory framework. These include:
• availability of aircraft parking positions;
• availability and punctuality of passenger and baggage transport systems; and
• availability and quality of restaurants and retailers.

The distinctive feature of the price-cap regulation at Hamburg is that it is based on a

sliding scale. For the five-year period the price cap is in place, the X value is fixed
at 2. However, if in any year growth in passenger numbers at the airport exceeds a
threshold number, then the X factor is increased the following year. For every
percentage point increase in passenger growth greater than three, X is increased
by 0.5. The objective of the sliding scale is to reduce the possibility of the airport
achieving ‘windfall profits’. There is no provision for the X value to fall if
passenger numbers decline (Niemeier 2001).

Capital expenditure is dealt with outside the cap, with proposed projects having to
pass a public review process.

G.4 Denmark
The system of economic regulation employed at Copenhagen Airport is another
variant of price-cap regulation. The approach has been to implement a system that
creates the incentive properties of price-cap regulation but on a more informal basis
than those used at airports in the United Kingdom and Germany.

Copenhagen Airport is owned and operated by a publicly listed company,

Københavns Lufthavne A/S, although the national Government holds 51 per cent of
the shares in the company. The airport is not subject to an explicit price cap, but it is
not permitted to increase charges without approval from the national Government.
If the airport operator wishes to increase charges, then it has to submit detailed
information to the Ministry of Transport, justifying the increase. If airlines wish to
dispute the proposed charges, then they can also make a submission to the Ministry.

The approach appears to be designed to provide similar incentives to those that exist
with a price cap (by providing airports with an incentive to reduce costs, and
airlines with decreasing airport charges in real terms), with the advantage of lower
compliance costs compared with more formal price-cap arrangements. However,
because any increase in airport charges requires approval from the regulator —
which will involve, to some extent, an assessment of the costs, revenues and
428 PRICE REGULATION
OF AIRPORT
SERVICES
profitability of the airport — the system will be subject to some of the problems
associated with cost-based regulation discussed above (for example, the information
asymmetry and compliance cost issues discussed in section G.1).

While there appears to be no formal criteria against which proposals are judged, the
Ministry of Transport has stated that the airport should be allowed to make a
reasonable profit (TRL 2000b). Since 1992, there has only been one price increase
at Copenhagen Airport — a 15 per cent increase in 1999. This increase was
associated with the abolition of the availability of duty-free goods on intra-
European travel in June of that year.

G.5 Summary of international experiences

As the above case studies illustrate, there is no universal approach to economic
regulation of privatised airports internationally. The mix of regulatory and market
mechanisms adopted by governments, with respect to the provision of aeronautical
services at major airports, differs substantially between (and sometimes within)
countries.

Some of the experiences of the UK approach to regulation at designated airports

highlight a number of difficulties with maintaining price caps in the long term. The
CAA has sought to address some of these issues, such as investment incentives, in
its current review. To this end, it has recommended moving to dual-till price caps
and encouraging direct contracting between airports and their users.

The regulatory systems introduced by governments in Germany (Hamburg Airport)

and Denmark (Copenhagen Airport) represent, in part, different attempts to deal
with some of the incentive problems that arise under long-term, price-cap
regulation. The sliding-scale approach at Hamburg Airport, where the X values are
increased during periods when airports are likely to earn ‘windfall’ profits, is an
attempt to ensure that airport operators share profits with airport users through
lower prices. In contrast, in Denmark, by not having any systematic review of the
price cap at Copenhagen Airport, airport operators are able to retain any profits as
long as nominal prices remain unchanged. This reduces the incidence of
intervention by the regulator, and airports are free to pursue efficiency objectives, as
the benefits can be retained.

The NZ approach is unusual in that there is currently no direct economic regulation.

Instead, the system relies on the countervailing power of airlines and the threat of
direct price regulation should airport operators abuse any market power they hold.
There has been considerable debate about the effectiveness of the NZ system.
Although it has resulted in a high degree of litigation, this may reflect its design
INTERNATIONAL 429
EXPERIENCES
rather than an ineffectiveness of light-handed regulation. Further, despite some
shortcomings in its application, price and rate-of-return outcomes do not appear to
have been excessive, and price agreements have been reached between some
airports and their major customers (chapter 11).

The procedural problems experienced in New Zealand do not appear to have

occurred in the light-handed system of regulation employed in the United Kingdom
for non-designated airports holding CAA permission. The attributes of the UK
system, such as having a credible threat of future price regulation (as demonstrated
with the price-cap regulation at designated airports), combined with the complaints
system available to airport users (and others), appear to have been reasonably
successful in preventing the abuse of any monopoly power by airports and in
achieving efficient outcomes.

430 PRICE REGULATION

OF AIRPORT
SERVICES
H Issues in allocating scarce airport
slots

In its draft report, the Commission suggested, in relation to Sydney Airport, that
airport charges that cleared the market for landing slots were likely to promote more
efficient outcomes compared with non-price allocation mechanisms, including the
current slot allocation scheme. This conclusion elicited comments and criticisms
from several participants.

In particular, Network Economics Consulting Group (NECG), in a report

commissioned by the ACCC (sub. DR55, attachment A), argued that the
Commission, in reaching this conclusion inter alia had erred in not analysing the
problem as a ‘congestion externality’ and also (it claimed), in presuming that it
would be preferable for scarcity rents to accrue to airport operators rather than
airport users. The ACCC commented:
Many factors are relevant to the choice of demand management instruments. The PC
[Productivity Commission] has recommended a pricing approach, but without
considering the alternatives or the merits of those alternatives. (sub. DR55, p. 17)

This appendix explores in more detail some issues relating to efficient rationing of
scarce airport slot capacity. Though allocation of slots has particular relevance to
Sydney Airport (where hourly aircraft movements are restricted and a curfew
applies), other Australian airports experience excess demand for some facilities at
certain times of day or more occasionally. Capacity constraints also may become of
greater importance to them in the future whether due to exogenously-imposed
movement limits (eg curfews), environmental concerns or possible consequences of
price regulation on new investment.

H.1 Airports and congestion

As noted, NECG criticised the draft report for failing to analyse excess demand for
slots at Sydney Airport as a congestion externality issue. The Commission,
however, does not accept that this is the most appropriate framework in which to
analyse scarce slot capacity when the number of available slots has been fixed, and
where this constraint is binding.

ALLOCATING 431
SCARCE AIRPORT
SLOTS
Congestion refers to a situation where use of a facility by one party imposes costs
(delays etc) on other users, and vice versa. NECG defined congestion in a
submission to a concurrent Commission inquiry (Radiocommunications Acts) as:
… the impedence users impose on each other in conditions where the use of the system
by means of which the service is provided approaches its capacity limits. The ensuing
outcome reflects a negative externality, namely that the attempted consumption of an
additional unit will lower the utility derived from total realised consumption.
(NECG 2001a, p. 8)

Thus, while congestion is a means of rationing capacity, it may impose efficiency

costs by expanding use beyond the socially-optimal level. (The efficient level is set
where the marginal opportunity costs (including marginal congestion costs) equal
the marginal benefits of using the service.) Essentially, congestion arises when the
services produced by a facility are not, or feasibly cannot be, priced,1 or where
transaction costs prohibit effective negotiation between the parties that generate and
are affected by the congestion. In the absence of these ‘market’ solutions, taxes
could be imposed so that users are faced with the costs they impose on others, or the
socially-optimal level of use could be fixed by a quantity limit. In practice, the
difficulty is knowing the optimum level, while the costs of applying any ‘solution’
must also be weighed against the benefits gained.

In the case of Sydney Airport, capacity is restricted to a maximum of 80 movements

per hour (during non-curfew times), imposed by regulation. This limit largely
reflects community concerns about noise, loss of amenity and so on.2 It does not
appear to be directed at reducing congestion costs though it probably has this effect.
This is not to say that the slot limit sets the socially-optimal level of use of the
airport, after taking into account all social costs and benefits. But, it is the outcome
of a political process and the Government has stated that it will not be altered
(Anderson 2000).

Once the number of airport slots is fixed for whatever reason, and the slot limit is
binding, the issue is not one of additional use of the airport generating congestion
externalities, because additional use is not permitted. There may still be some
congestion at the airport consistent with the level of use allowed by the slot limit
but, at the margin, the slot limit (when binding) will set the level of use, not average

1 In other words whether property rights can be adequately defined and enforced (that is, whether
the services are ‘public’ or ‘private’). The services of airports seem to meet most of the
requirements of private goods.
2 Noise levies also apply to certain jet aircraft; these roughly average about $2 per arriving
passenger (chapter 3).
432 PRICE REGULATION
OF AIRPORT
SERVICES
costs (including average congestion costs).3 Additional units of demand do not add
to congestion; rather, additional units of demand are competing for a fixed number
of slots (Forsyth 1997). The problem becomes one of allocating those scarce slots.

Congestion may ration use of other airport facilities where access is not controlled
or directly priced — for example, on landside access roads and inside terminals. But
this does not seem to be a significant matter for most of these facilities, and easing
such congestion is in the interests and, in general, under the control of the airports.

H.2 Price or quantity rationing?

If landing and take-off intervals or ‘slots’ are limited by regulation, then any excess
demand for a particular slot or cluster of slots will require some form of rationing to
equate demand with supply. Figure H.1 depicts the market for slots over, say, a one-
hour period. DD is the demand for slots over a one-hour period. As drawn, the
availability of slots in any one-hour period is restricted to Q0Qs. The market-
clearing price for these slots, Pe, exceeds the cost of supplying them.4 The regulated
airport charge is Pa, assumed equal to average costs.5

Figure H.1 Market for peak-hour slots

$
D Hourly slot limit

Pe a

Pa b c
D (peak hour)

Q0 Qs Qe Aircraft movements

As drawn, the imposed quantity constraint generates a loss of surplus equal to the
triangle abc; or the cost of limited capacity, relative to unconstrained capacity. This

3 That is, average costs (including congestion costs) will not lie above the point where demand
equals the fixed supply of slots. Marginal costs may equal, lie above or below this point.
4 The market-clearing price of course will vary with shifts in demand.
5 This charge might have been set such that the airport is expected to cover its average costs,
given total anticipated utilisation levels of the airport over the relevant regulatory period. For
example, it may be a charge estimated on the basis of a cost building-block methodology.
ALLOCATING 433
SCARCE AIRPORT
SLOTS
loss occurs whichever method for eliminating excess demand is used. Of course,
there may be no such loss, when all factors are considered, if the constraint were
imposed to address congestion or other externalities. For example, if the restriction
reduced congestion, promoted safety or reduced noise, there may be a net social
benefit in restricting use to Q0Qs.

At issue is which mechanism should be used to allocate Q0Qs slots, given excess
demand equal to QsQe at price Pa. There are two broad options for rationing excess
demand: price rationing and non-price rationing. Non-price rationing can involve
either formal mechanisms such as administered allocation, or informal mechanisms
(essentially queuing). (Non-price rationing will be combined with some price
rationing if price Pa prevails. That is, if slots were free, then demand would exceed
Qe.) Price rationing involves pricing slots in some way. It could involve a higher
charge for each time a slot is used and/or the sale or lease of scarce slots at specified
times for a specified period.

Rationing by queuing

If slots are neither fully priced nor systematically allocated to users, then rationing
will occur automatically by queuing. For example, planes could be placed in
holding patterns or some could delay their departure from other airports. Queuing in
regulated airspace could be viewed as a form of congestion, though there are rules
preventing encroachment on the airspace of another plane. The main ‘externality’
will be that the average time required to reach (or depart from) the airport (at times
of excess demand) will increase, with delays and disruptions to schedules.

This method of equilibrating the market involves real resource costs — the value of
time spent queuing, the opportunity cost of aircraft and crews, and additional direct
costs (including the cost of fuel incurred by airlines). In effect, the market-clearing
price Pe will be paid (figure H.1), but with part of this price paid in money (Pa) to
the airport and the remainder (Pe minus Pa) paid in queuing costs.6 Passengers will
tend to pay the equilibrium price either as the value of their time forgone and/or
through higher fares reflecting additional costs incurred by airlines. Thus the
rectangle of surplus PaPeab is likely to be eroded entirely by queuing costs. Airport
operators would continue to receive price Pa and thus revenue equal to their total
costs.

6 Strictly speaking, the demand curve will differ from the normal demand curve when goods and
services are allocated by queuing etc. This demand curve will reflect willingness to queue
rather than willingness to pay.
434 PRICE REGULATION
OF AIRPORT
SERVICES
In addition, it is possible that some airlines and consumers who value the slots at
less than their market-clearing value (Pe as drawn) gain access at the expense of
those prepared to pay at least that market-clearing value. The additional costs per
passenger of circling above the airport for an aircraft, for example, may be roughly
the same as the additional costs incurred by a larger aircraft. This could make it
worthwhile for the small plane (and its passengers) to join the queue and compete
for the valuable slot. To the extent that users with lower valuations of the slots gain
access at the expense of those with higher valuations, additional surplus could be
forgone.

Administered allocation

In Australia, access to limited slots at Sydney Airport is regulated by a non-price

‘demand management’ scheme administered by Airport Coordination Australia
(chapter 3, box 3.4). Airline requests for slots are dealt with according to rules of
priority and slot availability. Officially, airlines cannot buy or sell allocated slots,
though they are permitted to swap slots (barter) with each other. Airlines may
forfeit slots if they are not used to a specified minimum level over a ‘season’ but,
essentially, the airline is given an indefinite right of access to the airport and the
future revenue stream associated with that right. (The value of this revenue stream
will depend critically on the future level of airport charges and slot availability.)
These arrangements are typical of slot allocation systems worldwide and reflect
international guidelines established by the International Air Transport Association
(IATA).

In addition, at Sydney Airport, the Commonwealth Government quarantines slots

for regional airlines — the ‘regional ring fence’. This means that a certain
proportion of available slots is reserved for regional flights. With slots allocated to
airlines in this way, the price paid by airlines to use the airport continues to be Pa in
figure H.1. Rather than allowing the price to rise, supply and demand are
equilibrated by administratively allocating slots among airlines.

Restricted airport capacity translates into restricted aircraft capacity — airlines

ration seats at the constrained times by increasing prices above costs. Some airlines
receiving a slot may service a pool of passengers (for example, business travellers at
peak times) who would be prepared to pay a fare that would at least yield the
market-clearing price (Pe) for that slot. Airlines in this category can be expected to
use price to ration seats on aircraft such that, on average, they increase average fares
on the route at periods of excess demand such that a typical flight at the busy period
earns a revenue premium equivalent to PaPe. (The UK Civil Aviation Authority
(CAA 2001c) observed that the cheapest quoted return airfares (on the same carrier)

ALLOCATING 435
SCARCE AIRPORT
SLOTS
from Dublin to Heathrow are KLJKHU WKDQ IDUHV IURP 'XEOLQ WR *DWZLFN 7KLV
difference is not explained by differences in airport charges.)

However, with slots allocated rather than priced to clear the market, some slot
holders are likely to service markets in which passengers, on average, have lower
valuations of the slot (eg tourists who are less time sensitive). Slot holders in this
category will also tend to use price to ration seats but they will receive an average
fare premium somewhat smaller than PaPe.

A formal quantity rationing scheme (where secondary slot trading is not permitted)
could achieve the same allocation as achieved by price rationing, so airlines with
passengers with the highest valuations obtain access at price Pa. If this were to
occur, airlines would capture the area of surplus PePaba in figure H.1. However, in
practice, even if such an efficient allocation were sought, it is highly unlikely that it
could ever be achieved. This is because there is no mechanism for gauging the slot
valuations of the various user airlines. Typically, however, such schemes are
designed to bring about a somewhat different allocation, reflecting community or
other objectives. Nonetheless, any move away from an ‘efficient’ allocation will
reduce potential economic surplus, and this cost has to be weighed against the
benefits of overriding that allocation.

Rent-seeking also is likely to be a source of efficiency loss. As noted above, if some

slot recipients carry passengers who place, on average, a relatively low value on the
slot (giving yields of somewhere between Pa and Pe), some potential surplus will be
forfeited. An airline that could capture some of this potential rent (by providing
more flights to passengers with a higher willingness to pay) will have an incentive
to obtain premium slots by swapping with other airlines or by employing other
means (for example, under-the-table payments, or even via company merger or
takeover). Such efforts, to the extent they are successful, will promote a more
efficient (although unlikely to be optimal) allocation of slots. Whether successful or
not, they are likely to involve real costs, such as the cost of negotiating swaps, that
dissipate potential surplus.

Airlines with slot allocations that are more valuable to others also may use up
resources (eg persist with inefficient schedules and inappropriate plane size) to
retain allocations for their future use or to prevent new entrants. They have an
incentive to do this if the cost of holding the slot is low relative to the value of the
slot to potential competitors. It also is possible that airlines use some of the rents
that accrue from slot allocations to cover fixed costs of other flights.

436 PRICE REGULATION

OF AIRPORT
SERVICES
Price rationing

Airport slots can be price rationed in various ways, ranging from charges for using a
slot, to the sale of future access rights to, for example, a daily or weekly slot or to a
bundle of such slots. As discussed below, the prices that airlines are willing to pay
for future slot rights will be highly sensitive to user charges levied by airports and
potential changes in slot capacity. In other words, defining future rights and
conditions of airport access may be difficult unless constraints are imposed, or
agreements reached, on future airport pricing and investment.

Charges for airport use

Traditionally, airports have charged for aircraft landings and sometimes for take-
offs. Thus they charge for use of the runway and related infrastructure. As discussed
in chapter 7, a profit-maximising airport operator has a strong incentive to introduce
efficient peak charges to ensure that the airport extracts scarcity rents. As also
discussed in chapter 7, as well as higher charges at peak times, efficient pricing of
scarce runway capacity is likely to require charges to be levied in a way that is
invariant to plane size (for example, a fixed fee per landing, unless plane size
affects time spent landing and taking off). This will facilitate access at such times of
larger planes with higher payloads at the expense of smaller ones with lower
payloads.

In figure H.1, and assuming that peak demand is stable (implications of demand
volatility are discussed below), if user charges (rather than the sale of slots) were
used to allocate scarce runway capacity, then the price of using a slot at the peak
would rise to Pe to clear the market.7 At this price, the airport would retain scarcity
rents equal to PePaba.

Whether the airport earned higher profits overall would depend on charges levied at
off-peak times. If the latter fall to their short-run marginal cost, overall, the airport
may continue just to cover total costs.8 In other words, revenues generated by the

7 This assumes independent demands between peak and off-peak times. If demands are
interdependent, then the price rise required at peak times may be somewhat less than this.
8 NECG (ACCC, sub. DR55, attachment A) suggested that off-peak charges could be pushed to
zero to ensure that the revenue constraint is not breached by peak prices. If the revenue
constraint is fixed, such restructuring may promote efficiency, by providing scope for even
higher peak prices and a greater differential between peak and off-peak charges. However, it is
difficult to see why off-peak users should pay a price below the short-run marginal cost of
operating the airport at these times, just to ensure that the airport does not make profits in
excess of ‘normal’ levels. Moreover, unless costs fall as a result, an airport will have little
incentive to restructure charges if allowable revenue is fixed.
ALLOCATING 437
SCARCE AIRPORT
SLOTS
optimal peak/off-peak price structure feasibly may not exceed cost-related revenue
constraints. However, there is no a priori reason that efficient peak and off-peak
pricing will or should generate this result.

If demand fluctuates from day to day, the market-clearing user charge will need to
adjust to clear the market. As it is unlikely that a ‘spot’ market in slots would be
feasible or efficient, airports are likely to set an average peak charge, probably
somewhat below Pe. This, then, would require a complementary form of rationing to
clear the market. As discussed below, if the airport owns or controls slot rights, it is
likely to devise contracts and pricing structures to address this issue. If the airport
did not own or control slot rights, the market could be cleared by queuing, formal
slot allocation (and slot swapping), or slot trading.

With peak user charges, those airlines and passengers with the highest willingness
to pay will obtain access. Passengers not prepared to pay the peak price will shift to
flights at different times, use other transport and communications modes or not
travel at all. Airlines operating flights that carry insufficient passengers willing to
pay the higher charges will reschedule or reduce these flights. Thus peak charges
will encourage rationalisation of flight schedules so that each flight carries enough
passengers prepared to pay on average the peak charge.

Compared with free allocation of slots, airlines may be less likely to engage in
strategies to keep competitors out. Not only would they have to incur the costs of
flying the route, but also the higher airport landing charge. However, it is feasible
that one airline could buy all available slots with a view to monopolising the
industry if access to the slot-constrained airport were critical to network viability.

It also should be noted that allowing an airport to charge market-clearing peak

prices effectively would undermine any slot allocation scheme in place unless, for
example, some flights were quarantined from the higher charges. Only those airlines
operating flights with passengers willing to pay (on average) the higher charges
would wish to retain their slots.

Sale or lease of future slot rights

Under the existing slot allocation scheme operating at Sydney Airport and in most
other countries, slots can be swapped but, officially, not traded for cash. However,
slot trading increasingly has been mooted as one way of promoting more efficient
slot allocations. In addition, there has been discussion of selling rather than giving
slots to airlines.9

9 For a discussion of issues relating to the introduction of slot auctions and slot trading see
PC (1998a, chapter 8) and Ewers et al (2001).
438 PRICE REGULATION
OF AIRPORT
SERVICES
If airlines value certainty of airport access (for example, to facilitate scheduling),
and property rights for slots could be adequately defined (in the sense of defining
future rights of access to slots at an airport), slot rights could be leased or sold to
airlines.10 Airlines then would be prepared to pay the discounted value of
anticipated scarcity rents (that is, the additional expected fare revenue) for slots. As
with market-clearing user charges, slot sales generally would ensure that slots were
obtained by those who valued them most.

Future slot rights could be leased or sold to airlines by a body other than the airport
(so that, for example, scarcity rents accrued directly to the government). However,
in bidding for slots, airlines would take into account anticipated airport user
charges; any increases in user charges would reduce the value of a slot. They also
would need to take into account the potential for airport expansion, which also
could undermine the scarcity premium. In other words, as already noted, defining
and pricing rights for future airport slots is inextricably linked to expected future
airport pricing and investment behaviour.

If property rights for slots were vested in the airport, the airport operator might
introduce a form of two-part pricing, whereby airlines reserved access to the airport
at a certain time by purchasing a slot (with these access prices reflecting slot
scarcity), and pay a user charge (reflecting airport operating costs) when exercising
that access right. Again, how much airlines were prepared to pay for access at a
specified time would depend on the user charge and the potential for changes in slot
availability. Consequently, airports would have an incentive to sell a ‘package’
combining clearly-defined slot rights and future user charges. Whether particular
slots were reserved or leased (and for what period), sold outright or neither leased
nor sold, would depend on an assessment by airlines of the benefits and costs
associated with holding future slot rights versus not reserving access and paying on
the basis of airport use.

If property rights for future slots were vested with airlines (that is, if slots were
given to airlines on an indefinite basis or for a fixed period), and trading of these
slots permitted, the traded price of slots also would tend to reflect their scarcity
value and encourage transfer to their most efficient use. The incentive for an airline
to use a slot inefficiently would be reduced (compared to the situation where slot
trading is not permitted) because that slot could be sold to another airline. However,
incumbent airlines (which have been given slots) may retain an advantage over new
airlines (which must buy slots). Even though the opportunity cost of holding a slot

10 The ACCC (sub. DR55) refers to these allocation mechanisms as market-clearing mechanisms
rather than price mechanisms. But the question is not whether the market will clear. Rather, the
question is which variable adjusts to equilibrate supply and demand. In the case of slot sales,
prices for slots adjust to clear the market.
ALLOCATING 439
SCARCE AIRPORT
SLOTS
would increase (reflecting its cash value), incumbent airlines could use the rents
they derive from slots to drive down network prices in order to deter new airline
entry. In essence, slot quota rents could give incumbents deeper pockets than new
entrants. As noted above, slot acquisition could also be used as a monopoly
strategy.

H.3 Assessing the options

Desirable properties for an allocation mechanism include that it promotes efficient
use of a scarce resource and that it appropriately signals the need for, and
encourages, efficient investment to augment that scarce resource.

Queuing

Queuing generally can be ruled out as an efficient rationing mechanism because it

uses up real resources to clear the market. It also provides very poor signals about
the need for new investment because there is no way of gauging the willingness of
queuers to pay for the scarce good or service or for the airport to extract this value
from them when prices are regulated. Only their willingness to queue can be
ascertained.

Non-price administered allocation

A non-price demand management scheme (where slot trading is not permitted) has
advantages over queuing to the extent that the value airlines place on the slot
exceeds the regulated price. Thus, some of the potential surplus (represented in
figure H.1 by the area dPaba) will not be forfeited. Instead, it will tend to accrue to
airlines that hold the slots and that ration, through prices, available seats among
passengers. In principle, while an administered allocation could mimic the outcome
under price rationing such that all of this surplus is retained, in practice it is highly
unlikely that all scarce slots would accrue to those airlines and passengers who
value them most highly. This may be deliberate policy, but unless the allocation
addresses a market distortion, it comes at an efficiency cost. Also, as mentioned
above, as with any quota allocation scheme, slot allocation schemes create
incentives for wasteful rent-seeking behaviour by airlines in a bid to obtain or retain
valuable slots.

Moreover, with administered allocation (as with queuing), it is very difficult to

gauge the need for new airport investment. There is little direct indication to the
airport (or, indeed, the government or a regulator) about users’ willingness to pay

440 PRICE REGULATION

OF AIRPORT
SERVICES
for airport services and hence whether the benefits from investment will exceed the
costs. The existence of excess demand for an under-priced service says very little
about whether using up costly resources to expand the facility, or to build a new
facility, would be privately and socially efficient. The optimal ‘solution’ may be
just to ration existing capacity more efficiently.

Price rationing

Allocating scarce resources via the price mechanism generally will ensure that those
with the highest willingness to pay, obtain slots. This could occur whether pricing
occurs via charges for use of the airport or via slot sales and trading (provided
property rights for future airport access can be contracted).

As noted above, however, it is possible, if slots initially were given to incumbent

airlines while new airline entrants had to purchase slots, that the incumbent would
have ‘deeper pockets’, which could be used to frustrate market entry. Indeed, the
incumbent or some other party could buy up all slots. This could lead to inefficient
outcomes. On the other hand, the potential for a cash sale increases the opportunity
cost of holding a slot.

The ACCC suggested that price rationing using airport user charges is likely to be a
very difficult exercise:
One of the disadvantages of the approach to congestion pricing recommended by the
PC is the substantial information required to set efficient prices. This is not a trivial
issue. To the contrary it is central to establishing workable regulatory arrangements.
Furthermore the potential efficiency losses of getting prices wrong could be substantial.
(sub. DR55, p. 17)

Though pricing is a difficult exercise for most industries in the economy, producers
generally have an incentive to devise price structures that increase their profits.
They also are likely to have better information than most others (except the users)
about demand for their product. As discussed in chapter 7, an airport operator facing
an exogenously-imposed capacity constraint will have an incentive to devise
efficient prices to increase profits. Indeed, for this reason, an airport operator is
likely to devise an efficient peak/off-peak price structure (including marginal off-
peak users paying prices equal or close to short-run marginal cost). Implementing
this structure probably will require iterations from time to time (for example, by
adjusting discounts/rebates), but this need arises in any market.

That said, as discussed earlier, demand for slots is likely to change from day to day,
thus causing the market-clearing price to change. It is unlikely that clearing the
market for slots by continual iterations in airport user charges would be efficient. In

ALLOCATING 441
SCARCE AIRPORT
SLOTS
particular, many airlines are likely to value certainty of access to the airport and
predictable (maximum) future prices. Therefore, efficiency is likely to require user
charging combined with some form of slot reservation mechanism. This could be
achieved by long-term contracts and pricing structures devised by the airport or
airport user charges complemented by an independent slot trading mechanism.

Scope for flexible price structures could be a major advantage of airports

controlling slots. Some airlines (sometimes) may choose just to pay as they use the
airport, others may wish to enter a contract with the airport to reserve a slot (some
for a short period, others for a long period or indefinitely).

As for efficiency over time, the main requirements are that current prices signal the
strength of demand for a service, and that the provider has an incentive to invest
appropriately, including some expectation of making a reasonable return on the
investment required to meet that demand.

The CAA (2001c, p. xiv), while concluding that a ‘price cap at market-clearing
levels would be likely to improve the use of scarce capacity’, also noted that ‘while
market-clearing prices may provide good signals to the airports of where new
capacity is desired, they may not provide good incentives to actually deliver it at the
socially desirable time’.

In short, while higher airport charges may signal clearly the need for new
investment, that investment may not be forthcoming if the airport retains scarcity
rents. As discussed in chapter 7, under certain assumptions, an airport with market
power will under-invest to maximise monopoly profits over time. This does not
mean that an airport operator will never invest, just that investments will be delayed
to maximise profits. Nonetheless, to the extent that price discrimination is feasible
and/or additional throughput increases non-aeronautical profits, even an airport with
market power will have some incentive to invest efficiently in additional
aeronautical capacity. The threat of potential competition (from existing or new
airports) also may encourage more efficient investment. It is interesting to note that
physical capacity constraints at some airports overseas appear to have arisen with
regulated prices held below market-clearing levels and apparently below the
incremental costs of expanding capacity.

Where a capacity constraint is imposed exogenously to reduce aircraft noise levels,

for example, additional runway capacity at that airport may not be considered
socially desirable. This may be the case at Sydney Airport. In these circumstances,
market-clearing prices will provide information to government and prospective
investors about the viability of a second airport. They also will provide information
about the costs of the constraints and any policy that quarantines some airport users
from market-clearing charges.
442 PRICE REGULATION
OF AIRPORT
SERVICES
Where airlines, rather than airports, obtain scarcity rents, as they do under current
allocative arrangements, it is doubtful that investment signals and incentives
necessarily will be superior. Certainly when slots cannot be bought and sold, there
is virtually no public information about the scarcity value of those slots to airlines
— revenue yields on particular flights are not available to airports or
governments.11

If future slot rights could be bought and sold, there would be some indication of
scarcity rents and the willingness to pay for those slots. While this may assist in
assessing whether new investment would be socially desirable, the airport will
invest only if the regulated price will provide a reasonable return on that
investment. Also, just as airports may have an incentive to delay investment,
airlines holding slot rights would have an incentive to lobby against any expansion
of airport capacity that may undermine their value.

Alternatively, scarcity rents could accrue to governments through slot sales to

airlines or, as suggested by the Commission in relation to Sydney Airport, be
factored into the sale price for that facility if bidders knew in advance they could
charge market-clearing prices.12 In both cases, prices would give some indication
about the need for investment (in Sydney’s case, the need for facilities at another
site) and, provided the airport operator could obtain a reasonable return on
investment, efficient investment would be forthcoming. Nonetheless, issues of
sovereign risk could arise if the slot arrangements forming the basis for payments to
the government were subsequently changed.

11 In the USA, however, airlines are required to provide information to government about average
route yields.
12 The situation at Sydney Airport is somewhat unusual in that additional airport capacity is likely
to be developed at another facility. So key objectives for pricing at Sydney Airport are
promotion of efficient use of that constrained facility and signalling (to government) when a
second facility should be built. If bidders were advised they could set market-clearing prices
(and made aware that the government would choose when construction of the second airport
would commence), bidders would tailor their bids accordingly.
ALLOCATING 443
SCARCE AIRPORT
SLOTS
References

AAL (Adelaide Airport Ltd) 2000, Annual Report 1999-2000.

Abbott, M. and Wu, S. 2001, ‘Total factor productivity and efficiency of Australian
airports’, Melbourne, unpublished.
ACCC (Australian Competition and Consumer Commission) 1996, Statement for
the Public Register on Proposed Aeronautical Charges at FAC Airports,
September.
—— 1997a, Administration of Airport Price Cap Arrangements, January.
—— 1997b, Quality of Service Monitoring for Airports, Post-Leasing, February.
—— 1998a, Economic Regulation of Airports: An Overview, July.
—— 1998b, Fuel Throughput Levies: Report Pursuant to the Commission’s
Monitoring Functions under the Prices Surveillance Act 1983, December.
—— 1998c, Melbourne Airport Access Undertaking: Draft Determination, May.
—— 1998d, Perth Airport Access Undertaking: Draft Determination, May.
—— 1998e, Quality of Service Monitoring for Airports: Statement of the ACCC’s
Approach to Analysis, Interpretation and Publication of Quality Information,
February.
—— 1998f, Regulatory Information Requirements under Part 7 of the Airports
Act 1996 and Sections 21 and 27A of the Prices Surveillance Act 1983:
Guideline, Version no. 2, September.
—— 1998g, Section 192 of the Airports Act: Declaration of Airport Services —
Draft Guide, AusInfo, Canberra.
—— 1998h, Statement for the Public Register on Proposed Aeronautical Charges
at Sydney (Kingsford Smith) Airport.
—— 1999a, ACCC declares vehicle access services at Melbourne Airport, Media
Release no. 64/99, 17 May.
—— 1999b, Adelaide Airport: Proposal to Pass Through the Price Cap the Costs
of a Multi-User Integrated Terminal — Decision, October.
—— 1999c, Draft Statement of Principles for the Regulation of Transmission
Revenues, May.

REFERENCES 445
—— 1999d, Delta Car Rentals Request for Determination — Statement of Reasons,
April.
—— 1999e, Regulatory Report: Brisbane Airport, 1998-99, December.
—— 1999f, Regulatory Report: Perth Airport, 1998-99, December.
—— 2000a, Infrastructure Industries: Aviation, May.
—— 2000b, New Investment Costs Pass-through, Position Paper, April.
—— 2000c, Perth Airport: Proposal to Increase Aeronautical Charges to Recover
the Costs of Necessary New Investment — Final Decision, April.
—— 2000d, Regulatory Reports: Phase II, 1998-99, March.
—— 2000e, Submission to the Productivity Commission’s Review of the National
Access Regime, Sub. 25, December.
—— 2000f, Brisbane Airport, Proposal to Increase Aeronautical Charges to
Recover the Costs of Necessary New Investment — Decision, April.
—— 2000g, Melbourne Airport, Multi-User Domestic Terminal — New Investment
Decision, August.
—— 2001a, Melbourne Airport, Necessary New Investment: Taxi Charge Proposal
— Decision, May.
—— 2001b, Regulatory Report: Brisbane Airport, 1999-00, April.
—— 2001c, Regulatory Report: Melbourne Airport, 1999-00, April.
—— 2001d, Regulatory Report: Perth Airport, 1999-00, April.
—— 2001e, Regulatory Report: Sydney Airport, 1999-00, April.
—— 2001f, Regulatory Reports: Phase II, 1999-00, April.
—— 2001g, Review of Price Control Arrangements, February.
—— 2001h, Sydney Airports Corporation Ltd: Aeronautical Pricing Proposal —
Draft Decision, February.
—— 2001i, Sydney Airports Corporation Ltd: Aeronautical Pricing Proposal —
Decision, May.
—— 2001j, Virgin Blue Request for a Determination, Issues Paper, March.
—— 2001k, Melbourne Airport, Proposal to Increase Aeronautical Charges as per
Direction 24, Issued by Federal Minister, Joe Hockey, Final Decision,
November.
—— 2001l, Draft Determination, Pursuant to section 192 of the Airports Act 1996,
Application by Virgin Blue in Respect of Certain Terminal Services at
Melbourne Airport, October.
446 REFERENCES
—— 2001m, Regulatory Report: Sydney Airport, 2000-01, November.
—— 2001n, Sydney Airport charges, Media Release no. 199/01, 23 August.
ACT (Australian Competition Tribunal) 2000, Sydney International Airport
[2000] AcompT 1, 1 March, http://www.austlii.edu.au/au/cases/cth/ACompT/
2000/1.html (accessed 6 April 2001).
ADR (Aeroporti di Roma) (no date), Areas of Business, http://www.adr.it/en/soc/
societa-03.html (accessed 24 May 2001).
Aer Rianta (no date), Current Structure, http://www.aer-rianta.ie/htmlonly/ar/htm/
struct/index.htm (accessed 24 May 2001).
AIAL (Auckland International Airport Limited) 2001, Submission to Commerce
Commission Draft Report (July 2001) and Preliminary Recommendations —
Airfield Price Control Inquiry, August.
Anderson, J. (Deputy Prime Minister and Minister for Transport and Regional
Services) 2000, Sydney’s future airport needs, Media Release no. A197/2000,
13 December.
—— 2001a, Tesna Consortium Bid for Ansett Australia, Media Release no.
A257/2001, 27 November.
—— 2001b, Sub. to ACCC in response to Sydney Airport 2000 pricing proposal,
12 January, p. 1.
—— and Gosche, M. (New Zealand Minister for Transport and Minister
Responsible for Civil Aviation) 2000, Australia – New Zealand open skies
agreement, Joint Media Release no. A182/2000, 20 November.
—— and Kemp, R. (Acting Minister for Finance and Administration) 2001, Sale of
Sydney (Kingsford-Smith) Airport, Joint Media Release no. A11/2001,
29 March.
APAC (Australia Pacific Airports Corporation) 1998, Delta Car Rentals, Request
for Determination Under Section 192 of the Airports Act, Submission to the
ACCC, December.
—— 2000a, Annual Report 1999-00.
—— 2000b, Submission to the Productivity Commission’s Review of the National
Access Regime, Sub. 10, December.
—— 2001, Review of the National Access Regime: Comments on Position Paper,
Submission to the Productivity Commission’s Review of the National Access
Regime, Sub. DR60, May.
ASA (Airservices Australia) 2000, Charges for Facilities and Services: Standard
Contract Terms, 1 July, Canberra.

REFERENCES 447
Averch, H. and Johnson, L. 1962, ‘Behavior of the firm under regulatory
constraint’, American Economic Review, vol. 52, no. 5, pp. 1052–69.
BAC (Brisbane Airport Corporation) 2001, Airport announces release of new land,
Media Release, 22 February.
Ballantyne, T. 2001, ‘Terminal ailment’, Bulletin with Newsweek, 23 October,
pp. 84–5.
BARA (Board of Airline Representatives Australia) 2000, Submission to the ACCC
Regarding Sydney Airport Corporation Limited’s Draft Pricing Proposal,
November.
Barbeliuk, A. 2001, ‘Rush for cheap seats as Kendell takes flight’, (Hobart)
Mercury, 8 October, p. 9.
Barlow, C. 2001, No increase in international airline charges, Media Release,
Melbourne Airport, 5 November.
Bartholomeusz, S. 2001, ‘Domestic key to Qantas’ future’, Age — Business News,
6 October, p. 1.
Battersby, B. and Oczkowski, E. 2001, ‘An econometric analysis of the demand for
domestic air travel’, International Journal of Transport Economics, vol. 28,
no. 2, pp. 193–204.
Baumol, W.J., Panzar, J.C. and Willig, R.D. 1982, Contestable Markets and the
Theory of Industry Structure, Harcourt Brace Jovanovich, New York.
Betancor, O. and Rendeiro, R. 1999, ‘Regulating privatized infrastructures and
airport services’, World Bank Policy Research Working Paper no. 2180,
September.
BTCE (Bureau of Transport and Communications Economics) 1995, Demand
Elasticities for Air Travel to and from Australia, Working Paper no. 20, AGPS,
Canberra.
BTR (Bureau of Tourism Research) 2000a, International Visitors in Australia,
1999, Canberra.
—— 2000b, Travel by Australians 1999: Annual Results of the National Visitor
Survey, Canberra.
Button, K. and Stough, R. 2000, Air Transport Networks: Theory and Policy
Implications, Edward Elgar Publishing, Cheltenham, United Kingdom.
CAA (Civil Aviation Authority) 2000a, Easyjet Application for Designation of
Luton Airport, London, http://www.caaerg.co.uk (accessed 7 June 2001).
—— 2000b, Economic Regulation of Airports — General Guidance, London,
http://www.caaerg.co.uk (accessed 7 June 2001).

448 REFERENCES
—— 2000c, Issues for the Airport Reviews, Consultation Paper, London,
http://www.caaerg.co.uk (accessed 7 June 2001).
—— 2000d, Quality of Service Issues, Consultation Paper, London,
http://www.caaerg.co.uk (accessed 7 June 2001).
—— 2000e, The CAA Approach to Economic Regulation and Work Programme for
the Airport Reviews, Position Paper, London, http://www.caaerg.co.uk (accessed
1 June 2001).
—— 2000f, The ‘Single Till’ and the ‘Dual Till’ Approach to the Price Regulation
of Airports, Consultation Paper, London, December.
—— 2000g, The Use of Benchmarking in the Airport Reviews, Consultation Paper,
London, http://www.caaerg.co.uk/ (accessed 7 June 2001).
—— 2001a, Direct Contracting Between Airport Users: A Default Price Cap,
Consultation Paper, London, http://www.caaerg.co.uk (accessed 7 June 2001).
—— 2001b, Pricing Structures and Economic Regulation, Consultation Paper,
London, http://www.caaerg.co.uk (accessed 7 June 2001).
—— 2001c, Heathrow, Gatwick, Stansted and Manchester Airports’ Price Caps
2003–2008, Preliminary Proposals, Consultation Paper, November.
CC (Commerce Commission, New Zealand) 2001a, Price Control Study of Airfield
Activities, Critical Issues Paper, Wellington, March.
—— 2001b, Price Control Study of Airfield Activities at Auckland, Wellington, and
Christchurch International Airports, Draft Report, Wellington, July.
—— 2001c, Airport price control report to be finalised in 2002, Media Release
no. 2001/108, 2 November.
Chanticleer 2001, ‘Airport bids growth conundrum’, Australian Financial Review,
12 June, p. 72.
Cheesman, B. 2001, ‘Malaysia’s aviation goal up in the air’, Australian Financial
Review, 1 August, p. 9.
CIAL (Christchurch International Airport Limited) 2001, Response to Commerce
Commission Draft Report, August.
Commonwealth of Australia 1998a, Airport lease for Townsville airport.
—— 1998b, Crown lease, Airport lease for Canberra airport.
Costello, P. (Treasurer) 1997, Prices monitoring at privatised airports, Press
Release no. 69, 27 June.
—— 1998, Prices oversight of privatised airports, Press Release no. 55, 25 May.

REFERENCES 449
—— and Anderson, J. (Minister for Transport and Regional Services) 1999,
International air services, Joint Media Release no. A79/99, 3 June.
Cotterill, R.M. 1999, Experience of price caps in UK airport regulation, Speech
presented at the Incentive Regulation and Overseas Developments Conference,
ACCC, Sydney, 18–19 November.
CPA (Cairns Port Authority) 2000, Annual Report 1999-00, Cairns, Queensland.
Crew, M.A. and Kleindorfer, P.R. 2001, Regulation for Privatized Airports: Single-
Till Versus Multi-Till Pricing Methodologies for Sydney Airport, Paper
submitted to the ACCC’s review of Sydney Airport’s aeronautical pricing
proposal.
Davidson, K. 2001, ‘Essendon’s value is not as an airport’, Age, 19 April, p. 17.
Doganis, R. 1992, The Airport Business, Routledge, London.
DoTRD (Department of Transport and Regional Development) 1996, Pricing
Policy Paper, Canberra, November.
DoTRS (Department of Transport and Regional Services) 2000a, Air Transport
Statistics, International Scheduled Air Transport, 1999-00, AVSTATS,
Canberra.
—— 2000b, Air Transport Statistics: Regular Public Transport Services, Airport
Traffic Data, 1989-90 – 1999-00, AVSTATS, Canberra.
—— 2000c, Australian airport operations in the context of privatisation,
Presentation to the ICAO ANS conference, Montreal, June.
—— 2001, Domestic Airlines Monthly Activity Report, March 2001, AVSTATS,
Canberra.
Europe Economics 2001, Airport Cost Allocation: Report for the CAA by Europe
Economics, London, April.
Ewers, H. et al 2001, Possibilities for the Better Use of Airport Slots in Germany
and the EU: A Practical Approach, Commissioned by Hochtief AirPort GmbH,
Berlin, January.
FAC (Federal Airports Corporation) 1996, Annual Report 1995-96, Sydney.
—— 1997, Annual Report 1996-97, Sydney.
—— 1998, Annual Report 1997-98, Sydney.
Federal Court of Australia 2001a, Canberra International Airport Pty Ltd v
Australian Competition and Consumer Commission [2001] FCA 289, 23 March.
—— 2001b, Canberra International Airport Pty Ltd v Australian Competition and
Consumer Commission [2001] FCA 1172, 24 August.

450 REFERENCES
Ferguson, A. and Meyrick, M. 2001, ‘Ansett II hovering’, Business Review Weekly,
18 October, p. 37.
Forsyth, P. 1997, ‘Price regulation of airports: principles with Australian
applications’, Transportation Research-E, Great Britain, vol. 33, no. 4, pp. 297–
309.
—— 1999, Regulating access to airport facilities, Paper presented at the Air
Transport Research Group Conference, City University of Hong Kong,
6–8 June.
—— 2001, Privatisation and regulation of Australian and New Zealand airports,
Paper presented at 4th Hamburg Aviation Conference, ‘Regulation versus
Competition? Strategic Airport Management’, Hamburg University,
14–16 February.
Harris, P. (First Assistant Secretary, Aviation Policy, Department of Transport and
Regional Development) 1997, Post privatisation regulatory environment,
Synopsis of presentation to ‘Airports 97 Future Perspectives Conference’,
Melbourne, 11 August.
Heasley, A. 2001, ‘Railing against air chaos’, Age, 26 September, p. 4.
Hockey, J. (Minister for Financial Services and Regulation) 2001, Prices oversight
arrangements at airports, Media Release no. FSR/081, 5 October.
HoR (House of Representatives, Australia) 1995, Debates, vol. HR202, pp. 2793–
2800.
—— 1996, Debates, vol. HR206, pp. 1305–1311.
Huttner, D. 2001, Virgin Blue raises concerns about Gold Coast Airport charges
having an adverse effect on tourism, Media Release, Virgin Blue, Brisbane,
10 October.
IATA (International Air Transport Association) 2001, About Us: Mission and
Goals, http://www.iata.org/missgoal.htm (accessed 11 April 2001).
IC (Industry Commission) 1992, Intrastate Aviation, Report no. 25, AGPS,
Canberra.
ICAO (International Civil Aviation Organization) 1996, Manual on the Regulation
of International Air Transport.
Innis, M. 2001, ‘Lucky break’, Sydney Morning Herald, 10 October, p. 6.
Juan, J. 1995, Airport Infrastructure: the Emerging Role of the Private Sector —
Recent Experiences Based on 10 Case Studies, Cofinancing and Financial
Advisory Services Discussion Paper Series no. 115, World Bank, Washington.

REFERENCES 451
Kahn, A.E. 1988, The Economics of Regulation: Principles and Institutions,
vols 1–2, MIT Press, Cambridge, Massachusetts.
—— 1991, Statement on Behalf of United Kingdom of Great Britain and Northern
Ireland in US/UK Arbitration Concerning Heathrow Airport User Charges,
April.
—— 2001a Evidence on Behalf of Sydney Airports Corporation, NERA, New
York, January.
—— 2001b, Statement on Behalf of Auckland International Airport Ltd, August.
Kanafani, A. and Lan, L. 1988, ‘Development of pricing strategies for airport
parking — a case study at San Francisco Airport’, International Journal of
Transport Economics, vol. 15, no. 1, pp. 55–76.
Kemp, R. (Assistant Treasurer) 2000, Productivity Commission inquiry into prices
oversight arrangements at airports, Press Release no. 056, 21 December.
Kew, I. 2001, ‘Why airport car park fee increased’, Northern Territory News, Letter
to the editor, 6 November, p. 10.
King, S. 2000, ‘Access: what, where and how’, in Productivity Commission and
Australian National University, Achieving Better Regulation of Services,
Conference Proceedings, AusInfo, Canberra, pp. 63–93.
Kunz, M. 1999, ‘Airport Regulation: the Policy Framework’, in Pfähler, W.,
Niemeier, H. and Mayer, O.G. (eds), Airports and Air Traffic: Regulation,
Privatisation and Competition: [Economic Forum of the Free and Hanseatic
Hamburg], Peter Lang, Frankfurt, pp. 11–55.
—— 2001, Airport regulation: lessons from the UK, Paper presented at the
4th Hamburg Aviation Conference, ‘Regulation versus Competition? Strategic
Airport Management’, Hamburg University, 14–16 February.
Layard, P.R.G. and Walters, A.A. 1978, Microeconomic Theory, McGraw Hill
Book Company, New York.
Law Council of Australia 2001, Legislation Review of Clause 6 of the Competition
Principles Agreement and Part IIIA of the Trade Practices Act 1974, Submission
to the Productivity Commission’s Review of the National Access Regime,
Sub. 37, January.
Littlechild, S.C. 1983, Regulation of British Telecommunications Profitability: A
Report to the Secretary of State for Trade and Industry, Department of Trade
and Industry, London.
Melbourne Airport 2001, Introduction of Taxi Parking Charge, March.

452 REFERENCES
NCC (National Competition Council) 1997, Applications for Declaration of Certain
Airport Services at Sydney and Melbourne International Airports: Reasons for
Decision, 8 May.
—— 2001, Legislation Review of Clause 6 of the Competition Principles
Agreement and Part IIIA of the Trade Practices Act 1974, Submission to the
Productivity Commission’s Review of the National Access Regime, Sub. 43,
January.
NECG (Network Economics Consulting Group) 2000a, ‘Dual Till’ at Sydney
Airport, Report prepared for the ACCC, Final Report, May.
—— 2000b, Land Valuation at Sydney Airport, Report prepared for the ACCC,
Final Report, May.
—— 2001a, Submission to the Productivity Commission’s Review of the
Radiocommunications Acts and the Role of the Australian Communications
Authority, Sub. 73, November.
—— 2001b, Virgin Blue Request for Determination: An Economic Analysis, Report
prepared for the ACCC — Final Report - July, in ACCC 2001l, pp. 29–49.
NERA (National Economic Research Associates) 2001a, Global Energy
Regulation, Newsletter, Issue no. 24, http://www.nera.com (accessed 30 June
2001).
—— 2001b, Price Control Study of Airfield Activities: A Critique, Report for
Auckland International Airport Limited, Sydney, August.
Niemeier, H. 2001, Regulating airports, Paper presented at the 4th Hamburg
Aviation Conference, ‘Regulation versus Competition? Strategic Airport
Management’, Hamburg University, 14–16 February.
NTTC (Northern Territory Tourist Commission) 2000, Territory Tourism Selected
Statistics 1999/2000, http://www.nttc.com.au/statistics/g_method_of_entry.htm
(accessed 20 June 2001).
OECD (Organisation for Economic Co-operation and Development) 2000, Airline
Mergers and Alliances, Committee on Competition Law and Policy,
Competition Policy Roundtables no. 26, DAFFE/CLP (2001)1, Paris.
Ordover, J. and Saloner 1989, ‘Predation, Monopolization, and Antitrust’ in
Schmalensee, R. and Willig, R.D. (eds), Handbook of Industrial Organization
Volume 1, Elsevier Science Publishers B.V., Amsterdam, pp. 537–96.
Oum, T.H.O., Waters, W.G. and Yong, J. 1992, ‘Concepts of price elasticities of
transport demand and recent empirical estimates’, Journal of Transport
Economics and Policy, vol. 26, no. 2, pp. 139–54.

REFERENCES 453
PC (Productivity Commission) 1998a, International Air Services, Report no. 2,
AusInfo, Canberra.
—— 1998b, Regulation and its Review 1997-98, AusInfo, Canberra.
—— 2001a, Review of the National Access Regime, Report no. 17, AusInfo,
Canberra, (forthcoming).
—— 2001b, Review of the Prices Surveillance Act 1983 draft report public hearing
transcripts, 8 June, Melbourne.
—— 2001c, Review of the Prices Surveillance Act 1983, Report no. 14, Canberra,
(forthcoming).
—— 2001d, Telecommunications Competition Regulation, Report no. 16, Canberra.
—— 2001e, Structural Adjustment — Key Policy Issues, Commission Research
Paper, AusInfo, Canberra.
Pitchford, R. 2000, Sydney Airport Land Valuation: An Assessment, ACCC
Consultancy Report, http://www.accc.gov.au/airport/fs-air.htm (accessed 20
June 2001).
PSA (Prices Surveillance Authority) 1993, Inquiry into the Aeronautical and Non-
Aeronautical Charges of the Federal Airports Corporation, Report no. 48,
Melbourne.
—— 1995, Regulation of Airport Pricing — Is the New Zealand Approach
Applicable to Australia?, Discussion Paper no. 8, Melbourne.
Reed, R. 1999, ‘The Impact of Airline Deregulation on Costs and Efficiency’ in
Gaudry, M. and Mayes, R.R. (eds), Taking Stock of Air Liberalization (Centre
for Research on Transportation, University of Montreal, 25th Anniversary
Series, 1971–1996), Kluwer Academic Publishers, Boston, pp. 77–92.
SACL (Sydney Airports Corporation Limited) 2000, Sydney Airport Revised Draft
Aeronautical Proposal, September.
—— 2001, Sydney Airport, Aeronautical Pricing Proposal, Response to ACCC
Draft Decision, March.
Salazar de la Cruz, F. 1999, ‘A DEA approach to the airport production function’,
International Journal of Transport Economics, vol. 26, no. 2, pp. 255–70.
SATC (South Australian Tourist Commission) 2001, Tourism Research: Interstate
Travel to South Australia 2000, http://www.tourism.sa.gov.au/publications/
FastFacts/PDFDocument/Interstate_Data_Card_2000.pdf (accessed 5 December
2001).
Skulley, M. and Bolton, R. 2001, ‘Firms left to flounder but videophones boom’,
Australian Financial Review, 9 September, p. 8.

454 REFERENCES
Sommerfeld, J. 2001, ‘Airport site suits discount retailer’, Courier-Mail,
19 October, p. 39.
Starkie, D. 2001a, ‘Airport regulation and competition’, Journal of Air Transport
Management, vol. 7, no. 5, (forthcoming).
—— 2001b, ‘A new deal for airports’, in Robinson, C. (ed.), Regulating Utilities:
New Issues, New Solutions, Edward Elgar Publishing, Cheltenham, United
Kingdom.
—— 2001c, ‘Reforming UK airport regulation’, Journal of Transport Economics
and Policy, vol. 35, part 1, pp. 119–135.
—— and Thompson, D. 1985, Privatising London’s Airports, Institute of Fiscal
Studies, Report Series no. 16, London.
—— and Yarrow, G. 2000, The Single-Till Approach to the Price Regulation of
Airports, CAA, London, July, http://www.caaerg.co.uk (accessed 7 June 2001).
Stigler, G.J. 1969, The Theory of Price, 3rd edn, Macmillan, London.
Supreme Court of Victoria 2001, TXU Electricity Limited v Office of the
Regulator-General & Ors [2001] VSC 153, 17 May.
Taxis, M. 2001, ‘Airport will lose money’, Northern Territory News, Letter to the
editor, 25 October, p. 10.
Templeton, R. and Mills, A. 2001, ‘Higher tax hit upsets Virgin’, Gold Coast
Bulletin, 13 October.
Tirole, J. 1990, The Theory of Industrial Organization, MIT Press, Cambridge,
Massachusetts.
Toms, M.R. 1994, ‘Charging for airports: the new BAA approach’, Journal of Air
Transport Management, vol. 1, no. 2, pp. 77–82.
Tourism New South Wales 2001, Big Sky Country Region Tourism Profile, Year
end June 2000, Sydney.
Tourism Queensland 2000a, Fact Sheet — Brisbane, http://www.tq.com.au/
research/pdf/regional/brisbane.pdf (accessed 20 June 2001).
—— 2000b, Fact Sheet — Gold Coast, http://www.tq.com.au/research/pdf/
regional/goldcoast.pdf (accessed 20 June 2001).
—— 2000c, Fact Sheet — Queensland, http://www.tq.com.au/research/pdf/
regional/TotalQld.pdf (accessed 20 June 2001).
—— 2000d, Fact Sheet — Tropical North Queensland, http://www.tq.com.au/
research/pdf/regional/tnq.pdf (accessed 20 June 2001).

REFERENCES 455
Tourism Tasmania 2001, Tasmanian Visitor Survey 2000 Data Card,
http://www.tourismtasmania.com.au/research/dcard2000/ (accessed 20 June
2001).
Tretheway, M. 1996, ‘Giving airports an incentive’, Airport World, no. 3,
pp. 36–40.
TRL (Transport Research Laboratory) 2000a, Airport Performance Indicators 2000,
Crowthorne, United Kingdom.
—— 2000b, Review of Airport Charges 2000, Crowthorne, United Kingdom.
Vogelsang, I. 2001, A 20-year perspective on incentive regulation for public
utilities, Paper presented at the Regulation and Investment Conference, ACCC,
Sydney, 26–27 March.
—— and Finsinger, J. 1979, ‘A regulatory adjustment process for optimal pricing
by multiproduct monopoly firms’, Bell Journal of Economics, vol. 10, no. 1,
pp. 157–71.
Walters, A.A. 1978, ‘Airports: an economic survey’, Journal of Transport
Economics and Policy, vol. 12, no. 2, pp. 125–60.
WIAL (Wellington International Airport Limited) 2001, Submission to the
Commerce Commission on the Commission’s Draft Report and Draft
Recommendations Arising out of its Price Control Study of Airfield Activities,
Vol. 1: Main Submission and Appendices, August.
Willis, R. (Treasurer) 1994, Speech at Australian Labor Party function, Christie
Beach, Adelaide, 18 September, http://search.aph.gov.au/search/ParlInfo.ASP?
action=view&item=0&resultsID=FoKyY, (accessed 25 November 2001).

456 REFERENCES