Paper 3 03 TP Manual 2022

Tolley® Exam Training
ADIT
PAPER 3.03
Advanced International Taxation (Thematic)
TRANSFER PRICING OPTION

Study Manual
2022 Sittings
June and December
PLEASE READ “HOW TO PASS” ON THE ACADEMY BEFORE
STARTING YOUR STUDIES.
Published by
Tolley Exam Training

Lexis House
30 Farringdon Street
London
EC4A 4HH
Tolley Online Academy: http://www.tolleytraining.co.uk/
Printed in the UK by KnowledgePoint Limited
All rights reserved. No part of this publication may be reproduced, stored in a

retrieval system or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording or otherwise without the prior written
permission of the publisher.
This material contains general information only. Whilst every care has been taken
to ensure the accuracy of the contents of this work, no responsibility for loss
occasioned to any person acting or refraining from action as a result of any
statement in it can be accepted by the author or the publishers.
Parts of this manual have been prepared from original material supplied by
authors for the production of the Tolley publication: UK Transfer Pricing 2012/13.
Additional material was provided by:
Gareth Green
Paul Griffiths
Philip Howell
BDO
Tolley® Exam Training CONTENTS
CONTENTS
CHAPTER 1 FUNDAMENTAL SOURCES ......................................................................................1
CHAPTER 2 ASSOCIATED ENTERPRISES...................................................................................21
CHAPTER 3 THE ARM’S LENGTH PRINCIPLE ............................................................................27
CHAPTER 4 TRANSFER PRICING METHODS.............................................................................45
CHAPTER 5 FUNCTIONAL ANALYSIS.......................................................................................67
CHAPTER 6 PRACTICAL ASPECTS OF PREPARING A FUNCTIONAL ANALYSIS .....................77
CHAPTER 7 RELATING THE FUNCTIONAL ANALYSIS TO SELECTION OF TP METHOD ............95
CHAPTER 8 ENTITY CHARACTERISATION..............................................................................107
CHAPTER 9 COMPARABILITY ANALYSIS: OECD PROPOSED PROCESS...............................121
CHAPTER 10 COMPARABILITY ANALYSIS: AGGREGATION AND USE OF THIRD PARTY NON-

TRANSACTIONAL DATA .....................................................................................133
CHAPTER 11 COMPARABILITY ADJUSTMENTS INCLUDING PRACTICAL ISSUES ...................145
CHAPTER 12 SPECIFIC TRANSACTIONS: INTRA-GROUP SERVICES .......................................151
CHAPTER 13 SPECIFIC TRANSACTIONS: LOANS AND OTHER FINANCIAL TRANSACTIONS 175
CHAPTER 14 TRANSFER PRICING FOR FINANCIAL TRANSACTIONS .....................................189
CHAPTER 15 SPECIFIC TRANSACTIONS: INTANGIBLE PROPERTY ..........................................207
CHAPTER 16 COST CONTRIBUTION ARRANGEMENTS ...........................................................223
CHAPTER 17 SPECIFIC TRANSACTIONS: BUSINESS RESTRUCTURING ....................................229
CHAPTER 18 NON-RECOGNITION ISSUES..............................................................................243
CHAPTER 19 PERMANENT ESTABLISHMENTS ..........................................................................255
CHAPTER 20 ATTRIBUTION OF PROFITS TO PEs.......................................................................273
CHAPTER 21 PE IN THE UN MODEL DOUBLE TAX CONVENTION ...........................................285
CHAPTER 22 COMPLIANCE ISSUES.........................................................................................289
CHAPTER 23 AVOIDING DOUBLE TAXATION AND DISPUTE RESOLUTION I ..........................307
CHAPTER 24 AVOIDING DOUBLE TAXATION AND DISPUTE RESOLUTION II .........................333
CHAPTER 25 E-COMMERCE AND TRANSFER PRICING ..........................................................345
CHAPTER 26 OTHER ASPECTS OF TRANSFER PRICING...........................................................363
© RELX (UK) Limited 2021 I 2022 Sittings

APPENDIX 1 CASE LAW ..........................................................................................................375
© RELX (UK) Limited 2021 II 2022 Sittings

DETAILED CONTENTS
CHAPTER 1 FUNDAMENTAL SOURCES ......................................................................................1

1.1 Introduction .................................................................................................................. 1
1.2 Management Accounting Origins .............................................................................1
1.3 Taxation Context ..........................................................................................................2
1.4 Origin of the Arm's Length Principle ...........................................................................4
1.5 The OECD Guidelines...................................................................................................4
1.6 Fundamental Source: OECD TPG...............................................................................5
1.7 Fundamental Source: OECD Model Double Tax Convention (DTC) .....................6
1.8 Fundamental Source: Application of the OECD Transfer Pricing Guidelines
(TPG) by States ...........................................................................................................11
1.9 UN Practical Manual on Transfer Pricing for Developing Countries (UN
Manual) ................................................................................................................... 15
1.10 Alternatives to the Arm's Length Principle...............................................................16
1.11 Pacific Association of Tax Administrations (PATA)..................................................17
1.12 African Tax Administrators Forum (ATAF).................................................................18
1.13 World Bank Group International Transfer Pricing and Developing Economies:
From Implementation to Application ......................................................................18
1.14 COVID–19 and Transfer Pricing.................................................................................20
CHAPTER 2 ASSOCIATED ENTERPRISES...................................................................................21

2.1 Introduction ................................................................................................................ 21
2.2 The OECD and UN Model DTC and the Definition of Associated Enterprises .....21
2.3 State Practice and the Associated Enterprises Definition .....................................23
2.4 Conclusion .................................................................................................................. 26
CHAPTER 3 THE ARM’S LENGTH PRINCIPLE ............................................................................27

3.1 Introduction ................................................................................................................ 27
3.2 Why is the Arm’s Length Principle Needed?...........................................................27
3.3 The OECD TPG and Comparability ..........................................................................31
3.4 The Five Comparability Factors ................................................................................32
3.5 Recognition of the Accurately Delineated Transaction .......................................38
3.6 Losses ........................................................................................................................... 39
3.7 The Effect of Government Policy .............................................................................39
3.8 Location Savings and Local Market Features.........................................................40
3.9 Assembled Workforce ...............................................................................................41
3.10 MNE Group Synergies ................................................................................................41
3.11 UN Manual Chapter B2 on Comparability ..............................................................42
3.12 Application of Comparability Analysis ....................................................................42
CHAPTER 4 TRANSFER PRICING METHODS.............................................................................45

4.1 Introduction ................................................................................................................ 45
4.2 The OECD Guidelines.................................................................................................45
4.3 Comparable Uncontrolled Price (‘CUP’) Method .................................................46
4.4 Resale Price Method (‘RPM’)....................................................................................50
4.5 Cost Plus Method (‘C+’) ............................................................................................53
4.6 Transactional Profit Methods ....................................................................................55
CHAPTER 5 FUNCTIONAL ANALYSIS.......................................................................................67

5.1 Introduction ................................................................................................................ 67
5.2 Goal of Functional Analysis.......................................................................................67
5.3 Analysis of Functions, Assets and Risks .....................................................................69
5.4 UN Manual and Risk ...................................................................................................73
5.5 Summarising the Functional Analysis .......................................................................73
5.6 Functional Analysis and Entity Characterisation ....................................................76
© RELX (UK) Limited 2021 III 2022 Sittings

CHAPTER 6 PRACTICAL ASPECTS OF PREPARING A FUNCTIONAL ANALYSIS .....................77

6.1 Introduction ................................................................................................................ 77
6.2 Preparing for the Functional Analysis.......................................................................77
6.3 The Interview...............................................................................................................80
6.4 The Functional Analysis ..............................................................................................81
6.5 Functions ..................................................................................................................... 83
6.6 Assets ........................................................................................................................... 86
6.7 Risks .............................................................................................................................. 88
6.8 Output from Functional Analysis: Meeting Notes ...................................................91
6.9 Presenting the Functional Analysis ...........................................................................91
6.10 Maintaining the Functional Analysis ........................................................................93
6.11 Conclusion .................................................................................................................. 93
CHAPTER 7 RELATING THE FUNCTIONAL ANALYSIS TO SELECTION OF TP METHOD ............95

7.1 Introduction ................................................................................................................ 95
7.2 An Overview of the Methodologies.........................................................................96
7.3 Most Appropriate Transfer Pricing Method .............................................................96
7.4 Comparable Uncontrolled Price (CUP)...................................................................98
7.5 Cost Plus ...................................................................................................................... 99
7.6 Resale Price................................................................................................................. 99
7.7 Transactional Net Margin Method .........................................................................100
7.8 Profit Split ................................................................................................................... 100
7.9 Choice of Tested Party ............................................................................................100
7.10 Examples of Functional Profiles and Links to Pricing Methodologies .................101
7.11 The Financial Indicator Where a Transactional Profit Method is Selected ........101
7.12 Availability of Comparables ...................................................................................106
7.13 The Identification of the Significant Comparability Factors to be Taken into
Account .................................................................................................................... 106
CHAPTER 8 ENTITY CHARACTERISATION..............................................................................107

8.1 Choice of Tested Party ............................................................................................107
8.2 Entity Characterisation – Comparison of Simple and Complex Entities ............109
8.3 Sales Functions..........................................................................................................111
8.4 Manufacturing Entities .............................................................................................114
8.5 Support Service Activities ........................................................................................116
8.6 More Complex or Entrepreneurial Entities .............................................................117
8.7 Planning Aspects of Entity Characterisation.........................................................118
CHAPTER 9 COMPARABILITY ANALYSIS: OECD PROPOSED PROCESS...............................121

9.1 Introduction ..............................................................................................................121
9.2 Performing a Comparability Analysis.....................................................................121
9.3 Typical Process .........................................................................................................121
9.4 Choice of the Tested Party .....................................................................................124
9.5 Information on the Controlled Transaction ...........................................................124
9.6 Comparable Uncontrolled Transactions ...............................................................124
9.7 External Comparables and Sources of Information.............................................125
9.8 Selecting or Rejecting Potential Comparables ....................................................126
9.9 Comparability Adjustments ....................................................................................127
9.10 Arm’s Length Range ................................................................................................128
9.11 Timing Issues in Comparability ................................................................................129
9.12 Compliance Issues ...................................................................................................131
© RELX (UK) Limited 2021 IV 2022 Sittings

CHAPTER 10 COMPARABILITY ANALYSIS: AGGREGATION AND USE OF THIRD PARTY NON-

TRANSACTIONAL DATA .....................................................................................133
10.1 Introduction ..............................................................................................................133
10.2 Transactions: Aggregation and Unbundling.........................................................133
10.3 Set-offs ....................................................................................................................... 135
10.4 Segmentation of Comparable Data .....................................................................136
10.5 Sources of Information and Timing Issues ..............................................................137
10.6 Sources of Third Party Non-transactional Data.....................................................138
10.7 Commercial Databases ..........................................................................................138
10.8 Comparability: Lessons from DSG Retail................................................................142
10.9 Proprietary Databases and ‘Secret Comparables’ .............................................143
10.10 Using Databases.......................................................................................................143
10.11 Other Sources of Information..................................................................................143
10.12 Timing of Information on Comparable Transactions............................................144
CHAPTER 11 COMPARABILITY ADJUSTMENTS INCLUDING PRACTICAL ISSUES ...................145

11.1 Introduction ..............................................................................................................145
11.2 Adjustments for Accounting Items .........................................................................145
11.3 Capital Intensity Adjustments .................................................................................146
11.4 Other Adjustments ...................................................................................................147
11.5 Tax Authority Responses to Comparability Adjustments. ....................................148
11.6 The Arm’s Length Range .........................................................................................148
11.7 Compliance Issues ...................................................................................................149
11.8 Safe Harbours ...........................................................................................................149
11.9 Frequency of Review ...............................................................................................150
11.10 UN Manual ................................................................................................................ 150
CHAPTER 12 SPECIFIC TRANSACTIONS: INTRA-GROUP SERVICES .......................................151

12.1 Introduction ..............................................................................................................151
12.2 Determining Whether a Service Has Been Rendered..........................................152
12.3 Determining an Arm’s Length Charge ..................................................................160
12.4 Transfer Pricing Method...........................................................................................162
12.5 Low Value-Adding Intra-Group Services ...............................................................168
12.6 EUJTPF Report on Low Value-Adding Intra-Group Services ................................171
12.7 Cost Contribution Arrangements (CCAs)..............................................................172
12.8 The UN Manual .........................................................................................................172
12.9 Scoping of the Future Revision of Chapter VII ......................................................173
CHAPTER 13 SPECIFIC TRANSACTIONS: LOANS AND OTHER FINANCIAL TRANSACTIONS 175

13.1 Introduction ..............................................................................................................175
13.2 Loans.......................................................................................................................... 175
13.3 Thin Capitalisation ....................................................................................................180
13.4 Guarantee Fees .......................................................................................................183
13.5 Cash Pooling.............................................................................................................183
13.6 Captive Insurance ...................................................................................................185
13.7 Hedging..................................................................................................................... 185
13.8 BEPS Action Plan and Interest Deductions ............................................................186
CHAPTER 14 TRANSFER PRICING FOR FINANCIAL TRANSACTIONS .....................................189

14.1 Introduction ..............................................................................................................189
14.2 The OECD Final Report on Financial Transactions................................................189
14.3 Interaction with Chapter I of the OECD TPG ........................................................190
14.4 Treasury Function......................................................................................................193
14.5 Financial Guarantees ..............................................................................................198
14.6 Case Law and Guarantees ....................................................................................200
14.7 Captive Insurance ...................................................................................................202
14.8 UN Manual and Financial Transactions .................................................................205
© RELX (UK) Limited 2021 V 2022 Sittings

CHAPTER 15 SPECIFIC TRANSACTIONS: INTANGIBLE PROPERTY ..........................................207

15.1 Introduction ..............................................................................................................207
15.2 The Life Cycle of Intangibles ...................................................................................208
15.3 Development of Intangibles ...................................................................................208
15.4 Exploiting Intangibles: Principal Structure V Licensing Out..................................209
15.5 OECD TPG and Intangibles .....................................................................................210
15.6 Valuation of Intangibles ..........................................................................................214
15.7 Case Law on Valuation of Intangibles ..................................................................217
15.8 UN Manual and Intangibles ....................................................................................220
CHAPTER 16 COST CONTRIBUTION ARRANGEMENTS ...........................................................223

16.1 Introduction ..............................................................................................................223
16.2 OECD Guidelines and Cost Contribution Arrangements ....................................223
16.3 Structuring and Documenting a CCA ...................................................................226
16.4 Case Law on Cost Contribution Arrangements ...................................................227
CHAPTER 17 SPECIFIC TRANSACTIONS: BUSINESS RESTRUCTURING ....................................229

17.1 Introduction ..............................................................................................................229
17.2 The Rationale for Restructuring and the Role of Tax ............................................229
17.3 Typical Models Applied During Restructuring .......................................................231
17.4 The OECD Approach...............................................................................................233
17.5 UN Manual ................................................................................................................ 237
17.6 Tax Authority Response to Business Restructuring.................................................240
17.7 Conclusion ................................................................................................................ 242
CHAPTER 18 NON-RECOGNITION ISSUES..............................................................................243

18.1 Introduction ..............................................................................................................243
18.2 The OECD Transfer Pricing Guidelines (TPG) .........................................................244
18.3 Country Examples of Case Law and Other Guidance on Non-recognition ....249
CHAPTER 19 PERMANENT ESTABLISHMENTS ..........................................................................255

19.1 Introduction to Permanent Establishments ...........................................................255
19.2 Physical Presence PE ...............................................................................................256
19.3 Construction Site PE .................................................................................................258
19.4 Exclusions from Definition of a PE ...........................................................................259
19.5 Agency PE................................................................................................................. 261
19.6 Position of Subsidiary................................................................................................263
19.7 Related Party ............................................................................................................263
19.8 Services PE................................................................................................................. 263
19.9 E-commerce Issues ..................................................................................................265
19.10 The Multilateral Instrument (MLI).............................................................................265
19.11 Relevant Case Law ..................................................................................................266
19.12 Discussion Drafts and the BEPS Project ..................................................................271
19.13 Double Taxation Relief.............................................................................................271
CHAPTER 20 ATTRIBUTION OF PROFITS TO PEs.......................................................................273

20.1 Introduction ..............................................................................................................273
20.2 Rejection of Force of Attraction Principle .............................................................273
20.3 Business Profits (Article 7 OECD Model Treaty)......................................................274
20.4 2010 and 2017 Updates to the Model Convention..............................................275
20.5 Practical Application of the Transfer Pricing Process...........................................281
20.6 Attribution of Profit in Excess of the Total Profit of the Enterprise ........................283
20.7 Comparison of the Article 7 OECD Approach to Article 9 .................................283
CHAPTER 21 PE IN THE UN MODEL DOUBLE TAX CONVENTION ...........................................285

21.1 Introduction ..............................................................................................................285
21.2 Definition of a PE ......................................................................................................285
21.3 Article 7 Business Profits............................................................................................286
© RELX (UK) Limited 2021 VI 2022 Sittings

CHAPTER 22 COMPLIANCE ISSUES.........................................................................................289

22.1 Introduction ..............................................................................................................289
22.2 Why is Documentation Important? ........................................................................289
22.3 The OECD Guidelines on Transfer Pricing Compliance .......................................290
22.4 Domestic Law Approaches to Transfer Pricing Compliance..............................297
22.5 Unilateral or Multilateral Documentation?............................................................298
22.6 Non-documentation Considerations.....................................................................302
22.7 Safe Harbours ...........................................................................................................304
22.8 Scoping of Future Revision of Chapter IV of the OECD TPG...............................306
22.9 Summary ................................................................................................................... 306
CHAPTER 23 AVOIDING DOUBLE TAXATION AND DISPUTE RESOLUTION I ..........................307

23.1 Introduction ..............................................................................................................307
23.2 Transfer Pricing Audits ..............................................................................................308
23.3 Corresponding Adjustments ...................................................................................310
23.4 Secondary Adjustments ..........................................................................................312
23.5 Article 25 of the OECD Model Double Tax Convention (DTC) (MAP)................314
23.6 The Tax Treaty Arbitration Procedure.....................................................................320
23.7 BEPS Action Point 14.................................................................................................321
23.8 Dispute Resolution in the Multilateral Instrument (MLI) ........................................325
23.9 EU Arbitration Convention.......................................................................................326
23.10 EU Directive on Tax Dispute Resolution Mechanisms ...........................................329
23.11 Overview of Article 25 of the UN Model Tax Treaty..............................................331
CHAPTER 24 AVOIDING DOUBLE TAXATION AND DISPUTE RESOLUTION II .........................333

24.1 Introduction ..............................................................................................................333
24.2 What is an APA? .......................................................................................................333
24.3 Duration of an APA ..................................................................................................336
24.4 International Perspective and Trends ....................................................................336
24.5 APAs and the BEPS Action Plan ..............................................................................338
24.6 Conclusion ................................................................................................................ 342
CHAPTER 25 E-COMMERCE AND TRANSFER PRICING ..........................................................345

25.1 Introduction ..............................................................................................................345
25.2 Earlier Reports: The OECD Advisory Groups’ Work on Tax and E-commerce ...345
25.3 E-commerce and Permanent Establishments ......................................................347
25.4 The BEPS Project .......................................................................................................350
25.5 Implementation and Post BEPS Developments ....................................................351
25.6 Unilateral Action on the Digital Economy .............................................................358
CHAPTER 26 OTHER ASPECTS OF TRANSFER PRICING...........................................................363

26.1 Introduction ..............................................................................................................363
26.2 Transfer Pricing and Public Affairs ..........................................................................363
26.3 The Impact of Taxation on Business Decisions ......................................................364
26.4 Transfer Pricing as a Management Tool ................................................................365
26.5 Customs Duties and Transfer Pricing ......................................................................369
26.6 World Customs Organization (WCO) Guide to Customs Valuation and Transfer
Pricing ........................................................................................................................ 372
APPENDIX 1 CASE LAW ..........................................................................................................375
© RELX (UK) Limited 2021 VII 2022 Sittings

© RELX (UK) Limited 2021 VIII 2022 Sittings

Tolley® Exam Training ADIT PAPER 3.03 CHAPTER 1
CHAPTER 1
FUNDAMENTAL SOURCES
In this chapter we will look at some key aspects including:

– origins of Transfer Pricing
– the origin of the Arm’s Length Principle (ALP)
– the ALP in the OECD Guidelines
– the ALP in the OECD Model Double Tax Convention
– the ALP in the UN Practical Manual on Transfer Pricing
– application of the OECD Guidelines by states
– alternatives to the ALP
1.1 Introduction
In this chapter we are going to begin by looking at the general concept of transfer
pricing, not just in a tax context. As you will see, transfer pricing is not just used for
tax purposes. We will then go on to look at transfer pricing in a tax context, looking
at the role of double tax conventions and the OECD Transfer Pricing Guidelines.
This chapter lays the foundations for this course – we will go on to look in detail at
the key concepts outlined here in later chapters.
1.2 Management Accounting Origins
To a management accountant, transfer pricing is a fundamental process required

in order to draw up accounts for any organisation which operates through more
than one segment (ie, company, division, profit centre). In this context, a transfer
price can be defined as ‘the amount charged by one segment of an organisation
for a product or service that it supplies to another segment of the same
organisation’. (Charles T Horngren and Gary L Sundem ‘Introduction to
Management Accounting’, page 336, ninth edition. Prentice-Hall International
Inc.)
The economic reason for charging transfer prices is to be able to evaluate the
performance of the relevant segments of the organisation. By charging
appropriate prices for goods and services transferred within a group, managers of
group entities are able to make the best possible decision as to whether to buy or
sell goods or services inside or outside the group and they are able to make a
realistic judgement about the relative contribution being made by each entity to
the overall profits from a particular product or service.
So it is not all about taxation; the illustration below demonstrates the importance of
setting the right transfer price for a business.
© RELX (UK) Limited 2021 1 2022 Sittings

 Illustration 1
Entity 1
← Production cost
(Loss)
Product
↓
Entity 2
Profit
↑
Sale
Revenue
This is a simple scenario where entity 1 produces a product and it is sold to an

external customer by entity 2. If there were no transfer price for the product, all of
the sales revenue would be in entity 2 and all of the production cost would be in
entity 1, so entity 1 would show a loss and entity 2 would show a profit margin of
100% (or perhaps a little lower if it has sales and marketing costs).
Clearly, a transfer price is needed, so that the accounts of both entities better
reflect the economic contribution that they have made.
The above might seem obvious, but it bears repeating, because there is a
tendency, particularly in recent years, for the term ‘transfer pricing’ to be used (by
certain politicians, journalists and campaigners) in a pejorative sense, as if the term
has inherent connotations of deliberately charging inappropriate prices in order to
shift taxable income artificially from a company in a high tax country to a related
company in a country with a lower tax rate.
This is a misuse of the term. Transfer pricing is a process which every multinational
enterprise must necessarily carry out and it is inherently neither good nor bad. This
was a point made by the United Nations (UN) Secretariat in 2001. (‘Transfer Pricing
History-State of the Art-Perspectives’, a paper by the United Nations Secretariat
dated 26 June 2001) It is certainly possible to set transfer prices that are blatantly
inappropriate, in an attempt to avoid tax, and no doubt there are some
organisations that do so. This is, however, certainly far less common than some
campaigners would claim, and highly unlikely to be successful.
We will look at the role transfer pricing has in respect of the internal operation of
multinational groups in a later chapter towards the end of this manual dealing with
the strategic and managerial aspects of transfer pricing.
1.3 Taxation Context
As the taxable profits of the entities will normally be based upon the accounting
profits, which will necessarily reflect the transfer prices that have been used, the
transfer prices will affect the taxable profits in both countries. For this reason,
transfer pricing usually has important tax consequences, which has given rise to
specific tax legislation in relation to transfer pricing. This manual mainly relates to
transfer pricing in its tax context, rather than the wider management accounting
sense.
The following illustration shows the tax impact of transfer pricing.

 Illustration 2
Entity 1
Country 1
Production cost Transfer price

amount will affect
profit split
Product
↓
↓ 3rd Party Customer
Entity 2 ←
Country 2
↑
↑ Sales Product →
Revenue
To be specific, the main tax impact of transfer pricing is that although it would not
normally change the combined profit before tax made by the entities between
which the transfer price applies, transfer pricing does affect how that combined
profit is split between the entities. If those entities are taxpayers in different
countries, transfer pricing therefore affects the share of that combined profit that is
taxable by each country.
There is therefore potential scope for the two taxpayers to conspire to set the
transfer price in order to influence how the profits are split between them or for
other factors to lead the taxpayers to adopt transfer pricing that differs from what
they would have adopted if they were unrelated. Accordingly, transfer pricing
rules seek to ensure that transfer prices are set so that each country gets to tax its
fair share of the profit.
Understanding the arm’s length principle is key to your understanding of transfer

pricing. You need to ensure you are happy with this part of the chapter before you
continue your study.
The term transfer pricing is inseparable from the concept of the arm’s length
principle (we will look in detail at the arm’s length principle below).
Tax rules on transfer pricing generally authorise a tax authority to increase the
taxable profits of the taxpayer entity if the transfer pricing between it and another
related party is higher than or lower than the arm’s length price, and, as a result,
the taxable profits of the taxpayer entity have been understated.
The arm’s length price is the price that is paid in a comparable transaction
between unrelated parties.
The process of comparing the actual transfer price with the arm’s length price is
referred to as the arm’s length test.
The arm’s length principle is the principle that transfer prices between related
parties should meet the arm’s length test.

In some countries, the arm’s length principle is used in a slightly different way, as
the standard by which it is determined whether a company has given a
constructive dividend or hidden profit distribution to its parent. The result is that the
company is denied a deduction for the relevant expenses for tax purposes and
may sometimes also suffer withholding tax on the deemed dividend.
Although the arm’s length principle has been adopted extremely widely as the
appropriate standard by which it should be judged if a transfer price is
acceptable, there are some countries which refuse to accept the arm’s length
principle. In recent years, countries such as India and China have adopted the
arm’s length principle. Perhaps the most notable country which rejects the arm’s
length principle is Brazil, which currently sets its own rules about acceptable levels
of profits from intercompany transactions. See below for details of Brazil’s joint
project with the OECD.
1.4 Origin of the Arm's Length Principle
The arm’s length principle appears in two main settings. First, it appears in domestic
tax transfer pricing legislation of many countries. Second, it appears in double
taxation treaties (also known as double tax agreements or conventions) and
related guidance. We will look at the arm’s length principle in the context of tax
treaties below.
Based on an OECD survey, the arm’s length principle was first introduced in
domestic legislation in 1911, by Norway. It was followed by the UK in 1915. The USA,
which is often seen as the birthplace of transfer pricing, did not introduce this
obligation until 1935, although this would still make it one of the earliest adopters.
It is not just OECD member countries that have transfer pricing rules, many other
countries (eg Romania) now have transfer pricing rules. The main exceptions are
many of the tax havens, countries cut-off from international trade, such as North
Korea, and some of the least developed countries in the world.
The arm’s length principle was included in tax treaties concluded by France, the
UK and the USA as early as the 1920s. We now find this principle in Article 9,
paragraph 1 of the present OECD and UN Model tax treaties. (‘Transfer Pricing
History-State of the Art-Perspectives’) Article 9 is described further on in this
chapter. These days, the arm’s length principle is a ubiquitous feature of virtually
every fully-fledged tax treaty.
1.5 The OECD Guidelines
The OECD Guidelines on Transfer Pricing are an important part of this course. We
will be making reference to them throughout the manual. You may find it useful to
have a copy to hand so that you can look at them as you are studying.
In the early 1990s, the OECD recognised the need to update, consolidate and
expand its 1979 and 1984 reports on transfer pricing and multinational enterprises
in response to huge growth in international trade and the spread of multinational
enterprises. Further impetus was given by the fact that the US was pushing ahead
with its own, wide-ranging, unilateral views on how the arm’s length principle
should be applied and these views were not always congruent with the views of
other countries. The OECD took the view that it was important to bridge the
differences that were developing, as one of its main missions is to avoid double
taxation.

The result was a report entitled ‘Transfer Pricing Guidelines for Multinational
Enterprises and Tax Administrations’, which is almost universally referred to within
transfer pricing circles as the ‘OECD Guidelines’. As described below, the
Guidelines have evolved since they were created in 1995. In this manual we refer
to the OECD Guidelines as OECD TPG.
1.6 Fundamental Source: OECD TPG
The original OECD TPG were issued in 1995. In fact, they were released in
instalments, starting with Chapters I to V in 1995, covering the Arm's Length
Principle (Chapter I), Traditional Transaction Methods (Chapter II), Other Methods
(Chapter III), Administrative Approaches to Avoiding and Resolving Transfer Pricing
Disputes (Chapter IV), and Documentation (Chapter V). These were
supplemented by Chapter VI, Special Considerations for Intangibles, and Chapter
VII, Special Considerations for Intra-Group Services, in 1996. A year later, Chapter
VIII, on Cost Contribution Arrangements, was published.
After a relatively quiet period in terms of new guidelines from the OECD, 2010 saw
the culmination of several years work on two projects. The first project led to
revisions of the first three chapters of the OECD TPG. Symbolic of the removal of
the preference for the traditional transaction-based methods, all methods are now
dealt with in a single chapter, Chapter II. Additional guidance was provided in
relation to comparability analysis. Arguably, the discussion of the Transactional Net
Margin Method is more accepting of the fact that often this method can only be
applied using the overall profitability of comparable companies, and therefore
moves closer to the US Comparable Profits Method.
The second project related to the Transfer Pricing Aspects of Business Restructurings
and this gave rise to Chapter IX, the first chapter to be added to the OECD TPG
since 1997. (Discussed further in a later chapter).
In April 2013 Chapter V was revised to reflect the new stance on safe harbours. We
will also look at this in a later chapter.
In May 2016 the OECD formally adopted the proposed changes to the 2010 OECD
TPG put forward in the final reports of October 2015 for Action Points 8 to 10 of the
Base Erosion and Profit Shifting (BEPS) project (see section on BEPS below). This
means that they are encouraging both members and non-members to follow the
guidance set down in the October 2015 BEPS reports. The OECD TPG were reissued
in July 2017.
The executive summary for Action Points 8 to 10 (as they were dealt with together)
states that the work under the BEPS Action Plan ‘will ensure that transfer pricing
outcomes better align with value creation of the MNE’.
The final report issued in October 2015 amends part of Chapter I, adds additional
guidance to Chapter II and replaces in their entirety Chapters VI, VII and VIII.
Chapter I of the OECD TPG is where we find the guidance on the arm’s length
principle. The changes arising from the BEPS project will be looked at in detail as
you work through this manual, however at this point it is worth noting that the
changes focus on the accurate delineation of a transaction. That is to say,
ensuring that when we look at a controlled transaction we are looking at what is
really happening rather than what the paperwork might say. An important step in
this process will be an accurate functional analysis and the accurate assignment
of risk, as you will see as you work through this manual.

The impact of the recommendation to adopt the final report on Action Points 8 to
10 will vary from country to country; as mentioned below (see section on
Application of OECD TPG by States) countries take different approaches with
respect to whether and how they incorporate the OECD TPG into their domestic
tax systems. It may be that some countries will require formal administrative or
other action to incorporate a new version of the TPG into the domestic law, in
other cases it may be automatic. In addition, some countries may take the view
that the amendments to the TPG merely clarify pre-existing transfer pricing
principles, and consequently in practice could have retroactive effect.
The agreed amendments don’t just affect OECD members as they represent the
consensus of all the countries that were part of the OECD/G20 BEPS Action Plan.
These included all OECD Member States, plus Argentina, Brazil, the People’s
Republic of China, Columbia, India, Indonesia, Latvia, the Russian Federation,
Saudi Arabia and South Africa.
As mentioned above, the OECD TPG were reissued in July 2017. In addition to the
above changes, consequential amendments were made to other chapters, in
particular Chapter IX. In this manual, unless otherwise stated, all references are to
the 2017 Guidelines.
The OECD published its final report on financial transactions in February 2020. It
contained amendments to Chapter I of the OECD TPG and the text of a new
Chapter X. The final report is the end product of a work stream that came from
Action Points 4 and 8 to 10 of BEPS Action Plan. As such it is has come about from a
work stream involving more parties than the OECD members as during the BEPS
project the OECD extended collaboration to what became known as the
‘inclusive framework’ which allows interested countries and organisations to work
with the OECD and G20 members on developing standards on BEPS related issues
and to review and monitor the implementation of the BEPS Package.
Therefore, it is a final report from the inclusive framework (137 members) and as
such may be said to be a consensus document of the inclusive framework.
At the time of writing the OECD had not stated the procedure for adding the new
Chapter X to the OECD TPG or the date it would be effective from. The
announced amendments to Chapter I will add 20 new paragraphs to the OECD
TPG therefore where this impacts on the paragraph references contained in this
manual the revised paragraph reference once the changes are formally adopted
is shown in brackets. Where reference is made to the new paragraphs that will be
inserted, they are referred to as new paragraph 1.107 to 1.126 as appropriate.
Finally, the OECD is currently looking at the possibility of changes to Chapter IV

and VII of the OECD TPG. Consultation began in 2018; at the time of writing no
updates had been published.
1.7 Fundamental Source: OECD Model Double Tax Convention (DTC)
This is the agreement reached between Member States of the OECD that acts as
guidance when negotiating tax treaties. The convention consists of articles,
commentaries, position statements and special reports on evolving tax issues. Its
primary application is in guiding the negotiation of bilateral tax treaties between
countries. The OECD Model DTC has to be read in conjunction with the detailed
commentary on its interpretation. The Model and the Commentary are the work of
the Committee on Fiscal Affairs of the OECD, which is composed of senior
government officials drawn from the OECD members.

The aim of the OECD Model DTC is to provide certainty to international trade
transactions. The Commentary is used to provide guidance on treaty interpretation
and to try to provide conformity in international tax.
It should also be noted that the UN has also produced a Model Tax Convention
(UN Model DTC) that is used by developing nations when negotiating tax treaties.
The UN Model DTC is designed to aid developing states to tax a larger part of the
overseas investor's income than other models.
In this section we are going to look at some of the articles from the OECD Model
DTC that are important to your studies. The current model treaty was issued in 2017
following an update; reference to this update will be made where relevant.
As already mentioned, the arm's length principle has been incorporated in double
taxation agreements since as early as the 1920s. This led to it becoming Article 9 of
the Model DTC originally issued by the OECD in 1963, based on earlier model tax
treaty wording created by the League of Nations. Article 9 of the UN Model DTC
has similar wording. The arm's length principle also plays a role in several other
articles of the OECD Model DTC including Article 11, Interest, Article 12, Royalties,
and Article 7, Business Profits. These are discussed below.
Article 9 Associated Enterprises
Article 9 is the article which permits countries that have signed a double tax treaty
(with wording based on the OECD Model DTC) to adjust inappropriate transfer
pricing. If it were not for this Article, it could be protested that making a transfer
pricing adjustment should not be allowed, because the treaty generally sets out to
eliminate double taxation.
The article reads as follows:
Article 9
ASSOCIATED ENTERPRISES
1. Where
a. an enterprise of a Contracting State participates directly or indirectly in the

management, control or capital of an enterprise of the other Contracting
State, or
b. the same persons participate directly or indirectly in the management, control

or capital of an enterprise of the Contracting State and an enterprise of the
other Contracting State,
and in either case conditions are made or imposed between the two enterprises in
their commercial or financial relations which differ from those which would be
made between independent enterprises, then any profits which would, but for
those conditions, have accrued to one of the enterprises, but, by reason of those
conditions, have not so accrued, may be included in the profits of that enterprise
and taxed accordingly.
2. Where a Contracting State includes in the profits of an enterprise of that State –

and taxes accordingly – profits on which an enterprise of the other Contracting
State has been charged to tax in that other State and the profits so included are
profits which would have accrued to the enterprise of the first-mentioned State if
the conditions made between the two enterprises had been those which would
have been made between independent enterprises, then that other State shall

make an appropriate adjustment to the amount of the tax charged therein on

those profits. In determining such adjustment, due regard shall be had to the other
provisions of this Convention and the competent authorities of the Contracting
States shall if necessary consult each other.
The wording is somewhat tortuous, but if it is read slowly the meaning is clear.
Broadly speaking, paragraph 1 authorises a country that is a signatory to the tax

treaty to increase the taxable profits of an enterprise. The conditions for it to do so
are that those profits have been understated as a result of making or imposing
non-arm’s length conditions (which usually means prices that do not meet the
arm's length test) in its commercial or financial relations (which usually means
transactions) with another enterprise which is associated with the first enterprise.
It should be noted that it is generally accepted that Article 9 is intended to be

permissive; it allows Contracting States to apply the transfer pricing rules that form
part of their tax legislation. It is generally considered that, although not explicitly
stated in either the Model Tax Convention or the Commentary thereon, Article 9
does not create a stand-alone right for countries to make transfer pricing
adjustments that go beyond what is authorised by their own domestic rules. This is
because the basic purpose of a double taxation agreement is to relieve double
taxation; it would go way beyond this purpose if a double taxation agreement
imposed harsher tax treatment on a particular transaction between country A and
country B than would have applied if the same transaction had taken place
between country A and another country, with which country A has not concluded
a double taxation agreement.
Paragraph 2 relates to what are known as corresponding adjustments, although

this term is not specifically used in paragraph 2. (The term is used in the glossary to
the OECD TPG.) It requires that where a transfer pricing adjustment is made by
country A to increase the profits of the enterprise which is a taxpayer in that
country, country B must give consideration to reducing the profits of the other
enterprise (a corresponding adjustment) so that no profits are taxed in both
countries. Corresponding adjustments need to be distinguished from
compensating adjustments. The latter occur when a taxpayer reports a transfer
price for tax purposes which in the taxpayers opinion is arm’s length even though it
differs from the amount actually charged between the associated enterprises. This
adjustment would be made before the tax return is filed.
Country B is not automatically obliged to give a downward adjustment merely

because country A has made an upward adjustment. But it is obliged to give an
adjustment if it agrees that the profits of the enterprise in country B would have
been lower if its profits had been calculated based on prices which met the arm's
length test.
The Commentary makes it clear that if the two countries disagree about the
appropriate adjustment, the Mutual Agreement Procedure (MAP) provided for
under Article 25 should be implemented (we will look in detail at Article 25 in a
later chapter).
Article 9 does not contain any time limits. The Commentary makes it clear that this
is not because it necessarily feels there must be an open-ended commitment to
give a corresponding adjustment. The question of time limits is left for individual
countries to negotiate in their individual bilateral double taxation agreements,
based on the OECD Model DTC. If the relevant double taxation agreement is silent
on this matter, time limits will presumably follow domestic law.

Your syllabus includes knowledge of certain aspects of the UN Model DTC. The UN
Model DTC contains wording which mirrors Article 9 of the OECD Model DTC (also
at Article 9), but in 2001 a third paragraph was inserted, as follows:
3. The provisions of paragraph 2 shall not apply where judicial, administrative or

other legal proceedings have resulted in a final ruling that by actions giving rise to
an adjustment of profits under paragraph 1, one of the enterprises concerned is
liable to penalty with respect to fraud, gross negligence or willful default.
In other words, where a transfer pricing adjustment arises because of a failed

attempt at tax avoidance or tax evasion by a multinational group, the penalties
directly incurred as a result of that attempt will be compounded by subjecting the
multinational group to double taxation. The underlying philosophy seems to be
that relief from double taxation in relation to a transfer pricing adjustment is a
privilege which should only be extended in cases where the transfer pricing
misstatement arose despite good-faith efforts to comply with the arm's length
principle.
This is an important difference between the OECD Model DTC and the UN Model
DTC that you should note.
In March 2021 the OECD issued a public discussion draft which includes proposals
for changes to the Commentary on Article 9 and other related articles (7, 24 and
25). The proposed changes will clarify the position on interest deductions (including
restriction under Action Point 4 of BEPS) and the classification of interest as equity.
In addition they provide that domestic law limitations on deductibility of expenses
would not be considered to lead to economic double taxation subject to relief
under Article 9, as Article 9 does not address the deductibility of expenses or
computation of taxable income under domestic law, only allocation of profits.
There is also a proposed addition to the Commentary on the MAP to clarify when
access to MAP should be given for transfer pricing cases. The amendment to
Article 24 confirms that additional reporting requirements can be placed on
payments to non-residents. The deadline for comments on the proposals was set
as 28th May 2021.
Article 7 Business Profits
The other main part of the OECD Model DTC in which the arm's length principle
plays a core role is Article 7.
Article 9 deals with transactions between separate enterprises, one of which is

resident in one Contracting State and the other is resident in the other Contracting
State. Article 7 deals with a single enterprise resident in one Contracting State
which also operates in the other Contracting State through a permanent
establishment (PE). A typical example would be a company operating through an
overseas branch. We will look at the definition of a PE in a later chapter when we
look in detail at Article of 5 of the OECD Model DTC that sets down the conditions
for a PE to exist.
Article 7 sets out the consequences of having a PE and defines the extent to which
the business profits of an enterprise are taxable in a Contracting State within which
it has a PE (sometimes referred to as the ‘source’ state).
We will look at Article 7 in more detail in the later chapter on attribution of profits to
PEs. The provisions of the UN Model DTC will also be considered in this later
chapter.

In summary, the arm's length principle is brought into play by paragraph 2 of the
Article, specifically the requirement that the profits that are attributable to the PE
should be the profits that the PE might be expected to make if it were a separate
and independent enterprise engaged in the same or similar activities under the
same or similar conditions.
Article 11 Interest
Article 11 of the OECD Model DTC relates to interest arising in one Contracting
State and paid to a resident of the other Contracting State. The first five
paragraphs of the article apply regardless of whether the borrower and lender are
associated. However, paragraph 6 introduces a special rule which denies the
protection of the article to interest that does not meet the arm’s length test.
Take this opportunity to look at paragraph 6 of the OECD Model DTC.
The general effect of Article 11 is to limit the amount of tax that can be applied by
the country where the interest is sourced (generally referred to as withholding tax
(WHT)). The OECD Model DTC limits the tax to 10% of the interest, although many
double taxation agreements that are based on the OECD Model DTC adopt
different limits and in many cases source taxation is prohibited altogether. This
protection is removed for interest in excess of an arm’s length amount.
The effect is that a multinational enterprise can potentially be doubly penalised if it

is shown to have charged interest which exceeds an arm’s length amount. First, it
can be denied a deduction for the excess interest, by virtue of Article 9; second, it
can suffer greater tax at source (WHT) on the excess interest.
It will be noted that whereas Article 9 defines association in terms of direct or

indirect participation in management, control or capital of an enterprise, this
wording is not mirrored in Article 11, which uses the simple term ‘special
relationship’. This is because Article 11(6) is intended to apply more widely than
Article 9, for instance where the special relationship arises by way of family or
marriage ties between individuals or where there is ‘any community of interests’ –
see paragraph 34 of the Commentary to Article 11 – between the lender and
borrower other than the loan relationship itself.
The UN Model DTC includes similar provisions relating to interest.
Article 12 Royalties
Article 12, the royalties article of the OECD Model DTC, includes a special
relationships paragraph (in this case paragraph 4) which has a similar effect to
paragraph 6 of Article 11. That is, there is no protection from source taxation of a
royalty that exceeds an arm’s length royalty.
Take this opportunity to review paragraph 4 of Article 12 of the OECD Model DTC.
The UN Model DTC also deals with royalties in Article 12. In addition it has an Article
12A which covers technical services and Article 12B dealing with automated
digital services. All include provisions relating to beneficial ownership.

1.8 Fundamental Source: Application of the OECD Transfer Pricing

Guidelines (TPG) by States
The OECD has no authority to bind member countries to its Guidelines. Few of
them have, like the UK, incorporated the OECD TPG within their domestic
legislation. However, all member countries accept that the OECD TPG are
intended to represent the international consensus and that if there is a divergence
of practice, this heightens the risk of double taxation, which could discourage
international trade. In practice, they are highly influential amongst all OECD
member countries and increasingly amongst countries that are not OECD
members. They have become the international norm for how transfer pricing
analysis is conducted, except in the relatively rare cases where governments
explicitly reject the arm's length principle, the most prominent example being
Brazil.
Brazil has detailed rules on related entities and classes all entities in low tax
jurisdictions as related. The acceptable methodologies do not follow the OECD
TPG. For example, they do not include any profit based methodologies.
Brazil has an approach that sets out a maximum ceiling on the expenses that may
be deducted for tax purposes in respect of imports and lays down a minimum
level for the gross income in relation to exports, effectively using a set formula to
allocate income to Brazil. Taxpayers are allowed to use the method that gives the
lowest taxable income.
As noted earlier, in 2018 the OECD announced that they had launched a joint
project to examine the similarities and gaps between the Brazilian and OECD
approaches to valuing cross-border transactions between associated firms for tax
purposes, and to assess the potential for Brazil to move closer to the OECD’s
transfer pricing rules.
Brazil has been working with the OECD to align its transfer pricing rules with the
OECD standard, following its formal request in 2019 to become a member of the
OECD. In December 2019 a joint report by the OECD and Receita Federal, Brazil’s
federal revenue authority was published. The report states that Brazil has identified
a clear pathway for bringing its existing transfer pricing framework into alignment
with the international consensus and is weighing two options – immediate or
gradual implementation. Both options contemplate full adherence to the arm's
length principle, which is at the core of the OECD standard, while seeking to
preserve simplification, maintain ease of compliance and effectiveness of tax
administration, as well as cross-border tax certainty. The next phase will be to
prepare a blueprint and a roadmap that will outline the implementation process.
Part of the reason why most countries endeavour to abide by the OECD TPG is
that they can effectively be seen as forming part of the OECD Commentary on
how Article 9 of its Model DTC should be applied. The actual Commentary on
Article 9 cross-refers to the OECD TPG. The OECD TPG would therefore be a major
determinant of how any dispute between treaty partners is resolved under MAP
(that is, negotiations to try to reach a common position about the arm’s length
transfer price for the purposes of Article 9, so that any transfer pricing adjustment
made by one country is balanced by a corresponding opposite adjustment in the
other country, without which double taxation would arise). We look at the MAP in
detail in a later chapter.
If a country adopts a transfer pricing position that is contrary to the OECD TPG, it
knows that it is likely to find itself trying to defend this position in the MAP and few
countries are willing to allow the MAP to fail by insisting on an interpretation of the

arm's length principle that is inconsistent with the OECD TPG. This tends to
encourage countries to abide by the OECD TPG at all stages of applying their
It should also be noted that the OECD TPG are just that: guidelines. They are not
worded in the same way as legislation, in a completely definitive, prescriptive
fashion. Rather, they make observations about what would generally be
preferable and they frequently leave room for alternative interpretations. They are
the result of discussions amongst many people representing many countries and
other organisations and it is not always possible to reach full consensus. As a result,
the OECD TPG sometimes deliberately do not address certain contentious issues
(or they explicitly state that no consensus has yet been reached). An example
would be the discussions about business restructuring before the 2010 update to
the OECD TPG, which frequently became bogged down by disagreements about
the treatment of intangible assets. In order not to hold up the business restructuring
chapter of the OECD TPG any longer, intangibles were taken out of the scope of
the business restructuring work. They were carved out into a separate project.
At the time of writing the OECD member countries are: Australia, Austria, Belgium,
Canada, Chile, Columbia, Costa Rica, Czech Republic, Denmark, Estonia, Finland,
France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Korea
(South), Latvia, Lithuania, Luxembourg, Mexico, Netherlands, New Zealand,
Norway, Poland, Portugal, Slovak Republic, Slovenia, Spain, Sweden, Switzerland,
Turkey, the United Kingdom and the United States.
Generally speaking, these are highly developed countries, although some of the
more recent joiners, such as Turkey and Mexico, would generally be considered to
be developing countries, albeit at the richer end of the spectrum of development.
The OECD is accordingly seen by some as being a club that represents the
interests of rich countries.
Membership has gradually expanded. Costa Rica joined the OECD in May 2021. In
addition, the OECD has enhanced agreements with Brazil, China, India, Indonesia
and South Africa. Representatives from those countries participate in the OECD’s
deliberations about transfer pricing as observers.
The UN Model DTC is generally seen as being an alternative to the OECD Model
DTC and contains wording that is intended to be more favourable to less
developed countries. (However, the differences tend not to relate to transfer
pricing.) As noted earlier the UN Model DTC Article 9 (Associated Enterprises) refers
to the arm’s length principle in a similar manner to the OECD Model DTC.
The UN has not produced a set of guidelines of its own. Instead they have
published a Practical Manual on Transfer Pricing for Developing Countries. This
manual is designed for countries seeking to apply the arm’s length principle and
provides a practical approach to applying the OECD TPG. This manual is part of
your syllabus – again you should consider obtaining a copy to look at as you work
through this study manual. There is further detail on the UN Practical Manual on
Transfer Pricing below.
As mentioned above the OECD TPG are a guide to the member countries. The
OECD website includes country profiles at http://www.oecd.org/tax/transfer-
pricing/transfer-pricing-country-profiles.htm. If you go to this page you can click on
the various member countries and see how many diverge from certain aspects of
the guidelines, eg Mexico, Argentina or Germany.

The OECD Report on Base Erosion and Profit Shifting (BEPS)
Although the OECD TPG and the OECD Model DTC have now been updated for
the outcomes of the BEPS project, the project itself and some of the final reports
are still part of the syllabus. As a result, this section includes a recap on BEPS and a
summary of the final reports that are still included in the syllabus.
BEPS has wider implications than just transfer pricing. We will look briefly at some of
the wider aspects here and consider in detail the aspects relating to transfer
pricing throughout this manual whenever they are relevant.
The OECD published the BEPS Action Plan in July 2013. OECD secretary general
Angel Gurría said that the Action Plan marked ‘a turning point in the history of
international tax co-operation’. KPMG commentators noted that the scale of the
report and its ambitions for change represented ‘a potential seismic shift in the
international tax landscape’. At the G20 meeting which followed the publication it
was commented that the effective taxation of ‘mobile income’ was a key
challenge.
As noted in the report, one of the key problems is that national tax laws have not
kept pace with the globalisation of corporations and the digital economy, leaving
‘gaps that can be exploited by multinational corporations to artificially reduce
their taxes’. The digital economy offers ‘a borderless world of products and
services that too often do not fall within the tax regime of any specific country,
leaving loopholes that allow profits to go untaxed’.
The Action Plan offered a ‘global roadmap’ to allow governments to collect the
tax revenue they need and give businesses the certainty they need to invest and
grow. Fifteen specific actions were identified in the report which would give
governments domestic and international instruments to prevent corporations from
paying little or no taxes.
The principle of coherence was highlighted in the plan with the intention that there
should be neither double taxation nor double non-taxation. From a business point
of view, coherence means all business income should be taxed once and all
business expenditure should be deducted once.
It is notable that the Action Plan clearly rejected the introduction of unitary
taxation, or global formulary apportionment (discussed below). This is because of
‘the practical difficulties associated with agreeing to and implementing the details
of a new system consistently across all countries’. However, within the existing
transfer pricing regime the plan did give consideration to measures which may ‘go
beyond the arm's length principle’ in dealing with intangibles, risk and capital
allocation – which is likely to mean a wider use of profit split methods in place of
the reliance on comparable transactions.
The Action Plan contained proposals to:
• review the application of existing international tax rules to the digital economy
(ie online businesses);
• develop treaty anti-avoidance provisions to deal with hybrid entities and

hybrid instruments;
• develop recommendations for countries to deal with controlled foreign

companies, interest deductibility, treaty abuse and hybrids;

• improve information exchange, including spontaneous exchange in some

cases;
• improve dispute resolution mechanisms;
• broaden the definition of permanent establishment, with particular focus on

commissionaires;
• amend the transfer pricing rules, including broadening the definition of

intangibles, and documentation requirements;
• develop recommendations for reporting business restructuring and tax

planning arrangements; and
• develop a multilateral instrument under which countries could implement the

action plan proposals.
However, the Action Plan did not propose any radical change. Instead, the Action
Plan relied on amending and strengthening existing mechanisms such as transfer
pricing, controlled foreign company and permanent establishment rules.
The report can be downloaded from the OECD website.
The final reports from the Action Plan were delivered in October 2015. Their impact
on the OECD TPG and transfer pricing in general will be discussed throughout this
manual.
The final report on Action Points 8 to 10 used one report to cover the transfer
pricing aspects relating to the need to assure that transfer pricing outcomes are in
line with value creation in relation to intangibles, including hard-to-value ones, to
risks and capital, and to other high-risk transactions. The final report impacted on
many of the chapters in the OECD TPG beginning with Chapter I where
amendments were needed to the definition of risk and control of risk, the
guidance on comparability factors including synergies and market specific factors
and a new definition for when transactions can be set aside (non-recognition).
Specific paragraphs were added to Chapter II to deal with commodities. Chapter
VI on intangibles was substantially amended. A section on low value-adding
services was added to Chapter VII on services. Chapter VIII on Cost Contribution
Arrangements (CCA) was rewritten. In addition, the final report set down the need
for further work to be done in relation to the profit split method. We will work
through each of these amendments as we progress through the manual.
The final report on Action Point 13 led to one of the first BEPS outcomes, being the
revision of Chapter V on documentation and the implementation of the three-tier
approach including the now widely adopted country-by-country report (CbC).
The revised approach to documentation will be looked at in detail in a later
chapter.
As noted earlier, the work streams that came from Action Points 4 and 8 to 10 lead
to a final report on financial transactions in February 2020 containing amendments
to Chapter I and the introduction of Chapter X to the OECD TPG.
The final report on Action Point 14 which concerned making dispute resolutions
more effective led to changes in Article 25 of the OECD Model DTC and its
Commentary. It also included minimum standards and best practices – we will look
at these later when dispute resolution is covered in detail.

1.9 UN Practical Manual on Transfer Pricing for Developing Countries (UN

Manual)
The latest version of the UN Practical Manual on Transfer Pricing was issued in April
2021.
The aim of the UN Manual is to strengthen the application of the arm’s length
principle as set down in Article 9 of both the UN and OECD Model DTC and the
OECD TPG. The manual shows how the arm’s length principle can be applied by
developing nations who have particular problems trying to identify the market
price. These problems include a lack of comparables and local development of
intangibles not being taken fully into account.
The manual also includes areas of non-consensus with the OECD approach. It
includes chapters outlining the rules of Brazil, China, India, Mexico, South Africa
and Kenya.
The 2021 version of the manual includes some new content, in particular, on
financial transactions, profit splits, centralised procurement functions and
comparability issues.
A: Transfer pricing in a global environment
This section looks at the background and objectives of transfer pricing rules. It
includes a section on ‘value chain analysis’. Value chain analysis (VCA) looks at all
the processes that a business uses to produce a valuable product or service and
take it to the market place. The UN Manual gives the following example:
VCA looks at where in the process the value is added. Businesses use VCA to look
at every step in the process and identify where improvements can be made to
save cost or add value.

In transfer pricing the value chain is examined to determine where profitability

belongs within a MNE. VCA must be looked at to determine where the value is
created in a business; the Functional Analysis (which is looked at in later chapters)
is then correlated with this understanding.
You may find it useful to look at a copy of the UN Manual at this point.
B: Design principles and policy consideration
This is the main part of the UN Manual which includes a detailed discussion of
comparability analysis and practical examples of the five OECD approved
methodologies, including selection and implementation of each. (We will look at
the methodologies in a later chapter). There is also a section on what is referred to
as the ‘sixth method’ for commodity pricing. This ‘sixth method’ is used in Brazil and
Argentina and there is also a mention of it in the revised OECD TPG. There are
chapters on services, intangibles, Cost Contribution Arrangements, business
restructuring and financial transactions. We will look in more detail at Section B
throughout this manual.
C: Practical implementation of a Transfer Pricing Regime
This section includes practical guidance for implementing a transfer pricing regime
including practical advice for tax authorities trying to build capacity to deal with
transfer pricing issues. It also covers documentation, audit, risk assessment and
dispute avoidance.
D: Country Practice
As noted above the examples given are Brazil, China, India, Mexico, South Africa
and Kenya. Each country is an example of a regime that does not conform to the
OECD approach and as a result may not always adhere to the arm’s length
principle. Mexico is the new addition in the revised version of the manual.
1.10 Alternatives to the Arm's Length Principle
The arm’s length principle has always had its detractors. The original report by the
OECD in 1979, ‘Transfer Pricing and Multinational Enterprises’, devoted space to
discussing the alternatives to the arm’s length principle as a way to allocate taxing
rights in relation to the profits made by multinational enterprises.
The main alternative is global formulary apportionment, (as noted earlier this is
sometimes called unitary taxation) under which the total consolidated worldwide
profits of a group are allocated between the various jurisdictions where the group
carries out business activities. The allocation is based on a formula, usually
consisting of certain allocation keys, such as turnover, payroll and the value of
assets in each country. Variations on this approach are used within federal states,
such as the USA and Switzerland, in order to determine what portion of the profits
of individual entities in those countries are taxable in each State/Canton.
The OECD came to the firm conclusion that such methods should not be
endorsed, on the grounds that they are arbitrary, disregard market conditions,
ignore management’s own allocation of resources, do not bear a sound
relationship to the economic facts, and carry significant risk of double taxation.
This conclusion is reiterated in the OECD TPG.
The tide has long been in favour of the arm’s length principle and it has been
adopted by almost all major economies, whether or not OECD members, with a

particular surge of adoption since 1995. Even Brazil, which rejects the arm’s length
principle, does not use formulary apportionment. It uses instead a modified version
under which it insists on the right to specify acceptable profit margins for Brazilian
entities in relation to transactions with non-Brazilian related parties.
In recent years, particularly since the global financial crash that began in 2007,
voices opposing the arm’s length principle and championing formulary
apportionment have grown louder, fuelled by a growing perception (correct or
not) that transfer pricing is being widely abused by multinational companies to
avoid paying their fair share of taxes in the countries where they have significant
operations. A form of formulary apportionment is being considered by the
European Union if it moves to a common tax base for member countries (or a
subset thereof).
Now would be a good time to have a look at a copy of the OECD TPG. If you turn
to Chapter I you will see that it begins with an explanation of the arm’s length
principle. Reading through this will reiterate what we have discussed so far in this
manual. The next section of Chapter I looks at Article 9 and sets down the OECD’s
views on the arm’s length principle. You can see in paragraph 1.15, as mentioned
above, the OECD do not think a move away from the arm’s length principle would
be a good idea despite the recent debate in this area. Paragraph 1.15 states that
experience to date with the arm’s length principle should be used to elaborate
and refine its operation. You will also note that in paragraph 1.15 the OECD do not
see that there is a viable alternative to the arm’s length principle.
Finally Section C gives the OECD view on global formulary apportionment. As part
of the work on Action Points 8 to 10 of the BEPS Action Plan, the OECD considered
whether special measures needed to be introduced either within or beyond the
arm’s length principle. The final report concluded that ‘the goals set by the BEPS
Action Plan in relation to developing transfer pricing rules have been achieved
without the need to develop special measures outside the arm’s length principle’.
We will look at other aspects of Chapter I of the OECD TPG later in the manual.
1.11 Pacific Association of Tax Administrations (PATA)
PATA was formed in 1980 to combat income shifting, improve cross-border

information flows, and develop conciliatory relationships among the member
countries. One of their specific concerns was to identify and stop improper transfer
pricing.
In 2003 they adopted a documentation package that was voluntary and did not
exceed the requirements of local laws. The member countries are all currently in
the process of adopting domestic law to comply with and implement the OECD
three-tier documentation package that includes country-by-country reporting (we
will look at this in more detail in a later chapter). The PATA members include
Australia, Canada, Japan and the United States.
PATA has produced guidance on the MAP to facilitate cooperation amongst

members and bilateral Advance Pricing Agreements (APAs) to try to standardise
procedures.

1.12 African Tax Administrators Forum (ATAF)
The ATAF is a forum to promote and facilitate cooperation between African tax
administrators and other interested parties. The ATAF has been working with the
OECD to promote awareness of the need for transfer pricing rules in Africa. A
working group has been set up to look at transfer pricing issues including review of
the OECD TPG.
In 2017 ATAF formed a technical committee on exchange of information (EOI).

Training will be provided to individuals from some of the member countries.
Although not directly related to transfer pricing, this shows a commitment to
transparency which is important in combatting tax avoidance.
1.13 World Bank Group International Transfer Pricing and Developing

Economies: From Implementation to Application
There has been a lot of concern with the implications of transfer pricing for
developing countries. It has been recognised that developing nations are not just
in danger of losing tax revenue through the use of tax havens but also via transfer
pricing if the arm’s length principle is not observed.
We have seen that the UN has developed a Practical Manual on Transfer Pricing
for Developing Countries. The World Bank has produced a handbook to help
developing economies (Transfer Pricing and Developing Economies: A Handbook
for Policy Makers and Practitioners - you can access it at
https://openknowledge.worldbank.org/handle/10986/25095. The World Bank has
developed as a social institution since its inception in 1944. The handbook seeks to
provide an overview of transfer pricing and act as a framework for identifying and
addressing issues that arise when implementing and applying a transfer pricing
regime in a particular country. The handbook is not an alternative to the OECD
TPG or the UN Practical Manual. It includes guidance on drafting transfer pricing
legislation, applying the arm’s length principle, and developing a transfer pricing
audit system.
The importance of the investment environment is stressed. It is noted that this can
be protected and damage limited by ensuring that double taxation is avoided
and that the compliance regime is not burdensome, along with an efficient and
equitable dispute resolution mechanism and well trained staff. The World Bank
supports the use of the MAP found in double tax treaties (a subject we will look at
in detail in a later chapter). The compliance burden can be reduced by the use of
safe harbours; however the World Bank notes that if these are unilateral then they
may result in problems of double taxation and offer little solution. The handbook
gives guidance on the use of Advance Pricing Arrangements (again a subject we
will look at in detail in a later chapter). The handbook also includes illustrative
examples from actual country domestic law and guidance. The overall aim is to
provide technical guidance; as a result it can be useful to both tax administrations
and multinational enterprises.

The Platform for Collaboration on Tax (PCT)
Responding to a request by the Development Working Group of the G20, the

Platform for Collaboration on Tax – a joint initiative of the IMF, OECD, UN and World
Bank Group – has developed a toolkit designed to assist developing countries in
respect of transfer pricing. In June 2017 the Platform published the toolkit:
‘Addressing Difficulties in Accessing Comparables Data for Transfer Pricing
Analyses’; it specifically addresses the ways developing countries can overcome a
lack of data on ‘comparables,’ or the market prices for goods and services
transferred between members of multinational corporations. The toolkit also
includes a supplementary report on ‘Addressing the Information Gaps on Prices of
Minerals Sold in an Intermediate Form’. The toolkit can be found at
http://www.oecd.org/tax/toolkit-on-comparability-and-mineral-pricing.pdf.
In January 2021 the PCT issued a toolkit aimed at helping developing nations
design transfer pricing documentation rules. It can be found at:
http://www.oecd.org/tax/Practical Toolkit to Support the Successful Implementati
on by Developing Countries of Effective Transfer Pricing Documentation Requirem
ents - OECD
The toolkit compiles essential information on transfer pricing documentation and

analyses policy choices and legislative options. It includes sample legislation as
well as examples and practices from over 30 countries and additional approaches
to transfer pricing documentation.
The PCT has a website to promote the work it does. The website can be found at:
https://www.tax-platform.org/
Tax Inspectors Without Borders (TIWB)
The Tax Inspectors Without Borders (TIWB) project was launched in 2015 by the
OECD and the United Nations Development Programme (UNDP) as an innovative
attempt to address widespread tax avoidance by multinational enterprises in
developing countries and as a contribution towards financing the UN's Sustainable
Development Goals. A joint OECD and UNDP team links requests for help from
developing countries with the right experts from wherever they are located.
TIWB organises deployment of highly qualified tax experts to countries that request
assistance with ongoing audits of multinational companies. The projects focus on
revenue recovery and improving local audit capacity while sending a strong
message on the need for tax compliance.
Eleven countries, including the UK, Netherlands and Spain have deployed serving
tax officials as part of TIWB to provide assistance to auditors in developing
countries to strengthen their effectiveness in taxing multinational enterprises. The
programmes deal mainly with transfer pricing and international taxation. More
information can be found at www.tiwb.org

1.14 COVID–19 and Transfer Pricing
The ADIT Transfer Pricing exam is a very practical exam. To be successful in the
exam requires not only a good technical knowledge of transfer pricing, but also
the ability to apply that knowledge to the practical situations given in the exam
questions. The questions set tend to be very topical therefore it is worth noting
some of the transfer pricing considerations that will arise as a result of the global
pandemic caused by COVID–19.
Some MNEs are changing the way they work; in addition to staff working from
home as a result of lockdowns, around the world business models are changing. All
these changes need to be documented and the transfer pricing implications
considered. The changes may lead to shifts in functions between group
companies in order to maintain supply chains; again any such shifts need to be
documented and reflected in the intra-group pricing. In some cases, there will be
a shift from physical to digital platforms; this may require the development of new
software tools such as bots to automate replies and deal with standard enquiries.
The costs in developing these tools and their ownership will need to be considered.
The manual will not refer to the impact of COVID–19 in every chapter, however it is
something that you should consider as you work through each chapter as there
will be many consequences. Where relevant, references to the impact of the
pandemic may strengthen an answer in an exam question.
The OECD have published guidance on the OECD website on the transfer pricing
implications of COVID–19.

CHAPTER 2
ASSOCIATED ENTERPRISES
In this chapter we are going to look at Associated Enterprises, an important but somewhat
ill-defined concept, in particular looking at:
– The OECD and UN Model DTC and the definition of Associated Enterprises
– State practices and the Associated Enterprise definition
2.1 Introduction
As we saw in the previous chapter, Article 9 of the OECD Model DTC permits
Contracting States to apply domestic law transfer pricing legislation only in respect
of transactions between ‘associated enterprises’. As such, defining the scope of
what is, and is not, an associated enterprise is of considerable practical
importance. There are definitional difficulties with this term as it appears in the
OECD and UN Model DTC. Moreover, there is very extensive variation between the
scope of different countries’ domestic associated enterprises rules which may give
rise to the possibility of double taxation even in cases where tax treaties apply.
2.2 The OECD and UN Model DTC and the Definition of Associated
Enterprises
The Glossary to the OECD TPG define associated enterprises as follows:
‘Two enterprises are associated enterprises with respect to each other if one of the
enterprises meets the conditions of Article 9, sub-paragraphs 1a) or 1b) of the
OECD Model tax Convention with respect to the other enterprise’.
However, the above ‘definition’ is not illuminating. Article 9 of the OECD Model
DTC is titled Associated Enterprises, but the text of the article does not explicitly use
the word associated. You will recall that Article 9(1) reads as follows:
Where
a. an enterprise of a Contracting State participates directly or indirectly in the

management, control or capital of an enterprise of the other Contracting
State, or
b. the same persons participate directly or indirectly in the management, control

or capital of an enterprise of a Contracting State and an enterprise of the
other Contracting State,
and in either case conditions are made or imposed between the two enterprises in
their commercial or financial relations which differ from those which would be
made between independent enterprises, then any profits which would, but for
those conditions, have accrued to one of the enterprises, but, by reason of those
conditions, have not so accrued, may be included in the profits of that enterprise
and taxed accordingly.
Article 9(1) of the UN Model DTC is worded identically.

There are many circumstances where there will be no doubt that two enterprises
are associated with each other. The two most obvious cases are two companies in
a parent/100% subsidiary relationship or two companies under 100% control by a
third common parent company.
 Illustration 1
A Inc is a body corporate resident in State A; B Limited is a body corporate

resident in State B. A Inc holds 100% of the voting shares in B Limited and no person
other than A Inc has an interest in the management, control or capital of B
Limited.
A Inc
State A
B Limited
State B
 Illustration 2
As Illustration 1, except A Inc also holds 100% of the voting shares of C SA resident
in State C and no person other than A Inc has an interest in the management,
control or capital of C SA.
A Inc
State A
B Limited C SA
State B State C
All states which have domestic transfer pricing legislation include such blatant
control relationships within the scope of that legislation, so that if conditions are
made or imposed between the above which deviate from arm’s length terms,
then such domestic legislation that may exist in States A, B or C would be
supported by a double tax convention which included wording based on Article
9(1) of the OECD Model DTC.
Enterprises
The Commentary on Article 9 does start with the following statement:
‘This Article deals with adjustments to profits that may be made for tax
purposes where transactions have been entered into between associated
enterprises (parent and subsidiary companies and companies under common
control) on other than arm's length terms’.
Article 9 is phrased to apply to ‘enterprises’. This is in contrast to most of the rest of

the OECD Model DTC, which generally applies to residents. The phrase ‘enterprise’
is also used for Article 5, which deals with Permanent Establishments, and Article 7,
which deals with the Business Profits of Permanent Establishments. The phrase
‘enterprise’ is not defined in the body of the Model DTC, but is commonly thought
of as analogous to ‘business’.

It might be thought that the word ‘enterprise’ was deliberately used instead of the
word ‘company’ in order to ensure that the article was not restricted to
companies, despite the suggestion quoted in the Commentary above. It is clearly
possible that taxable entities other than companies (such as partnerships and
individuals) could be sufficiently associated with other taxable entities that they
might not be dealing on an arm’s length basis.
In practice, the distinction is of limited importance because it is relatively rare for

there to be a transfer pricing issue on transactions between persons at least one of
which is not a company. Many countries explicitly limit their domestic transfer
pricing legislation to transactions between companies, although this is not the
case in the UK, which phrases its transfer pricing legislation in terms of provisions
between persons.
2.3 State Practice and the Associated Enterprises Definition
Overview
It is not clear from the OECD Model DTC or its Commentary what constitutes
participation in management, control or capital. It is left to individual states to
define (usually in their domestic legislation) exactly how much participation in ‘the
management, control or capital of an enterprise’ is sufficient to bring transfer
pricing rules into effect.
There is an enormous variation in state practice. At one extreme are countries such
as Denmark which apply a relatively narrow de jure requirement of at least 50% of
share capital or voting rights. As an example of the other extreme, Australia’s
legislation is engaged in a wide variety of de facto circumstances including both
parties having common directors or being members of a cartel.
It is instructive to examine the required relationship threshold for two countries, the
UK and India, in a little more depth.
United Kingdom
TIOPA 2010 s.157 to s.163 defines participation in the management, control or

capital of a person for the purposes of this legislation and sets out rules that
attribute rights and powers to a person when considering whether that person
controls a company or partnership.
The term ‘person’ includes a body of persons. So, for example, a partnership can
control a company even if, individually, none of the partners control the
partnership or company.
In the first instance, TIOPA 2010 s.217 determines that control is defined by
reference to CTA 2010 s.1124 which considers matters such as voting power, power
given by the Articles of Association and the actual ability of a person to direct the
affairs of the company in the absence of the visible signs of such rights.
We can see from the above that it is very important to underline that control does
not only manifest itself where one enterprise is the majority shareholder in the
other. Control exists whenever one enterprise has the power to ensure that the
business of another enterprise is managed to achieve the other enterprise’s goals.
There are detailed rules setting down when voting rights and control should be
attributed to a person. However, the attribution rules need to be considered in
relatively few cases.

While the legislation prevents abuse where trusts are interposed in a control chain,
it does not reproduce ITA 2007 s.993(4), with the result that persons are not
connected simply by virtue of being members of the same partnership.
The control rules in TIOPA 2010 s.160 contain an important feature. This is the
inclusion of a provision deeming a 40% participant in a joint venture to control that
joint venture where there is one other participant who owns at least 40% of the
venture. Hence, the transfer pricing rules also apply to joint ventures; however, the
rules only apply to transactions between at least one of the joint venture parties
and the joint venture itself, not between the two joint venture parties themselves.
Nevertheless, if the transaction meets the conditions of TIOPA 2010 s.157 transfer
pricing rules still apply.
UK transfer pricing rules, in relation to financing transactions only, also apply where
persons have ‘acted together’ in relation to the financing arrangements of a
company or partnership. This concept of ‘acting together’ is much more widely
drawn than the above condition; it connotes a ‘community of interests’.
 Illustration 3
The ordinary shares in D Limited are owned respectively 45% by A Inc, 38% by B
GmbH and 17% by C SA in a contractual joint venture. D Limited is a UK resident
company. D Limited’s shareholders are otherwise unconnected by virtue of legal
control.
D Limited enters into the following separate transactions with its shareholders:
i. A Inc sells trading stock to D Limited in the normal course of its trade.
ii. B GmbH and C SA jointly provide trade finance to D Limited.
The sale of trading stock to D Limited is unlikely to be within the scope of the UK
participation condition. Although A Inc owns more than 40% of D Limited, there is
no other participant that also owns at least 40% of D Limited.
However, the provision of financing facilities will fall within the extended ‘acting
together’ definition and the tax return of D Limited must reflect arm’s length terms
in relation to that transaction.
There are a number of cases where associated enterprises are exempt from the
transfer pricing rules, including certain small- and medium-sized enterprises (SMEs)
transacting with states that have a comprehensive double tax convention with the
UK.
India
In India the transfer pricing law is found in Income Taxes Act 1961 s.92 onwards. An
associated enterprise is ‘defined’ analogously to Article 9(1) of the OECD Model
DTC. However, an extensive list is then provided of situations in which two
enterprises shall be regarded as associated. These include:
• Direct or indirect holding of at least 26% voting power.
• A loan advanced from one enterprise constituting at least 51% of the book
value of the assets of the borrower.
• One enterprise guarantees at least 10% of the total borrowings of the other
enterprise.

• Appointment by an enterprise of more than half of the board or the

appointment of executive directors.
• Complete dependence of an enterprise on intangibles (can be referred to as

intellectual property or intangible property) licenced to it by the other
enterprise.
• Substantial purchases or sales of raw materials or manufactured goods at

prices and conditions influenced by the other enterprise.
• The existence of prescribed mutual interest relationships.
Thus a very broad view of direct or indirect control is taken by the Indian
legislation.
Indian tax law does not have exceptions for SMEs as in the UK rules outlined above.
There are certain simplifications for small transactions where the aggregate value
of the international transaction does not exceed 10 million INR – the simplification
relates to documentation requirements.
The Consequences of Variation in State Practice: Corresponding Adjustment
An important practical implication of the differences in state practices described

above is the possibility that a corresponding adjustment under the applicable
treaty equivalent of Article 9(2) of the OECD Model DTC might be denied if the
Contracting State that would otherwise be obliged to grant the adjustment does
not regard the two enterprises as associated but the other Contracting State does.
 Illustration 4
Fledgling Enterprises Pvt Limited, an Indian company, is owned 30% by Big Farm
Danmark A/S, a Danish manufacturer of specialised agricultural machinery. The
other 70% of the Indian company is owned by a number of individual members of
a wealthy family. The Indian company is in the business of importing and
distributing agricultural machinery into India. Over 90% of its stock is procured from
Big Farm Danmark A/S which has a dominant market position and superior
bargaining power to the Indian company. The Indian tax authorities succeed in
asserting that the prices paid for inventory by Fledgling Enterprises are in excess of
an arm’s length amount and an increase in Indian tax is imposed accordingly.
The two companies are associated enterprises within the meaning of the Indian
transfer pricing rules because there is greater than 26% voting power and the
Indian company is dependent on the Danish company for nearly all its inventory.
However, the two companies are not associated within the meaning of the Danish
rules because there is less than 50% control by share capital or voting rights.
The India/Denmark DTC includes an Article based on Article 9(2) of the OECD
Model DTC.
‘Where a Contracting State includes in the profits of an enterprise of that State

and taxes accordingly profits on which an enterprise of the other Contracting
State has been charged to tax in that other State and the profits so included
are profits which would have accrued to the enterprise of the first-mentioned
State if the conditions made between the two enterprises had been those
which would have been made between independent enterprises, then that
other State shall make an appropriate adjustment to the amount of the tax
charged therein on those profits. In determining such adjustment due regard

shall be had to the other provisions of this Convention and the competent
authorities of the Contracting State shall, if necessary, consult each other.’
It is far from clear that Denmark is obliged to make a corresponding adjustment to

Danish tax on the sale of inventory from Big Farm Danmark A/S to Fledgling
Enterprises Pvt Limited under these circumstances. The Danish Tax Authority might
be expected to argue that ‘participation in control’ is to be construed by
reference to Danish legislation and that the relationship falls short of the requisite
degree of control.
2.4 Conclusion
The concept of associated enterprises is important because it concerns the control

thresholds which bring into play domestic transfer pricing rules which often include
onerous documentation requirements and the risk of penalties for non-compliance
(we will look at these issues in a later chapter). Whether or not two enterprises are
associated with each other will, in many cases, be uncontroversial in both
Contracting States. However, the lack of a clear definition in the OECD Model DTC
and Commentary has permitted a wide variation in state practice which could
give rise to double taxation in practice.

CHAPTER 3
THE ARM’S LENGTH PRINCIPLE
In this chapter we are going to look at:

– why the arm’s length principle is needed
– the OECD Guidelines and comparability
– the five comparability factors
– application of comparability analysis
3.1 Introduction
The purpose of this chapter is to explore the arm’s length principle, and to
understand how it should be applied by taxpayers and tax authorities (namely,
through comparability analysis). The arm’s length principle itself is very simple:
pricing between related parties for any transaction should reflect pricing that
would be agreed between independent parties (ie. parties operating at ‘arm’s
length’ from each other). However, the complexity of applying this in practice is at
the heart of all uncertainty and controversy within transfer pricing.
Therefore, this chapter will:
• Explain why the arm’s length principle is needed;
• Demonstrate how the arm’s length principle should be defined and

interpreted;
• Give an overview of how comparability analysis should be used to apply the

arm’s length principle.
3.2 Why is the Arm’s Length Principle Needed?
Although this is seemingly a fairly basic question, it is worth considering as it informs

us as to why there are so few alternatives and why authorities persist with the arm’s
length principle as the foundation for transfer pricing on a global basis.
Consider a company manufacturing and selling products in its local market. If that
company becomes successful, it will likely seek to grow its market, and at some
point that will involve seeking to sell its products overseas. To achieve this, the
company would have the choice of either using third party distribution channels in
other countries, or selling directly to customers itself. Using third parties may have
the short-term advantage of selling to companies with an existing customer basis
and local market knowledge, however it would require sharing some of the value
chain profit with another party. Therefore, the manufacturer may choose to sell
directly. With small scale sales, this may be achievable whilst maintaining only a
sales force in the home country. Nevertheless, at some point it would most likely
seek to establish a sales force overseas.
Once this happens, for a raft of administrative reasons, the company is likely to
seek to establish a subsidiary overseas to employ the sales force (the same effect
could be achieved through establishing a branch, but this would basically have
the same outcome for transfer pricing purposes). When this company begins to
purchase products from the local country manufacturer to sell to its local
customers, we have created the need for a transfer price – a price that is

reflected in the sales ledger of the local manufacturer and in the cost base of the
accounts for the overseas sales company. Thus, the transfer price exists for
accounting purposes rather than explicitly for tax purposes.
In the absence of any tax constraints, the company is in theory free to choose any
price for the products. If the sales company were located in a country where the
corporate tax rate is currently say 40%, the company would be incentivised to set
a relatively high transfer price. For a given cost of production and end sales price,
this would allow the bulk of the profit to be earned in the home country, where a
corporate tax rate of say 23% would be applied. With a high transfer price, little
profit would be earned in the overseas country and thus little tax would be paid at
the higher rate, minimising the group’s overall tax liability.
Conversely, if the sales company were located in a country with a corporate tax
rate of say 12.5%, the company would be incentivised to set a low transfer price.
This would allow more of the profits to be earned and taxed in the overseas
country, and less in the home country, reducing the overall tax liability.
The above is illustrated below using Country A as the home country.
 Illustration 1
Final sales price 500.
Cost of production 200
Transfer Country A Country B Country C Total Profit

Price Profit Profit Profit
23% 40% 12.5%
400 200 100 300
300 100 200 300
In each case the total profit is 300. However where Country A has the lower tax
rate, more of the profit has been left there to be taxed at 23%. Where the tax rate
is lower in Country C most of the profits have been moved there.
Let us focus on sales in Country B. With a transfer price of 400 to Country B the total
tax payable by the group is 46+40 = 86 leaving after tax profits of 214.
If we amend the transfer price from 400 to 300 as used for Country C the total tax
bill would become 23+80 = 103 giving after tax profits of 197. We can see how the
lower transfer price has resulted in a larger tax bill overall on the sales in Country B.
This setting of the transfer price based on tax rates is sometimes referred to as tax
arbitrage.
This inherent ability of multinationals to set transfer prices based on tax rate
arbitrage is what drives the need for transfer pricing rules. In the example above, a
Country A manufacturer selling to third party distributors in Country B and Country
C would not have flexibility to choose any price, but rather would have to
negotiate. Furthermore, it would not care about the tax liability of the
counterparties, and instead would only be interested in setting a price that
maximised its long term profits.
To protect against this, tax authorities need a basis for determining an appropriate
transfer price that is independent of the tax rate of counterparties. From a short-
term tax yield perspective, tax authorities may be inclined to use an approach
that maximises the taxable profit in their territory, irrespective of the fact pattern.

The problem with this approach is that it would be highly unlikely to be accepted
by the tax authority in the territory of the counterparty. This would lead to two
different transfer prices used in the calculation of taxable profits in each country
and the double taxation of group profits.
Such an outcome would be undesirable on many levels. Ultimately it would lead

to a significant reduction in cross-border trade, impacting jobs and economic
prosperity. Such an outcome would be unpalatable for most governments. The
result is that compromise is required, and this comes in the form of the arm’s length
principle.
Under this principle, prices are to be set on an objective basis to reflect the price
that would have been agreed if the two parties couldn’t collude to produce a
better post tax outcome. In short, it says the price should be fair to both parties,
and by inference, to both tax authorities.
As you will have noted when you studied Chapter 1 of this manual, the OECD TPG
recognise the limitations of the arm's length principle as ‘the separate entity
approach may not always account for the economies of scale and interrelation of
diverse activities created by integrated businesses. There are, however, no widely
accepted objective criteria for allocating the economies of scale or benefits of
integration between associated enterprises.’ (See OECD TPG Chapter I, B, 1.10).
However imperfect, the arm’s length principle has been adopted by most tax
jurisdictions and no other alternative has yet been recognised; hence, the
comparability analysis is key in ensuring that the transfer pricing can be supported
in case of a tax audit. The OECD TPG state at paragraph 1.15:
‘A move away from the arm's length principle would abandon the sound
theoretical basis described above and threaten the international consensus,
thereby substantially increasing the risk of double taxation. Experience under
the arm's length principle has become sufficiently broad and sophisticated to
establish a substantial body of common understanding among the business
community and tax administrations. This shared understanding is of great
practical value in achieving the objectives of securing the appropriate tax
base in each jurisdiction and avoiding double taxation. This experience should
be drawn on to elaborate the arm’s length principle further, to refine its
operation, and to improve its administration by providing clearer guidance to
taxpayers and more timely examinations. In sum, OECD member countries
continue to support strongly the arm’s length principle. In fact, no legitimate or
realistic alternative to the arm's length principle has emerged.’
This brings us to the importance of comparability analysis. As no alternative has yet

been acknowledged, comparability analysis is in all effects the only available test
for determining whether two related parties are transacting at arm's length.
Notwithstanding the opportunity for associated enterprises to manipulate prices, it

should not be assumed that prices will not be arm’s length. The OECD TPG observe
that, ‘Associated enterprises in MNEs sometimes have a considerable amount of
autonomy and can often bargain with each other as though they were
independent enterprises. Enterprises respond to economic situations arising from
market conditions, in their relations with both third parties and associated
enterprises.’ (See OECD TPG Chapter I, A, 1.5).
Although taxpayers cannot rely upon the defence that prices have been
negotiated for compliance purposes, it may still very well be the case that such
prices are in fact arm’s length. Prices within a business are often set by commercial
rather than tax departments, with performance incentives for both counterparties

meaning that there will be a natural tension tending towards an arm’s length
outcome (we will look at this in a later chapter). Furthermore, commercial teams
will generally have a good understanding of their own industry and a reasonable
idea of what constitutes arm’s length arrangements.
A further complication is that taxpayers will often enter into transactions and
arrangements that simply do not exist between unrelated parties.
‘Such transactions may not necessarily be motivated by tax avoidance but

may occur because in transacting business with each other, members of an
MNE group face different commercial circumstances than would
independent enterprises. Where independent enterprises seldom undertake
transactions of the type entered into by associated enterprises, the arm’s
length principle is difficult to apply because there is little or no direct evidence
of what conditions would have been established by independent enterprises.
The mere fact that a transaction may not be found between independent
parties does not of itself mean that it is not arm’s length.’(See OECD TPG
Chapter I, B, 1.11).
It has long been established by economists that large firms tend to exist to take
advantage of economies of scale and scope not available to smaller firms. The
concept of comparability analysis can sometimes be difficult to apply as
businesses part of a larger multinational enterprise (MNE) often exchange services
and products, which are often not the ‘finished article’.
For example, multinational groups may centralise certain ancillary activities to be

more efficient. These activities may include human resources, IT support and
finance. Although it is possible to find independent companies providing these
services, it is important to note that for the independent companies, these services
are core activities. As such, they will also need to perform their own sales activity
and commercial management and generally act in an entrepreneurial manner.
Where these activities are performed in the context of a related party service,
there is generally no such entrepreneurial element. We will look at services in more
detail in a later chapter.
A further example is found in the context of business restructuring (discussed in

more detail in a later chapter). Companies may choose to centralise many of the
key economic activities and business risks, such that operations in local markets
may have a very limited role. For example, distributors may have only a very
specific role in sales management, and manage few risks. Such distributors are
unlikely to be found operating independently. Nevertheless, provided the structure
is not purely for the purpose of tax avoidance, the transaction should be
respected and the arm’s length principle should still apply. In such cases, more
thorough analysis would be required to evaluate the arm’s length price.
The natural point that follows from this is to note that the arm’s length principle
does not require taxpayers to behave in an arm’s length manner.
It is simply the case that pricing for transactions should be consistent with an arm’s
length consideration. For example, companies are not required to negotiate as
independent parties would, or limit access to commercial information.

3.3 The OECD TPG and Comparability
The guidance on comparability is found in Section D of Chapter I of the OECD

TPG. As noted in an earlier chapter this section was amended as part of the
update to the OECD TPG following the adoption of the recommendations in the
final report on Action Points 8 to 10 of the BEPS Action Plan.
Section D of Chapter I is called ‘Guidance for Applying the Arm’s Length

Principle’. The first subsection is entitled ‘Identifying the commercial or financial
relations’. It emphasises the importance of identifying the commercial or financial
relations between the associated enterprises and the conditions attaching to
those relations in order that the controlled transaction is accurately delineated;
this is seen as a very important first step in the comparability analysis.
As mentioned in an earlier chapter, when the OECD refers to the delineation of a

transaction they are referring to accurately drawing out what is taking place
rather than just following what the available documentation might say. The steps
to achieve this will involve a detailed functional analysis and assignment of risk as
set out below. Part of the aim is to stop cash rich enterprises with low functionality
from receiving too much reward for the actual risk that they take on.
The OECD suggest that features of the parties, such as capabilities and actual
contributions, can affect the options realistically available to the parties. Therefore,
the process of identifying the economic circumstances of the commercial and
financial relations should include consideration of the capabilities of the parties,
how under the arm’s length hypothesis such characteristics affect options
realistically available, and whether similar capability is reflected in potentially
comparable arm’s length arrangements.
Where there are written contractual agreements, those agreements provide the
starting point for identifying the commercial and financial relations between the
associated enterprises. The OECD states that where no written terms of a
transaction exist, or where the conduct of the parties indicates that the
contractual terms are ambiguous, incorrect or incomplete, the delineation of the
transaction should be deduced, clarified, or supplemented through the analysis of
the commercial or financial relations as deduced from the actual conduct of the
parties.
Section D includes examples to illustrate the concept of clarifying and

supplementing the contractual terms based on the identification of the actual
commercial or financial relations.
In order to apply the arm’s length principle, it is necessary to undertake

comparability analysis (a subject that we will be looking at in detail later in this
manual).
Section D (paragraph 1.33) states that there are two steps to a comparability
analysis:
i. the identification of the commercial or financial relations between the

associated enterprises and the conditions and economically relevant
circumstances attaching to those relations in order that the controlled
transaction is accurately delineated; and
ii. a comparison between the conditions and the economically relevant

circumstances of the controlled transaction as accurately delineated, with the
conditions and the economically relevant circumstances of comparable
transactions between independent enterprises.

The first step is dealt with in Chapter I of the OECD TPG, while Chapters II and III
focus on the second aspect.
The economically relevant circumstances can be referred to as the comparability

factors. The OECD TPG identify five comparability factors. Although the
comparability factors are in essence as before the update following the final
report on Action Points 8 to 10 of the BEPS Action Plan, their order within the
chapter has been adjusted and the amount of detailed guidance, in particular in
relation to contractual terms and risk in the functional analysis, has been
expanded.
3.4 The Five Comparability Factors
This is the first time we are going to look at the five comparability factors; however
we will look at them again in more detail in a later chapter.
The OECD makes reference to consideration of economically relevant

characteristics when undertaking comparability analysis, and specifically identifies
five comparability factors. These factors should all be taken into account when
undertaking comparability analysis:
• Contractual terms;
• Functional analysis;
• Characteristics of property or services;
• Economic circumstances; and
• Business strategies.
The 5 comparability factors have two key roles. Firstly, they are used to accurately
delineate the controlled transaction, hence they are discussed in Chapter I of the
OECD TPG. Secondly, they are a used as part of the comparability analysis, as set
down in Chapter III of the OECD TPG, in order to determine the arm’s length price
of the controlled transaction. The relationship between the comparability factors
and comparability analysis is very important to understanding transfer pricing; they
are not the same thing, neither are they completely distinct. They are combined to
find the arm’s length price. In summary, we can say that the comparability factors
have to identify potential comparables and that comparability analysis tests if they
are indeed good comparables that can be used to test the controlled transaction
and decide if it meets the arm’s length standard.
This section will look at the comparability factors. There is an overview of

comparability analysis later in this chapter however a subsequent chapter
contains more detail on comparability analysis.
The comparability factors are fundamental to choosing the right comparables,

however their relative importance varies depending on the transfer pricing
method chosen to price the transaction under review. Transfer pricing methods will
be considered in detail in this manual.

To best understand the role of comparability factors, and the reason they impact
on the arm’s length price, it is helpful to consider them in the context of a simple
example. Consider a Japanese-owned multinational company, manufacturing
televisions in Japan and selling them into the UK market through a related party
distributor.
MNE based in Japan
Connected Distributor
UK sales
The sole transaction under review is the sale of televisions, and the transfer price
that we therefore need to establish is the price per unit of the televisions. The five
comparability factors will tell us what we need to consider when evaluating
objective data that may be available.
Contractual Terms
The contractual terms define how the parties to a transaction are going to divide
the risk and responsibilities between them. This may be done explicitly or implicitly.
In some cases there may not be a formal contract as we are dealing with
associated enterprises. Here it will be necessary to look at correspondence and
other communications to determine the contractual terms.
Another complication with associated enterprises is that they may not hold each
other to the terms of a contract. In a third party situation we would normally
expect to see each party wanting to maximise their benefits and reducing their risk
– this can be achieved by imposing the strict terms of the contract. This may not
happen where the companies are associated.
There may be practical difficulties in obtaining the details of contractual terms for
third parties. The impact of this will vary according to the type of transaction we
are looking at.
Contractual terms and conditions should always be reviewed when using the CUP
method (see later in this chapter) as differences between the third party and the
related party contracts could result in different pricing (eg transfer of stock and
forex risk from one distributor to another). At one level, the contractual terms may
simply consider the size of and nature of the transaction. A transaction where the
distributor is seeking to import 50,000 units of a product will likely attract a very
different price to a contract for 500 units, even if the products are identical.
At a deeper level, the contractual arrangements may confer rights and

obligations on the parties that need to be reflected in the transfer pricing. For
example, the distributor may have exclusive rights to distribute the televisions within

the UK market and would therefore expect to pay a premium for such right. It may
also be obligated to undertake a certain level of advertising and promotion
activity that would need to be factored into the price. It may also contractually
bear or pass on some of the risks identified in the functional analysis. An
understanding of the contractual relationship between the parties is therefore
necessary to enable the identification of sufficiently comparable third party
relationships.
Section D states that the starting point will be any written agreement between the
parties. It goes on to emphasise that the conduct of the parties and the other four
comparability factors will also be important in establishing if the economic reality
matches the written agreement. The conduct of the parties will be important as
will the options that would have been available in terms of how (or whether) to
conduct the transaction. There is more emphasis on economic reality than legal
form.
The OECD TPG (see paragraphs 1.42 to 1.45) now explicitly authorise the recasting
of contractual terms to reflect the commercial or financial relations that actually
exist between the parties based on their conduct and the economically relevant
characteristics of the transaction, including options realistically available to the
parties to the transaction.
Functional Analysis
Functional analysis is going to be looked at several times in this manual as it is a key

part of transfer pricing.
Functional analysis is the area of most significant focus in undertaking

comparability analysis. It involves consideration of the key economic activities of
the parties, not only in terms of their functions, but also how risks are managed and
how assets are developed and owned. Each of these should be considered in
turn.
Functions are the most easily identifiable of the three areas, and this area requires
an assessment of the relative activities of both counterparties, specifically in
relation to the transaction under review. In the case of our example, there may be
a range of activities undertaken by the UK entity. At one end of the spectrum, it
may have very limited functions. There may be only a handful of employees
engaged in account relationship management. Orders by the customers may be
shipped directly to them by the manufacturer with the distributor taking legal title
for only a split second.
At the other end of the spectrum, the distributor may have a full range of activities.
It may operate a warehouse and engage in a full suite of logistics services, or it
may have an extensive marketing team developing, advertising and producing
marketing materials to promote the products directly to customers. Other things
being equal, it would be expected that a distributor undertaking more of the
value chain activities would purchase goods at a lower price in order to be able
to fund those activities, and would typically expect to earn more profits.
Comparability analysis needs to consider whether the counterparts in the third
party data have a similar functional profile.
In relation to risks, it is noted that a party bearing more risk would typically expect
to earn more profit (albeit that the actual level of profit earned may fluctuate
depending on whether those risks are realised). In the case of our distributor, it
would be expected that the distributor would earn a higher margin if it bore
inventory risk, warranty risk and customer credit risk, than if those risks were passed
on to the manufacturer. It should be noted that consideration needs to be given

to the behaviour of the parties and not just the contractual relationships. Under
OECD principles, risk (and the reward associated with it) should be attributed to
the party that manages that risk, not just the party that contractually bears it.
Assets are important too, and in particular, intangible assets can be a critical
determinant of transfer prices. If the Japanese manufacturer owns a globally
recognised brand, and attaches that brand to the televisions, that will result in a
very different transfer price to the case where the manufacturer simply produces
unbranded products. In the latter case, the UK distributor may have developed its
own brand through marketing activities, and therefore would expect to pay less
for the products, even if the technical capabilities were the same.
In undertaking comparability analysis, third party data involving companies with

similar functional, risk and asset profiles to the tested party is required. In practice, it
is impossible to find companies with identical profiles. Therefore, broader analysis
of the industry and the company are required to determine which are the
significant determinants of profit for the company, and which are routine.
We will look at functional analysis in more detail in a later chapter. In this chapter
we will concentrate on section D 1.2.1 of Chapter I of the OECD TPG. This section
looks at risk analysis which is a key part of the functional analysis. In open market
conditions risk and reward go together; any increase in the assumption of risk
would normally lead to the expectation of more reward. It follows that in order to
make a comparison with an uncontrolled transaction we have to analyse the
sharing of risk between the parties in a controlled transaction. Paragraph 1.60
gives a six step process for analysing risk. This can be summarised as follows:
1. Identify economically significant risk with specificity.
2. Determine the contractual agreements on these risks.
3. Use a functional analysis to determine the potential impact of those specific

risks. Identify which enterprises assume and manage the risks.
4. Does 3 agree with 2? Confirm via the functional analysis, establish who
exercises control over risk and that they have the financial capacity to do so.
5. Reallocate the risk to the enterprise(s) that has/have control over the risk and
the financial capability to assume the risk.
6. Price the accurately delineated transactions, taking account of the

consequences of risk allocation.
This would be a good time to look at paragraph 1.60 in your copy of the OECD
TPG.
Control of risk is an important concept. Clearly risk can never be fully controlled,
however it can be managed. The OECD TPG look at risk management in detail.
Paragraph 1.61 of the OECD TPG sets down the three elements that are required
for risk management:
i. The capability to make decisions to take on, lay off, or decline a risk-bearing
opportunity, together with the actual performance of that decision-making
function.
ii. The capability to make decisions on whether and how to respond to the risks
associated with the opportunity, together with the actual performance of that
decision-making function.

iii. The capability to mitigate risk - that is the capability to take measures that
affect risk outcomes together with the actual performance of such risk
mitigation.
Financial capacity refers to an enterprise's capability to access funding when

managing risk as well as absorbing the consequences of risk in the event of an
unfavourable outcome (see paragraph 1.64).
Points i and ii above relate to the control over risk as noted in paragraph 1.65
where it is further stated that in order to have control over risk, an enterprise is not
required to perform the risk mitigation activities itself, but it is required to be
actively involved in the decision process when outsourcing these activities.
The above requirements ensure that ‘cash box’ companies within a MNE (ie
capital rich entities without any other relevant economic activities) cannot be
allocated high returns unless they have control of risk as defined.
The OECD final report on financial transactions (see later chapter) added new
paragraphs 1.107 to 1.126 to Chapter I Section D1.2.1 on the analysis of risk which
gives guidance on when a funder should receive merely a risk-free return rather
than a risk-adjusted return. This guidance will be looked at in more detail in a later
chapter on the functional analysis.
The six-step analysis of risk and control of risk are an important part of
understanding correct delineation of controlled transactions. You should look at
this section of the OECD TPG before you continue through this manual. You may
also find it useful to read through Sections 3.4.4.21 to 3.4.4.45 (section B) of the UN
Manual where you will find a further discussion, a flow chart and worked examples
on risk analysis.
Characteristics of Property or Services
Here we are going to mention some of the approved methodologies such as

Comparable Uncontrolled Price (CUP) and Transactional Net Margin Method
(TNMM).
Differences in the specific characteristics of property or services can account for

differences in their value. If we are going to compare two transactions then we
must look at the potential impact of differences in the product or service.
Paragraph 1.107 (1.137) of Chapter I of the OECD TPG sets down some guidance
on factors to consider as follows:
In the case of transfers of tangible property:
• the physical features of the property;
• quality and reliability;
• the availability and volume of supply;
In the case of the provision of services:
• the nature and extent of the services;
In the case of intangibles (referred to as intangible property):
• the form of transaction (eg licensing or sale);
• the type of intangible property (eg patent, trademark, or know-how);

• the duration and degree of protection, and the anticipated benefits from the
use of the intangible property.
Differences in the characteristics of the property or service may impact on the

transfer pricing method selected as the most appropriate. We will explain each of
the transfer pricing methods in more detail in a later chapter. For now we will just
illustrate the interaction of the characteristics of the product or service and the
chosen transfer pricing method by looking at a transaction involving televisions.
(Don’t worry about understanding this illustration fully now – you can come back
to it once you have studied the methodologies in a later chapter).
If we are able to identify market data for the wholesale price of televisions, we
may be able to use a CUP method. In this case the specification of the products in
question would be very important. Factors such as screen size, 4K resolution,
internet access and others would all impact the price. In order to use the third
party data to set or test the transfer price, the third party products would need to
be almost identical. If the third party data in question related to televisions that
had 4K resolution and the tested party products did not, one would expect the
third party products to be materially more expensive. As such they could not be
used as comparable data to apply the CUP.
However, if we were looking at distribution companies and the TNMM appeared

to be the most appropriate method considering all the comparability factors, the
degree of comparability required would be much less. As we will see later, the
TNMM considers the net margin earned by parties for their activities undertaken.
Unless there is robust evidence to the contrary, there would be no reason to
believe that a distributor of televisions would earn a different operating margin on
one type of television compared to another. Indeed, it is likely that the margins
earned on distribution of all types of durable consumer electronics would be
similar. Therefore, comparability requirements are much less onerous in applying a
TNMM approach.
Economic Circumstances
Economic circumstances relate to the broader context in which the transaction

takes place. There are many factors that are beyond the control of the related
parties that would influence the arm’s length price for any given transaction. The
most obvious example is the state of the economy. If we consider the connected
distributor purchasing televisions against a backdrop of very low economic growth
and poor consumer confidence, this will have a significant impact on the demand
for televisions and other consumer products. This will impact the overall profitability
for the group, and level of profitability that the distributor might expect to make.
Another factor may be the degree of competition amongst customers. For the
connected distributor, there may be a large number of significant customers and
a high degree of competition at the retail level in its home country. This would
enable the distributor to earn a higher margin. In other territories, there may be
much less competition, with one or two retailers accounting for the vast majority of
sales. Those customers would have much more buying power, driving down the
profit potential for the connected distributor.
In undertaking comparability analysis, it is therefore necessary to consider whether

the third parties face the same economic circumstances as the tested party, and
if not, whether those differences are material. Comparable data from different
markets, geographic or otherwise, should not automatically be excluded but do
require careful evaluation.

Business Strategies
The OECD TPG acknowledge that business strategies will play an important role in
determining the arm’s length price. In practice, this is most commonly considered
in the context of market penetration. The Japanese group and its products may
be new to the UK. It may therefore be the case that it incurs abnormally high set-
up costs and additional marketing and promotional costs to make consumers
aware of the new product, whilst at the same time being unable to command the
same market premium as more established market participants. Under such
circumstances, it may therefore be acceptable for the UK distributor to earn lower
profits (or even losses) than comparable companies whilst still paying an arm’s
length price.
However, it should be cautioned that there should be consistency between the

business strategy and the functional analysis. If in the long term the UK distributor is
to be considered a low-risk distributor, with very little responsibility for managing
market risks in the UK, then it would not be expected to incur the costs of starting
up the UK business and establishing the brand. Under those circumstances, the
transfer price would need to be lowered to allow the UK distributor to earn
sufficient profit from a very early stage.
3.5 Recognition of the Accurately Delineated Transaction
Section D2 begins by stating that where the characteristics of a transaction are

different to the contracts the transaction must be delineated in accordance with
the actual conduct of the parties.
Every effort should be made to determine the price for the accurately delineated
transaction.
Tax authorities should only disregard a transaction (non-recognition) if the

exceptional circumstances set down in paragraphs 1.122 to 1.125 (1.142 to 1.145)
of the OECD TPG apply.
The OECD TPG recognise the fact that associated enterprises may have the ability
to enter into a much wider variety of arrangements than their unrelated
counterparts, and therefore calls for an analysis of whether the arrangement
possesses commercial rationality that would be seen should such a transaction
take place between unrelated parties. An accurately delineated transaction can
be disregarded and where appropriate replaced with an alternative when the
arrangements made in relation to the transaction when viewed in their totality
differ from those that would have been entered into by independent enterprises
behaving in a commercially rational manner (see paragraph 1.122 (1.142)).
For non-recognition, the key question is whether the actual transaction possesses
the fundamental economic attributes of arrangements between unrelated
parties, not whether the same transaction can be observed between
independent parties. The non-recognition of a transaction that possesses the
fundamental attributes of an arm’s length arrangement is not an appropriate
application of the arm’s length principle. An arrangement should be considered
to include these fundamental economic attributes if it offers each of the parties a
reasonable expectation to enhance or protect their commercial or financial
positions on a risk-adjusted basis, compared to other opportunities realistically
available to them (including the alternative of not entering into the transaction) at
the time of entering into the arrangement. It is also relevant to consider whether
the multinational group as a whole is left worse off on a pre-tax basis (see
paragraph 1.123 (1.143)).

Where non-recognition takes place, the structure will be replaced by the

alternative transaction that corresponds closely with the facts of the actual
transaction and affords the parties the opportunity to come to a price acceptable
to both of them at the time the transaction was entered into (see paragraph 1.124
(1.144)).
Section D2 finishes with a couple of illustrative examples which are worth reading
through as part of your studies. We will look again at Section D2 in a later chapter
when we look at non-recognition.
We will also consider the Cameco case (Cameco Corporation [2020 FCA 112]
which considered the issue of non-recognition (referred to as recharacterisation in
this case which we will explain later).
3.6 Losses
Independent enterprises sometimes make losses so on first principles a loss

sustained by an associated enterprise may not in itself be an indicator that prices
are not arm’s length. However an independent enterprise would not be able to
continue to make losses in the long term. The OECD TPG at paragraphs 1.129 to
1.131 (1.149 to 1.151) address the subject of losses and the impact for the arm’s
length principle. It is noted that the existence of losses may lead to scrutiny by the
state where the associated enterprise is based.
It is recognised that in the short term losses can be justified as a way to break into
a new market or launch a new product but in the long term such an approach
would not be maintained by an independent enterprise.
In the case of an MNE it may be that there are business reasons why some non
profit making products are kept; it may be that the MNE wants to be able to say it
supplies the full range of products. The OECD TPG point out that in this case if the
associated company making the non profit making product was an independent
enterprise it would only produce the product if it was compensated in the form of
a service fee so that in the end it did make a profit. The OECD TPG suggest that
this may be the approach that needs to be taken within an MNE – a service fee to
the associated enterprise that would otherwise be loss making.
3.7 The Effect of Government Policy
In some jurisdictions we see government interventions into the market in the form
of price controls, subsidies, interest rate controls, exchange controls, anti dumping
policies etc (see paragraph 1.132 (1.152)). The question then arises as to how these
interventions should be taken account of in price setting between associated
enterprises and as a consequence their impact for the arm’s length principle.
As always this question is to be answered by looking at what would happen

between independent enterprises. However this is not a straightforward question.
The government intervention may affect the final price to the consumer but this
does not mean that it is the final seller that should take the impact. Independent
enterprises will want to maximise profits so there may be negotiation within the
supply chain. The final seller will seek to pass on some of the impact of the
intervention to the intermediate supplier; their ability to do this will be affected by
many factors.
The OECD TPG paragraphs 1.134 – 1.136 (1.154 – 1.156) devote some time to
considering the impact for the arm’s length principle of blocked payments. What
will happen if the government intervention means that payment cannot be

received (for example if there are exchange controls that prevent the funds
leaving the country)? Would an independent enterprise enter into a transaction
where there was a real risk that they would not receive payment? As always we
have to look at all the terms of the transaction; an independent enterprise may
seek a higher profit level to compensate for such risk or they may seek some kind
of service fee.
Where the policy blocking payments applies equally to independent enterprises as

it does to associated enterprises then there will be information available to see
how independent enterprises deal with the policy. For example, do they continue
to enter into transactions and what terms do they build into the agreements?
If the policy only affects associated enterprises it follows that it is harder to establish
what independent enterprises would do in the same circumstances. The OECD
TPG suggest that where independent enterprises would not enter into such
transactions then the party who cannot receive the payment should be treated as
performing a service for the MNE. In addition the OECD TPG put forward an
alternative of preventing the party that cannot receive the payment from being
able to deduct related expenses until the payment is received. The concern will
be with protection of the tax base and in the final analysis an enterprise should not
be able to treat a blocked payment from an associated enterprise any differently
than when from an independent enterprise (see OECD TPG paragraph 1.136
(1.156)).
3.8 Location Savings and Local Market Features
Location savings are the cost savings made by moving to a different geographical
location. A typical example would be a lower wages bill due to cheaper labour.
There may also be increased costs such as the cost of transport. Location savings
are the net of the savings and increases.
Section D contains some guidance on location savings and local market features.
It begins by explaining that the points made on the subject in Chapter IX on
Business Restructuring also apply in a wider context (we look at Chapter IX in a
later chapter).
To the extent that location savings exist and are not passed on to customers or
suppliers it is necessary to determine how they would be allocated in an arm’s
length situation. This can be done by searching for a reliable local comparable. In
the absence of a local market comparable, the analysis needs to be done based
on all the facts and the functional analysis as set down in Chapter IX.
There is a recognition that there can be other local market features such as a
skilled workforce or expansion or contraction in a market that can give rise to
comparability concerns not related to location savings. Here if local market
comparables can be found there will be no requirement for comparability
adjustments as the same features will apply to the comparable. In the absence of
a local comparable it is necessary to identify if an advantage exists, to what
extent any advantage remains with the MNE and assess how it would be
allocated in an arm’s length transaction.
The section on other local market features states that there may be intangibles,
such as contractual rights and government licences that should be considered.
Cross reference is made to Chapter VI of the OECD TPG for further guidance and
examples are provided to distinguish between licences and rights that can “affect
the manner in which the economic consequences of local market features are

shared between parties to a particular transaction” and those that are available
to many, if not all, of those in the market.
3.9 Assembled Workforce
The 2010 version of the OECD TPG did not have a section on an assembled
workforce. One was added into the 2017 version following work on intangibles to
clarify the OECD position. The opening paragraph states:
‘Some businesses are successful in assembling a uniquely qualified or

experienced cadre of employees. The existence of such an employee group
may affect the arm’s length price for services provided by the employee
group or the efficiency with which services are provided or goods produced
by the enterprise. Such factors should ordinarily be taken into account in a
transfer pricing comparability analysis. Where it is possible to determine the
benefits or detriments of a unique assembled workforce vis-à-vis the workforce
of enterprises engaging in potentially comparable transactions, comparability
adjustments may be made to reflect the impact of the assembled workforce
on arm’s length prices for goods or services.’ (See OECD TPG, Chapter I, 1.152
(1.172))
It is also stated that if a workforce is transferred from one entity to another as part
of a restructuring, which will reduce the time, effort and expense that would have
been incurred hiring a new workforce, then it is possible that a comparability
adjustment might be required in respect of the arm’s length price.
3.10 MNE Group Synergies
This section is also an addition to the OECD TPG. With respect to group synergies it
is put forward that arm’s length remuneration is only required for deliberate
concerted synergies. Paragraph 1.161 (1.181) states:
‘Where corporate synergies arising from deliberate concerted group actions

do provide a member of an MNE group with material advantages or burdens
not typical of comparable independent companies, it is necessary to
determine
– the nature of the advantage or disadvantage.
– the amount of the benefit or detriment provided, and
– how that benefit or detriment should be divided among members of the

MNE group.’
The guidance refers to the example of centralising purchasing power as a

deliberate action whereas an entity receiving more favourable terms simply
because the supplier hopes to attract business from others in its group is deemed
coincidental.
Five examples are given to demonstrate the principle of group synergies. We

recommend that you review them.

3.11 UN Manual Chapter B2 on Comparability
Part B3 in the UN Manual helps explain the approach taken in Chapter I of the
OECD TPG on accurately delineated transactions as the starting point for a
transfer pricing analysis.
As noted above it has a section on risk and control over risk.
Sections B3.4.5.6–B3.4.5.17 cover location savings, expanding on the OECD

definition and adding the term location specific advantages (‘LSAs’) which are
the sum of location savings plus specific local factors such as a highly specialised
skilled manpower, or proximity to growing markets.
LSAs can lead to what the UN Manual calls location rent which is defined as the
incremental profit derived from LSAs. Even if LSAs exist there may be no location
rent eg where the market is competitive and LSAs have to be passed to third party
customers through lower prices. In other cases the whole of the location rent may
pass to the group in the short term until competition erodes the profit.
The allocation of location rent depends on the competitive factors in the market.
Some examples are given:
• The group could have production intangibles that allow it to manufacture at a

lower cost than competitors. At arm’s length, the owner of the intangible
would be entitled to the rents associated with this cost saving.
• The group low cost producer may be the first to operate in the low cost
jurisdiction and there are no comparable low cost producers in its or other
jurisdictions. Therefore, the low cost producer can take advantage of the
location rents.
The question is then how to split these profits. The guidance is fairly limited but does
refer to the relative bargaining position of the group as one solution to how to do
this split.
We will look at this topic again in more detail in a later chapter on business
restructuring.
3.12 Application of Comparability Analysis
Comparability analysis will be examined in more detail in a later chapter. We need

to mention it here as to apply the principles above, it is typically necessary to
undertake a comparable search. This may be a search for comparable
transactional data, but more typically this involves searching for companies
involved in comparable activities to the related party chosen to be the tested
party.
Even though there are a number of criteria for choosing comparables that are
easy to follow (eg industry, location, time, etc.), the decision whether to accept or
reject a potential comparable bears a certain level of subjectivity which is often at
the centre of challenges by the tax authorities (eg does the business description
give enough details to decide if the company should be added to the set of
accepted comparables? And even if the business operates in the same industry,
market and location how can we establish differences in strategy, commercial
goals, etc.?).
In theory, the larger the sample of comparables, the greater the likelihood of
identifying a tighter and more defensible range, since the impact of outliers will be

reduced. However the availability of comparables that can be used for transfer
pricing studies has become more of a concern in recent years.
A third party business can only be used as a potential comparable if it meets strict
independent criteria. The lack of independent comparables has made the
comparability analysis process even more difficult. One reason for this is that
globalisation and the last financial crisis have led to increased competition and
smaller players being taken over by the larger groups; hence, the number of
independent parties in all industries has decreased.
A further complication is the increase in information available through the internet.

Whereas many years ago, comparability would have predominantly been
determined through a short business description on publicly available databases,
and through financial statements, a larger number of companies now have their
own websites providing much more information about activities undertaken.
Ironically, this additional information usually serves to highlight the lack of
comparability between the tested party and the potential comparables.
Regardless of the guidance provided by the OECD TPG to enhance

comparability, the element of subjectivity remains. Running sensitivity analysis on
the comparable set (eg varying the acceptance criteria to either include or
exclude comparable enterprises) can be very valuable.
For example, if the tax authorities are challenging the comparability of some of the
businesses in the final set of comparables (ie the set used to build the arm’s length
range), having run sensitivity analysis and choosing a profit margin (or other profit
indicator depending on the transfer pricing method being chosen) that is within
the range for different sets of comparables could lower the risk of the tax
authorities demanding an adjustment.
When using transfer pricing methods (eg TNMM) where third party comparables
are used to generate the arm’s length range for the comparison with the tested
party, it can be often difficult to identify comparables, which deal with
comparable services or products and operate in the same industry. Running two
analyses based on functional and industry comparability then comparing the
results can help make the analysis more robust and support the pricing with the tax
authorities.

CHAPTER 4
TRANSFER PRICING METHODS
In this chapter we are going to look at the methodologies set down by the OECD
Guidelines, in particular:
– Comparable Uncontrolled Price Method (CUP)
– Resale Price Method (RPM)
– Cost Plus
– Transactional Profit Methods
4.1 Introduction
Selecting the appropriate transfer pricing method is key both during planning for a
new transaction/product/service and when putting in place documentation to
support the current transfer pricing.
It is preferable to look at transfer pricing methods prior to setting up intra-group

transactions as it limits the risk of exposure in case of an audit.
Testing an existing pricing policy by choosing one of the transfer pricing methods
does not always guarantee that the current pricing will be supportable (ie at arm’s
length).
The OECD TPG deal with transfer pricing methods in Chapter II and provide a
description of all the acceptable methods and when they should or could be
applied. The methods are broken down into two types; traditional transaction
methods (CUP, RPM, Cost Plus) and transactional profit methods (TNMM,
transactional profit split method).
It is important to understand all the methods; however, it is even more important to

understand how to apply them and when they should be chosen.
4.2 The OECD Guidelines
Up until the 2010 edition, the OECD TPG presented a hierarchy of methods;
therefore, the choice was dictated mainly by the availability of data and the
taxpayer had to start with the preferred method and work their way down if the
higher ranking method could not be applied.
In 2010 the OECD TPG introduced a change in the way the method should be
chosen. The hierarchy no longer exists and the choice of method is based on the
optimal method and best fit method. In other words, the taxpayer should choose
the method that best describes the transaction under test and that also reflects
the functional and risk profile for the transaction.
We find the methodologies in Chapter II of the OECD TPG. Note Chapter II was
updated following the final report on the BEPS Action Points 8 to 10. The updates
are noted below.
We will look first at each method and how they should be applied and then
understand how the appropriate method should be chosen and go through a few
examples.

4.3 Comparable Uncontrolled Price (‘CUP’) Method
‘The CUP method compares the price charged for property or services
transferred in a controlled transaction to the price charged for property or
services transferred in a comparable uncontrolled transaction in comparable
circumstances. If there is any difference between the two prices, this may
indicate that the conditions of the commercial and financial relations of the
associated enterprises are not arm's length and that the price in the
uncontrolled transaction may need to be substituted for the price in the
controlled transaction’. (OECD TPG Chapter II, 2.14.)
We can subdivide CUPs into two types – internal CUPs and external CUPs. An
internal CUP is available when an enterprise sells the same product or service to a
third party as it does to an associated enterprise. An external CUP is a transaction
between two unconnected parties. As a general rule internal CUP will give rise to
more reliable data.
The CUP method makes reference to the basic arm's length principle under which
related parties should interact as if they were not related. Therefore, if an
enterprise sells a particular product to a third party for a certain price, the same
price can be used to sell the same product to a related party.
However, as the OECD text highlights, there might be differences between the
third party transaction and the related transaction. For example, a manufacturer
might sell a product to third party distributors for a certain price, which also
includes the sales and marketing efforts of the manufacturer to become the
product supplier of the distributor. However, when the same manufacturer sells to
a related party, the sales and marketing efforts are almost non-existent (as the
related party distributor is more likely to buy and distribute a product
manufactured by the group it belongs to). Therefore, using the third party price
might result in overcharging the related party (in this example).
Some adjustments might be required to ensure the CUP can be used to price the
intra-group transaction. The main comparability criteria to take into account when
deciding if a CUP is applicable or can be applied (after being adjusted) are as
follows:
• Functional profile (ie does the controlled party carry out the same functions to
deliver the service or product in the uncontrolled transaction as in the
controlled transaction?);
• Risk profile (ie does the enterprise bear the same risk in both the controlled and
uncontrolled transaction?);
• Cost base (ie does the enterprise bear more or less cost in the uncontrolled
transaction?); and
• Contractual terms (ie are there any differences between the contract for
uncontrolled and controlled transactions? For example, payment terms, return
policy, cancellation terms, etc).
Depending on the number of transactions and products/services to be tested the

use of CUPs can become very onerous for the taxpayer. Furthermore, differences
in contractual arrangements can make the use of CUPs difficult.
As explained before, CUPs can be adjusted; however, implementing several

adjustments can result in making this method too artificial.

Product comparability should be closely examined in applying the CUP method. A

price may be materially influenced by differences between the goods transferred
in the controlled and uncontrolled transactions, although the functions performed
and risks assumed (eg marketing and selling function) are similar so as to result in
similar profit margins. The CUP method is appropriate especially in cases where an
independent enterprise sells products similar to those sold in the controlled
transaction.
Although product comparability is important in applying the CUP method, the

other comparability factors should not be disregarded. Contractual terms and
economic conditions are also important comparability factors.
When looking at CUPs (whether internal or external), in the absence of a perfect

CUP, we can identify close CUPs and inexact CUPs. These are the result of
uncontrolled transactions that are adjusted to take account of material
difference. Reliable adjustments may be possible for difference regarding the
source of the products, difference in delivery terms, volume discounts, product
modifications and risk incurred.
Reliable adjustment may not be possible for trademarks. Adjustments also cannot
be made to account for material product differences – the CUP method may not
be the appropriate method in such a case.
Difficulties resulting from performing reasonably accurate adjustments to remove

the effect of material differences on prices should not automatically prevent the
use of the CUP method. One should try hard to perform reasonable adjustments.
If reasonable adjustments cannot be performed, the reliability of the CUP method

is decreased. Another transfer pricing method may then be used in combination
with the CUP method or considered instead of the CUP method.
The CUP method used to be the number one method in the hierarchy. Although
the hierarchy no longer exists, some tax authorities and tax inspectors in general
tend to always look for the existence of CUPs. Therefore, it is good practice to
always consider the CUP method and either provide an explanation why it cannot
be used for a specific transaction or to use this method (even when it is not the
most appropriate) as a complementary method.
The CUP will often be used where the controlled party enters into a similar
transaction with an unconnected party on similar terms. The CUP can also be used
for fungible products; some examples would be chemicals, pens, paperclips. CUP
can be used for animal products such as offal and harvested crops like oats and
barley. It is possible to use CUP for branded products that are fungible such as
contact lenses. These are just some specific examples – remember the key is
comparability so CUP can be used whenever we have sufficient comparability
taking account of all the comparability factors and their relative importance.
Commodity trading is seen as important for developing countries; there is concern

that it is an area that could lead to base erosion and profit shifting. Additionally
there is a concern that some countries have issued domestic approaches to price
commodity transactions (eg the so-called sixth method in the Latin America
region).
Additional paragraphs added to Chapter II following the final report on BEPS

Action Points 8 to 10 clarify that the CUP method would generally be the most
appropriate transfer pricing method for determining the arm’s length price for
controlled commodity transactions, and that, under the CUP method, the arm’s
length price for the controlled commodity transaction can be determined, not

only by reference to comparable uncontrolled transactions, but also by reference

to a quoted price (see paragraph 2.18).
This is of course subject to the overall requirement of the OECD TPG that the most
appropriate method should be used.
The OECD TPG define commodities as referring to ‘physical products for which a
quoted price is used by independent parties in the industry to set prices in
uncontrolled transactions’ (see paragraph 2.18).
The guidance states that quoted or public prices can be used, provided the
conditions of the controlled transactions are comparable to the conditions of the
quoted prices. A relevant factor in determining the appropriateness of using
quoted prices is the extent to which such price is widely and routinely used in the
industry to negotiate prices between third parties (see paragraph 2.19).
Paragraph 2.20 stresses the importance of comparability and discusses when it will
be necessary to make adjustments for differences in comparability.
Paragraph 2.22 states that when using quoted prices the pricing date is
particularly relevant. Paragraph 2.22 also specifies clearly when the pricing date
can be disregarded and explicitly requires the tax authority to suggest any
alternatives based on market data, as opposed to choosing dates that maximise
the tax base.
The guidance puts a lot of requirements on taxpayers in terms of understanding,

analysing, and documenting complex commodity supply chains and pricing
mechanisms. Further paragraph 2.21 states that taxpayers should release this data
as part of transfer pricing documentation.
Where information is not available from the taxpayer the tax authority may deem
the price based on the available evidence. This may be the date of shipment
based on the bill of lading for example. The result would be that the price would
be based on the average price for the day of shipment, subject to comparability
adjustments. The point is then made that in these cases it will be important to
ensure that cases can be resolved via the MAP procedure. (See paragraph 2.22.)
UN Manual and ‘The Sixth Method’
The use of quoted commodity prices is included in part of the explanation of the
application of the CUP method in the OECD TPG as noted above.
Chapter B3 of the UN Manual covers the selection and application of the five
OECD approved methodologies. It also includes guidance on the commodity rule
which it lists as the ‘sixth method’. This is described as a method used by
developing nations when they lack sufficient documentation from the controlled
transaction to determine if it is comparable with uncontrolled transactions. The tax
authorities in these cases will rely on the quoted prices of the commodities market
to price commodity transactions between associated enterprises.
To date, practical experience with the rule exists in mainly in Argentina, Brazil,
Ecuador and Uruguay. However, the method is not applied unequivocally the
same in all these countries.
It is recognised that the workings of the sixth method may resemble the CUP
method, however no attempt is made to reconcile them or state the impact of
the sixth method for the arm’s length principle. Some countries consider the sixth
method an (imperfect) application of the CUP method.

It is noted that the sixth method could be considered as an anti-abuse rule, an

abuse-deeming rule or even presumed a form of safe harbour (see later chapter).
The sixth method can be used to prevent MNEs from shifting profits to low tax
jurisdictions. This approach can be taken where a commodity seller issues an
invoice for the sale of a commodity to an associated enterprise (the trading
company). However the goods are not shipped to that trading company but
instead are shipped directly to a third party purchaser who has entered into a
contract to buy the commodity with the trading company. This is illustrated below.
If the selling price of the commodity is $1.30 per pound then by invoicing the
trading company at a price of $ .090 per pound a large amount of the profits from
the sale of the commodity will go to the trading company that is in the low tax
jurisdiction.
The tax authorities of the state producing the commodity are losing tax revenue.
The sixth method can be used to prevent this tax avoidance. In some cases, the
sixth method will be used whenever the above structure is in place; in others only
when the trading company doesn’t have substance. It follows that the sixth
method could be classed as an anti-abuse rule rather than a transfer pricing rule.
Compensating adjustments via Article 9 of a double tax treaty will not be
available (unless specific provision has been made in the treaty for the use of the
sixth method) as this deals with transfer pricing adjustments rather than
adjustments as a result of anti-abuse rules. Double taxation may arise. The same
risk of double taxation will be in point if the sixth method is classed as a safe
harbour.
The UN Manual, like the OECD TPG, puts a lot of emphasis on the importance of
timing for the pricing. It is noted that in cases using intermediaries the goods are
frequently shipped after the invoice date and tax authorities will often use the
commodity price at the date that the goods are shipped.
The sixth method has the benefit of legal certainty and relative ease of application
for the tax authorities and its tax collection efficiency. However, it also has the
disadvantage that it is not one of the OECD approved methods and thus may not

be recognised by the country of the associated enterprise at the other end of the
transaction; this could lead to double taxation. There are also other disadvantages
outlined in this section of the UN Manual.
The conclusion in the UN Manual is that if the taxpayer is allowed to present

evidence that any intermediaries have substance and if the legislation permits
comparability adjustments then it is possible that the sixth method will be in line
with the arm’s length principle. In addition increased sophistication with the use of
the sixth method will provide legal certainty for the tax payer and transparency for
the tax authorities.
4.4 Resale Price Method (‘RPM’)
Note RPM is sometimes referred to as resale minus method.
‘The resale price method begins with the price at which a product that has
been purchased from an associated enterprise is resold to an independent
enterprise. This price (the resale price) is then reduced by an appropriate gross
margin on this price (the “resale price margin”) representing the amount out
of which the reseller would seek to cover its selling and other operating
expenses and, in the light of the functions performed (taking into account
assets used and risks assumed), make an appropriate profit. What is left after
subtracting the gross margin can be regarded, after adjustment for other
costs associated with the purchase of the product (eg customs duties), as an
arm’s length price for the original transfer of property between the associated
enterprises.
This method is probably most useful where it is applied to marketing

operations’. (Paragraph 2.27)
known
Associated ALP Associated 3rd party
→
Enterprise 1 → Enterprise 2 buyer
price
£
Selling price Known ↓
Transport costs (X)
Advertising (X)
Other costs (X)
Resale price margin* (X)
Arm’s length price X
*This is a known.
For the RPM method the starting point is the sales company (Associated Enterprise
2 in the illustration above). The focus here is on the gross margin. The product
would be looked at to determine the gross margin that should be earned by the
sales company. The focus on gross margins means that there is less emphasis on
product comparability than the CUP method. However, although product
differences are less important, if there are significant differences then this may
result in a loss of comparability (paragraph 2.31).
Gross profit is the net price for the products sold less the cost of goods sold.

The gross margin is calculated as Gross Profit/Revenue.
We can illustrate this with a simple example.
 Illustration 1
£
Sales 500,000
Cost of sales 350,000
Gross Profit 150,000
Here the gross profit margin is 150,000/500,000 = 30%
Accounting consistency is important in determining the cost of goods sold. It is

possible to make adjustments once we know that there are differences in
classification of costs or for example if we know that different valuation methods
are being applied to inventory.
 Illustration 2
Company X makes computer accessories which are unbranded. The accessories

are sold to a related Company Y who is the group distributor for Europe. It has
been determined that the appropriate gross profit in the European market for the
distribution of computer accessories is 25% of total sales. Company Y has final sales
to third parties of 1 million. Its cost of sales to third parties is 150,000. To determine
the transfer price for the goods from Company X to Company Y we take 1 million
as our starting point. Multiplying it by 25% gives us the gross profit on total sales of
250,000. We then deduct the 150,000 of costs to third parties, leaving the transfer
price as 600,000.
Selling price (total sales) 1,000,000

Cost to third parties (150,000)
Gross Margin (as established 25%) (250,000)
Transfer price balancing figure 600,000
If we turn this around we have
Selling price (total sales) 1,000,000

Cost to third parties (150,000)
Transfer price balancing figure (600,000)
Gross Margin (as established 25%) 250,000
The RPM is as the OECD TPG describe it the most appropriate method when testing
sales and marketing activities. This method is similar to the way a wholesaler
calculates the price to a distributor (ie by taking out its cost plus a margin and
arriving at a discount on the resale price for the distributor).
As we are looking at gross margins then accountancy consistency will be very

important and it will have a big impact on comparability.
For the RPM method to be the most appropriate method, functional comparability
is very important – more important than product similarity although as stated
above major differences in product can have an impact.
What we are looking for is the impact the product has on the distribution method.
Going back to illustration 2, computer accessories take many forms but we don’t
need to look at each individual product line when doing the RPM analysis.
However we would start to lose comparability if we were comparing computer

accessories with let’s say cycle accessories which may require different transport
considerations.
The risks taken, contractual terms and the market place will also impact on the
suitability of the RPM as the most appropriate method. We will return to this topic
again in a later chapter when we look at entity characterisation (sometimes
referred to as entity classification).
What happens if there are no internal comparables (ie the wholesaler does not sell
products the same way to third parties as it does to related parties) and looking for
external comparables does not return significant results?
How can the taxpayer estimate the level of discount to the distributor ensuring the
overall gross margin retained by the wholesaler is at arm’s length?
If we look at the overall supply chain for the product we can identify the following
steps:
• Procure;
• Research and develop;
• Make;
• Market and advertise;
• Sell; and
• Support and customer relationship.
If a manufacturer sells directly to customers it is likely to own (or outsource at a

cost) each step of the supply chain. In the case where a manufacturer sells to a
related party distributor the last three steps in the supply chain are owned by the
distributor. Therefore, there is a split of functions and risks between the
manufacturer/wholesaler and the related party distributor.
The end price to the customer (ie the resale price) can be adjusted (ie
discounted) to the distributor by adjusting the price based on the functional and
risk allocation. Cost can be a useful allocation key to calculate the discount in
combination with the risk profile for each of the functions no longer carried out by
the manufacturer.
For example, if a manufacturer sells directly to customer a product for 100 and its
cost base (covering the entire supply chain) is 80 it makes a 20 profit.
However, when it sells via a related party distributor all the marketing, advertising,
selling and customer service efforts fall on the distributor. Therefore, if the cost of
these functions to the manufacturer when selling directly to customer is 16 and we
use cost as the main indicator or allocation key, 20% (ie 16/80) discount should be
applied to the resale price when selling to the related party distributor.
As noted above it is important to ensure that the cost base is homogenous (eg
comparing cost of labour for each of the supply chain steps and including where
necessary amortisation of assets when the assets add considerable value to the
process) and that the risk associated with each function is also taken into account.
For example, the cost of R&D might not be as high as the manufacturing cost, but
if the R&D process generates a very valuable intangible (eg a design that makes

the product much more sellable) the cost of R&D should be adjusted to reflect its
true value in the supply chain.
The RPM as you can see from the above is often used for distributors and
marketing companies. For example, the gross profit for a distribution company
distributing cycles will be largely unaffected by product differences. In this context
the distribution of tennis racquets compared to squash racquets will not impact on
the expected gross profit margin even though the goods are not interchangeable
and may not have the same final sales price.
The RPM can be used where we have a fully-fledged manufacturer that owns the
intangibles selling to a related sales company that then sells to third parties. We
can also use the RPM when there is a commissionaire agreement to determine the
arm’s length commission that the commissionaire will receive.
Again in determining whether the RPM is the most appropriate method the five
comparability factors need to be considered as discussed in paragraphs 2.27 to
2.45 of the OECD TPG.
4.5 Cost Plus Method (‘C+’)
‘The cost plus method begins with the costs incurred by the supplier of
property (or services) in a controlled transaction for property transferred or
services provided to an associated purchaser. An appropriate cost plus mark-
up is then added to this cost, to make an appropriate profit in light of the
functions performed and the market conditions. What is arrived at after
adding the cost plus mark-up to the above costs may be regarded as an
arm's length price of the original controlled transaction. This method probably
is most useful where semi finished goods are sold between associated parties,
where associated parties have concluded joint facility agreements or long-
term buy-and-supply arrangements, or where the controlled transaction is the
provision of services’. (See paragraph 2.45).
The cost plus method uses cost as the main driver to arrive at an arm’s length
margin. The mark-up applicable to the cost reflects the return the enterprise aims
to achieve and should reflect the value added by the enterprise bearing the cost.
In the case of goods, we are looking at gross profit mark-up as the financial ratio.
Gross profit is defined as net sales minus cost of goods sold. The cost of goods sold
should be the cost of producing the goods sold – this will normally include direct
labour costs, material costs and the overheads from the factory/premises used to
produce the goods (a similar approach is taken for services). Clearly accountancy
consistency will be important as cost needs to be divided into direct and indirect.
Functional comparability will also be important if cost plus is to be used as each
additional function will have its associated costs. In applying this method we need
to decide whether to use actual costs or budgeted costs. Third parties will often
use budgeted costs with an adjustment clause in certain circumstances.

 Illustration 3
Company T assembles computers and sells them to a related Company S who will
distribute them. It is established that the appropriate gross mark-up is 30%. The
costs incurred in assembly are 300,000. The formula we use is 1+ mark-up (30%)
which gives a transfer price of 390,000 so that Company T will have:
Sales (TP in this case) 390,000

Cost of sales (300,000)
Gross Profit 90,000
The gross profit 90,000 gives us a mark-up of 30% of the cost incurred.
Per the OECD TPG, this method is particularly useful when looking at semi-finished
products or when looking at services, which do not constitute or are only part of
the finished product or service.
However, although cost can provide a good basis for valuing the input of an
enterprise delivering a service or a product, the mark-up associated with the cost
can exhibit great variance depending on the value added by the enterprise.
To use a similar example to the OECD TPG, if the cost is used as a basis to
calculate an arm's length return for manufacturing a product, depending on the
value added at the manufacturing stage the return should be different. It might
help to use premium luxury goods as an example. Goods might be manufactured
in country A, but designed in country B where the brand is owned.
In the case of luxury goods, the brand allows the distributor to sell the product at a
much higher price than non-branded goods; therefore, even though the
manufacturing cost might exceed the cost of designing and branding the
product, it adds far less value to the finished product. That is the cost basis cannot
always be used as a proxy for determining the value and margin to be retained.
Another important factor to consider when applying the cost plus method is
determining the cost base for the mark-up. Also, should all cost be marked up?
In general terms and for the purposes of transfer pricing, third party cost can be
recharged without a mark-up as the mark-up should indicate that the enterprise
charging the related party has added some value. The concept of adding value is
very important as when subject to a tax audit and in particular in some jurisdictions
(eg Belgium) cost plus recharges can be challenged by tax authorities if there is no
clear value added.
Furthermore, applying the cost plus method presents a number of difficulties in

relation to how cost is managed and what strategy drives spending and
investment patterns in a business as the OECD TPG highlight in the extract below.
‘The cost plus method presents some difficulties in proper application,

particularly in the determination of costs. Although it is true that an enterprise
must cover its costs over a period of time to remain in business, those costs
may not be the determinant of the appropriate profit in a specific case for
any one year. While in many cases companies are driven by competition to
scale down prices by reference to the cost of creating the relevant goods or
providing the relevant service, there are other circumstances where there is
no discernible link between the level of costs incurred and a market price (eg
where a valuable discovery has been made and the owner has incurred only
small research costs in making it)’. (OECD TPG Chapter II, 2.49)

This issue ties back to the cost versus value argument, where R&D cost might be
much smaller than manufacturing cost, but it is the R&D that makes the product
more sellable and allows the enterprise to charge a premium price.
However, making adjustments to the cost base can make applying an

appropriate mark-up difficult as when looking for comparable transactions it might
not be as easy to identify the cost base in the comparable (especially if this is a
third party comparable). This issue is also encountered when applying the cost plus
method under a TNMM.
Examples of when cost plus is used include contract manufacturing, toll

manufacturing (terms we will look at later when we look at entity characterisation)
and intra-group provision of services. In all cases a full analysis of the five
comparability factors needs to be undertaken to ensure that the cost plus method
is the most appropriate. See paragraphs 2.45 to 2.61 of the OECD TPG for a
discussion of the cost plus method.
4.6 Transactional Profit Methods
Action Point 10 of the BEPS Action Plan sought to clarify the application of the
transfer pricing methods, in particular the transactional profit split. The final report
on Action Point 10 scopes out the work that needs to be done, including
clarification of the circumstances in which transactional profit splits are the most
appropriate method for a particular case, and describing what approaches can
be taken to split profits in a reliable way.
There are two categories of profits methods – the Transactional Net Margin
Method (TNMM) and the Profit Split Method which can be divided further into
contribution analysis (PSM) and Residual Profit Split (RPSM) as you will see as you
work through this section. We will begin with an overview of the transactional profit
methods then look at each one in turn.
‘A transactional profit method examines the profits that arise from particular
controlled transactions. The transactional profit methods for purposes of these
Guidelines are the transactional profit split method and the transactional net
margin method. Profit arising from a controlled transaction can be a relevant
indicator of whether the transaction was affected by conditions that differ
from those that would have been made by independent enterprises in
otherwise comparable circumstances’. (OCED TPG Chapter II, 2.63)
Transactional profit methods used to be regarded as a ‘last resort’ approach in

previous versions of the OECD TPG, when the hierarchy of methods was in place.
Transactional profit methods essentially test the arm’s length nature of a

transaction based on the overall profitability (using a number of profit indicators)
of a related party when compared to a third party.
Finding detailed comparable transactions for each product or service to enable

the taxpayer to use the standard methods previously described can be very
difficult especially when looking at services which are not linked to the main line of
business (eg an electronic component manufacturer is provided management
services by a related entity).
The TNMM and PSM are probably the most commonly used methods in transfer
pricing as apart from transactions which occur with both third parties and related
parties, the majority of large MNEs exchange a number of services and products
for which it is not possible to identify specific CUPs.

The following extract from the OECD TPG details when it is appropriate to apply
each of the two methods.
‘A transactional net margin method is unlikely to be reliable if each party to a

transaction makes valuable, unique contributions, … In such a case, a
transactional profit split method will generally be the most appropriate
method, … However, a one-sided method (traditional transaction method or
transactional net margin method) may be applicable in cases where one of
the parties makes all the unique contributions involved in the controlled
transaction, while the other party does not make any unique contribution. In
such a case, the tested party should be the less complex one’. (OECD TPG
Chapter II, 2.65).
The extract above highlights the concept of ‘complexity’ and makes reference
once again to ‘value’.
Transactional Net Margin Method (TNMM)
Here the focus is on the net profit margin relative to an appropriate base such as
sales or assets. These are referred to as profit level indicators (PLIs).
Here are some examples of PLIs together with an example of when each one may
be used – it is not an exhaustive list:
Example of use
Berry Ratio Berry Ratio Gross Intermediate
profit/operating activities. Where
expenses profit is linked to
operating costs.
Operating Margin OM Operating Where sales is the
profit/sales main driver of profit
eg distribution
companies
Return on Capital ROCE Operating Industries that require
Employed profit/capital a large capital
employed investment eg
Telecoms
Return on Assets ROA Operating Manufacturing or
profit/operating asset intensive
assets industries eg
construction industry
Return on Total ROTC Operating Where costs are the
Costs profit/total costs main driver of profit
eg services
It is possible to use other net profit indicators. The OECD TPG at paragraph 1.05 lists
some; they include floor area for retail spaces and employee numbers. It is
important that the denominator can be accurately measured and is appropriate
for the transactions being looked at.
It is often said that TNMM looks at net profit. As you can see from the above table
with the exception of the Berry ratio the PLIs are based on operating profit.
Operating profit can be defined as sales less cost of goods sold, less labour costs
and other costs connected with the business. It can include depreciation, but will
be before deductions are made for interest and tax, sometimes referred to as
Earnings Before Interest and Tax (EBIT).

To continue our earlier illustration where we calculated gross profit, here we will
calculate the operating margin.
 Illustration 4
Sales 500,000
Cost of sales 350,000
Gross Profit 150,000
Operating costs 100,000
Operating profit 50,000
Here the operating margin is 50,000/500,000 = 10%
The thinking behind the TNMM is that companies in the same industry will tend
towards the same net profit levels over time. If one company had higher returns
than the rest then it would be able to expand at the expense of the others and
eventually the less efficient companies would go out of business.
The ideal comparison would be to compare the net margin of the controlled
transactions to the net margin earned by the same company in uncontrolled
transactions.
As with CUP, RPM and cost plus only one party to the transaction is tested when
using TNMM. This will be the least complex party as generally it is easier to get
information for the least complex party.
As the focus is on net profits there is less emphasis on product comparability.
The TNMM looks at third party comparables that carry out similar activities to the
tested party and measures their profitability. Therefore, if the transaction to be
tested is complex and involves several parties contributing to the overall value
being created, the TNMM cannot be easily applied.
The independence criteria together with the overall globalisation trends (ie fewer
totally independent third party comparables are available for comparison), make
it more difficult to find a potential comparable third party company that fits the
specific functional and risk allocation of a tested party entering into a complex
transaction.
In cases where the TNMM can be applied it is very important to choose the right
profit indicator to test. It is possible to convert from one profit indicator to another;
however, each profit indicator is subject to a number of sensitivities, which might
lower the comparability and generate non-arm’s length results as detailed in
OECD TPG Chapter II 2.82.
The TNMM compares the net profit margin (relative to an appropriate base) that
the tested party earns in the controlled transactions to the same net profit margins
earned by the tested party in comparable uncontrolled transactions or
alternatively, by independent comparable. For example, return on total costs,
return on assets, and operating profit to net sales ratio (as shown in the above
table).

 Illustration 5
Company A manufactures 500,000 designer ski boots. It has been decided that
TNMM is the most appropriate method to apply. It has been identified that the
company operates in an asset intensive industry where utilisation of assets is very
important to remain profitable and stay in business. A comparability study applying
the five comparability factors has identified that companies in the same industry
with the same functional profile make a return on assets (ROA) of 10%. This has
been measured based on the original cost of assets used.
It is established that Company A employs assets with an original cost of 10m. From
this we can calculate that a 10% return means it would expect to have profit of
1m.
We are told that the company produces 500,000 boots. We can now calculate
profits per boot as 1m/500K = 2.
It is established that the costs of production are 10 per boot. Therefore we can say
that the price for each boot needs to be 12. As the sale is to a related party that
means the transfer price is 12.
 Illustration 6
Company B is a distribution company. It distributes cycling accessories such as

helmets, gloves and overshoes. The accessories are all purchased from a related
party. It has been decided that TNMM is the most appropriate method.
Return on sales has been identified as the appropriate PLI. The comparability
analysis identifies 3.5% as the expected return on sales (sometimes referred to as
the operating margin).
We are told that the final sales income for Company B will be 800,000 and that
operating costs will be 72,000.
We know we want an operating margin of 3.5%. We can calculate that the net
profit needs to be 800,000 x .035= 28,000. Therefore the transfer price will be
700,000, calculated as 800,000 – (72,000+28,000) = 700,000.
 Illustration 7
Company X has been asked to provide invoicing services to the rest of the
companies in the group as part of a centralisation exercise.
It has been established that there are comparable independent companies that
can offer the same service and that these companies receive a return on total
costs of 12.5%.
Company X has the following cost relating to the invoice service:
Labour cost 200,000

Office cost 110,000
Operating expenses 40,000
Total cost 350,000
With a return on total cost of 12.5% we can calculate that the profit will be 43,750
(350,000 x 12.5%) giving a total transfer price of 393,750 (350,000+ 43,750).

As such, the TNMM is a more indirect method than the cost plus/resale price
methods that compare gross margins. It is also a much more indirect method than
the CUP method that compares prices, because it uses net profit margins to
determine arm's length prices.
One should bear in mind that many factors may affect net profit margins, but may
have nothing to do with transfer pricing.
The TNMM is used to analyse transfer pricing issues involving tangible property,
intangible property or services. However, it is more typically applied when one of
the associated enterprises employs intangible assets, the appropriate return to
which cannot be determined directly.
In such a case, the arm’s length compensation of the associated enterprise not
employing the intangible asset is determined by determining the margin realised
by enterprises engaged in a like function with unrelated parties.
The remaining return is consequently left to the associated enterprise controlling

the intangible asset; the return to the intangible asset is, in practice, a ‘residual
category’ being the return left over after other functions have been appropriately
compensated at arm’s length.
This implies as stated above that the TNMM is applied to the least complex of the
related parties involved in the controlled transaction. The tested party should not
own valuable intangible property. The application of the TNMM is similar to the
application of the cost plus method or the resale price method, but the TNMM
involves comparison of net profit margins.
For example, in the case of a related party distributor applying the resale price
method to establish an arm's length transfer price, the market price of products
resold by the related party distributor to unrelated customers (ie sales price) is
known, while the arm’s length gross profit margin is determined based on a
benchmarking analysis. The transfer price or cost of goods sold of the related party
distributor is the unknown variable.
The determination of an arm’s length transfer price based on the TNMM is similar.
The main difference with a gross margin analysis is that operating expenses are
considered in calculating back to a transfer price. In applying the TNMM on the
tested party distributor, the resale price and the operating expenses of the related
party distributor are known, while the arm’s length net profit margin (ie net profit to
sales ratio) is found on the basis of a benchmarking analysis. The cost of goods sold
and the gross profit are the unknown variables.
In the case of a manufacturer, applying the cost plus method to establish an arm’s
length transfer price, the cost of goods sold of the related party manufacturer is
known. The arm’s length gross profit mark-up is based on a benchmarking analysis.
The transfer price or sales revenue of the related party manufacturer is the
unknown variable.
In applying the TNMM to the tested party manufacturer instead of the cost plus
method, the cost of goods sold and the operating expenses of the related party
manufacturer are known. A benchmarking analysis will determine the arm’s length
net profit of the related party manufacturer using a profit level indicator such as
the ratio of net profit to total cost. The sales price and the gross profit are the
unknown variables.
The TNMM can be used in setting transfer prices; it can also be used in testing the
transfer prices set under another methodology. We will see TNMM being used

where the parties are involved in continuing transactions and one holds valuable
intangibles. The tested party is likely to be a distribution company or contract
manufacturer. We will also see TNMM being used in respect of intra-group services.
The TNMM is not so reliable when factors other than the transfer price impact on
net profit margins. Factors such as the cost of finance, the age of machinery and
efficiency of management are hard to build into the comparability analysis and
thus can lead to TNMM being less reliable than say cost plus where these factors
may have little or at least less impact on gross margins. See the OECD TPG
Chapter II Section B.3 for further details.
TNMM is similar to the Comparative Profits Method (CPM) under the US regulations.
Transactional Profit Split Method
This section of the OECD TPG was revised in July 2018; the references here are to
the updated version.
We will begin with an overview of the profit split method then go on to look at how
profits may be split under contribution analysis (simply PSM) or residual analysis
(RPSM).
‘The transactional profit split method seeks to establish arm’s length outcomes
or test reported outcomes for controlled transactions in order to approximate
the results that would have been achieved between independent enterprises
engaging in a comparable transaction or transactions. The method first
identifies the profits to be split from the controlled transactions—the relevant
profits—and then splits them between the associated enterprises on an
economically valid basis that approximates the division of profits that would
have been agreed at arm’s length. As is the case with all transfer pricing
methods, the aim is to ensure that profits of the associated enterprises are
aligned with the value of their contributions and the compensation which
would have been agreed in comparable transactions between independent
enterprises for those contributions. The transactional profit split method is
particularly useful when the compensation to the associated enterprises can
be more reliably valued by reference to the relative shares of their
contributions to the profits arising in relation to the transaction(s) than by a
more direct estimation of the value of those contributions.’ (OECD TPG
Chapter II, 2.114)
Take a look at the OECD TPG paragraph 2.126 for further guidance.
The guidance defines ‘unique and valuable contributions’ as follows:
‘Contributions (for instance functions performed, or assets used or

contributed) will be “unique and valuable” in cases where (i) they are not
comparable to contributions made by uncontrolled parties in comparable
circumstances, and (ii) they represent a key source of actual or potential
economic benefits in the business operations.’ (OECD TPG Chapter II, 2.130)
Contributions include intangibles.
Turning to highly integrated activities, paragraph 2.133 states that:
‘A high degree of integration means that the way in which one party to the
transaction performs functions, uses assets and assumes risks is interlinked with,
and cannot reliably be evaluated in isolation from, the way in which another
party to the transaction performs functions, uses assets and assumes risks.’

In deciding if activities are highly integrated, a key area to consider is the extent to
which the same (or closely related) economically significant risks are shared
amongst the parties (see OECD TPG paragraph 2.136). It is also important to look
at the question holistically as a unique contribution by one party may have a
significantly greater value when considered in combination with the particular
unique contribution of the other party.
OECD TPG paragraph 2.128 makes it clear that a lack of closely comparable
uncontrolled transactions will not be enough by itself to justify the use of the PSM.
The availability of reliable CUPs does indicate that PSM will not be the most
appropriate method (see OECD TPG paragraph 2.143).
All previous methods we have discussed deal specifically with one transaction and
require a specific functional and risk profile. That is, the other methods necessitate
a specific transaction, which can be identified, compared and weighed (ie
identify the risk profile to assess whether it should generate a routine or a non-
routine return).
The PSM looks at the overall value generated by the efforts of all the transacting
related parties and provides an arm's length apportionment of the profit based on
the value each party contributes to the business on the basis of its functional and
risk profile.
‘The main strength of the transactional profit split method is that it can offer a
solution for cases where both parties to a transaction make unique and
valuable contributions (eg contribute unique and valuable intangibles) to the
transaction. In such a case independent parties might effectively price the
transaction in proportion to their respective contributions, making a two-sided
method more appropriate. Furthermore, since those contributions are unique
and valuable there will be no reliable comparables information which could
be used to price the entirety of the transaction in a more reliable way, through
the application of another method. In such cases, the allocation of profits
under the transactional profit split method may be based on the contributions
made by the associated enterprises, by reference to the relative values of
their respective functions, assets and risks. See section C.2.2 on the nature of
the transaction’ (OECD TPG Chapter II, 2.119)
The extract above further highlights the strengths of the PSM. The presence of high
value intangibles is rapidly increasing in the context of large multinational groups.
The globalisation trends and increasing competition require businesses to come up

with unique ways to sell their products and services and gain (plus keep) the
interest of new and existing customers.
Technical intellectual property, business and industry know-how and brands are
often becoming the main driving force for large businesses. With brands valued in
excess of tens of billions, ensuring that intellectual property is properly accounted
for when allocating profit is key.
One of the inherent properties of an intangible is its uniqueness as businesses gain

by differentiating themselves from the competition. However, the unique nature of
an intangible results in more difficulties in finding comparable transactions when
pricing the intangible contribution for transfer pricing purposes.
The PSM provides a solution to the comparability problem. The PSM based on
contribution analysis and the Residual PSM based on residual analysis (‘RPSM’) are
often used to price both value and the profit portion contributed by intangibles.

It is important to understand how the PSM and RPSM work in practice.
As stated above, the profit split method seeks to eliminate the effect on profits of
special conditions made or imposed in a controlled transaction by determining
the division of profits that independent enterprises would have expected to earn
from engaging in a transaction or a series of transactions.
The profit split starts with identifying the profits to be divided between the
associated parties from the controlled transactions. Subsequently, these profits are
divided between the associated enterprises (see below for more detail). External
market data (eg, profit split percentages among independent enterprises
performing comparable functions) should be used to value each enterprise's
contribution when possible, so that the split of combined profits between the
associated enterprises is in accordance with that of third party enterprises
performing functions comparable to the functions carried out by the related party.
However, not all functions can be priced using comparables as we already briefly
discussed for intangibles.
As noted above two main methods to split the profits amongst the associated
enterprises can be used:
• Contribution analysis; and
• Residual analysis.
With contribution analysis (or simply PSM), the aggregated profits from the related
party transactions are allocated amongst the associated parties on the basis of
the relative value of functions performed and risk borne by the associated
enterprises engaged in the controlled transactions.
Comparable market data should (when possible) be used to calculate the portion
of the profit due to each of the related parties based on their functional and risk
profile and as detailed in the functional analysis conducted for the purposes of
putting in place transfer pricing documentation.
If the relative value of the contributions can be calculated directly, then

determining the actual value of the contribution of each enterprise may not be
required. The combined profits from the controlled transactions should normally be
determined on the basis of operating profits. However, in some cases it might be
proper to divide gross profits first and subsequently subtract the expenses
attributable to each enterprise. Furthermore, when services are exchanged which
contribute to the overall value proposition, other methods can be used to
apportion the overall profit, such as the cost plus method.
If we compare contribution analysis to TNMM we see that TNMM depends on the

availability of external market data comparables to measure efficiently the value
of contribution of each of the related parties, while the contribution analysis can
still be carried out even when it is not possible to measure directly each party's
contribution to the overall profit base of the multinational group.
The contribution analysis and TNMM are difficult to apply in practice and therefore
not often used, because reliable external market data necessary to split the
combined profits between the associated enterprises are often not available.
How does the Residual PSM (‘RPSM’) differ from the simpler PSM, which we have
just analysed? The RPSM model differs from a standard PSM as it involves a two-
step approach. We have already mentioned how certain value contributions

cannot be easily priced by means of comparable benchmarking (eg intangibles).

The RPSM first allocates comparable functions’ profits, which then leaves a residual
profit to be split amongst the more difficult to price functions such as intangibles.
In the first step an allocation of arm’s length profit to each related party is
implemented to provide a basic compensation for routine contributions (ie
functions where the risk profile can be regarded as low – eg support services,
limited risk distribution, toll manufacturing, etc.).
The routine profit allocation does not account for any possible valuable intangible
assets owned by the associated party. The routine compensation is determined
based on the returns earned by comparable third party enterprises, which (ideally)
work in a similar industry or that (at least) carry out comparable functions and
exhibit a similar low risk profile. The TNMM is usually employed to determine the
appropriate routine returns for the first step in the RPSM.
Once all routine functions have been remunerated, the residual profit is then split
to account for non-routine activities, which are usually associated with a higher risk
profile. It is interesting to note that as risk bearing functions, the residual does not
necessarily translate in profit allocation; as the routine functions take priority in
allocating the profit, it might be that the residual profit is negative.
Hence, the non-routine functions might end up being allocated a portion of the
loss. However, it is also true that when large profits are collected within a
multinational group, the RPSM is more likely to allocate the majority of the profits to
the risk taking functions within the group. These trends follow the standard
economic trends for risk and reward (ie the higher the risk, the higher the potential
for profit, but also losses).
The residual analysis is usually applied in cases where both sides of the controlled
transaction own valuable intangible properties.
The profit split put forward needs to be aligned with Chapter I of the OECD TPG
which sets down how transactions are to be accurately delineated. This may
require the segregation of financial information. Harmonisation of financial
information will also be required where the results of two or more related parties
are aggregated.
Where closely related economically significant results are shared then it is actual
profits that should be split according to the guidance. Where a party makes
unique and valuable contributions without sharing economically significant risk
then it is forecast profit that should be used (see paragraph 2.160). It is clearly
stated that hindsight must not be used (see paragraph 2.161).
Operating profit is put forward as the most appropriate point at which profit is to
be split, however it is recognised that in some cases the split can be made at gross
profit. Section C5 makes it clear that profit splitting at any level is acceptable so
long as it is objective, verifiable and supportable. Section C5.1 includes a non-
exhaustive list of profit splitting factors such as percentages, asset based or cost
based measures.
A profit splitting factor can be a percentage (eg a 30%-70%) or a variable (e.g.

relative value of participant’s marketing contributions or other possible factors)
which could be calculated on the basis of a single profit splitting factor or a
weighting of multiple factors. See OECD TPG paragraph 2.170.
The OECD TPG give some guidance on possible profit splitting factors that can be
used. They include relative value of expenditure on assets, capital employed,

marketing spend, incremental sales, employee remuneration, head count and

time spent. Cost is also listed as a potential profit splitting factor although it is
recognised that it is a poor indicator of value for intangibles. Relative costs
incurred by the parties can still be used as a proxy for share of relative value. The
list provided is not an exhaustive list. (See paragraphs 2.169 to 2.172).
The local file and master file (which we look at later in relation to compliance
issues) are cited as good sources of information for determining the appropriate
profit splitting factors (see paragraph 2.173).
The OECD TPG do not refer to specific allocation keys to be used when allocating
the residual profit, but it is good practice to investigate a number of allocation
keys and run sensitivity analysis to ensure that the split returns arm's length results.
Below are a number of examples to illustrate how the residual can be split
depending on the availability of data, comparables and allocation keys.
Third party market benchmarks can be used to assess the fair market value of the
intangible property or other non-routine function to be allocated as part of the
residual profit.
The capitalised cost of developing the intangibles and all related improvements
and updates adjusted to account for the useful life of the asset and its future
potential in providing the added value (ie advantage) can be used. However,
using cost as the base to allocate the residual profit might not provide the correct
allocation as some non-routine functions might incur lower cost, but generate high
value.
Another way to allocate the residual profit is to look at the development

expenditures in recent years and identify a trend (ie if these costs have been
constant over time).
 Illustration 8
Company P develops a valuable intangible which it licenses to a 100% owned

Company E in Europe. Company E uses the intellectual property (IP) from the
parent alongside its own IP which has been developed to adapt the product for
the local market.
Company E markets the adapted product in Europe under its own brand name. In
year XXXX Company P has no direct costs related to the license. The tax authorities
for Company P determine that a Residual Profit Split is the most appropriate
transfer pricing method. It is also determined that the first step of the profit split will
be undertaken based on a return to assets employed. It is established that the
assets employed by Company E are 500,000 and that the appropriate return is
15%, which gives a return of 75,000 to Company E. It is established that the pre-
royalty profits for Company E are 400,000. After we deduct the 75,000 of profits for
Company E for its contribution this leaves 325,000 of residual profit to be split.
It is determined that the residual profit will be split based on expenditure on

intangibles as shown in the balance sheet of each company (capitalised).
Investigations determine that Company P has 7.5M in its balance sheet and
Company E has 2.5M in its balance sheet.
This means that the residual profit of 325,000 will be split 7.5/10 to Company P,
being 243,750, and 2.5/10 to Company E, being 81,250.

The RPSM is becoming more popular following the recent restructuring trends of
large multinational groups, which are centralising some of the non-routine
functions and creating structures which present a complex setup and would not
lend themselves to the standard transfer pricing methods. The RPSM provides a
good alternative in such cases, as the residual approach splits up a complex
transfer pricing problem into two more manageable steps and allows the use of a
number of allocation methods to benchmark, value and weigh the non-routine
component to be transfer priced. Secondly, potential conflict with the tax
authorities is reduced by using the two step residual approach since it reduces the
amount of profit split in the potentially more controversial second step.
The list below highlights some of the strengths and weaknesses of the PSM and the
RPSM.
Both the PSM and RPSM are suitable for highly integrated operations for which a
one sided method may not be appropriate.
The PSM and RPSM are also useful when third party benchmarks cannot be
identified.
The PSM and in particular the RPSM are most useful when looking at non-routine
functions, highly integrated operations and intangible property, which cannot be
easily defined using the standard transfer pricing methods due to their uniqueness
(ie lack of comparables in the market that match the functional and risk profile).
However, both the PSM and in particular the RPSM require a higher level of
reviewing, testing and sensitivity checking when using allocation keys, which do
not necessarily generate arm's length results.
Both the PSM and the RPSM are highly dependent on having access to quality
information and data from group affiliates. The information and data have to be
reviewed and compared to ensure consistency, which is sometimes lacking in
large multinational groups, which have just undergone or are undergoing
restructuring or acquisitions.
The PSM can be used in cases involving highly interrelated transactions that
cannot be analysed on a separate basis. This means that the PSM can be applied
in cases where the associated entities engage in several transactions that are
interdependent in such a way that they cannot be priced on a separate basis
using any of the traditional transaction methods. The transactions are thus so
interrelated that it is impossible to identify distinct comparable transactions. Due to
this particular strength, the PSM and the RPSM are suitable for use in complex
industries such as financial services.
The RPSM (in particular) is often used in complex cases where both sides to the
intra-group transaction own valuable intangible property (eg technical IP, patents,
trademarks, and trade names). If only one of the associated enterprises owns
valuable intangible property, the other associated enterprise would have been
the tested party in the analysis using the cost plus, resale price or TNMM. However,
if both sides own valuable intangible property for which it is impossible to find
comparables, then the PSM is more likely to be the most reliable method.
As part of your study you should look at the worked examples that are provided in
the guidance on the PSM.
Below is a table setting down some of the main strengths and weakness of the 5
methodologies that we have looked in this chapter – it is not an exhaustive list.

Summary table
Method Strength Weakness
CUP Reliable when the same Difficulty in finding

product is sold in the same comparable transactions
circumstances by independent between independent
parties. enterprises.
Adjustments are possible for Too many adjustments make
certain differences such as CUP unreliable.
product volume.
RPM Not heavily reliant on product Differences in functional
characteristics being the same. analysis will affect reliability.
As it is looking at gross margins Based on gross margins so
there is less possibility of reliant on accountancy
variables not related to the TP consistency.
affecting the comparables.
C+ Uses data from internal cost Based on gross margins so
which will be available. reliant on accountancy
Again, not heavily reliant on consistency.
product characteristics being May result in there being no
the same. motive to control cost.
TNMM Net margins are less affected Margins may be affected by
by accounting inconsistencies. factors not related to the TP
More tolerant of functional such as cost of finance or
differences. management efficiency.
There may be difficulties in
isolating the information
relevant to the controlled
transaction.
PSM/RPSM Suitable for highly integrated It may be difficult to access
activities. Also where both information on foreign
parties have unique and affiliate.
valuable IP. Difficult to identify the
Not reliant on the availability of appropriate profit splitting
comparables. factors.
As mentioned in an earlier chapter the UN Manual has been updated to include a

revised section on the profit split method. The revisions bring the UN Manual in line
with the post BEPS approach and provide more worked examples.

CHAPTER 5
FUNCTIONAL ANALYSIS
In this chapter we look at:

– The goal of functional analysis
– An introduction to the analysis of functions, assets and risk
– Summarising the functional analysis
– Functional analysis and entity characterisation
5.1 Introduction
Functional analysis plays a critical part in establishing arm’s length transfer pricing.
It involves gathering information and analysing the businesses engaged in the
controlled transaction to ensure that the parties to the transaction and the
transaction itself are understood. This enables an understanding of the
economically significant factors on which the pricing and its analysis will be based.
The OECD TPG place great stock on functional analysis as a pre-requisite for an
appropriate assessment of the comparability of a controlled transaction (of the
‘tested’ party – we will look at this in more detail in a later chapter), the selection
of a transfer pricing method and for establishing the appropriate pricing by
reference to comparability, including where necessary, any adjustments.
It is the normal starting point for any examination of an enterprise’s transfer pricing
and also the means by which businesses and tax authorities can form a high level
view of value chains and the role and reward of transfer priced entities within
them.
As you can see functional analysis is very important and has many roles to play in
transfer pricing.
As noted in an earlier chapter, the section in Chapter I of the OECD TPG on

functional analysis was revised to take account of the final report on BEPS Action
Points 8 to 10.
5.2 Goal of Functional Analysis
The goal of functional analysis can be summarised as the identification of the

economically relevant functions, assets and risks of a party to a transaction to
enable the accurate assessment of comparability with an uncontrolled
transaction in the setting of a transfer price.
In order to fully understand the importance of the functional analysis, we must look
at the role it plays in determining comparability between the controlled
transaction under review and uncontrolled transactions. We looked briefly at
comparability in an earlier chapter; if we now look at paragraph 1.33 of the OECD
TPG we can see that it sets out the essence of comparability in transfer pricing:
‘Application of the arm’s length principle is generally based on a comparison

of the conditions in a controlled transaction with the conditions in transactions
between independent enterprises.’

It then goes on to state that there are two steps to a comparability analysis:


The guidance from the OECD on the first step can be found in Section D1 of
Chapter I of the OECD TPG. In particular paragraph 1.36 sets down the
economically relevant characteristics or comparability factors that need to be
identified for Step 1.
We looked at these comparability factors in an earlier chapter. You will recall that
they may be summarised as:
• Contractual terms
• Functional analysis
• Characteristics of property or services
• Economic circumstances
• Business strategies
The functions (taking into account also the assets used and risks assumed)
performed by an enterprise which is a party to a controlled transaction is one of
the key comparability factors to be understood. This is established by way of a
functional analysis.
Have a look now at paragraph 1.51 of the OECD TPG. It summarises the
importance of the functional analysis and its impact on arm's length pricing.
An important aspect to this guidance is contained in the word ‘each’ (in the first
sentence of the paragraph). Whilst the eventual transfer pricing method selected
may be essentially ‘one sided’ (ie. it tests and supports a price or targeted margin
for one of the parties to the transaction), a functional analysis should consider
factors relevant to both parties engaged in the transaction (for instance, those
relevant to establishing their relative bargaining power) as otherwise there might
be a limited basis for comparability which may in turn raise doubts about the
appropriateness of the selected method and the robustness of the support for the
pricing of the controlled transaction.
Increasingly tax authorities are taking such a ‘two sided’ view on examination of
pricing with their starting point being to corroborate the results of the application
of the selected transfer pricing methodology of the tested party with the results of
the counterparty to the transaction.
Whichever view is taken, the analysis of the economically significant functions,

assets and risks remains key.

5.3 Analysis of Functions, Assets and Risks
The role of the functional analysis was looked at briefly in an earlier chapter and
will now be looked at in more detail. As noted at the start of this chapter the role
of the functional analysis is a key part of the syllabus hence it will also be revisited
in later chapters.
The functional analysis is the factual basis of transfer pricing and the right effort
and focus should be placed on capturing accurate and relevant information
concerning functions, assets and risks, which will be critical in determining the
economically relevant characteristics for comparison with an independent party
situation and therefore minimising adjustments to transfer pricing policies that have
been implemented. Many transfer pricing issues arise due to a lack of clarity on
the factual position.
As we have seen the OECD TPG, in paragraph 1.51, as set out above, reinforce the
need for the functions, assets and risks to be identified and the TPG further expand
on functions, assets and risk in paragraphs 1.42 to 1.106.
‘The functional analysis should consider the type of assets used, such as plant
and equipment, the use of valuable intangibles, financial assets, etc., and the
nature of the assets used, such as the age, market value, location, property
right protections available, etc.’ (paragraph 1.54)
As mentioned in an earlier chapter the analysis of risk has a key role to play thus
we see at paragraph 1.56 of the OECD TPG:
‘a functional analysis is incomplete unless the material risks assumed by each

party have been identified and considered since the actual assumption of
risks would influence the prices and other conditions of transactions between
the associated enterprises.’
Paragraph 1.60 sets down a six step process for analysing risk. There are two key
concepts related to risk allocation, being control over risk and financial capacity
to assume risk. The six step process is as follows:
1. Identify economically significant risks with specificity;
2. Determine how those specific risks are allocated in contractual arrangements;
3. Use a functional analysis to determine the potential impact of those specific

risks. Identify which enterprises assume and manage the risks;
4. Does 3 agree with 2? Confirm via the FA, establish who exercises control over
risk and that they have the financial capacity to do so;
5. Reallocate the risk to the enterprise(s) that has/have the control over the risk
and the financial capability to assume the risk;
6. Price the accurately delineated transaction taking account of the allocation

of the risk.
This six step process has formed the basis of a past exam question so you should
review it in the OECD TPG.
Step 1 in effect gives a materiality approach as we only look at economically

significant risk.

In order to have control over risk, an enterprise is not required to perform the risk
mitigation activities itself, but it is required to be actively involved in the decision
process when outsourcing these activities. Financial capacity refers to an
enterprise's capability to access funding when managing risk as well as absorbing
the consequences of risk in the event of an unfavourable outcome (see
paragraph 1.64).
Paragraph 1.61 states that risk management has three elements:
‘(i) the capability to make decisions to take on, lay off, or decline a risk-
bearing opportunity, together with the actual performance of that decision-
making function; (ii) the capability to make decisions on whether and how to
respond to the risks associated with the opportunity, together with the actual
performance of that decision-making function; and (iii) the capability to
mitigate risk; that is the capability to take measures that effect risk outcomes
together with the actual performance of such risk mitigation’.
The following paragraphs outline some key elements of control over risk, risk
management and risk mitigation. Some risk management functions can only be
undertaken by the party performing the functions and using assets to pursue the
commercial opportunity. Others can be undertaken by a separate party. Risk
management is not to be thought of as a separate function. In some cases where
a service provider is used, for example with R&D, elements of risk mitigation may
be included in the service and reflected in the pricing for the service.
Financial capacity to assume risk can be defined as access to funding to take on

or lay off risk. Provision of funding in itself does not mean the provider has taken on
the financial risk. It is possible to outsource the day to day mitigation of risk. In these
cases control of risk will require the capability to determine the objective of the
outsourced activity to decide to outsource and the ability to assess whether the
objectives are being adequately met by the outsourced party and, where
required, the capability to terminate the contract. The capability to perform
decision making functions related to risk will involve an understanding of the risk
based on the information available.
Paragraph 1.71 defines risk for transfer pricing purposes as the ‘the effect of
uncertainty on the objectives of the business’. Some examples of risks are provided
in paragraph 1.72:
a. Strategic risks or marketplace risks.
b. Infrastructure or operational risks.
c. Financial risks.
d. Transactional risks
e. Hazard risks.
The section provides examples of circumstances in which the legal owner of an

asset may not be taking control over risk. It is made clear throughout the section
that capital-rich entities without any other relevant economic activities (often
referred to as ‘cash boxes’) will not be entitled to any profits beyond those that
appropriately remunerate their contributions.
The next step is to look at the contractual terms as these will normally set out the
intended assumption of risk. Some risk will be explicitly assumed, such as inventory
risk, others will be implicit in the terms.

The OECD TPG say that for the contract to be relied on there must be evidence of
a commitment to assume risk prior to the risk materialising and that the tax
authority will need to see this evidence.
Care needs to be taken with the written contracts as in some cases written
contracts may be inconsistent or may not be followed in practice.
Risk will have a material impact on pricing, thus in a controlled situation it cannot
be concluded that the pricing arrangement alone sets the assumption of risk. The
facts will need to be looked at to determine how the parties manage and control
risks (see paragraph 1.81).
A successful functional analysis will draw out the functions, assets and risks in a
manner that will enable comparison to uncontrolled transactions. It is inevitably a
simplification of the complexity of the value drivers in a business and will rarely be
capable of being exhaustive.
As such, a successful functional analysis will identify and draw attention to the most
important and relevant factors. It should allow a reader unfamiliar with the
specifics of the industry to understand the functions, assets and risks of the
enterprise in sufficient detail for them to understand the key relevant economic
characteristics. It will also typically bring to light aspects of the other comparability
factors, for example characteristics of the service.
A functional analysis will often be performed by way of an interview with key

stakeholders in an enterprise. As a tool, the functional analysis interview is an
effective way to explore the full range of comparability factors.
The representation of the functional analysis in documentation is important. It

typically forms a core part of a transfer pricing report and can be the subject of
scrutiny by a tax authority or other interested party (for instance a minority
shareholder) many years after writing. As such, it needs to be a full explanation
that stands alone in a fashion that is not reliant on reference back to source or
detailed supplementary materials.
Step 4 takes the information that has been gathered via steps 1 to 3 and analyses
it.
In cases where the contractual assumption of risk is fully supported by the parties’
conduct, including an alignment with the exercise of control and financial
capacity to assume the risk, the analysis will be straightforward.
If this is not the case it will be necessary to move to step 5.
Step 5 states that the party which does exercise control over the risk and has the
relevant financial capacity should be allocated the risk. If multiple associated
enterprises are identified that both exercise control and have the financial
capacity to assume the risk, then the risk should be allocated to the associated
enterprise or group of enterprises exercising the most control.
Paragraph 1.99 points out that in exceptional cases it may be the case that there
is no party that both exercises control and has the financial capacity to assume
the risk. Such a scenario would rarely occur between independent enterprises and
therefore a thorough analysis should attempt to identify the reasons for this. An
assessment of the commercial rationality of such a transaction based on Section
D2 of Chapter I of the OECD TPG may be necessary (we looked at this section in
an earlier chapter when looking at business strategies; you will recall that this sets

down the circumstances in which a tax authority can disregard transactions

entered into; see paragraphs 1.19 to 1.125 (1.19 – 1.145)).
Once step 5 is complete then step 6 involves using the approved methodologies
to price the transaction to ensure that the party assuming the risk is compensated
for the risk via the price received.
It follows that the party that both assumes and mitigates risk will have more reward
than a party that only assumes risk or only mitigates risk (see paragraph 1.100).
A commitment to assume risk prior to the materialisation of the risk outcomes will
give rise to an entitlement to receive unanticipated profits (or bear losses) that
result from that risk (see paragraph 1.106).
The next section of the OECD TPG (paragraphs 1.107 to 1.126) was added at the
same time that Chapter X on financial transactions was added. It expands on step
6 of the six step process above; it introduces guidance on risk-free and risk-
adjusted returns (as briefly mentioned in an earlier chapter).
Where the accurate delineation of a transaction shows that a funder lacks the
capability, or does not perform the decision-making functions, to control the risk
associated with investing in a financial asset, it will be entitled to no more than a
risk-free return (see paragraph 1.108 of the revised Chapter I).
A risk-free rate of return will take account of the funder’s borrowing cost; generally
the funded party should be able to claim a deduction for the arm’s length
borrowing cost (subject to any specific rules which we will look at in later
chapters). The difference between the arm’s length amount and the risk-free rate
of return will be allocated to the party making the funding decisions. This rule is to
stop the use of cash box companies with little or no substance to divert profits.
Paragraph 1.110 of the amended Chapter I states that:
‘a funder lacks the capability, or does not perform the decision-making

functions, to control the risk associated with investing in a financial asset, it will
be entitled to no more than a risk-free return’..
The guidance then goes on to note the steps that can be taken to eliminate
currency risk such as using bonds issued in the same currency as the funder’s
functional currency. Other comparability factors such as the time of issue and the
length of the bond will also need to be taken account of.
New paragraph 1.115 of the revised Chapter I puts forward potential measures of
a risk-free rate of return:
• highly rated government securities,
• interbank rates,
• interest rate swap rates or repurchase agreements of highly rated government

issued securities.
Turning to the risk-adjusted rate of return, the guidance refers to paragraph 6.61 of
the OECD TPG which is found in Chapter VI (see later chapter for a full discussion
on Chapter VI). Paragraph 6.61 states that a party providing funding for
intangibles that doesn’t assume any risk other than financial risk will only be
entitled to a risk-adjusted rate of return on its funding. This approach is also laid
down in the examples at paragraph 1.85 to 1.103 in Chapter I of the OECD TPG.

It follows that it will be important to identify the financial risk assumed by the funder
and the operational risk assumed by the borrower.
New paragraph 1.120 states that:
‘In general, the expected risk-adjusted rate of return on a funding transaction

can be considered to have two components, ie the risk-free rate and a
premium reflecting the risks assumed by the funder’..
Assuming financial risk means that the return will be greater than the risk-free rate
of return discussed above.
It is suggested that the risk adjusted rate of return can be found by looking at a
realistic alternative with the same economic characteristics via a functional
analysis.
Possible approaches are put forward for finding the risk-adjusted rate of return.
Comparable Uncontrolled Price (CUP) can be used (which is explained in full in an

earlier chapter). Potential CUPs are bond issuances or loans which are
uncontrolled transactions.
Alternatively, a premium can be added to the risk-free return, based on the

information available in the market on financial instruments issued under similar
conditions and circumstances (see new paragraphs 1.123 and 1.124 of the revised
Chapter I).
Finally, the cost of funds approach can be used which adds a profit margin to the
costs incurred by the lender to raise the funds advanced to the borrower. That
mark-up should be proportionate to the risk assumed by the lender and
calculated according to the guidance set down on intra-group loans in Chapter X
of the OECD TPG (see later chapter on the OECD guidance on financial
transactions).
5.4 UN Manual and Risk
The UN Manual flow chart of the six step approach to analysing risk along with
illustrative examples was mentioned in an earlier chapter. You may find it useful to
look at this again now (see B3.4.4.21 for the flowchart, the examples begin in
B3.4.4.45).
5.5 Summarising the Functional Analysis
While a full text explanation of the functional analysis is critical to robust and
effective transfer pricing documentation, a summary showing the key functions,
assets and risks and their location in the group is often useful. This will be a benefit
to the reader, who will be required to take in a lot of information, and for when the
preparer comes to characterise each of the entities involved.
This summary should work through the supply chain in a logical order. Any lack of
clarity in its preparation will identify insufficient understanding in the functional
analysis review.

 Illustration 1
Here we have a functional analysis for the H Group which consists of HO Ltd (the
parent company), MO Ltd and distribution companies.
Activity HO Ltd MO Ltd DO Ltd

Parent Co Manufacturer Distributors
Functions
R&D – management, review, X
budget
R&D – performance X
Procurement X
Production X
Logistics and shipping X
Marketing & business X
development
Sales (inc customer X
contract)
After sales service X
Insurance X
Strategic management X
Day-to-day management X X
Back office support X
Assets
Raw materials stock X
Design intellectual property X
Production equipment X
Manufacturing know-how X
Stock of finished goods X
Trade mark and brand IP X
Customer lists X
IT support systems X
Website X
Risks
New product development X
Warranty X
Market X X
Foreign exchange X
Stock X
Regulatory X
Here we can see that the H group is what we would describe as a group with
devolved activities. The parent company just carries out head office activities with
the result that all the manufacturing functions, assets and risk are within MO Ltd
and the functions, assets and risks relating to distribution are in each of the DO Ltd
companies.
We can contrast this to the following illustration.

 Illustration 2
Here we have a functional analysis for the C Group.
Activity CP Ltd CM Ltd CD Ltd

Parent Co Manufacturer Distributors
Functions
R&D – management, review, X
budget
R&D – performance X
Procurement X
Production X
Logistics and shipping X
Marketing & business X X
development
Sales (inc customer X
contract)
After sales service X
Insurance X
Strategic management X
Day-to-day management X X X
Back office support X
Assets
Raw materials stock X
Design intellectual property X
Production equipment X
Manufacturing know-how X
Stock of finished goods X
Trade mark and brand IP X
Customer lists X
IT support systems X
Website X
Risks
New product development X
Warranty X
Market X
Foreign exchange X
Stock X
Regulatory X
The C group would be described as a group with centralised activities. We can

see from the table that the parent company CP Ltd is responsible for many group
functions. As a result it also holds a lot of the groups assets and carries a lot of the
risk.
If we compare the functions of the parent company in the C group to the parent
in the H group we can see that functions such as procurement are undertaken by
CP Ltd rather than by the manufacturing company. As a result it holds stock as an
asset and has the risk associated with holding stock.

5.6 Functional Analysis and Entity Characterisation
As we said at the start of the chapter, the functional analysis has many important
roles; one such role is to help identify the key characteristics of an enterprise so
that it may be categorised.
Characterisation of a controlled party is an important part of the transfer pricing

analysis. It allows other entities with the same characteristics to be identified as
part of the comparability analysis.
The functional analysis along with information from the industry can be used to
characterise the entities. If we take a manufacturing company for example,
common characterisations include full blown manufacturer, contract
manufacturer or toll manufacturer.
Entity characterisation can be a helpful high level tool to assist in examining often
complex value chains (that is, how a business derives value from the various
activities involved in the production of a product or service) and the manner in
which component parts of the value chain relate to each other from a transfer
pricing perspective.
This allows both an initial determination of transfer pricing interactions in existing

value chains, as well as providing a valuable basis of examination of the transfer
pricing consequences for those value chains which are being transformed.
We will look at entity characterisation in more detail in a later chapter.

CHAPTER 6
PRACTICAL ASPECTS OF PREPARING A FUNCTIONAL ANALYSIS
In this chapter we are going to look at the practical aspects of preparing a functional
analysis, including:
– the audience and purpose
– the sponsor
– the interviews
– the functional analysis
6.1 Introduction
This chapter focuses on the practical aspects of preparing a functional analysis,

from considering the audience for the functional analysis through to conducting
interviews and preparing and updating its documentation.
When preparing a functional analysis, the aim is to document the key functions,
assets and risks of the business. In many respects a functional analysis is the most
interesting aspect of a transfer pricing project. It involves meetings with employees
at all levels of the business, from the people at the coal face, the operational
team that support them through to the executives that develop, drive and
implement its long term vision and strategy.
This chapter has been prepared on the basis that a third party is preparing the
functional analysis, however the guidance can be equally applied by a company
preparing its own functional analysis.
6.2 Preparing for the Functional Analysis
Audience and Purpose
Before beginning, consider the audience for the functional analysis. Is it a basic
document to support an uncontroversial tax filing position? Is there a dispute with
a tax authority and is it therefore a defence document setting out the company's
position on the issues? Is it to support a proposed transaction; maybe the
company is proposing to close manufacturing facilities in higher cost countries (in,
for example western Europe or North America) and move them to a lower cost
country in Asia? Such a restructure would need to be supported from a transfer
pricing perspective, clearly documenting the change in functions, assets and risks.
Whatever the purpose of the document, the effort required will depend on the
nature, size and complexity of the parties and transactions involved.
Chapter V of the OECD TPG points out at paragraph 5.4 that the overarching
consideration in determining the rules for documentation is to balance the
usefulness to tax authorities of data for risk assessment with the cost to the
taxpayer of providing the information.
The length and style of the functional analysis will need to reflect the complexity
and materiality of the arrangements it will support. For a basic, uncontroversial
cross border arrangement, a simple tabular functional analysis like the example
that we will look at later in this chapter would be appropriate. In contrast, a tax-

advantaged principal/sales agent arrangement may require a detailed functional

analysis to demonstrate the limited role played by the sales agent.
Tax authorities also have different approaches to functional analysis. For example,
some tax authorities will regularly conduct their own functional analysis where
there appears to be inconsistencies between the taxpayer's own functional
characterisation of the business and the returns achieved.
The Importance of a Sponsor
To prepare a functional analysis a number of interviews with employees of the

business will need to be conducted. In order to do this, suitable people must be
identified, and time found in their schedules for the interviews. An internal sponsor
within the business is invaluable for facilitating completion of the transfer pricing
project (and not just for organising functional analysis interviews).
A good sponsor will help to arrange interviews, understand the internal politics of
the business, and provide guidance on how to best introduce the project to the
interviewees. Ideally they should be at every interview. A sponsor may also help to
keep interviews focused and on track, and assist in building a rapport with the
interviewee.
Importantly, a sponsor can clear interviewees to discuss problems and past

business failures which may provide invaluable examples to demonstrate the
functions, risks, and assets of the business. No one likes to talk about their business's
failures to strangers but often these are the best source of information about which
party bears the risks of the business.
There are several attributes of a good sponsor:
• Have a good understanding of the business;
• Well connected in the organisation;
• Well respected in the organisation;
• Engaged in the process;
• Able to convey the importance of the project to a variety of interests and

people in the business; and,
• Able to provide insights on contradictory facts that crop-up in the interviews,

and/or identify the right people to get to the bottom of the issue.
Preparation for the Interviews
It is important to consider in what order to interview people for the functional

analysis. It is helpful to have the sponsor identify a ‘soft’ opening interview:
someone who can provide a good overview of the business, ideally someone who
is open and friendly.
In reality the order of interviews is likely to be determined by people's availability.

However, the ideal option would be to start with someone that is likely to
understand transfer pricing and how it relates to the business, for example the
Finance Director or Chief Financial Officer, or Head of Tax. Next, consider an
operational head, then drill down into greater operational detail if required, and
finally meet with senior members of staff, such as the CEO, for the strategic
overview. While it is good to get a senior view of where the business is going, keep

in mind that it may be more difficult to schedule interviews with senior people, and
they are likely to be available for a shorter length of time.
If the functional analysis involves interviewing many people, think carefully about
how to do this. It is not uncommon to go for a ‘big bang’ approach, with a whole
day of back-to-back interviews. This has lots of advantages: it is efficient, quickly
giving an understanding of the company, and if any issues arise or there are
conflicting facts, it may be possible to clarify these in the next interview. However
this approach also comes at a cost: the interview team will get tired, and material
covered in the interviews will start to blend together. The big bang is particularly
hard on the note taker and even harder on them when they have to type up the
meeting notes. If the functional analysis interviews have to be back-to-back, try to
organise a 10-15 minute break between each interview to reflect on what has
been said and how this will impact the following interviews.
If possible, try to meet at least the first few interviewees in person even if time,
geography, schedules and budget constraints may mean that some or all of the
functional analysis interviews have to be conducted by telephone.
The Functional Analysis Interview Team
For each functional analysis interview, it is advisable to have one person

responsible for conducting the interview and another person whose sole task is to
take notes of what the interviewee has said. It is very difficult to conduct an
interview and take adequate notes at the same time.
Ideally, in addition to the interviewee, it is helpful to have three participants:
• the functional analysis interview leader;
• the sponsor; and,
• the note taker.
Each has their own role to play.
The Functional Analysis Interview Leader
• Completes the introductions (if there is no sponsor or the sponsor chooses not
to do this), and provides context for the interviewee (for example, what
transactions and issues they anticipate examining, the nature of the
questionnaire, etc.).
• Leads the interviewee through the functional analysis questions, identifies any
interesting areas to explore, and moves the conversation in these directions.
• Considers what the interviewee is saying in the context of the project, the
business, and what the team understands from other information sources or
interviewees.
• Needs to be flexible in their questioning style, and recognise when a

conversation tangent is helpful and when the conversation needs to get back
on track.
• Spots issues and is responsible for ensuring the interview covers all the areas
required in the time scheduled.

The Sponsor
• Introduces the interview participants and provides context for the interviewee
(for example, the framework for the exercise it proposes and its importance to
the business).
• Is helpful in clarifying issues/solutions.
• Helps the functional analysis interview leader to conduct the interview.
• Can act as a ‘tie breaker’ when contradictory facts are raised at the
interview.
The Note Taker
• Is the busiest person in the room: if people are talking they probably should be
writing.
• Captures the detail (names, dates, terms of contracts, product names,

company names…) so the functional interview leader can concentrate on
what they hear, and how this fits into the bigger picture.
• Keeps track of actions, ‘to-do’ lists, and follow up points.
• Prepares the meeting notes/draft functional analysis.
6.3 The Interview
Conducting the Interview
There is undoubtedly an art to conducting a good functional analysis interview but

as a rule, as indeed with most things, the more preparation that is done ahead of
the meeting the better the interview is likely to go.
The interviewee has made the time to talk to the team, so it is important to respect
that time and make the most of it. An agenda and a list of questions will give
structure to the meeting and will help with this. As a functional analysis interviewer's
experience increases, they will be able to prepare for interviews more efficiently,
but even the most experienced functional analysis interview leader can forget
critical questions to ask, so a list of key questions is a must.
It is good practice to begin by explaining to the interviewee the purpose of the

interview, what types of transactions/issues are the highest priority and how the
information will be used. It is surprising how much more relaxed interviewees are
when they find out that the information will only be used in a document for tax
purposes. Keep in mind that the interviewee may be defensive: the questions they
are being asked are trying to get to the core of the business. They are in essence
being asked what they do, why it is important and how they add value. As such, it
never hurts to tell the interviewee that in the majority of cases the document that is
being prepared is unlikely to ever be read by anyone outside of the business's tax
department. Transfer pricing documentation, like insurance, is best when you
never have to use it.
It is good practice to start the functional analysis interview by asking the

interviewee to describe their current role, the team they work in, and their history
with the company. Open questions are key to getting the information needed,
such as ‘can you please explain how…’ or ‘describe the process involved in…’. Be
prepared to deviate from the planned agenda and absorb answers to questions
that may not be in the planned order.

In general it is fine, in fact helpful, to profess ignorance of the business, particularly

with the operational side. This will encourage the interviewee to explain the
business in layperson's terms. Ultimately, the functional analysis is going to have to
explain the business in a simple and understandable way. Most businesses have
three letter acronyms, technical phrases and other abbreviations; don't be afraid
to ask the interviewee to explain these.
While it is important to strike a balance and make efficient use of the interviewees’
time, if they offer a tour of their operations, warehouse, or factory, it is an
opportunity that should not be refused. The more real the business is to the
interviewer the easier it will be to write about it.
When trying to understand the unique attributes of the company it can be useful
to ask why would a customer use this company instead of a competitor, or when
trying to understand risks, ask what would happen if there was a catastrophic
incident; a key factory burning down for instance, or the product causing the
hospitalisation of a number of customers.
Active, critical listening in conjunction with open questions is crucial to conducting

a good functional analysis interview. While staff members may have a detailed, in-
depth understanding of the functions of their own department, a functional
analysis needs to take a balanced wider view. It is important to listen to
interviewees critically, and consider how their job, business unit or division
contributes to the functions, assets and risks of the broader issue being
documented.
When the interview has finished, take some time to summarise the key points and
issues. It is surprising how often members of the interview team understand critical
facts mentioned in the interview differently.
It is helpful to confirm with the interviewee if they are happy to be contacted

directly to clarify any issues that arise.
6.4 The Functional Analysis
As noted in earlier chapters a functional analysis is the fact finding process of

researching and documenting the relevant functions performed, assets owned
and risks borne by the business. All companies will have a slightly different mix of
functions, assets and risks, and so the following sections are intended as a guide
only. In some instances, a functional analysis is prepared to support a single
transaction; in other cases it will support multiple transactions, in multiple
jurisdictions. Accordingly, the time it takes to gather the facts and conduct
interviews, and the length and detail included in a functional analysis will vary
greatly.
From a practical perspective, it is necessary to consider how much information to

include about each function, asset, and risk and how to present the information.
Usually, a functional analysis will include the functions of a number of entities in the
document. When preparing a summary functional analysis table, such as the one
included in this chapter, this is relatively straight forward. However, a functional
analysis often involves a longer narrative, addressing each key function, asset and
risk in turn. There are a number of ways to do this. One approach is to do it on an
entity-by-entity basis, which sets out the functions, assets and risks of each entity
separately. An alternative would be to lay it out on a functional basis, which
introduces each function, asset and risk and then describe the entities to which
these apply under each relevant heading.

Throughout this chapter there are examples to help describe key concepts and
issues when preparing a functional analysis. There will be a variety of examples
throughout the chapter, however the following is a Core Example to which we will
often refer.
Core Example: Otaki Group
 Illustration 1
OTAKI CENTRAL
(New York) 3rd PARTY
Pricing, Logistics, Marketing Contract
Design selection, Stores Manufacturers
Trade marks
(Asia)
OTAKI OTAKI
Design Manufacturing
(Milan) (Philippines)
Develops designs 20% of products
Otaki Group is a leading clothing retailer designing, manufacturing and retailing

clothes throughout the world. Otaki Group is headquartered in New York (‘Otaki
Central’) and Otaki Central is responsible for selecting designs, determining pricing
strategy, store layout and location, organising logistics, and developing marketing
campaigns, and it owns all Group trademarks.
To ensure that its products are leading the market, Otaki Central has set up a
dedicated design house in Milan (‘Otaki Design’) with 100 top designers. Otaki
Design puts on four fashion shows a year, one for each season, and Otaki Central
buyers select garments they like for manufacture. Otaki Design is free to develop
any garment designs it likes, and typically, only 1 in 5 garments are selected.
Otaki Group operates on a high volume, low margin model for 80% of its sales, and
to keep costs down utilises third party contract manufacturers to produce its
garments.
Otaki Central selects its designs, colours, materials, and manufacturing quantities
12 months in advance and invites bids from third-party Asian manufacturing
plants.
Recently the Otaki Group established its own factory (“Otaki Manufacturing”) in
the Philippines to produce limited volume (20% of total sales), higher margin ‘fast-
fashion’ garments, with a short delivery timeframe (6 weeks). These garments have
been very successful, and sell out in hours.
If Otaki Central selects the wrong garments from Otaki Design, or orders the wrong
volumes of garments from the manufacturers, it bears the costs of these failures.
The finished garments are sold in stores (‘Otaki Retail’) designed by Otaki Central,
using store layout, staff training systems, and staff scheduling systems also
developed by Otaki Central.

6.5 Functions
When conducting a functional analysis the aim is to distil what are the most
important activities undertaken by the business, and also to convey an
understanding of the relative importance of each function as compared to the
other functions performed within the group.
The various functions should be addressed in an appropriate order: the list below
includes some common functions along with some further thoughts for
consideration, but please note that this is not an exhaustive list and that any
functional analysis will need to be tailored to the specific project to which it
relates.
Research and Development (R&D)
R&D can cover a wide spectrum of activity, and depending on the industry can
be a core value driver in a business. For example, in computer processor chip
manufacturing R&D is key to developing smaller, more efficient and faster chips
and this area would likely be a significant area of focus in the functional analysis.
But in other industries, for instance making Champagne, a long established

process must be followed, and the importance of R&D is likely to be less than other
factors such as owning land in the right appellation (an asset).
R&D can lead to valuable intangible property, which is discussed further below. In
general, if a business is undertaking R&D it is important to determine what is the
R&D being performed, which party directs the R&D at a strategic level and on a
day-to-day basis, who determines the budgets, who pays for the R&D, what party
owns the R&D, and what happens if the R&D goes wrong. Are multiple entities
within the group undertaking the R&D function? For example, does the R&D team
work on a technology platform that has been created, owned and maintained by
another entity within the group? Answering these questions will assist in preparing
the functional analysis and categorising the business or transaction.
In the Otaki Group's context, an R&D function is being performed by Otaki Design
which designs garments for Otaki Central. This is undoubtedly a valuable function,
but it needs to be viewed in conjunction with the functions performed and risks
born by Otaki Central. Otaki Central chooses which garments will be produced,
how many will be produced, organises manufacturing and logistics, prepares
marketing materials, and determines how the garments will be displayed in-store.
In this regard, Otaki Design could be viewed as a contract R&D house with Otaki
Central ensuring the benefit of the R&D and being entitled to any valuable
intangible property deriving from it.
Procurement
Procurement is the acquisition of goods or services. Within a global group, this

function might be performed separately by many entities or, in some cases, by a
dedicated business whose sole function is to arrange procurement for members of
a group. By centralising procurement in this way a group may hope to produce
efficiencies and potentially economies of scale.
An entity performing a procurement function may be expected to have expertise

in sourcing products and services, negotiating prices with suppliers and
contracting. If procurement is an important function in the group, consider what
exactly the procurement group is contributing. For example, if a multinational
coffee retailer set up a dedicated procurement centre in Singapore to source all

its coffee bean purchases from one coffee wholesaler, would the resulting
discount in bean prices be due to the negotiation skills of the procurement team
or the volume discount of the business (or perhaps a combination of the two?).
In our example, Otaki Central is responsible for organising the manufacturing of the
garments and the delivery of the final items to Otaki Retail. However, it would not
be unreasonable for this activity to be undertaken by a separate entity in the
group. In the garment industry, it is not uncommon to use purchasing agents which
will identify factories to produce garments and ensure that these are delivered on
time to the agreed destination, in exchange for a percentage of the purchase
price of the goods.
But it should be noted that these agents are bearing substantial risks relating to the
delivery of the garments.
Services
Many of the functions set out in this chapter relate to products in one way or
another, but many successful companies do not sell products; they sell services,
and some companies that once could have been described as selling products
are re-creating themselves as service companies (for example ‘Software as a
Service’).
Service transactions incorporate many of the functions described in this chapter,

including sales, and they are subject to many of the same risks. When undertaking
a functional analysis for a service company, consider which party is performing the
service, who won the work, and what would happen if the services are not
delivered as contracted.
Manufacturing
Manufacturing can range from low value, low skill functions like manufacturing
toys for a Christmas cracker, through to manufacturing a one-off, extremely high-
value item like a communications satellite. It should be noted that the value of the
item produced does not always relate to the value of the manufacturing function.
For example, having machinery and processes that can produce extremely high
volumes of plastic trinkets for Christmas crackers may be a highly valuable, critical
function for a company.
The functions, risks and assets of a manufacturer will also vary considerably. When
considering a manufacturer, it is important to understand exactly what functions
are performed by the manufacturer. Helpful questions to ask, and factors to
consider when categorising the manufacturer, are provided in a later chapter.
The Otaki Group relies mostly on third-party manufacturers for its sales, but its own
factory produces high margin ‘fast-fashion’ garments which have been very
successful. Part of this success is the flexibility to identify a trend and exploit it
quickly. Once again, this is down to Otaki Central’s ability to pick the right
garments. If there was something unique about the manufacturing know-how, or
technology developed by Otaki Manufacturing, then this would require further
consideration in the functional analysis.
Warehousing and Logistics
A warehouse may be used to store inventory that has not yet been sold. The
company may use its own facilities or a third party’s warehouse. It may also be
storing goods for other group members. Potential issues to consider include: which
party has title to the goods, when is title passed, and which party is responsible for

logistics. What happens when products are damaged in transit or at the

warehouse?
Sales and Distribution
The importance of sales and distribution to the success of the business varies from
industry to industry. This is related to the type of products/services being sold, for
instance whether they are generic or highly technical, requiring sales people with
specialist skills. For example, a distributor of medical devices used in surgery may
require sales people with a different or higher skill set than a distributor of stationery
products. The former may involve a more in depth sales process involving specialist
medical or technical knowledge and include meetings with medical specialists
and doctors.
The functions, assets and risks of a distribution entity vary considerably, as

considered in more detail in a later chapter.
Marketing
Marketing, particularly localised marketing performed by a local sales company

may result in the creation of intangibles, and some tax authorities have taken a
firm stance that this activity is not routine and requires additional reward. If
marketing is important to a company, for example it is in the fast moving consumer
goods market, it is advisable to pay particular attention to this area of the
In classifying marketing activity, it is important to consider what value is likely to

flow to the local company and possibly the wider group. For example, is the
company merely taking global marketing materials (pamphlets, brochures etc)
and translating these into the local language, or are they engaging in bespoke
advertising for the local market? If so, is this marketing above and beyond the
level of marketing undertaken by its competitors and is it likely to create an
intangible for the Group? For example, if a local territory was to sponsor a team in
a globally televised event, such as Formula 1 motor racing, should this cost be
shared with other group members?
After-Sale Customer Support Services
The level and importance of after-sale customer support services varies from
industry to industry. For example, a company providing foreign exchange trading
software to a multinational bank may require substantial, experienced technical
support to be available to the customer 24 hours a day, 365 days a year. This
service may be critical to completing a sale, and an important function to
document. In contrast, a fast food restaurant would typically not require these
services. It is important to also consider who bears the cost of warranties, whether
the products require repair or replacement, when, how and by whom.
Strategic Management
Strategic management services can be centralised to create efficiencies in

controlling entities that may be spread across broad geographic areas. Examples
may include setting the global marketing plan which subsidiaries can then adapt
to their own local market.

Intra-group Services
Many multinational groups have services provided by one or more member to

others, often centralised and/or combined with head office roles. These services
may include back-office support such as human resources, recruitment, IT, and
financial reporting. It is important to consider how beneficial these are, and
whether they are unique or could be sourced from a third-party. This is especially
important for high value services such as marketing.
Financing
Some multinational companies may centralise expertise within an entity to provide

a financing or treasury management function for the group. Examples of this may
include providing capital to subsidiaries in the form of intercompany loans,
providing hedging of foreign exchange exposures, cash pooling, and sweep
accounts. It is important to identify whether the role of a finance entity is to
provide intra-group services, such as advice in respect of hedging, or whether it
enters into transactions and if so to what extent it is exposed to risk.
Flows of debt around the group and how they vary can have an impact on
characterisations, such as whether a lender to a cash pooling arrangement is
effectively making a short term deposit or a long term loan. Depending on the role
of the individual(s) interviewed, the functional analysis interview may be able to
provide an insight into the character of loans and the relative lending risk of
various group members more easily than through examining spreadsheets.
The OECD guidance on risk-free and risk-adjusted returns was looked at in an

earlier chapter. You should refer back to this now and when you look at the OECD
guidance on financial transactions in a later chapter.
6.6 Assets
The type of assets to be included in a functional analysis is broad. As noted in

paragraph 1.54, the OECD TPG highlight the importance of considering assets as
part of the functional analysis:
‘The functional analysis should consider the type of assets used, such as plant
and equipment, the use of valuable intangibles, financial assets etc., and the
nature of the assets used, such as the age, market value, location, property
right protections available, etc.’
As with the functions listed in the earlier section, the following list of assets should
not be considered exhaustive. It is important to keep an open mind when
considering a company’s assets. Often, the most important asset is not initially
obvious. For example, some people may consider the trademark of their mobile
telecommunications supplier to be one of the company’s most important assets,
but in order to provide the service the company first had to acquire a licence. For
instance, British Telecom spent just over £4bn to acquire its licence.
Both tangible and intangible assets should be considered. It is important to identify

assets at both ends of the transaction, making it clear which party owns the assets
(legally and economically) and which party uses the assets and how the owner is
compensated.

Tangible Assets
When considering tangible assets focus should be on the significant items: there is
no need to cover the routine items like office fittings. Some tangible assets which
may be relevant are described below.
Cash
Does the company have a large amount of cash? Perhaps this cash is on deposit
in a related party interest bearing account. Some groups may pool cash in a
central treasury account. If this is the case, consideration will need to be given to
whether the interest rate applied to the deposits can be supported from a transfer
pricing perspective.
Trade and Receivables
Does the company have a large amount of receivables when compared to

payables? If so what are the payment terms offered to related parties and third
parties? Is there a significant difference, and does this impact on the working
capital of the company under review? For example, if a distribution company
receives payment from third party customers after 60 days, but must pay for stock
purchased from related parties within 7 days, this may result in a cash flow issue.
Inventory
Inventory can be in the form of raw materials or finished goods that a company
has not yet sold. It is important to understand whether the company is responsible
for managing its level of inventory and how much risk is associated with this. For
example, if a company is acting as a distributor of products for a related party and
is selling the products to a third party, it may be required to carry a certain level of
inventory to meet the customers' demands. In contrast, a distributor selling
exclusively to related parties may not have to maintain a large inventory as orders
are more certain and predictable, which means purchasing can be clearly
planned.
Property
Does the company own significant amounts of property or have leases, and is this
normal for the industry? In the UK, particularly in the retail sector, holding a large
number of leases may be problematic, particularly if there have been significant
changes in where people shop. For example, a fast food restaurant may have
signed up for a 20 year lease, only to find that a new shopping mall has opened
nearby and footfall has significantly reduced. The fast food restaurant may then
have to open a new site in the mall, and bear the costs of the old premises unless
they can be sub tenanted.
Referring back to our Otaki Group example, what would happen if Otaki Central
made a decision to open a store three times the normal store size in an expensive
high-end mall as a flagship store? Should Otaki Retail have to bear the higher cost
of the store?
Intangible Assets
There is increasing emphasis in transfer pricing on determining which party is

responsible for the development, enhancement, maintenance and protection of
intangibles. The functional analysis should clearly state which party is responsible
for these activities and which party is entitled to the rewards arising from the
intangibles.

Legally Protectable Intangibles
Legally protectable intangibles include trade names, trademarks, patents, and

copyrights. Often these are very valuable to the company, as the company has
gone through the effort and expense to seek legal protection. However, focus on
intangibles should not be limited to just legally protectable intangibles: many
companies choose not to apply for patents as they do not want to share their
intangibles with competitors or the intangibles concerned (often hugely valuable)
may not be eligible for legal protection in the form of a trade name or mark, a
patent or copyright, knowhow, for example.
Manufacturing Intangibles
In some industries, manufacturing intangibles are an important factor in a

company's success. In computer chip production, the complexity and cost of
building a manufacturing plant to produce the next generation of microprocessors
has been a key factor in consolidation in the industry. It should be noted that even
when a company uses third-party manufacturers to produce its products and does
not have any production facilities of its own, it may still possess and have
developed manufacturing intangibles.
Marketing Intangibles
The definition of marketing intangibles in the OECD TPG is fairly wide. The definition
was updated as part of the BEPS Action Plan. It can include trademarks, trade
names, customer lists, customer relationships and proprietary market and customer
data that is used in or aids in marketing or selling goods (see paragraph 6.16).
As noted elsewhere in this chapter, marketing intangibles can be a contentious

area, with some tax authorities asserting that these are more valuable than
taxpayers may consider reasonable. Caution should be exercised when
documenting these intangibles in the functional analysis.
6.7 Risks
In the past risk was often the most neglected area in a functional analysis, and in
many ways it is the most important area to focus on. The OECD and tax authorities
are placing increased attention on risk as demonstrated in the increased
guidance on risk in Section D of Chapter I following adoption of the final report on
BEPS Action Points 8 to 10. You will recall from an earlier chapter that Chapter I of
the OECD TPG now includes a six step analytical approach to risk allocation. The
list of risks set out below includes the common risks that people consider, but it is
critical to investigate what other risks are unique to the business during the
As noted above there is now OECD guidance on the risk-free and risk-adjusted
rates of return. (The guidance is summarised in an earlier chapter and can be
found in the new paragraphs 1.107 to 1.126 of Chapter I of the OECD TPG.)
When conducting functional analysis interviews, it can be helpful to ask what

would happen should a catastrophic disaster occur. In the Otaki Group example,
what would happen if it turned out that a third party contract manufacturer used
toxic chemicals in dying a batch of t-shirts that resulted in 50 customers being
hospitalised? How would this be different if it was Otaki Manufacturing that used
the toxic chemicals?

It is important to review any contracts that underpin the intercompany

transactions under review. It is surprising how often a company is not following the
terms of a contract, and may have unnecessarily borne costs or become liable for
risks that another group member is responsible for.
As noted earlier, while companies do not like to discuss failures or significant issues
that they have experienced in the past, it is often these examples that are the
most illuminating when trying to determine which party bears risk in the wider
group when issues arise.
Market Risk
Market risk is the risk that a downturn in overall market trading conditions affects
either the turnover or profitability of a company operating in that industry. Using
Otaki Group as an example, assume that the company expanded rapidly in
China, positioning itself as a desirable new western brand in expensive upmarket
retail malls in major Chinese cities. If China was then to experience a downturn,
and sales dropped to a point that the Chinese shops could not cover rent or
wages, which party would pick up these costs? If a number of Chinese stores
closed, which entity would bear the costs of the closures?
Regulatory Risk
Regulatory risk arises when an industry is particularly subject to compliance with

government regulations. While this is a barrier to entry that can protect businesses
meeting the requirements, changes to these requirements creates risk that either
additional cost will be incurred or that new competition will be allowed into the
market. For example, assume that an energy sector construction company
entered into a contract to build a new refinery in an emerging market. The
contract is signed, and contains a clause that states that the refinery will meet
local emission laws. Half way through construction the developing nation reacts to
non-government pressure to improve its poor environmental record, and halves
the allowable emissions from all commercial sites including refineries.
As a consequence, the construction company will incur an addition £50m in

construction costs to install new scrubbers and other equipment to meet the
standards.
Contractual Risk (or Warranty/Performance Risk)
Contractual risk is the risk which an enterprise exposes itself to under contractual
arrangement with its customers, for example for the proper performance or
function of contracted services or products. Where remedy is required, enterprises
risk additional costs from fulfilling warranties, providing replacements and potential
compensation. The ability to secure future contracts within the industry can also be
at stake.
The refinery example above is also an example of contractual risk, as the energy
company failed to include protective language in the contract stipulating either
what the eventual emissions would be, or stating that it would comply with
emission standards at the date the contract was signed.
Procurement Risk
Procurement risk arises where an enterprise is responsible for securing its source of
goods or raw materials for processing and/or sale. During a recent construction
boom, a construction company failed to meet its delivery deadlines as it required
a very large crane to assemble a number of modules that had been constructed

off-site. The company had failed to procure the right tools at the right time to meet
its obligations.
Inventory Risk
Where an enterprise holds stock, inventory risk manifests around the maintenance
of required stock levels, and the cost this entails, together with the potential sunk
cost from unsold (or unsellable) stock retained in the inventory.
In the Otaki Group example, if the company had predicted that florescent
coloured wetsuits would be the next high street fashion trend and had
commissioned large volumes of stock from its third party manufacturers, it may
have to substantially discount the garments to move them out of inventory should
this trend not occur.
New Product Development Risk
In many industries, continued success relies on the ongoing development of new

or improved products. Some examples of this may be the development of new
technology, improving an existing technology or a new design for an existing
product. New product development risk arises where the enterprise is primarily
responsible for successfully maintaining this development cycle. For instance,
Apple has been very successful in developing new products. Products such as the
iPad and iPhone have captured significant amounts of market share at the
expense of other companies, and in the case of the iPad created a new market.
However, none of Apple’s competitors have developed a tablet computer to

date which has exceeded the sales volumes of the iPad. Developing new
products can be very risky, and many companies, such as Polaroid and Kodak,
have failed when their products have failed to keep pace with changes in the
market when consumers moved to digital, and mobile phone cameras.
Employment Risk
Staff risk is the risk of employing, retaining and replacing sufficient numbers of
employees who are experienced or qualified enough to perform the tasks of the
business. This includes meeting the costs of retention or replacement payments
which may be required when this risk is realised. Staff may develop specialist
technical knowledge and it is important the company is able to transfer this
knowledge through the organisation so if the staff member is lost, the knowledge
remains and is able to be effectively utilised by the business.
Credit Risk
Where an enterprise is responsible for credit control and cash collection from its
customers, this risk manifests where there is non- or late-payment and steps both to
recover amounts due and maintain cash flow are required.
This will differ by the customer base, which should have been addressed in the
industry analysis and the specifics of the group’s customers. For example, are there
many small customers or a few large ones, and is the industry as a whole in
difficulty?
Foreign Exchange Risk
Foreign exchange risk arises where an enterprise is exposed to currency

fluctuations on contracts. The risk arises when expenses and revenue are
denominated in different currencies.

Exchange rates can be quite volatile and generally, unless the subject is a
financial company, the company’s core competence will not be in financial
markets.
6.8 Output from Functional Analysis: Meeting Notes
There are varying opinions on what materials should be prepared at the

conclusion of the functional analysis interviews: whether it is best to prepare
detailed meeting notes, or focus solely on the functional analysis document.
Detailed meeting notes can be helpful, as a week or two after the interview it can
be difficult to remember the detail of discussions and these will be the key record
of the meetings. However, meeting notes are not a substitute for a functional
analysis, which will still need to be prepared. Ultimately, the decision to prepare
meeting notes will be a function of time, availability of resources, and budget. If it
is considered necessary to prepare meeting notes, it is helpful to structure the
meeting note in the same way as a functional analysis, where the meeting is
condensed under the relevant function, asset and risk headings.
Keep in mind that the purpose of the meeting notes is not to provide a
stenographer's record of what was said, but rather to be shaped for its specific
purpose.
6.9 Presenting the Functional Analysis
The form and content of the final functional analysis will be determined by the
underlying transaction/business that is being documented. The more complex and
contentious a tax authority is likely to find the arrangement(s), the more substantial
the effort that is likely to be required.
A simple table summarising the key functions, assets and risks is useful as it provides
a snap-shot of the business under review. This is helpful when characterising the
respective entities (see later chapter) and when identifying comparables to
benchmark the arm's length nature of the business. However, for more complex or
contentious arrangements, it is common to include a narrative description for
each significant function, asset and risk. When writing a narrative for the functional
analysis, it is important to strike a balance between being too brief to convey the
detail required and providing too much information. All in all, it is beneficial to be
succinct, including the most economical amount of information required in order
to demonstrate the point that needs to be conveyed.

Table 1: Functional Analysis Summary: Otaki Group
Company Otaki Otaki Otaki Otaki Retail

Central Design Manufacturing
Functions
Manufacturing X
Design XX
Product selection XXX
Sales & Distribution X
Warehousing X
Strategic XX
Management
Financing X
Tangible Assets
Trade receivables X
Inventory X X X
Property X X XX
Plant/equipment X X
Intangible Assets
Trademark XXX
Manufacturing X
Intangibles
Marketing XXX
Risks
Market XX X
Credit X
Inventory XX
Product selection XXX
Contract XX
Staff X X X X
Key: XXX Key

XX Important
X Routine
If the functional analysis is incorporating a table, it is beneficial to include some

form of weighting so that the reader can quickly determine which functions, assets
and risks are important to the business. If the weighting for a function, asset or risk is
not immediately apparent then additional information should be given; this can
be particularly relevant for intangibles. For example, if Otaki Central outsourced
management and protection of its trade marks to another group member then
value and importance would need to be attributed between value building
activities by Otaki Central and management activities undertaken elsewhere.
When preparing the functional analysis, always keep in mind that it will need to be
updated in the future. In the case of factual information, such as the number of
employees in each division, consider that it will be necessary to locate this
information every time the functional analysis is updated. Consider how easy it will
be to get the information next time and whether it is necessary to include this
detail at all. Where information is presented in a reduced or summarised form,
such as in a table as above, it may be helpful to hold more detailed records
separately so the conclusions can be understood later for updating or in the event
of an enquiry.

It is best to avoid including names of individuals responsible for division/business

units in the documentation. If it is necessary to make reference to an individual, it is
preferable to refer only to their title. In rare circumstances, it may be necessary to
mention an exceptional individual by name, but this should be by exception.
Providing a list of names will increase the time that it takes to update the
functional analysis in the future as there are likely to be changes in personnel in the
intervening time.
6.10 Maintaining the Functional Analysis
Businesses never sleep, and the functional analysis will need to be reviewed
periodically when the transfer pricing documentation is being updated. An
efficient way to do this is to send the relevant sections to the respective
interviewees (or their successors) and have them review it ahead of the update
meeting. It is also not uncommon to have them just update the word document
using track changes. If the business undergoes a significant restructure it will be
critical to document how the functions of each entity have changed post-
restructure, and support the transfer pricing policy with robust benchmarking
analysis.
The functional analysis needs to be aligned to how the company portrays itself. If,
for example, the company is categorised as a low risk distributor, this will be
challenged by the local tax authority if the marketing spokesperson or CEO is
interviewed and claims the success of the group is down to the unique skills and
contributions of the local company. This is a difficult area to manage, but media-
facing company staff need to understand that their message must be aligned to
the functions, risks and assets of the company. Many audits have been started or
unnecessarily prolonged by a five minute interview in the media making grand
statements that are inconsistent with how the company actually operates.
Also, make sure that the categorisation of the company is consistent with its
website. Keep in mind that websites, press releases and other publicly available
information will be reviewed by tax inspectors. This publicly available information
should be in alignment with the functions, assets and risks of the company.
6.11 Conclusion
A good functional analysis will provide a succinct summary of the business'

functions, assets and risks, and the relative importance of these elements to the
business. Once the functional analysis has been established, the next step is to use
this information to characterise the business, which is considered in detail in a later
chapter.

CHAPTER 7
RELATING THE FUNCTIONAL ANALYSIS TO SELECTION OF TP

METHOD
In this chapter we are going to look briefly at how the functional analysis is used when
selecting the transfer pricing method, in particular looking at:
– An overview of the methodologies
– Most appropriate transfer pricing method
– Comparable Uncontrolled Price Method
– Cost Plus
– Resale Price Method
– Profit Split
– Transactional Net Margin Method
– Choice of tested party
– Some examples of profiles and links to transfer pricing methodologies
– The financial indicator where a transactional profit split method is selected
– Availability of comparables
– The identification of the significant comparability factors to be taken into account
7.1 Introduction
We have seen that the identification of the functions, assets and risks performed
and controlled by the enterprises which are parties to the transaction being tested
is the precursor to assessing and establishing the comparability of the transaction
under review to an uncontrolled transaction.
The functional analysis has the following aims:
• To identify and understand the intra-group transactions;
• To identify economically significant risk and who is able to manage and

control it;
• To ensure that controlled transactions are correctly delineated;
• To enable a choice of tested party (where needed);
• To form the basis for comparability;
• To determine any necessary adjustments to the comparables;
• To select the most appropriate transfer pricing method;
• To determine the correct profit level indicator;
• To ensure that the functional and risk profile of the tested party is reflected in
the chosen comparable.
Chapter III of the OECD TPG sets out a 9-step process to a comparability analysis in
paragraph 3.4. Step 3 of the OECD process describes the relevance of factual and
functional analysis to establishing comparability. We look at the 9-step process in a
later chapter.

‘3. Understanding the controlled transaction(s) under examination, based in

particular on a functional analysis, in order to choose the tested party (where
needed), the most appropriate transfer pricing method to the circumstances
of the case, the financial indicator that will be tested (in the case of a
transactional profit method), and to identify the significant comparability
factors that should be taken into account.’
An understanding of the relevant functions, assets and risks is therefore critical.

However experience has shown that certain transfer pricing methods present
strengths and weaknesses that lend themselves better to certain transaction types.
For example, a cost based method is usually more useful for determining an arm’s
length price for services and manufacturing, and a resale price based method is
usually more useful for determining an arm’s length price for distribution/selling
functions. These are discussed in an earlier chapter, you may wish to look at them
again before continuing with this chapter. Although we revisit the methodologies
in this chapter, it is at a higher level assuming you are familiar with each one.
7.2 An Overview of the Methodologies
As stated in an earlier chapter, the OECD sets out five methods, together with a
provision for ‘other’ methods where none of those listed are appropriate. The five
methodologies were looked at in detail in an earlier chapter however a brief
summary of each is provided below for completeness. You should refer to the
earlier chapter for more detail on each.
• Comparable uncontrolled price (‘CUP’)
• Cost plus
• Resale price
• Profit split
• Transactional net margin method (‘TNMM’)
As mentioned earlier the TPG no longer contain a hierarchy for selection of the
transfer pricing method, however some countries continue to do so.
7.3 Most Appropriate Transfer Pricing Method
You will recall that the overarching guidance is that the aim should be to find the
most appropriate method for the particular case and that the selection of the
method be considered in the context of the nature of the controlled transaction
determined, in particular, through a functional analysis. It can often be the case
that a decision is made as to the type of entity which is the party to a transaction
(‘entity characterisation’) that will influence the selection of method and, where a
transactional profit method is selected, the financial indicator to be used.
‘The selection of a transfer pricing method always aims at finding the most
appropriate method for a particular case. For this purpose, the selection
process should take account of the respective strengths and weaknesses of
the OECD recognised methods; the appropriateness of the method
considered in view of the nature of the controlled transaction, determined in
particular through a functional analysis…;’ (paragraph 2.2)
It is important to stress again how the selection of transfer pricing method is

dependent on the functional analysis – the facts and circumstances of the
transaction. Case law provides examples of where this has not been adhered to.

Dixons (DSG Retail Ltd and Others v HMRC [2009] UK FTT 31 (TC))
This was a UK case. A UK retailer (DSG) of white goods was selling warranties for
products bought in store. The UK retailer contracted with a third party (Cornhill) on
the understanding that it would pass on most of the profits to a group company in
the Isle of Man (low tax). Later on, changes in the law resulted in amendments so
Cornhill was no longer involved. As a result of the agreements in place, a large
proportion of the profits accumulated tax-free in the Isle of Man subsidiary, DISL.
The retail company in the UK didn’t receive a fee for providing a business facility to
the IOM Co.
The issues contested included:
• Whether there was a provision between DSG and DISL for the purposes of UK
law?
• Did the actual provision differ from the provision that would have been made
between independent enterprises?
• Did the provision confer a potential advantage in relation to UK tax?
• What adjustment was required?
• Whether CUP was the correct methodology as suggested by the taxpayer?
The highest court deciding the case held that there was clearly a provision
between DSG and DISL. Furthermore, the court rejected the various CUPs put
forward. It was decided that DSG’s profits should be adjusted to that of an arm’s
length arrangement based upon the profit split method and that the adjustment
should take the form of a commission payable by DISL to the UK retailer.
The decision in this case reinforces the importance of performing a sound

economic analysis of commercial agreements and that HMRC will construe the
arrangements as a whole, and if it can show that the third party was engaged by
a group entity only on the understanding that the third party would, in turn, enter
into a contract with another entity from the group, the transfer pricing legislation
will apply.
This is an important case – you will find more details on it in the case law appendix.
Roche Products Property (2008 AATA 261)
This case involved the judges looking at the testimony of expert witnesses who do
not agree with each other.
Roche Australia is a subsidiary of Roche Holdings Ltd of Basel, Switzerland. Globally,

the Roche Group carries on the business of producing, selling and supplying
pharmaceutical and diagnostic products. Roche Australia comprised three
operating divisions over the audit period which was 11 years up to 2002. The
Australian Tax Office (ATO) questioned the transfer pricing methodology used by
the company. The court disagreed with the evidence put forward by the ATO
criticising a number of ‘presumptions’ made by a number of the expert
economists. The court thought the economists were too US-focused in their
approach and as a result they did not provide analysis that specifically addressed
Australia’s transfer pricing provisions.

The key points made in the case included the following:
• That arm’s length prices be determined for each separate year under
consideration, rather than a multiple-year average.
• The ruling acknowledges the difficulty in finding available comparable data,

and uses a uniform gross margin to price the transfers of all pharmaceutical
products; a preference was expressed for transactional methods over profit
methods, such as the profit-based transactional net margin method (TNMM).
Following the case the ATO released as a statement that in their view Roche is
confined to ‘to the facts of the case’.
There is more detail on this case in the Case Law Appendix.
Maruti Suzuki India Limited v ACIT (2010)
Suzuki Motor Corporation owned over half of Maruti Suzuki India Ltd and provided
the Suzuki name for the company to co-brand cars (alongside the Maruti name)
for the Indian market. A royalty was paid to Suzuki for use of the name. The Indian
tax authorities successfully challenged the value of the Suzuki name in the Indian
market, looking closely at local marketing spending to conclude that Suzuki had,
in their view, ‘piggy-backed’ a better known local brand. This shows the
requirement to understand the functional analysis from both sides and
perspectives, as value may be perceived differently in different territories.
It is worth noting that the challenge in these cases has been to the nature of the
underlying transaction rather than to the method itself. However, in almost every
instance where a transaction is not appropriately identified, the resulting transfer
pricing method will likewise be inappropriate.
7.4 Comparable Uncontrolled Price (CUP)
The CUP method is often referred to as the most objective method. If we look to
the glossary in the OECD TPG we see that it is defined as a method that compares
the price for property or services transferred in a controlled transaction to the price
charged for property or services transferred in a comparable uncontrolled
transaction in comparable circumstances.
In cases where comparable uncontrolled transactions can be found, the CUP

method is a direct and sound method to determine whether the conditions of
commercial and financial relations between associated enterprises are at arm's
length. It is often useful to consider the CUP method first when looking at possible
internal comparables and external comparables.
As noted earlier the CUP method is often chosen when:
• One of the associated enterprises involved is engaged in comparable

uncontrolled transactions with an independent enterprise (ie an internal
comparable is available). In such a case, all relevant information on the
uncontrolled transactions is available and it is therefore probable that all
material differences between controlled and uncontrolled transactions will be
identified;
• The transactions involve commodity type products, but only those in which
product differences are adjustable; and
• We are looking at the interest rate charged for an intercompany loan.

If the CUP method cannot be applied another transaction or transactional

method can be chosen. In these cases it is good practice to specifically exclude
the CUP as an appropriate method in the transfer pricing documentation’s
‘selection of method’ section.
7.5 Cost Plus
This method takes the direct and indirect costs of the controlled transaction and
adds the appropriate mark-up so that a profit is made on the controlled
transaction.
The cost plus method is often most appropriate where the cost of the product or
services provision, rather than sale price, is the key value driver. This will be
determined through the functional analysis.
For example, the cost plus method is typically applied in cases involving the
intercompany sale of tangible property where the related party manufacturer
performs limited manufacturing functions and incurs low risks, because the level of
the costs will then better reflect the value being added and the market price.
The cost plus method is often used in transactions involving a contract

manufacturer, a toll manufacturer or a low risk assembler which does not own
product intangibles and incurs little risk. The cost plus method is usually not a
suitable method to use in transactions involving a fully-fledged manufacturer
which owns valuable product intangibles, as it is difficult to locate independent
manufacturers owning comparable product intangibles.
The cost plus method can also be used to price charging for services (eg legal,
accounting, information technology, marketing, tax, etc.) if the services can be
considered to provide a benefit to the service recipient. However, for services,
often in practice TNMM is most commonly chosen with a cost based profit level
indicator (PLI). PLIs were covered in the earlier chapter on methodologies and are
also looked at again later in this chapter.
It is important to have good quality data and to ensure that the comparable
transactions are indeed comparable and a close match to the controlled
transaction.
7.6 Resale Price
This methodology is often described as going backwards from the sale price to
find the transfer price. The final selling price is reduced by the cost of getting the
product to market, eg transport costs and an appropriate profit margin.
The Resale Price Method is normally used in cases which involve the purchase and
resale of tangible property in which the reseller does not add substantial value to
the tangible goods by way of physically modifying the products before resale or in
which the reseller does not contribute substantially to the creation or maintenance
of intangible property, for example a local marketing intangible.
In a typical intercompany transaction involving a fully-fledged manufacturer

owning valuable patents or other intangible properties and affiliated sales
companies which purchase and resell the products to unrelated customers, the
Resale Price Method is a method to use in respect of the sales company if the CUP
method is not applicable and those sales companies do not own valuable
intangible properties.

In the case of distribution activities, where the distributor takes ownership of the
goods being sold, the Resale Price Method lends itself best to test the arm's length
nature of the transaction (again, in the absence of a CUP).
7.7 Transactional Net Margin Method
The TNMM compares the net profit margin (relative to an appropriate base) that
the tested party earns in the controlled transactions to the same net profit margins
earned by the tested party in comparable uncontrolled transactions or
alternatively, the net profit margins earned by independent comparable parties.
For example, return on total costs, return on assets, and operating profit to net
sales ratio.
In all cases where individual products or services cannot be priced separately the
use of TNMM provides the optimal solution as it compares the profitability of a third
party with the related party entity. However, when using the TNMM it is very
important to choose the right profit indicator (ie cost plus for services, resale price
minus discount for distributors, etc.).
Another key issue when applying the TNMM is ensuring that the functional and risk
profiles match those of the selected third party comparables. However, when
running benchmarking studies it is important to understand that tax authorities can
always challenge the choice of comparables; therefore, running sensitivity analysis
on the set of comparables can be valuable as it shows how the arm’s length
range vary. If the transfer pricing is chosen in a manner that makes it less sensitive
to changes in the set of comparable third parties the overall risk of adjustments by
the tax authorities can be reduced.
7.8 Profit Split
The transactional profit split method and the transactional net margin method are
known as the transaction profit methods, as they focus on the outturn of the
transaction rather than the price of the sale of goods or services themselves.
The profit split method takes the total profit for all the associated enterprises and
splits it amongst them in a way that reflects how it would have been split between
unconnected parties.
In general the profit split method should be applied when transactions cannot be
benchmarked using internal or external comparables or when the transaction to
be benchmarked involves the input of several parties, which might also be
contributing intangible assets to generate the overall value for the business.
The section on the profit split method in an earlier chapter includes considerable
detail on when the method would be used. You should refer to that chapter as
necessary.
7.9 Choice of Tested Party
Paragraph 3.18 of the OECD TPG indicates that it is usually the least complex party
to the transaction that should be tested, as this will allow for the greatest reliability.
‘...The choice of the tested party should be consistent with the functional
analysis of the transaction. As a general rule, the tested party is the one to
which a transfer pricing method can be applied in the most reliable manner
and for which the most reliable comparables can be found, ie it will most
often be the one that has the less complex functional analysis.’

7.10 Examples of Functional Profiles and Links to Pricing Methodologies
The following table shows some common examples of functional profiles together
with the transfer pricing methods that may be appropriate to a particular profile.
Functional profile Description Pricing method

Group This is the entity containing Residual profit after
entrepreneur/intangible the decision makers, taking rewarding the other
owner the investment risks (eg entities in the supply
research, new markets and chain for their
innovation). The group functions.
entrepreneur can take
several forms. For example it
can be a manufacturer, the
group researcher or the
group product designer.
Contract manufacturer A contract manufacturer CUP/Cost plus
produces goods under the method/
direction and using the Transactional net
technology of the group margin method
principal (usually by
reference to a contract). Its
risks are primarily limited to its
efficiency and ability to
retain the group
manufacturing contract. In
its most limited risk form it will
be a toll manufacturer with
the principal supplying and
retaining ownership of all
materials
Service provider A service provider supplies CUP/Cost plus
services to other group method/Transactional
companies usually by net margin method
reference to a contract. Its
provide contracted services
at budgeted costs
Distributor A group distributor distributes CUP/Resale price
goods supplied by its method/Transactional
principal. Its risk profile can net margin method
vary dependent on the
structure of the operation.
7.11 The Financial Indicator Where a Transactional Profit Method is

Selected
On working through the selection of a method, it may be decided that a

transactional profit method is the most appropriate to test and support a transfer
price. Whilst there are variations, these methods essentially compare a profit
measure resulting from a transaction between controlled parties to a profit
measure earned on a comparable uncontrolled transaction.
The TNMM (similar to the Comparable Profits Method ‘CPM’ under the US Regs, if
not in concept, in terms of application) is a popular pricing methodology. It relies
on comparing a margin earned from a transaction/function with comparables

that may either be internal or external (which typically rely on proprietary

databases of financial data).
Where this method is selected, and external data used, a decision needs to be
made on the types of company that the tested party will be compared against
(which the functional analysis will inform) and the appropriate profit level indicator
(‘PLI’) to make a comparison against. The selection of the PLI will usually be
determined by reference to the appropriateness as judged against the
transaction and entity, informed by the functional analysis.
For example:
• sales transaction – consider measuring profit against revenues (sales) of sales

entities;
• service provision – consider measuring profit against costs incurred by service

providers providing the comparable services.
Other transactional profit methods ie profit split, whilst not necessarily reliant on a
financial indicator, will be reliant on the functional analysis of the parties to the
transaction.
Again, case law can also provide examples of how functional analysis, selection of
method and PLI are considered together.
GAP International Sourcing (India) PvT. Limited v CIT (2012)
GAP International Sourcing provides procurement services for its group in India. The
Indian tax authorities sought to challenge the company’s transfer pricing policy of
a mark-up on value added expenses, preferring a commission of 5% of the Free on
Board price. The taxpayer’s position was upheld as the Tribunal found no evidence
of local intangibles that would move its transfer pricing method away from cost
plus and that any location savings would be passed on to customers by a third
party.
NB: ‘Free on Board’ is a transportation term that indicates that the price for goods
includes delivery at the Seller’s expense to a specified point and no further.
LG Electronics India Pvt. Limited v ACIT (2013)
This was an Indian case. LG India manufactured and distributed LG Korea’s

products under license, for which it paid a royalty of 1% for the use of industrial
property rights, designs, and technical know-how to manufacture the LG Products.
LG Korea also provided LG India with certain marketing subsidies, such as funds
toward sponsorship of a global cricket event, to assist it with this presumably global
branding event. There were no intercompany agreements for these subsidies or
any provisions for brand-building expenses.
LG India was to market and exploit LG products in the Indian market.
LG India was quite profitable, and considered more so than its perceived
comparable companies, notwithstanding that it incurred significant Advertising
Marketing and Promotion (‘AMP’) expenses. For the years in issue, its AMP
expenses were 3.85% of its gross sales, this was higher than comparable
companies in India.
The Indian tax authorities contested the payment stating that the AMP costs were
too high. The issue being that the Indian subsidiary was incurring cost that directly

added to the brand value of the trademark owned by the Korean parent. A
‘bright-line’ (see below) was applied. The Court questioned whether the costs
were being incurred on behalf of the parent and whether a mark-up should be
added for providing marketing services.
The Court decided that the AMP expenses that LG India incurred on behalf of LG
Korea to build its brand were more than what independent third-party entities
would incur proportionately in a similar transaction.
The Court rejected the parties’ various arguments that the bright-line test was not
a sanctioned transfer pricing method, but rather merely saw the test as a tool in
the cost allocation analysis to determine routine versus non-routine brand-building
expenses.
The Court also rejected the taxpayer’s argument that LG India's operating profit
was higher than that of the comparable firms and that any additional profit from a
marketing service fee was, therefore, unnecessary and duplicative.
The broader implications that arise from this case include:
• The Court identified several factors have a bearing on the question of

determination of the cost/value of the international transaction of brand/logo
promotion through AMP expenses
• The Court focused on selling expenses and correctly noted that they should
not be considered AMP expenses
• The Court stated that each transaction must be considered separately

notwithstanding the overall level of profitability
• Looking at the non-routine brand-building expenses, the Court considered this

a marketing service by LG India on behalf of LG Korea and agreed that the
cost-plus method was appropriate
• The Court considered Maruti Suzuki and noted that the principles observed in
that case are still held to be correct: (1) brand promotion expenses are an
‘international transaction’; (2) AMP expenses incurred on behalf of a foreign
enterprise should be compensated at arm's length; and (3) identification of
the factors required to be considered for determining an arm’s length price
You will find more details on this case in the case law appendix.
NB: the ‘Bright line Test’ was first put forward in a US case (DHL). The judge in this
case identified that test which notes that, while every license or distributor is
expected to spend a certain amount of cost to exploit the items of intangible
property with which it is provided, it is when the investment crosses the ‘bright line’
of routine expenditure into the realm of non routine that economic ownership,
likely in the form of a marketing intangible, is created.
SNF (Australia) Pty Ltd v FCT (2010) FCA 635
An Australian company purchased products from related companies outside

Australia.
The Australian company had used the comparable uncontrolled price using the
pricing of transactions between the suppliers and their arm's length customers. The
company incurred losses, in part due to commercial issues (including a low level of
sales per salesperson, competition in the Australian market, excessive stock levels,

and poor management) and partly due to a strategy to penetrate the Australian
market.
The Australian Taxation Office argued that the transactional net margin method
should be used, where a ‘benchmark operating profit’ should be determined with
reference to the operating profits achieved by other ‘functionally comparable’
distributors.
The Federal Court accepted the company's pricing, and held that it could not be
concluded that the prices were artificially inflated.
This case is also interesting as it considered the importance of the OECD TPG. The
commissioner had put forward a strict definition of comparability defending it as
being in line with the OECD TPG. The courts agreed with his interpretation of the
OECD TPG but further went on to say that it was examining Australian domestic
law, not a treaty, and it was not obvious how the OECD TPG came to be relevant
to domestic law. However the case did involve countries where Australia had
concluded treaties containing Article 9 in line with the OECD Model DTC and
hence the treaties need to be considered when looking at domestic law.
It noted that the OECD TPG were not part of the Commentary and were
expressed to be ‘guidelines’. They went on to say that under the Vienna
Convention, the OECD TPG might be examined if they reflected subsequent
agreement or practice of the parties to the treaty which ‘establishes the
agreement of the parties regarding its interpretation’; on the facts they decided
this was not the case. This meant that the OECD TPG could not be used to interpret
the meaning of Australia’s domestic transfer pricing legislation.
SNF (Australia) Pty Ltd v Commissioner of Taxation (Full Federal Court Decision)
(2011) FCAFC 74
Following on from the above case, the Full Court concluded that SNF Australia was
not required to quantify and provide evidence to establish the correct arm’s
length price, but only had to establish that the Commissioner’s assessments were
excessive.
In summary the narrow approach was rejected; it was held that the taxpayer had
demonstrated that the prices paid were arm's length or less than arm's length
prices and thus it had absolved itself of the burden of proof.
In relation to the fact that SNF Australia had incurred sustained trading losses over
the years in question, the Federal Court said that this was irrelevant to applying the
test of whether arm’s length consideration had been paid in respect of the
acquisitions.
The case provides a lot of detail on how Australian law should be interpreted. The
starting point in Australia is not to treat domestic law as international, however in
cases of doubt an interpretation based on Australia’s international obligations can
be appropriate.
In Australia, the OECD TPG are not a legitimate aid to the interpretation of DTAs or
Australia’s domestic transfer pricing provisions.
Under the law in Australia so long as appropriate comparable transactions are

available, an analysis of relevant comparables (eg the ‘comparable uncontrolled
pricing’ method) is an appropriate method to determine an arm’s length price for
transfer pricing purposes.

If there is a global market for the relevant product or service, taxpayers may use
comparables from countries and regions other than Australia in determining an
arm's length price.
When identifying appropriate comparables to determine an arm’s length price for

transfer pricing purposes, the selected entity comparables should carry out similar
activities to the taxpayer (ie have ‘functional comparability’ with the taxpayer)
Unilever Kenya (Income Tax Appeal) 753 of 2003
Unilever UK manufactured various products for Unilever Kenya. Unilever Kenya

charged lower prices to Unilever Uganda than it did to domestic buyers and to 3rd
party importers. The transfer pricing documentation included the following based
on the capital used for production of the goods:
• A charge on working capital will be made on average working capital

employed.
• The price will not be less than full variable cost + 10% profit mark-up plus actual
transport costs.
• Insurance will be for the account of the buyer.
• Prices may be adjusted taking into account increases or decreases in prices of

raw materials, exchange rate, import duty on raw materials and changes in
prices of other inputs.
The Kenyan Tax Authorities tried to impose a CUP. Two potential CUPs were put
forward: the average price of sales in Kenya by Unilever Kenya and the average
price of export sales by Unilever Kenya. No attempt was made to adjust the
potential CUPs for functional or other differences. Unilever said that there was no
CUP pointing out the functional differences in the third party sales compared to
the sales to Unilever Uganda. An important difference was that for third party sales
Unilever Kenya undertook the marketing function and hence incurred costs. This
was not the case with the sales to Unilever Uganda, allowing a sale at a lower
price. The Court agreed that there wasn’t a CUP.
It was held that cost plus was an acceptable method.
In reaching its decision the Court endorsed the use of the OECD TPG in the
absence of detailed guidance from the Kenyan Revenue Authority (KRA). The lack
of guidance resulted in a shift of the burden of proof from the taxpayer to the KRA.
The government's response to this judgment was the introduction of Transfer Pricing
rules, which are largely based on the OECD TPG.
The main implications from this case include that a tax authority that lacks
experience can take the wrong approach when dealing with complex transfer
pricing cases and that if CUP is to be used then it cannot be applied where there
are a wide range of product and contract differences.

7.12 Availability of Comparables
For all transfer pricing methods, access to information on comparables is

necessary and it may be that due to difficulty in getting access to reliable data on
comparables a different method is then chosen.
Although independent unrelated comparables are usually used for transfer pricing
purposes, in practice it is often observed that for certain countries it is not possible
to identify comparables or reliable company data that meet the comparability
requirements. In such cases, practical solutions must be sought in good faith by
taxpayers and the tax administration. A possible solution may include searching for
comparables in other geographical regions that share certain key similarities with
the country in which a company conducts its business (eg depending on the
industry, for manufacturers established in, for example, Africa, a search for
comparables could be carried out in Asia or Eastern Europe).
Alternatively an industry analysis (publicly available or internally conducted by the

company) could be used to identify profit levels that can reasonably be expected
for various routine functions (eg production, services, distribution, etc.).
7.13 The Identification of the Significant Comparability Factors to be

Taken into Account
As we have seen the functional analysis will indicate, out of the functions, assets
and risks identified, which are the significant ones that will be critical in a
comparability exercise.
It is sometimes the case that adjustments will be needed in establishing

comparability of the controlled transaction with uncontrolled transactions. The
functional analysis is the means by which the need for a comparability adjustment
is identified, determining the nature of the adjustment itself and the basis for
determining whether the adjusted data is sufficiently comparable.
We will look at this topic in more detail in a later chapter.

CHAPTER 8
ENTITY CHARACTERISATION
In this chapter we look at how functional analysis is used for entity characterisation and
how characterisation can affect the chosen tested party, in particular looking at:
– An overview of entity characterisation
– Entity characterisation comparing simpler and complex entities
– Sales functions
– Manufacturing entities
– Support service activities
– More complex or entrepreneurial entities
– Planning aspects of entity characterisation
8.1 Choice of Tested Party
One of the most helpful outcomes from a functional analysis review is to enable
entities in a group to be classified, based on each entity's functions, assets and
risks, from ‘simpler’ entities to the more ‘complex’. We will see in this chapter how
this helps with the choice of tested party.
When the transfer pricing method employed is a cost plus, resale price or
transactional net margin method, it is necessary to select the party to the
transaction for which a financial indicator is tested (the ‘tested party’). That
financial indicator will be, in each of those cases respectively, the mark-up on
costs, the gross margin, or the net profit indicator.
The OECD TPG indicate that, as a general rule, the tested party will be the one for
which a transfer pricing method can be applied in the most reliable way and for
which the most reliable comparables can be found, which will usually be the party
for which the functional analysis of the transaction is less complex. This arises from a
practical perspective, as comparable companies (and their financial data) are
easier to identify where there are fewer differentiating functions, assets and risks
involved.
The OECD TPG give an example under which company A manufactures product 1
and product 2, and sells both to overseas associated company B.
PARENT CO
COMPANY A Product 1 → COMPANY B

Manufacturer Intangibles
Simple functions Technical
specifications
↑
Tested Party

Product 1 is manufactured using valuable intangibles owned by company B and

following technical specifications set by B. Company A performs only simple
functions, and does not make any valuable, unique contribution in relation to the
transaction. The tested party would most often be company A.
PARENT CO
COMPANY A Product 2→ COMPANY B

Manufacturer Distributor
Intangibles Simple functions
↑
Tested Party
By contrast, in the case of product 2 company A uses its own valuable and unique
intangibles while company B only acts as a distributor, performing simple functions.
The tested party for this transaction would most often be company B.
Despite this guidance, in practice the choice of tested party will often be driven
by the structure of the relevant legislation in the jurisdiction in question. This may
specifically require consideration of the position of the resident taxpayer, rather
than of any other party. Even where the focus of the legislation is on the terms of
the transaction rather than the position of the taxpayer, so that in theory arm's
length terms could be established by considering either party, in practice there
may be an expectation by the tax authorities that the question should be
approached by reference to the party whose tax liability is at issue. In addition,
that will normally be the party in relation to which most information is available
both to advisers and to the tax authorities; in practice, obtaining sufficient
information from related companies in other jurisdictions may be difficult.
A complex entity might own, manage and develop intellectual property and
make key strategic decisions. A simpler entity would normally undertake more
routine tasks with lower risk such as contract manufacturing or support service
provision. Simpler entities would typically not own valuable intellectual property.
Take the example of the pharmaceutical industry. A complex entity would be that
which manages the development of new drugs, and owns and manages the
intellectual property relating to existing drugs. If this entity sold the drugs to a
related party distributor in another country, the provision from a transfer pricing
perspective would be the price of these drugs as between the two entities.
From a transfer pricing perspective, it is difficult to quantify the arm’s length return
to be made by a pharmaceutical company for selling drugs intra-group. The value
of drugs will be dependent on a number of factors such as the treated condition,
whether the drug is seen as revolutionary, the number of competitor products and
local market conditions (for example whether the main buyer of drugs is a single
national health service or whether there are multiple private providers).
Given the difficulty of looking at an arm’s length provision from the perspective of
a more complex entity, transfer pricing work tends to focus on the simpler entities:
in this case, the local distributor of the drugs. From a transfer pricing perspective,
distributors of drugs should make relatively similar economic returns regardless of

the pharmaceutical company or the type of drug being distributed. This is due to
the fact that a pure sales activity requires essentially the same skill set and
practices for all pharmaceutical products – the functions, assets and risks of a
distributor is likely to be broadly the same.
In order to support the pricing between the parent and the distributor, if no
comparable uncontrolled price (CUP) is available it would be normal to look at
the profit margins earned by the distributor as a result of the purchase of the drugs
and compare these margins with those achieved by independent companies (for
instance, companies acting as distributors for third parties) performing the same
activities in that market. This is normally carried out by a benchmarking study using
an economic database that draws on data from different markets, for example
North America, Europe or Asia.
Some local differences may exist. For example, in markets such as the US it may be
necessary for pharmaceutical distributors to spend significant sums targeting
doctors and patients with promotions and advertising which arguably could
create a local marketing intangible. However, it could still be possible to
benchmark a local distributor’s returns, taking into account this marketing
intangible by comparing them against the returns of other independent entities in
the market that bear similar costs. Effective functional analysis will guide these
types of decision.
While the transfer pricing analysis will normally be performed on the simpler entity,
there is still work to be carried out using the functional analysis to assess the precise
rewards of the tested party: there is a sliding scale between ‘simple’ and
‘complex’. Some simpler entities will have a much higher level of functionality and
risks assumed than other entities, and this could have a crucial impact on the
arm’s length transfer pricing provision. It is also important not to generalise a
specific fact pattern into a generic characterisation; a service function that
includes key business risks – for example outsourced analysts who perform quality
control of deliverables going direct to a client – might not be appropriately
rewarded as a routine, low-risk function.
The next section looks at entity characterisation based on the functional analysis
and its impact on the reward achieved by different entities that are party to a
provision.
8.2 Entity Characterisation – Comparison of Simple and Complex Entities
Comparing simpler and more complex entities gives an initial understanding of

how to direct the transfer pricing analysis. However, further work is needed to
understand the precise value contributed by each entity in order to measure an
appropriate arm's length reward.
Once a detailed functional analysis is performed, it is possible to obtain an

understanding of the economic purpose of each entity within the group and the
contribution that they make to the overall ‘value chain’ of the group for providing
a particular product or service and the component transactions from which this is
made.
A value chain relates to the steps needed to deliver a product or service and
measures the contribution (in value terms) of each step or process to the overall
value chain. By understanding the value chain and in particular the contribution of
each entity, it is possible to classify each entity for transfer pricing purposes.

Entity characterisation is perhaps one of the most important and controversial

areas of transfer pricing. It requires groups to take a dispassionate look at the
outputs of their functional analysis and to assess the activities that each entity
performs and the value they add, and then to classify each entity accordingly.
The entity characterisation will feed directly into the selection of transfer pricing
method and economic analysis.
The diagram below shows how the entity characterisation, which is derived from
the functional analysis, can affect the transfer pricing policy and, through it, the
level of local profitability.
Fundamentally there is a direct correlation between the profit potential of a

company and the type of activity it conducts, the risks that it assumes and the
assets, especially intellectual property, that it owns. More complex entities possess
a higher degree of functionality, risks and assets, and so have the potential to
generate the greatest profit margins. On the other hand, increased functionality
and risk gives rise to the potential for much greater fluctuations in profitability,
including the possibility of financial losses. These more complex entities are more
“entrepreneurial” where they drive the key strategic and critical decisions for a
group and take on the associated risk. The right or wrong decisions in this regard
will have a direct impact on the financial performance of the group.
In some cases, multinational enterprises may seek to centralise key strategic or

high value functions and risks in a single entity to avoid duplication and simplify
management. Where implemented effectively, this further reduces the functions,
assets and risks of local activities in favour of those in the entrepreneur or
‘principal’ company.
The key to successful entity characterisation is to draw evidence directly from the
functional analysis outputs without imposing an oversimplified view which is neater
but which may not reflect the variation of local activities. Where the latter occurs,
tax authorities are increasingly identifying and challenging the position, particularly
when in practice ‘limited risk’ operations are less limited than they are presented.

To achieve this, the table in the functional analysis, like the one we saw for Otaki
Group in an earlier chapter, summarising the location of key functions, assets and
risks is often valuable as this shows each entity’s relative complexity.
Simpler or Routine Entities
Many group entities often carry out more simple or ‘routine’ functions, especially
where key risks are managed centrally. These are normally the focus for the
economic analysis as the ‘tested party.’
These entities perform functions which, while part of the value chain, do not
contribute materially to the value added functions, assets and risks which
differentiate the business. They will normally undertake activities which can be
easily replicated (ie they are not protectable) and as such are neither unique nor
dependent on proprietary intellectual property. This might include distribution
activities, contract manufacturing or contract research and development, or the
provision of services, but as always the characterisation of these activities will
depend on the fact pattern concerned.
It is important to remember that a characterisation can extend to the whole

activity of an entity (for example a contract manufacturer) or be limited to
activities in respect of specific transactions, such as the provision of IT services.
The type of entity characterisation that is applied depends on the nature of the
activity involved.
8.3 Sales Functions
Whilst a sales function on its own is normally considered routine, its type of activity
and level of risk can vary widely. At one end of the spectrum is the example of a
full risk sales entity or a licensed distributor, taking stock risk or licensing a brand or
other intangibles. At the other, an entity may provide sales support, research the
market and facilitate logistics but not enter into customer contracts or take title to
products. Some of the terms commonly given to the spectrum of sales function
and the potential for profit (and profit volatility) associated with them are
illustrated below.

Taking each in turn:
Sales Support Entity
The information from the functional analysis may reveal evidence that a local sales
operation is in reality a sales support function. For example, it does not enter into
contracts directly with customers (these may be concluded between customers
and the entrepreneurial entity over the internet) or take product title or stock risk. It
may facilitate distribution logistics and/or local marketing under the close direction
of the entrepreneur in the value chain.
This may suggest that value driver in this sales support entity is primarily cost rather
than sales. This may then support the selection of a transfer pricing method based
on cost, such as cost plus, as the most appropriate method.
Sales Agent or Commissionaire
Where an entity acts as a sales agent for a principal, more substance will be
shown through the functional analysis as the entity will negotiate with customers
within outlined parameters set by the principal. The entity will not take legal title,
and so not be exposed to stock, warranty or litigation risks, again leaving local
entity risk at a low level. How this is reflected through intercompany agreements
will be an important point for the functional analysis to confirm a characterisation
of sales agent (under common law) or commissionaire (under civil law).
As the name suggests, a commissionaire will receive a commission for the sales it
secures, suggesting that a transfer pricing method based on sales, such as the
resale price method may be most appropriate.
Below is a simple diagram of how a commissionaire structure would work.

As shown by the diagram, in a commissionaire structure the title of the goods

passes directly from the principal company to the customer. The customer has
entered into a contract to buy the goods from the commissionaire company.
The commissionaire has a contract with the principal to buy the goods and pay for
them. The commissionaire company then receives a fee/commission. There is no
contract between the principal and the customer. The customer may not know
that the principal exists.
Limited Risk Distributor
An entity may be a limited risk distributor (‘LRD’) where it is entering into contracts
on its own behalf, making sales in its own name, and performing local
implementation of a central marketing strategy. A LRD's title to a product may only
be brief, with ‘flash title’ passing at the moment of sale. Local risk will still be limited,
with key risks such as inventory, warranty, currency and bad debt risks borne by
the entrepreneurial entity, which also provides strategic management.
The most appropriate method for rewarding an LRD will depend on the fact
pattern and available data. For example a target operating margin may be
assessed under TNMM which is then implemented through the product price,
supported by periodic adjustments.
Licensed Distributor
It is common to see third parties buy in intellectual property through a license or

franchise arrangement. Here, the third party franchisee will not own the intellectual
property but will have a local right to exploit it and maximise their returns.
In these cases, the functional analysis may show a significant level of local decision
making and risk, for example in relation to market strategy, pricing and inventory in
addition to the other risks assumed by an LRD. There will usually be a significant
profit potential from these activities, but also the downside risk of potential losses,
as results can fluctuate with this level of local risk.
The functional analysis becomes even more important in determining the most
appropriate method as, even with these fluctuations, profit or loss will need to be
allocated appropriately in the value chain. It may be that the entity's underlying
distribution functions are essentially routine and may be tested in a similar way to
those of an LRD (although at a higher level in the resulting pricing range), with any
balance of profit attributable to the licensor or franchisor. If the licensor is shown by
its functional analysis to be more passive or the licence of a limited or measurable

value, it may be that the licence transaction becomes tested with the balance of
profit or loss remaining in the distribution entity.
Full Risk Distributor
If the functional analysis shows a sales activity which begins to take on elements of
a smaller-scale replica of the whole entity, ‘full risk distributor’ may be appropriate.
This may be similar to a licensed or franchised distributor, but either replacing or
enhancing any bought-in intellectual property with local value adding assets. This
could include local marketing intangibles such as sub-brands, as well as greater
levels of local management and risk.
As its characterisation suggests, a full risk distributor may not be a simpler entity in
its relationship with other group entities – as always this will depend on the fact
pattern. The method will need careful consideration as profit split or TNMM may be
most appropriate, or it may be that counterparties such as the group’s
manufacturer or brand owner should themselves become the tested party.
For all these arrangements, it is important that the functional analysis review shows
the risk borne by both entities, including the more complex. Key questions to
consider will include:
• Who takes the risk of unsold stock?
• Who bears the cost of a market shortfall when demand is insufficient to cover
a distributor’s costs?
• Who is liable for proper performance of customer contacts, for example in the
event of late delivery or warranty issues?
8.4 Manufacturing Entities
Many groups outsource their manufacturing operations to third party contract

manufacturers, particularly in low cost territories such as China and Eastern Europe,
to enhance profit margins and to be price competitive. Groups may decide that it
is more effective to set up their own manufacturing subsidiary – for instance, to
have more control over the manufacturing process or to protect their intellectual
property.
A functional analysis can identify different types of manufacturing entity. Again,

some of the terms commonly given to the spectrum of manufacturing function
and the potential for profit (and profit volatility) associated with them are
illustrated below.

Toll Manufacturer
The lowest risk entity is a toll manufacturer. Here the complex entity retains title to
both the raw materials and goods throughout the manufacturing process. The
complex entity buys the raw materials or sub-assembled goods, although the
physical flow of goods will be directly to the manufacturer itself. As a result, the
complex entity bears all the inventory and sales risk, while the toll manufacturer (or
‘toller’) is primarily responsible for the management and effective utilisation of its
assets in the production process.
As this fact pattern suggests, an appropriate method to reward a toller will be one
based on cost, such as cost plus or a return on assets employed.
Contract Manufacturer
A contract manufacturer is the first step up from a toll manufacturer. In addition to

owning plant and machinery and employing a skilled labour force, it will also own
the raw materials through the production process and have title to the end
product. However it will not own any of the design-related intellectual property
(despite having know-how relating to its production processes) and will usually
manufacture set volumes to order.
Functional analysis will show to what degree this exists: for example, a contract
manufacturer may perform its own procurement and may retain title to the
finished goods, or both of these could be centralised elsewhere in a group. It will
also show which entity has responsibility for increased unit costs from
undercapacity: these might be set out in an intercompany agreement showing
order volumes, or this risk may be assumed entirely by the entrepreneurial entity.
Again, the most appropriate method may be to apply a cost plus on product
pricing, or target a margin under TNMM which is then implemented through the
price of manufactured product sold to group entities.

Licensed Manufacturer
As with their equivalent distributors, a manufacturer may license (either explicitly or

in effect) design intellectual property from another group entity. This may be a
long term arrangement. The manufacturer will then have the responsibility to
exploit this effectively, potentially investing in developing intangibles used in the
production process. Again, there is a sliding scale of functionality. It may
undertake its own procurement or draw from a central group function, and it may
hold stocks of both raw materials, semi-finished and finished product. It is also
much more likely to carry local risk for under-utilisation of its production capacity.
Again, careful functional analysis is important to identify all the relevant

transaction types – does it contract out procurement, for example? This will allow
an understanding of whether the licensed manufacturer remains the simpler entity
or whether the other assets and functions on which it relies – including group
distributors – are in fact themselves the simpler parties in respect of the transactions
involved.
Full Risk Manufacturer
Where a functional analysis shows a manufacturer assuming significant levels of

functions and risks in respect of its processes, similar to or greater than a licensed
distributor, then it should be considered to be ‘full risk’. This will include
management decisions on the use of production capacity and related risks such
as procurement and warranty issues, together with pricing of its output.
As a result, profit levels are expected to be more volatile as aspects such as

capacity and input raw materials cost are taken into account. This will need to be
reflected through the transfer pricing method either for the overall activities of the
manufacturer (if a profit-based method is used) or for component transactions
with other group entities. If the latter is used, consideration should again be given
to whether the full risk manufacturer remains the simpler party and the appropriate
focus of testing. Key questions for the functional analysis may include:
• Who owns design intellectual property and/or how is this accessed?
• Who takes volume/capacity risk?
• Who takes inventory risk for raw materials and finished goods?
• To what extent does the manufacturer carry product warranty risk?
• Who takes procurement risk – ie securing appropriate raw materials at the right
price?
8.5 Support Service Activities
Groups often choose to centralise support services to avoid the cost of

duplication. These often comprise ‘back office’ functions such as human
resources, IT, finance and legal services. The functional analysis will show where
these are performed and their level of associated risk and business value –
frequently these activities are necessary but not a business differentiator. This gives
them the characteristics of a routine service more appropriately rewarded by a
return on their cost than an indicator based on, say, sales value.
These functions may be performed centrally with personnel employed by an

otherwise more complex entity within a group. However the services may

themselves be routine and appropriately charged on this basis, for example at

cost plus.
An effective functional analysis will also provide supporting evidence for the
benefit these activities provided to recipients in the group; this is something the
recipients’ local tax authorities increasingly seek in tax audits.
Specialist Contract Services
Third parties often outsource the performance of services to centres with specific
skill sets while retaining overall direction and control in-house. These could include
engineers, chemists or software developers. Where this fact pattern exists, the
service provision – for example contract research and development – may be
rewarded on a routine basis, usually at cost plus.
For this to be appropriate, the functional analysis would need to show that the
complex entity or entrepreneur is making the key decisions on the scoping and
management of the projects, and that all key risks are borne by the entrepreneur.
It is also important to demonstrate that the activity concerned could potentially
be contracted out to a third party. Where the research and development activity
performed is cutting edge, it may be difficult to find a third party contract
research and development company able to do the work, or in practice much of
the development may be devolved to local specialists. Under those
circumstances, treatment of the activity as a routine service may not be
appropriate. As such, the research and development company may be deemed
to be a more complex entity from a transfer pricing perspective, and share in the
entrepreneurial rewards for its research and development effort.
8.6 More Complex or Entrepreneurial Entities
The functions, assets and risks not found in the simpler entity will sit elsewhere in a
group. In summary, these will be the value added, differentiating elements of a
business such as intangible assets, strategic management, and research and
development. They may also extend to other functions, depending on the industry.
As discussed above, the more complex entity will not be the tested party as its
attributes may well be unique and so not comparable in practice. Typically it will
receive the balance of profit or loss on a transaction or share in the group’s system
profit. This reflects its greater profit potential which comes from its greater share of
business risk. In some cases where business risk has been deliberately centralised,
this may be particularly noticeable.
The functional analysis may not provide a clear cut answer, however. Where a
group's functions are dispersed, such as between a manufacturer, distributor and
an intellectual property company, there may be no obviously ‘more complex’
entity at first glance. Here, the functional analysis and evidence such as legal
agreements should show where key business risks fall and the extent of
management in each entity (ie where the significant people in the business may
be found). Conversely, in groups which have grown by acquisition, more than one
complex entity may exist. Once identified, careful consideration must be given to
how these are rewarded when the economic analysis is performed.

8.7 Planning Aspects of Entity Characterisation
From a planning perspective, group companies could have a financial motivation

to locate entrepreneurial activities in low tax jurisdictions and routine activities in
higher tax jurisdictions. There is therefore a risk that groups could let their preferred
choice of entity characterisation bias the way in which the functional analysis
review is undertaken and documented, whereas the functional analysis should
always come first and be used as the tool to drive entity characterisation.
Tax authorities are aware of the potential motivation for groups to operate through
limited risk entities in higher tax locations. Where an entity is purported to have
limited risk, a tax authority may raise detailed questions about the group functional
analysis or re-perform their own functional analysis during an investigation by
written correspondence, through meetings with management or reviewing
primary documentation such as board minutes, emails or mobile phone records.
Where a tax authority is able to show that the entity characterisation that has
been chosen is inconsistent with the functional analysis and the value chain of the
group, they may seek to recharacterise the entity or not recognise one or more of
the transactions, which could lead to a transfer pricing adjustment arising. This is
one of the major causes of transfer pricing adjustments in the current environment,
which underlines the importance of a robust functional analysis review.
Both the OECD and tax authorities are focusing more closely on key personnel
who are capable of making key decisions, and where they are located. Whilst an
entity could have its costs reimbursed and indemnified against any economic and
financial risk it still may not be successfully supported as limited risk based on the
functional analysis. A critical issue will be whether the entrepreneurial activity is
capable of directing the activities of the limited risk entity.

 Illustration 1
In order to bring all these concepts together, the following is a practical illustration:
Consider the case of the latest smart phone purchased by a customer.
The three key differentiators of a smart phone to customers typically will be the
brand, the particular operating system (its user interface and availability of apps
etc.) and the hardware (its features and look). This market is highly competitive
and requires a significant investment in research and development and branding
in order to be successful. There are many different ways in which a mobile phone
manufacturer could structure itself. The diagram below shows one potential
approach illustrating the concepts above.
Brand & IP
management company
(complex entity)
Wholesales to Software/hardware/brand Central services

retail stores development company
In this example, there is a complex entity directing the group (in some groups this
may be located in a lower tax territory). The role of this entity is to own and
manage the group’s brand and intellectual property. In order to be able to
operate on a global basis it needs to make use of professionals to assist in
designing the brand message, the software and hardware. These professionals will
typically be located in certain areas in the world such as Silicon Valley, California.
This will be typically achieved by limited risk development entities.
Many groups will outsource their manufacturing operations to third party contract
manufacturers in China, but equally they could use their own contract or toll
manufacturers. Typically the group would also centralise back office functions to
avoid duplication.
In order to facilitate a global distribution network, groups will need a presence in

the major territories around the world where they expect to do business. They will
normally have an entity whose role is to wholesale the smart phones to retail stores
and to market the product in the local marketplace. This could be achieved
through a limited risk structure.
The above is a simple example. When implemented properly, it could result in a

tax efficient supply chain with ‘supernormal’ profits associated with the group's
intellectual property accruing offshore. However, a group’s lack of robustness in
the implementation process could undermine the effectiveness of the structure.
Typically this will be due to commercial and personnel issues.
For example, it could be difficult to operate a brand management company

offshore as brand people from a commercial perspective may wish to work in
places such as London, California, etc where there is a pool of talent and key
advertising agencies to develop campaigns. There are also personnel issues in that

many individuals for personal reasons such as family, education and quality of life
would prefer not to work in an offshore location.
A detailed and regular functional analysis needs to be carried out to identify the
location of key personnel within the group and the decisions they make and assess
whether or not this is supportive of the overall structure from a transfer pricing
perspective. This personnel issue is one of the key practical issues that groups will
face and ultimately is the most important factor in a practical functional analysis
review.

CHAPTER 9
COMPARABILITY ANALYSIS: OECD PROPOSED PROCESS
In this chapter we look at the OECD guidance on how to perform a comparability analysis
including:
– the process laid down in the OECD TPG
– external comparables and sources of information
– selection of comparables
– comparability adjustments
– the arm’s length range
– timing issues
– compliance issues
9.1 Introduction
In this chapter we will look at comparability analysis (mentioned in an earlier

chapter) in detail. This chapter follows closely Chapter III of the OECD TPG. You
may find it useful to highlight key parts of your copy of the TPG as you work
through the chapter.
9.2 Performing a Comparability Analysis
The OECD’s suggested process for the comparability analysis is discussed in

Chapter III of its TPG. It is important to distinguish between the search for
comparables and the comparability analysis itself, the two should neither be
confused nor separated.
The search for information on potentially comparable uncontrolled transactions

and the process of identifying suitable comparables are dependent on prior
analysis of the taxpayer’s controlled transaction and of the relevant comparability
factors, which as we have seen earlier are identified in D.1.2 of the OECD TPG as:
• Contractual terms;
• Functional analysis;
• Characteristics of property or services;
• Economic circumstances; and
• Business strategies.
9.3 Typical Process
The OECD TPG set out in Chapter III Section A.1 (see paragraph 3.4 et al) a typical
process for the comparability analysis in nine steps. They indicate that this process
is accepted as good practice but is not compulsory; other methods that lead to a
reliable result are equally acceptable. The steps are as follows.
Step 1: Determination of years to be covered.
Step 2: Broad-based analysis of the taxpayer’s circumstances.

Step 3: Review of the controlled transaction and choice of the tested party.
Step 4: Review of existing internal comparables, if any.
Step 5: Identifying and assessing available sources of external comparables.
Step 6: Selecting the most appropriate transfer pricing method and (depending on
the method) determining the relevant financial indicator.
Step 7: Identifying potential comparables. This will involve determining the key
characteristics that need to be met by an uncontrolled transaction in order to be
potentially comparable, based on the analysis at step 3 and the five
comparability factors set out above.
Step 8: Making comparability adjustments as appropriate.
Step 9: Interpreting and using the data collected, in order to arrive at the arm’s
length price.
The OECD TPG note that these steps will not necessarily be applied by a simple
progression from Step 1 to Step 9. In particular, Steps 5 to 7 may need to be
repeated a number of times if the sources of information initially identified at Step
5 do not prove adequate to enable the process to be completed.
Comparability analysis is a popular subject with the examiner as it is a key element

of transfer pricing. You should ensure that you are familiar with the process set
down in this chapter and the discussion in an earlier chapter which looked at the
application of comparability analysis. To score well in the exam you need to be
able to distinguish between when a question requires you to discuss the five
comparability factors as set down in paragraph 1.36 of the OECD TPG and when it
requires you to discuss comparability analysis as set down in Chapter III of the
OECD TPG.
Broad-Based Analysis of the Taxpayer’s Circumstances
Step 2 of the typical process involves an analysis of the industry, competition,

economic and regulatory factors and other elements that affect the taxpayer and
its environment, in a wider context than that of the specific transactions for which
an arm’s length price is sought. The results will aid in the understanding both of the
controlled transactions and of the uncontrolled transactions with which they will
be compared later in the process.
Review of the Controlled Transaction and Choice of the Tested Party
Step 3 of the typical process is a review of the controlled transaction, in order to

identify the factors that are relevant to the choice of: the tested party; the most
appropriate transfer pricing method; the financial indicator that will be tested if a
transactional profit method is chosen; the selection of comparables and the
determination of comparability adjustments if these are required.
Evaluation of a Taxpayer’s Separate and Combined Transactions
Ideally, the arm’s length principle should be applied to individual transactions.

Sometimes, however, transactions are so closely linked that they must be
considered together. The OECD TPG cite the following examples of where
aggregation of separate transactions might be acceptable: (See Chapter III
Section A.3.1 paragraphs 3.9-3.12.)

a. goods or services under long-term contracts;
b. rights to use intangible property;
c. closely linked products where determining a price for each would be

impractical;
d. manufacturing know-how sold together with vital components;
e. goods or services routed through an associated enterprise;
f. supplies under a ‘portfolio’ business strategy that aims to yield an appropriate

return over the portfolio rather than on each product or service, and which
may therefore involve some products or services being supplied at a low level
of profit or even at a loss, in view of the expected ‘aftermarket’ sales. For
example, jet engines may be sold at a low profit margin under contracts that
involve maintenance work and the supply of spare parts over their 25-year life.
Portfolio approaches must be reasonably targeted; they cannot justify applying

the chosen transfer pricing method at a company-wide level to transactions that
have differing economic logic. Neither can such an approach justify a situation
where low profits in one company of an MNE group are balanced by high profits in
another.
The converse situation may also arise, where a contract gives a single price for a
package of supplies, but it is not feasible to apply transfer pricing methodology to
the package as a whole. In this case the elements of the package will initially
need to be considered separately, but it will be appropriate to consider whether
any adjustment is needed when they are ‘rebundled’, in order to arrive at an
arm’s length result. For example, a computer may be sold with included items of
software, each of which has an easily identifiable arm’s length price. However, in
considering the price of the package as a whole it may be appropriate to adjust
for the fact that the supply of anti-virus software below cost will be a sensible
commercial choice if this will encourage the customer to renew the subscription
automatically at the end of a trial period rather than seeking alternatives.
Where it is appropriate for transfer pricing purposes to determine an arm's length

price on a package basis it may still be necessary to make a split of the price for
other tax purposes; for example, where part of the consideration constitutes a
royalty subject to withholding tax.
Intentional Set-offs
Associated enterprises may intentionally incorporate a set-off into the terms of the
transaction under examination, on the basis that the totality of the arrangement
gives an arm's length result. So, for example, two enterprises may each allow the
other to use their intellectual property on the basis that (with a balancing payment
where necessary) this leaves neither side worse off. Such arrangements may vary
in complexity from a simple case where each side makes supplies of the same
value to the other at an equally favourable price, to a situation where all supplies
of goods and services in either direction over a period are aggregated and a
single payment is made by one party to reflect the perceived net benefit it has
received.
It may be necessary to evaluate the transactions separately to see whether the

arm’s length principle is met. If they are to be considered together care will be
needed in selecting comparable transactions, and the discussion of package
deals above will be relevant.

The terms of set-offs relating to international transactions may not be fully

consistent with those relating to purely domestic transactions because the set-off
may be treated differently under different national tax systems, or under rules in
bilateral tax treaties.
9.4 Choice of the Tested Party
This is a subject that we have looked at in an earlier chapter and you should
review this if required before continuing with your studies.
9.5 Information on the Controlled Transaction
In order to select and apply the most appropriate transfer pricing method,
information is needed in relation to the transaction on the five comparability
factors, and in particular on the functions, assets and risks of all the parties,
including the foreign associated enterprise. While one-sided methods such as cost
plus, resale price or transactional net margin only require a financial indicator or
profit level indicator for the tested party, some information on the comparability
factors of the transaction, and in particular on the functional analysis of the non-
tested party, will still be needed in order to justify the choice of that method.
Where the most appropriate transfer pricing method in a particular case is a

transactional profit split method, detailed financial information will be required on
all the parties to the transaction, and it will be reasonable to expect the domestic
enterprise to provide relevant information as regards the foreign associated
enterprise. Such information will be needed not only to demonstrate that this is the
appropriate method, but also to determine the amount of combined profits to be
split, and what constitutes an appropriate split.
In the case of a one-sided method (for example TNMM), financial information as

regards the tested party will be required in order to apply the method
appropriately (as well as the information needed to justify the choice of method
as above). If the tested party is a foreign entity then this will require information
relating to that entity, but if the tested party is the domestic taxpayer the tax
administration generally has no need to ask for financial data relating to the
foreign associated enterprise.
9.6 Comparable Uncontrolled Transactions
In General
The identification of comparable uncontrolled transactions forms Steps 4 and 5 of

the typical process. They may be transactions between one party to the
controlled transaction and an independent party (‘internal comparables’) or
between two independent enterprises, neither of which is involved in the
controlled transaction (‘external comparables’). Other controlled transactions
within the same or another MNE group are irrelevant to the application of the
arm's length principle and therefore should not be used by a taxpayer to justify its
transfer pricing policy or by a tax administration to justify an adjustment. The
presence of minority shareholders may be a factor leading to controlled
transactions being closer to arm's length, depending on the level of influence of
the minority shareholders, but this is not determinative in and of itself.

Internal Comparables
The review of existing internal comparables is Step 4 of the typical process. Where
such transactions exist information is likely to be more complete and less costly to
assemble, and the financial analysis may be easier and more reliable because it
relies on identical accounting practice. Nevertheless, the comparability factors in
paragraph 1.36 of the OECD TPG must still be met, and comparability adjustments
may still be required as at A.6 of Chapter III. For example, an internal comparable
that relates to the same goods as the controlled transaction but to very different
quantities may not give a reliable comparison unless an appropriate adjustment
can be devised.
9.7 External Comparables and Sources of Information
The search for external comparables is Step 5 of the typical process. It may not be
necessary to embark on this step if reliable internal comparables have been
identified.
Databases
Commercial databases of filed company accounts can sometimes provide a

cost-effective way of identifying external comparables, but they have a number
of limitations:
a. They are not available in all countries.
b. Within a country they may include different types of information, because

disclosure and filing requirements vary for different types of company.
c. They are not primarily compiled for transfer pricing purposes, with the result
that the information may be insufficiently detailed. In particular, it relates to
the results of companies rather than the results of transactions. Accounts are
therefore unlikely to be of any assistance in applying transfer pricing methods
based on comparable prices, as opposed to levels of profit. In addition, in
owner managed businesses (OMBs) margins may be affected by policy
decisions such as whether to reward owners by means of salary or dividend,
and it may be difficult to identify and separate out these factors.
d. Comparable companies are often identified using Standard Industrial

Classification (SIC) codes. However, whether the SIC code is chosen by the
company or the compilers of the database, it may not always be reliable. This
means that companies identified in initial searches may need to be excluded
because they are not comparable (and also, of course, that companies
which would have provided valid comparables may fail to be identified).
e. Companies identified for comparison may have significant transactions with

related parties, without this being apparent from their accounts.
f. The quality of the accounts may vary considerably, both between jurisdictions
and within any one jurisdiction. However comparable a company's activities, it
may not always be possible to extract reliable information to assist with transfer
pricing calculations.
The results of database searches may therefore need to be refined by reference

to other publicly available information. Some advisory firms maintain their own
databases, but these may be based on a more limited portion of the market than
commercial databases. In the next chapter we will see how databases can be
used in practice.

Foreign Source or Non-domestic Comparables
Non-domestic comparables are not automatically to be rejected, but their

reliability will need to be assessed on a case-by-case basis and by reference to the
normal five comparability factors in paragraph 1.36. It may be appropriate to
make one regional search for comparables for several subsidiaries of an MNE
group operating in different countries of that region, depending on the
circumstances, but it will be necessary to take account of market differences and
different accounting standards.
Information Undisclosed to Taxpayers
While tax administrations may have relevant information available to them from
dealing with the affairs of other taxpayers (‘secret comparables’) it would not be
appropriate for them to use this unless, unusually, they were permitted by their
domestic confidentiality requirements to disclose it to the taxpayer involved in the
controlled transaction.
Use of Non-transactional Third Party Data
Third party data relating to results at company or segment level may sometimes
provide reliable comparables for controlled transactions, where those results
represent the aggregate of a number of similar transactions. Where the results are
for a segment, this may raise issues as to the way in which expenses have been
allocated.
Limitations in Available Comparables
As a practical matter, it may not be possible to identify uncontrolled transactions

that are exactly comparable to the controlled transaction. It may therefore be
necessary to select comparables where the business strategy, business model or
economic circumstances are somewhat different; or where the transactions are in
the same industry but a different geographical market; or in the same
geographical market but a different industry.
In some circumstances it may be appropriate to apply a transactional profit split

method without comparables, where the absence of comparable data arises
because each party contributes valuable and unique intangibles to the
transaction. However, even where the comparables are scarce and imperfect this
does not alter the fact that the transfer pricing method selected should be
consistent with the functional analysis of the parties to the controlled transaction.
9.8 Selecting or Rejecting Potential Comparables
Identifying potential comparables, which is one of the most critical aspects of the
comparability analysis, is Step 7 of the typical process. (Step 6, which is the
selection of the most appropriate transfer pricing method, is dealt with in Chapter
II of the OECD TPG (and earlier in this manual)). There are two methods, which in
practice may be operated in combination.
The ‘additive’ approach involves drawing up a list of third parties who are
believed to carry out potentially comparable transactions, and then collecting
information on those transactions to see whether they provide acceptable
comparables on the basis of pre-determined comparability criteria. This may be
used as the sole approach where the taxpayer has knowledge of a few
appropriate third parties. It may also be combined with the use of internal
comparables.

The ‘deductive’ approach starts with a wide set of companies (typically obtained
from a database search) that operate in the same sector of activity, perform
similar broad functions and do not have obviously different economic
characteristics. This set is then refined using selection criteria and publicly available
information, informed by the guidance on identifying and assessing comparables.
The choice of selection criteria has a major influence on the outcome of the
comparability analysis and should reflect the most significant economic
characteristics of the transactions compared.
Examples of qualitative criteria are product portfolios and business strategies, and
tests to exclude special situations such as start-up companies or insolvent
companies. The most common quantitative criteria are:
a. figures for sales, assets or number of employees, and the size of the transaction
either in absolute terms or in proportion to the activities of the parties;
b. criteria related to intangibles, such as the ratio of the net value of intangibles
to the total value of net assets, or the ratio of R&D to sales, as compared to
the figures for the tested party. These criteria might, for example, exclude from
the potential comparables companies that had significant intangibles or R&D
expenditure, where those were not features of the tested party’s business;
c. the ratio of export sales to total sales; and
d. the absolute or relative value of inventories.
The deductive approach has the advantage of being more reproducible and
transparent than the additive approach, and easier to verify. However, its
outcome depends on the quality of the search tools on which it relies, which may
be a practical limitation in some countries. In practice in most countries there are
likely to be significant difficulties arising both from the fact that accounts often do
not disclose the necessary information and that, even where they do, it may not
be accurately reflected in the databases.
9.9 Comparability Adjustments
Because the comparables identified are unlikely to match the controlled

transaction exactly, adjustments may be appropriate, depending upon the costs
and compliance burden that this would involve. This is Step 8 in the typical process.
Different Types of Comparability Adjustments
Adjustments may be made to eliminate differences in accounting treatment

between the controlled and uncontrolled transactions, to exclude significant non-
comparable transactions included within composite data for the uncontrolled
transactions, and to take account of differences between companies in capital,
functions, assets and risks. In practice these factors may also affect decisions as to
where to place a company within the range of arm’s length results as discussed
below. For example, a transaction that exposes a company to higher than
average degree of risk may justify placing it in that part of the range which will
produce a higher profit.
Sometimes a working capital adjustment will be appropriate to reflect the fact

that where a company carries high levels of debtors and inventory the cost of
doing so may (theoretically at least) be reflected in the price it charges; similarly,
the benefit of a high level of creditors may be reflected in a reduced price.

Purpose of Comparability Adjustments
Comparability adjustments are only appropriate where they increase the reliability
of the results, bearing in mind the materiality of the difference for which the
adjustment is intended to compensate, the quality of the data subject to
adjustment, and the purpose and methodology of the adjustment itself. There is no
point in adjusting for small matters if there are major issues where adjustment is not
practical; nor in multiplying adjustments to give a spurious impression of accuracy
that is not justified by the underlying data. The need for numerous adjustments
may, indeed, indicate that the transactions in question are not in fact sufficiently
similar to provide valid comparables.
Reliability of the Adjustment Performed
The OECD TPG warn that ‘it is not appropriate to view some comparability
adjustments, such as those for working capital, as “routine” and uncontroversial,
and to view certain other adjustments, such as those for country risk, as more
subjective and therefore subject to additional requirements of proof and
reliability’. It is not entirely obvious what this means. Clearly, as the paragraph in
question goes on to say, ‘the only adjustments that should be made are those that
are expected to improve comparability’. Nevertheless, it must remain the case
that some adjustments will be more subjective than others, and tax authorities can
hardly be blamed for paying particular attention to these.
Documenting and Testing Comparability Adjustments
Taxpayers will need to be in a position to explain how adjustments were

calculated and why they were considered appropriate, and to supply
appropriate documentation as discussed in Chapter V of the OECD TPG (and later
in this manual).
9.10 Arm’s Length Range
Step 9 of the typical process involves interpreting and using the data collected, in
order to arrive at the arm’s length price.
In General
In some cases the process may arrive at a single arm’s length figure (whether that
be a price or a margin). In other cases a range of equally reliable figures may
result, either because of approximations employed in the process or because
independent parties would indeed vary in the price they charged for the goods or
services in question. It may then be possible to narrow the range by excluding
some of the uncontrolled transactions that have a lesser degree of comparability.
In practice single figures are perhaps more likely to result from identifying a
comparable uncontrolled price or from using other traditional transaction
methods, where specific comparable data exists; ranges are more likely to result
when dealing with more complex transactions that require the use of transactional
profit methods.
That may leave a range of figures based on comparables that (given the inherent
limitations of the process) are considered to include defects that either cannot be
identified or cannot be quantified. In this case, if the range includes a sufficiently
large number of observations it may be appropriate to apply statistical tools that
narrow the range around the centre; for example, by taking the interquartile
range (a practical illustration of this is given in a later chapter). As indicated above
in the context of comparability adjustments, factors such as different degrees of

risk may also be used to justify placing of the particular company in question at the
higher or lower end of the range.
A range of figures may also result where different methods are used to evaluate a
controlled transaction. In such cases it may be appropriate to locate the arm’s
length price where the ranges overlap, or to reconsider the accuracy of the
methods if there is no overlap; all depends on the reliability of the different
methods and the quality of the information they use.
Selecting the Most Appropriate Point in the Range
If the relevant condition of the controlled transaction (eg the price or margin) is
within the arm’s length range determined as above no adjustment should be
made. If it falls outside the range that the tax administration contends is
appropriate, the taxpayer should have the opportunity to present the case for a
different arm’s length range. If that cannot be done successfully, it remains for the
tax authorities to determine the point in the range which it will treat as the arm’s
length figure.
Where the range is made up of results of equal and high reliability, a case can be
made for any point within it. Where there are remaining comparability defects
within the results as discussed above, it may be appropriate to use measures of
central tendency such as the median, the mean, or weighted averages.
Extreme Results: Comparability Considerations
Extreme results in one of the potential comparables may indicate a defect in

comparability, or exceptional conditions that only apply to an otherwise
comparable third party. While extreme results may be excluded on the ground
that they bring to light previously overlooked defects in comparability, they should
not be excluded merely because they are extreme.
In general, all relevant information should be used. There is no general principle

that loss-making comparables should either be included or excluded, because it is
the circumstances of a company taken as a whole that determine whether it is
comparable, rather than its financial result. The existence of a loss will, however,
normally trigger further investigation of whether the comparable is valid. It should
be excluded where the loss does not reflect normal business conditions, or where it
reflects a level of risk that does not exist in the controlled transaction. Similar factors
apply as regards abnormally large profits.
It will also be helpful to take into account the different ways in which extreme
results may affect different statistical measures. For example, in general they will
have more effect on the arithmetic mean than on the median. Where those
measures diverge significantly it may be helpful to identify the results that account
for this difference and consider how they should be dealt with. Where there are
extreme results it will always be necessary to consider the features of specific
transactions rather than simply applying mathematical formulae to narrow the
range of results.
9.11 Timing Issues in Comparability
Various issues arise as regards the time of origin, collection and production of
information on comparability factors and on uncontrolled transactions for use in
the comparability analysis.

Timing of Origin
Ideally, comparisons would be made with contemporaneous uncontrolled

transactions, which reflect the same economic circumstances. In practice,
however, that may not always be practical.
Databases of company accounts will always be out of date by a minimum period

of about one year, once allowance is made for the time taken to produce
accounts and then to incorporate them into the database. Sometimes it will be
helpful to consider a number of years together in order to detect any underlying
trends.
Timing of Collection
Some taxpayers will establish transfer pricing documentation at the time they
undertake their controlled transactions, based on the information available at that
time (which necessarily relates to past transactions) together with information on
subsequent market and economic changes (‘arm’s length price setting’).
However, independent parties in similar circumstances would not base their
pricing decisions on historic data alone.
Other taxpayers will test their controlled transactions after the event, typically in
connection with the preparation of their tax return for the period concerned
(‘arm’s length outcome testing’), or the two approaches may be combined. This
raises issues concerning the use of hindsight. It is legitimate to use external data
such as other companies’ accounts as evidence of what pricing decisions those
companies reached at the same time and in comparable circumstances. It is not,
however, appropriate to suggest that the taxpayer whose affairs are in issue
should have been aware of those decisions, and should have taken them into
consideration in its own pricing decisions, before the accounts or other data were
publicly available.
The purpose for which the transfer price is set may affect the timing of information
gathering. We will look at this and further issues on timing of collection in the next
chapter.
Valuation Highly Uncertain at the Outset and Unpredictable Events
Where valuation uncertainties existed at the time of the controlled transaction, it is

necessary to ask whether the uncertainty was so great that independent parties
would have incorporated a price adjustment mechanism in their agreement.
Similarly, where unpredictable events affecting the value occurred after the time
of the controlled transaction it is necessary to consider whether these were so
fundamental to the value that independent parties would have renegotiated the
transaction. In such cases an arm’s length price should be determined on the basis
of the agreed price-adjustment mechanism or a hypothetical renegotiation, as
the case may be. In other cases it will not be appropriate to make calculations
using the benefit of hindsight that would not have been available to independent
parties.
Data from Years Following the Year of the Transaction
While care must be taken to avoid using hindsight, data from years after the year
of the transaction may be relevant in certain circumstances; for example, in
comparing product life cycles of controlled and uncontrolled transactions in order
to determine whether the uncontrolled transaction is an appropriate comparable.
Subsequent conduct of parties to a controlled transaction may also be relevant in
determining the actual terms and conditions operating between them.

Multiple Year Data
The use of multiple year data may be useful in certain circumstances, but it is not
required as a matter of principle. Information from prior years may disclose facts
that have influenced, or should have influenced, the price in the current year; for
example, it may help to establish whether a loss in the current year is part of a
history of losses, the result of exceptional costs in the previous year, or the result of
a product nearing the end of its life cycle. This may be particularly useful where a
transactional profit method is applied.
Multiple year data may also be useful in providing information about the business
and product life cycles of the comparables, and perhaps identifying significant
variances from the comparability characteristics of the controlled transaction
which make the use of those particular comparables inappropriate. Similarly, if
economic conditions in an earlier year, which did not apply to the controlled
transaction, affected the comparable enterprise's pricing or profit for the current
year, it may not provide a reliable comparison.
The use of multiple year data does not necessarily imply the use of averages, but
this may be appropriate in some circumstances.
In practice the use of multiple year data may be more useful in stable industries,
where it can help to establish trends. Where there is a high level of volatility in the
market or in general economic conditions, the prices or profit of one year will be
much less useful in giving any indication of what the terms of trade are likely to
have been in the very different circumstances of another year. In particular,
figures derived by averaging widely varying results of different years are likely to be
unhelpful or positively misleading.
9.12 Compliance Issues
Comparability analysis may impose a significant cost and compliance burden on

taxpayers. It is not necessary to conduct an exhaustive search for all possible
relevant sources of information, but simply to exercise judgement to determine
whether particular comparables are reliable.
It is good practice for taxpayers to set up processes to establish, monitor and

review their transfer prices, taking into account the size of the transactions, their
complexity, the level of risk involved and whether they take place in a stable or a
changing environment. Where transactions are small, simple, relatively risk-free
and conducted in stable circumstances it will be reasonable to devote less effort
to finding information on comparables, and perhaps not to perform a complete
comparability analysis every year. Practical issues with review procedures will be
looked at in a later chapter.

CHAPTER 10
COMPARABILITY ANALYSIS: AGGREGATION AND USE OF THIRD

PARTY NON-TRANSACTIONAL DATA
In this chapter we will look at the practical aspects of implementing Comparability

Analyses including:
– aggregation and unbundling
– set-offs
– segmentation of comparable data
– sources of information and timing issues
– sources of third party non-transactional data
– commercial databases
– comparability lessons from DSG Retail
– proprietary databases and ‘secret comparables’
– using databases
– other sources of information
– timing of information on comparable transactions
10.1 Introduction
In this and the next chapter we will examine the practical aspects of implementing
the OECD TPG on comparability. Some reference will be made to UK practice
together with examples of differing positions adopted by other jurisdictions to
highlight some of the compliance difficulties thereby caused. We will mention
again some of the issues set down in the previous chapter where we examined the
comparability process.
Comparability requires a comparison of the economically relevant conditions in a

controlled transaction with the conditions in an uncontrolled transaction.
It is useful at this point to remind ourselves of a section of the OECD TPG which we
have looked at previously. Section D (paragraph 1.33) states that there are two
steps to a comparability analysis:


10.2 Transactions: Aggregation and Unbundling
Article 9(1) of the OECD Model DTC permits the profits of an enterprise derived
from a controlled transaction to be adjusted for tax purposes to the level that
would have accrued had the two enterprises been independent and dealing at
arm's length.

In fact, the Article does not specifically refer to ‘transactions’ but the OECD TPG
assert that ‘ideally’ the arm's length principle should be applied on a transaction-
by-transaction basis.
The CUP method can only be applied by using transactional data, whereas all
other pricing methods rely on non-transactional data.
Examples of transactional and non-transactional data
Transactional data Non-transactional data

Publicly available price information Company-wide profitability data
(commodities exchanges) derived from published financial
statements
Royalty rate for patent licence derived Segmented profitability data derived
from US SEC filings 10K from published financial statements
Interest rate and covenants for a loan Trade association data
agreement derived from commercial
databases
Price lists
As we saw in the last chapter the OECD recognises that, in practice, it may be
unrealistic to evaluate each transaction separately.
In any comparability analysis it is essential to identify all controlled transactions,

whether or not specifically priced or documented, and then assess whether
individual transactions should more properly be evaluated separately or together.
Examples of some less obvious transactions might include:
• Secondment of staff to an affiliate for which no charge is made
• Ownership of a brand name exploited by an affiliate for no charge
• Capital contributions which may amount to hidden marketing support
• Provision of employee share options to or by an affiliate
• Cash pooling
• Debt factoring
A example of inappropriate aggregation of controlled and uncontrolled

transactions can be found in the United States case of Microsoft Corp. v. Office of
Tax and Revenue (D.C. Office of Admin Hearings, Case No. 2010-OTR-00012 (May
2012)) in which the IRS contract auditor applied a CPM analysis comparing profit-
to-cost ratio of Microsoft with the profit-to-cost ratio of businesses chosen as
comparables. However, the auditor aggregated both controlled and uncontrolled
transactions of Microsoft. The judge found that there was no justification for such
aggregation which rendered the analysis ‘arbitrary, capricious and unreasonable.’
Exceptionally, the OECD TPG suggest that ‘unbundling’ of aggregated

transactions priced as a package is appropriate if the separate transactions are
quite different in character, for example the licence of a patent and the rental of
office facilities. (See paragraph 3.11)
Let us consider the Canadian Case of GSK Canada [2010 FCA 2017] also known as
The Queen v Glaxo Smith Kline. Central to this case is a failed attempt by the
Canadian tax authority to segregate two inextricably linked transactions: the

purchase of an active pharmaceutical ingredient and the licence to manufacture

and sell the drug.
The appellant, Glaxo Canada, was a member of a UK-parented pharmaceutical

group. Glaxo Canada manufactured and marketed Zantac, a branded drug
developed and patented by the UK parent. Glaxo Canada bought the active
ingredient from a Swiss affiliate. However, unrelated pharmaceutical companies
sold generic versions of the drug in Canada but paid substantially less for the
active ingredient than Glaxo Canada did.
The tax authority disallowed a deduction to Glaxo Canada for the price
differential between the prices paid by Glaxo Canada and the generic
manufacturers, asserting that the latter was a CUP. This was upheld by the Tax
Court. The Federal Court of Appeal rejected this, holding that lower court should
not have disregarded the License Agreement which gave the Canadian
distributor access to Glaxo’s trademark which gave the company access to the
premium prices paid for the product over its generic competitors. The lower court
had failed to consider the business reality of the situation: although an arm’s
length purchaser could always buy the active ingredient at market prices from a
willing seller, the question is whether that arm’s length purchaser would be able to
sell this under the valuable trademark.
The Supreme Court of Canada (SCC) upheld Glaxo Canada’s appeal that the
licence agreement must be taken into account in examining the purchase price
for ranitidine.
The GSK case is a key case and some important conclusions can be drawn from it.
As part of its decision the SCC stated that an arm’s length price can fall within an
acceptable range of prices which has also been seen as significant, and
consistent with the OECD TPG. The SCC’s guidance suggests that Canadian courts
must keep an eye on the bigger picture in making their transfer pricing
determinations, and the determination of arm’s length pricing is quite distinct from
the narrower concept of fair market value. Following this case changes were
made to the detail of Canadian legislation, eg removing ‘in the circumstances’
from the reasonableness test.
This case shows how difficult and controversial the application of transfer pricing
principles can be. The taxpayer should carefully consider all the implications when
making decisions on contractual arrangements for intra-group purposes when the
transactions are particularly complex.
We will look at this case again in later chapters and you will find more detail in the
case law summary
10.3 Set-offs
A ‘set-off’ occurs if affiliate A provides goods or services to affiliate B and affiliate B

provides different goods or services to affiliate A. The OECD TPG address
intentional set-offs, where the taxpayer deliberately evaluates the overall
economic effect of both transactions. The OECD TPG accept that intentional set-
offs may be found at arm’s length and hence may be acceptable for similar flows,
although not for overall balancing of different transaction types. (See paragraphs
3.13-3.17.)
As part of the comparability analysis, it is clearly desirable for taxpayers making

intentional set-offs to gather evidence that similar transactions are found at arm’s

length. There is a good deal of material available regarding ‘package pricing’ in

some industries eg web hosting/development, broadband/telephony.
 Illustration 1
In this illustration we will look at the approach of the UK tax authorities to

aggregation. The Berg group of companies manufactures and distributes memory
chips for use in smartphones and tablets. The chips are manufactured by Berg
(Polska) sp. z.o.o in Poland and sold to Berg plc in the UK. Berg plc distributes the
chips to third party business customers in the UK and Europe. Berg S.A provides
consultancy services to computer manufacturers in South America.
BERG PLC
UK PARENT
BERG S.A. BERG (Polska) sp. z o.o

Brazilian subsidiary Polish subsidiary
During the year ended 30 June XXXX, Berg plc purchased 53 different chip
configurations from its Polish subsidiary, with volumes in the thousands of units for
each configuration. The 53 different chip configurations have similar attributes,
cost of production, cost of marketing and sale price per unit of memory.
Berg plc also granted a loan at a fixed interest rate of 18% to its Brazilian subsidiary
to support the development of the latter's consultancy business.
Looking at the UK approach to aggregation HMRC are likely to accept that all of
the purchases by Berg plc from its Polish subsidiary may be aggregated. On
enquiry, HMRC produces evidence the price paid in aggregate by Berg plc is
excessive. Berg plc claims that the interest received from its Brazilian subsidiary is
greater than an arm's length rate; with the effect that overall its return from
affiliated transactions approximates an arm’s length amount. Berg plc is unable to
demonstrate a linkage between the purchase of inventory and the making of a
loan, and its claim to set off the two transaction types is unlikely to succeed.
10.4 Segmentation of Comparable Data
The transactional nature of transfer pricing requires that if a taxpayer carries on

diverse activities which cannot legitimately be aggregated, the separate
transactions need to be evaluated separately. For profit methods, this usually
requires ‘segmentation’ of the taxpayer’s financial data. The structure of the
taxpayer’s accounting system will have a bearing on whether this can be
achieved, but even if segmentation is superficially possible, great care will need to
be taken to ensure that expense, revenue, asset and liability allocations between
segments are reliable.

Obtaining segmented financial data for comparable companies carrying on

diversified functions is even more challenging. Some companies may be required
to disclose segmented financial information – the applicable standards include
FRS 102 in the UK, SFAS 131 in the United States and IFRS 8 for companies reporting
under International Accounting Standards. However, the following limitations need
to be borne in mind:
• IFRS 8 and SFAS 31 only require disclosure for listed companies which may be
prima facie unsuitable potential comparables for small and medium size
taxpayers;
• IFRS 8 and SFAS 31 disclosures include earnings before interest, tax

depreciation and amortisation (EBITDA) and total assets per reported segment
as reported internally to management; such internal reporting need not be
consistent with the basis of consolidated financial reporting.
10.5 Sources of Information and Timing Issues
The OECD TPG distinguish between internal and external comparables. You will
recall that an internal comparable is a comparable transaction between one
party to the controlled transaction and a third party; an external comparable is a
comparable transaction between two independent enterprises.
 Illustration 2
Copland Machines, Inc. is a United States manufacturer of advanced construction

machinery. Its UK subsidiary, Copland Machines (UK) Limited, acts as the exclusive
distributor of the parent company’s products in the UK. In Canada, Copland
Machines, Inc. distributes its product via an unrelated distributor, Delius Inc. An
unrelated UK company, Hindemith plc, is also identified which distributes similar
machinery manufactured by two other companies, both unaffiliated with
Copland, to third party customers in UK and Europe.
Whether the price paid for the sale of inventory from Copland Machines Inc. to
Delius Inc. passes muster as an internal CUP depends on an analysis of the five
comparability factors discussed in earlier chapters and, if appropriate, whether
sufficiently reliable comparability adjustments may be made (discussed in more
detail in the next chapter).
It is very unlikely that Copland will have access to detailed price data in relation to
the purchase of inventory by Hindemith and, consequently, Hindemith is unlikely to

be suitable for evaluation as a potential external comparable company at an

aggregated level.
Again using the UK as an example, HMRC guidance expresses the importance of

evaluating potential internal comparables. It cautions against the mechanical
dismissal of internal comparables merely because they are not identical to the
controlled transaction. Indeed, HMRC take the position that a taxpayer who
adopts a TNMM analysis but ignores a very clear internal comparable without
justification may be exposed to ‘deliberate inaccuracy’ penalties. Many other tax
authorities also express a preference for internal comparables. To some extent, this
may still be supported by some of the wording from the 1995 TPG that has been
retained in the 2017 TPG. (See, for example, paragraph 2.64 in the context of
TNMM.) However, it is submitted that internal comparables are not automatically
elevated above external comparables.
10.6 Sources of Third Party Non-transactional Data
In the event that sufficiently reliable transactional data cannot be identified, the
practitioner and tax official will need to look to other sources of data and
information. Such sources include:
• Commercial databases
• Proprietary databases
• Industry and trade association publications
• Articles from industry press and specialist financial newspapers/journals
• Court documents (for example, anti-trust cases, ie cases where parties are
accused of limiting free competition)
• Investment research
10.7 Commercial Databases
Commercial databases, permitting search, filtering and analysis of company

information reported to national company registries and similar institutions, play a
major part of transfer pricing practice. The OECD TPG note the practical benefits
that such databases may bring, but urge caution against potential misuse. In
particular, databases should not be the default option if reliable information is
available elsewhere, regard must be had to the extent and quality of the source
data, and the emphasis must be on quality over quantity.

Leading commercial databases include:
Name Type Coverage

Amadeus (Bureau van Dijk) Company financial Europe
statements
Orbis (Bureau van Dijk) Company financial Global
statements
Oriana (Bureau van Dijk) Company financial Asia-Pacific
statements
Thomson Reuters Company financial Global
Fundamentals statements
Thomson Reuters European Company financial Europe
Comparables statements
Compustat (Standard & Company financial North America
Poor's) statements Global
ktMINE Royalty Rate Finder Intangible property SEC Global (with US
filings emphasis)
Thomson Reuters Loan Loan pricing market Global
Connector information
Some of the databases may also be used with a commercially available ‘front
end’ software tool designed to assist the user with efficient evaluation of the vast
amount of information that can be contained in the underlying databases. The
potential user of such databases will wish to make his or her own cost/benefit
assessment of such factors as:
• Number of companies/transactions covered
• Geographical scope: country, regional or global
• Recognition by the national tax authorities concerned
• Scope of the source data and reliability of reporting of that data
• Extent of specialist adviser support

A standard framework for performing a database search, which should of course

be properly informed by a prior industry, economic and functional analysis can be
summarised diagrammatically as follows:
Initial Search
This will typically include filters for:
• Standard Industry Code/principal activities keywords
• Geographical scope
• (Lack of) independence
• Periods covered
• Persistent loss making companies
• Missing financial information
• Dormant companies
• Start-up companies
Standard Industry Codes (SIC) can certainly be a powerful search tool, but caution
is required. It is essential to understand the structure of the particular classification
system used in the database or software front-end.
There are a number of free lookup and conversion tools online. Companies vary
considerably in the diligence with which they disclose SIC codes; for example
many sophisticated businesses return a ‘other business activities’ SIC code. SIC
codes can present a particular challenge when searching for transactional data
on intangible property.
Similarly, creativity and lateral thinking is sometimes required to make the best use
of keyword searches. Companies often do not classify themselves in the same way
as transfer pricing practitioners and a thorough grasp of industry jargon may assist

with a targeted search. As a simple example ‘value-added reseller’ may yield

better results than ‘distributor’ in the computer software industry.
Independence screening should also be approached with care. For example, one
might reject all companies which are subsidiaries of a parent company. Some
writers take the view that it might be preferable to evaluate whether consolidated
financial statements of the parent (if they exist), which should eliminate intra-group
transactions, could in fact be used to assess an arm’s length position.
The initial search may yield very few or very many hits. An iterative process can
then be applied to expand, narrow or vary the initial search criteria.
Bulk and Subsequent Stage Rejections
Second stage quantitative filters are typically applied to further refine the search
strategy. These may typically include number of employees, turnover or assets, or
financial ratios. Maxima, minima or a range can be set to mirror the characteristics
of the tested party.
Among the financial ratios applied are:
Name Formula Use

R&D Intensity R&D expenditure/net sales A measure of relative
importance of R&D
Days sales of (Inventory/cost of sales) × A measure of how long it
inventory 365 takes to convert
inventory to sales –
suitable for assessing
distribution activities
Profit per employee Operating profit/number of A measure of whether
employees profit is primarily driven by
productivity or by size of
workforce
Qualitative Evaluation
A detailed evaluation of financial statements, company websites, analysts’ reports

and any other available information relating to the remaining comparables is key
to the assessment of their suitability.
For example, the Companies Act 2006 requires financial statements of companies
(other than small companies) to include a business review with a description of the
main risks facing the company. This can be a useful tool for comparing the risk
profile of the comparable with that of the tested party, although the depth and
quality of the analysis varies widely.
Financial statements of listed companies should also address trends and factors
facing future development, performance and position of the business and
information about environmental matters, employees and social and community
issues.

10.8 Comparability: Lessons from DSG Retail
Decisions handed down by the courts in a number of countries illustrate the

standard for comparability is unlikely to be attained with mechanistic database
searches.
Of prime relevance in the United Kingdom, the DSG Retail case contains important
lessons on the likely approach of the appellate tribunals to comparables. The
decision is described in detail in the Case Law Appendix, but the key points in
relation to comparables were as follows:
• The DSG group comprised the largest UK consumer electronics retailers.

Extended warranties were offered to customers at the point of sale. DSG sales
staff acted as agent for a third party warranty insurer, Cornhill, which reinsured
95% of risk with a DSG subsidiary, DISL, resident in the Isle of Man.
• The Special Commissioners were troubled by the mismatch between the

profitability of DISL and its limited functions and staff as a consequence of
which it had limited bargaining power in contrast to DSG.
• DSG advanced a number of potential CUPs and a TNMM analysis. All were
rejected; a selection is summarised in the following table:
Potential Description Reason for rejection

comparable
offered
Orion 1982 extended warranty Too old: market changes and
arrangement between better claims experience at the
a large electrical retailer time of the controlled
and a third party transactions
National Satellite An insurer acted as an The goods were not comparable:
Services (NSS) agent for NSS in selling in the controlled transaction, a
insurance contracts to large diversified range of goods
customers of NSS who were sold. The contractual terms
purchased or repaired were different
satellite equipment. The
insurance was sold to
customers at the time of
installation or repair
Office of Fair A competition authority The anonymous data could not
Trading (OFT) report containing be tested; commission rates were
Report anonymised data on very fact-specific
extended warranty
commission rates
Domestic & A third party provider of D&G provided insurance to
General (D&G) domestic appliance smaller businesses and therefore
breakdown insurance. A had an infrastructure far in excess
TNMM analysis was of that of DISL
advanced by
comparing the return on
capital achieved by
D&G with that achieved
by DISL

10.9 Proprietary Databases and ‘Secret Comparables’
Some large accountancy and other advisory firms have developed in-house
pricing databases using proprietary data or proprietary data mining techniques.
The OECD TPG (paragraph 3.34) do not proscribe such databases; they urge
caution that the potentially more limited data, compared with commercial
databases, may support the tax administration being granted access to the
private database in the interests of transparency.
The flip side of proprietary data belonging to the taxpayer/adviser is that ‘private’
data is almost certainly possessed by the tax authority. Publicly unavailable
material collected from tax returns and tax audits of other taxpayers potentially
provides a very powerful platform on which taxpayer transfer pricing may be
challenged. The OECD TPG adopt an even-handed approach: such ‘secret
comparables’ are unfair unless disclosed to the taxpayer. (See paragraph 3.36.)
Most OECD member countries also respect the OECD TPG with respect to secret
comparables, although the position for non-member countries is variable. For
example, the authorities in China and Japan have sanctioned the use of such
data.
10.10 Using Databases
Loss-making Companies
National practices vary as to whether loss-making companies are likely to be

acceptable as potential comparables. Chapter I of the OECD TPG at paragraphs
1.129 (1.49) to 1.131 (1.51) recognises the legitimacy of losses in controlled
transactions in start-ups and other circumstances but asserts that, in the long term,
losses would not be sustainable at arm's length.
A balanced approach is required to the question of loss-making comparables. For

example, when performing a multi-year search, it may be appropriate to exclude
companies reporting three consecutive years of loss, but include those reporting
one or two years subject to scrutiny of financial statements and company websites
for other indicators of particular reasons that are not relevant to the tested party.
Local, Regional or Global Comparables?
Comparable company information is plentiful in some countries and almost non-

existent in others. Geographical scope of databases is key: the greater the
granularity of data local to the tested party, the more likely the database is to
provide a good starting point, but equally obtaining country-specific information in
every tested party location, even if that is possible, may be prohibitively expensive.
A common compromise is to use pan-European, pan-American or pan-Asian
regional data sets.
10.11 Other Sources of Information
It is misguided to place exclusive focus on database and software solutions. An

ever-growing quantity of information is freely available online, although great
caution has to be exercised as to the authoritativeness and age of such
information.

10.12 Timing of Information on Comparable Transactions
Transfer pricing studies may be prepared either for the purposes of setting the
prices and other conditions of prospective transactions to secure compliance with
the arm's length principle, or for the purposes of testing compliance after the end
of the accounting period. Typically the price setting process may need to be
carried out several months before the start of the accounting period, whereas the
testing process may not need to be completed until the corporate income tax
return is filed, perhaps a year after the end of the accounting period and more
than two years after a price setting process.
It is clear that different data will be available at different times to inform the
comparability analysis. National tax rules have widely varying requirements
regarding the time at which data must be compiled or made available to the tax
authorities. For example, in Vietnam the timing of documentation requirements has
always been very tight. In 2017 the requirements were updated to take account of
the post BEPS three-tier approach recommended by the OECD (see later
chapter). Companies are required to prepare their documentation before their
tax returns are submitted. This means they have only 90 days from the year end to
prepare their documentation. Many other countries refer to ‘contemporaneous’
documentation but often this amounts to a requirement to create the
documentation by the filing date for the tax return.
The OECD TPG recognise the different timing of data for the setting versus the
testing approach. There is recognition that double taxation may arise in controlled
transactions where two tax authorities take a different view on timing of
comparable data.
The OECD TPG also recognise that data from years following the year of the
transaction may also be relevant to the analysis of transfer prices, but care must
be taken to avoid the use of hindsight. The most notorious example of the use of
hindsight is the 1986 United States ‘commensurate with income’ regulations for
intangible property transfers which require periodic after the fact revisions.
The UN Manual contains much the same wording on timing issues as the OECD
TPG. Paragraph B3.6.2.8 states that:
‘Data from years following the year of the transaction may also sometimes be
relevant to the analysis of transfer prices, but care must be taken to avoid the
use of hindsight, perceiving the significance of facts and events with the
benefit of knowledge accruing after they have occurred.’

CHAPTER 11
COMPARABILITY ADJUSTMENTS INCLUDING PRACTICAL ISSUES
In this chapter we will look at comparability adjustments, in particular, looking at:

– adjustments for accounting items
– capital intensity
– other adjustments
– tax authority responses to comparability adjustments
– the arm’s length range
– compliance issues
– safe harbours
– frequency of review
– treatment in the UN Manual
11.1 Introduction
This chapter considers what ‘reasonably accurate adjustments’ (see paragraph

3.47) might be made to reduce the effect of differences between economically
relevant conditions in a controlled transaction and the conditions in an
uncontrolled transaction.
Comparability adjustments may be made to either the tested party or to the

potential comparable transaction/company. They may be made under any
transfer pricing method and may be particularly useful where there is an internal
comparable which can accurately inform the adjustments that need to be made
to the controlled transaction. They are also seen frequently in profit methods,
perhaps most commonly in the form of working capital adjustments.
The OECD TPG emphasise that comparability adjustments need to be reliable,

objective, transparent and documented. They should not be applied
mechanically or used to create an impression of precision where they are in fact
unwarranted. (See paragraphs 3.50-3.54.)
11.2 Adjustments for Accounting Items
Adjustments may be made to neutralise the effect of accounting items reflecting

differing functions, assets and risks that are present in the potential comparable
but not in the tested party, or vice-versa. Common examples are exceptional
items and non-operating income.

 Illustration 1
Stockhausen Ltd is a UK contract manufacturer, for its German parent, of

communications systems for aeronautical applications. A transfer pricing study
identifies TNMM as the most appropriate pricing method with a profit level
indicator (PLI) of operating margin (Operating profit (being earnings before
interest and tax)/sales). One of the seven potential comparables remaining after
all bulk and second stage rejections is Simpson Ltd, a UK manufacturer of
communications systems for marine applications.
The financial results of both companies for the year ended 31 March XXXX is as
follows:
Stockhausen Ltd Potential

comparable
(Tested party) Simpson Ltd
£’000 £’000
Sales (S) 2,750 3,876
Cost of sales (750) (1,121)
2,000 2,755
Distribution costs (776) (855)
Administration costs (1,059) (1,530)
Exceptional costs _______ (311)
Earnings before interest and tax = 165 59
operating profit
Operating margin 165/2,750 = 6% 59/3,876 = 1.5%
Examination of the financial statements of Simpson Ltd reveals that the

exceptional item of £311k relates to closure costs of a fabrication plant. Company
press releases state that this closure is due to consolidation of manufacturing
operations in one plant. This is a one-off circumstance which is not replicated in
Stockhausen Ltd and may therefore be adjusted for:
Operating profit 59
Add: Exceptional costs 311
Revised profit 370
Revised operating margin 370/3,876 = 9.5%
11.3 Capital Intensity Adjustments
Working Capital
Working capital adjustments are perhaps the most common type of adjustment
seen in practice. These are intended to equalise the tested party with the potential
comparable by adjusting for differences in capital actively employed in the
business. Working capital is defined as:
Trade debtors (receivables) + Stock (inventory) − Trade creditors (payables)
The need for adjustment is founded on the concept that, for two otherwise
comparable businesses, the profit will increase in line with working capital
employed: the company with the higher net capital will have higher sales revenue
because prices reflect the facility of extending credit terms to its customers, it will
have more stock and can benefit from volume discounts.

 Illustration 2
This is a simplified illustration of the type of working capital adjustment featured in

the OECD TPG. (See Annex to Chapter III.) The OECD TPG are at pains to stress that
this form of adjustment is merely illustrative and this type of adjustment is not
binding on taxpayers or tax authorities.
In this illustration, the transfer pricing method is TNMM with a PLI of operating
margin, that is, earnings before interest and tax/sales.
The process outlined is:
1. Compute the working capital as a proportion of sales for both the tested party
and the comparable company;
2. Calculate the difference applied to a notional interest rate to reflect the time
value of money;
3. Adjust the result to reflect the working capital difference – here the adjustment
is made to the comparables results, but conceptually it could also be made to
the tested party or to both parties.
Tested Comparative
party company
£ million £ million
Sales 600 800
Earnings before interest and tax (EBIT) 6 24
EBIT/sales (%) 1.00% 3.00%
Working capital:
Trade debtors (receivables) (D) 40 80
Stock (Inventory) (S) 45 90
Trade creditors (C) 22 27
Working capital: 63 143
Working capital/sales 10.5% 17.8%
Working capital adjustment

Working capital/sales – tested party 10.5%
Working capital/sales – comparable company 17.8%
Difference (D) -7.3%
Interest rate (i) 5%
Adjustment D × i -0.37%
EBIT/sales (%) adjusted 3.00%−0.37% = 2.63%
Fixed Assets
Also referred to as ‘property, plant and equipment’ (PP&E) adjustments, these are
used to equalise the PLI between a tested party and a potential comparable with
different levels of productive assets. Again, an imputed interest rate is used. One
example of this is a start-up situation for the tested party which requires a large
initial investment in fixed assets but the comparables have lower fixed assets. The
validity of such adjustments is potentially controversial.
11.4 Other Adjustments
Less common forms of comparability adjustment include geographical risk and

economic volatility.

11.5 Tax Authority Responses to Comparability Adjustments.
There are widely differing country practices and tax authority responses to the use
of comparability adjustments. In the UK, HMRC guidance is extremely cautious
(See INTM 485110). Their position is that adjustments can quickly become
meaningless, and the more fearsome the algebra, the more suspect they
become. Comparability adjustments are much more embedded in US practice.
11.6 The Arm’s Length Range
The OECD TPG emphasise that a range of ‘relatively’ equally reliable results may
be a legitimate reflection that independent companies engaged in comparable
transactions to the controlled transaction in reality do impose different prices and
conditions to each other. However, every effort must be made to eliminate results
that have a lesser degree of comparability; excessively wide ranges stemming
from large deviations among data points may indicate inadequate comparability
of some data points.
 Illustration 3
Schoenberg Scooters Limited carries out contract research and development in

the United Kingdom for its Austrian parent company. The group commissions a
transfer pricing study which is intended to corroborate the pricing policy adopted
of an operating margin of 5%.
Eleven companies are evaluated as potential comparables, as follows:
Comparable company Operating margin

1 -17.0%
2 -0.5%
3 2.7%
4 3.9%
5 4.1%
6 4.4%
7 8.8%
8 9.0%
9 9.8%
10 13.4%
11 17.3%
On the basis that the median of the above data points is 4.4%, and the mean is
5.08%, it is asserted that the pricing policy is thus supported. However, this is a very
wide range of results which may call into question the validity of some of the
companies used. It is noteworthy that eliminating companies 1 and 2 increases the
median to 8.8%. Any data set should be evaluated for the effect of eliminating or
including potential data points; volatility may call into question the comparables
or, more fundamentally, the pricing method.
Statistical tools
The OECD TPG contain a statement to the effect statistical tools that take account
of ‘measures of central tendency’ may usefully be applied to larger data sets
where every effort has been made to ensure comparability or adjust for non-
comparability. (See paragraph 3.57.) The example of such tools cited is the
interquartile range or other percentiles.

The interquartile range includes the 50% of middle values from a sample and
removes the influence of the top and bottom 25% of values. The interquartile
range has long featured in US regulations and it is very commonly used in other
countries, although it should be noted that even the US regulations stipulate that
the interquartile range is only required where the data points are not sufficiently
equally reliable. (US Treasury Regulation Sec 1.482-1(e)(2)(iii)(B)) In the UK, HMRC
guidance summons little enthusiasm for the use of the interquartile range,
preferring a more qualitative assessment of where the tested party should be
placed in the (entire) arm's length range. A similar stance is taken by the tax
authorities in Canada and New Zealand.
11.7 Compliance Issues
This is a subject that we will look at in more detail in a later chapter.
The OECD TPG acknowledge the need for a risk-based and pragmatic approach,
particularly for SMEs. It is emphasised that there is no need for an exhaustive
search of all possible information sources (see paragraphs 3.2, 3.81). It may not be
necessary to perform a detailed comparability analysis each year for simple
transactions.
Although the acknowledgment of potential compliance burden in the OECD TPG

is welcome, it must be said that there is no specific guidance on practical
approaches, no doubt as reflection of reality that such “soft law” cannot achieve
consistency of approach among all OECD member country tax administrations.
However, the OECD has placed a renewed emphasis on the need for
simplification measures as part of its review of the administrative aspects of transfer
pricing.
11.8 Safe Harbours
The IBFD Online Glossary defines a safe harbour as:
‘An objective standard or measure, such as a range, percentage, or absolute

amount, which can be relied on by a taxpayer as an alternative to a rule
based on more subjective or judgmental factors or uncertain facts and
circumstances.’
In the specific context of transfer pricing comparability analysis, a safe harbour

provides simplified compliance obligations, such as an acceptable arm’s length
range, for transactions with limited mispricing risk. A consultative document issued
by the OECD in June 2012 recorded a number of member country examples as at
1 January 2012, most commonly:
• Safe harbour arm’s length range for ‘low value-adding’ intra-group services;
• Safe harbour interest rate for loans.
Such safe harbours are certainly welcome, but they are all unilateral with widely
varying rules for similar transaction types.
We will look in detail at the OECD approach to safe harbours in a later chapter. It
is sufficient to note here that the OECD TPG recognise that bilateral or multilateral
safe harbours can provide certainty and reduce the compliance burden for
taxpayers whilst releasing valuable resources for tax authorities.

 Illustration 4
Kodaly Concepts Limited is the UK parent of a group of marketing and

communications consultants. It provides accounting, human resources and IT
support to its operational subsidiaries in Australia, Austria and Japan. The company
qualifies for the SME exemption from UK transfer pricing rules and has brought
forward trading losses. It identifies the direct and indirect costs of providing the
above services and charges them at a 12.5% mark-up.
The safe harbour position in the three subsidiary companies for low value-adding
intra-group services is:
Australia Cost plus 7.5% to 10%

Austria Cost plus 5% to 15%
Japan Cost only
The group’s policy will not therefore satisfy two countries’ requirements and it is
likely that a conventional transfer pricing analysis will have to be carried out.
11.9 Frequency of Review
Another common practical issue is how often it is necessary to update the

comparability analysis. The OECD TPG accept that it may not be necessary to
perform a detailed comparability analysis each year for simple transactions in a
stable environment (see paragraph 3.82). It is sometimes advocated that an
annual ‘sense check’ could be used to ensure that previous comparability analysis
has not been invalidated by a significant change in the industry, market or
economic environment, together with a full review every three years. Again,
country expectations differ here. The danger is that incremental changes to the
enterprise and its operating environment may have a material cumulative effect.
11.10 UN Manual
You will recall from an earlier chapter that Chapter B3 of Section B of the UN
Manual looks at comparability analysis. It expands on the discussions in this and
earlier chapters of this manual on the process and application of comparability
analysis with worked examples. This would be a good time to look at sections B3.2,
covering comparability analysis in operation, and B3.6, covering further issues.

CHAPTER 12
SPECIFIC TRANSACTIONS: INTRA-GROUP SERVICES
In this chapter we will look at what the OECD TPG say in respect to services, in particular
looking at:
– determining whether a service has been rendered
– the arm’s length charge
– the transfer pricing method
– low value-adding intra-group services
– EUJTPF Report on low value-adding intra-group services
– UN Manual
– scoping of the future revisions to Chapter VII
12.1 Introduction
Almost all groups will have intra-group services of some kind. Often, the services
arise because it is more efficient and economical to centralise certain activities.
For instance, it is particularly common for various back-office services to be
provided, including IT, legal, finance, human resources and so on. These are
typically carried out by the parent company or by a group service centre or by a
regional HQ.
Some types of intra-group services will be apparent to the customers of the

business, such as one company carrying out warranty repairs of equipment sold by
a related company in another country, or one company carrying out sales as
agent of a related company.
This chapter primarily focuses on the main issues that arise in determining the arm’s
length price for services that have been rendered as part of an intra-group
transaction.
Some countries have specific legislation, regulations or guidelines on this, but in

most cases the only guidance is the OECD TPG. These include a specific chapter,
Chapter VII, dealing with intra-group services. This was published in 1995 and has
been updated as a result of the final reports on Action Points 8 to 10 of the BEPS
Project. Further guidance was added on shareholder activities together with a
new Section D on low value-adding services which provides for an elective 5%
mark-up for such services as discussed below.
In the analysis of transfer pricing for intra-group services, the OECD TPG mainly
concentrate on two issues:
• First of all, it is important to understand whether a service has actually been

rendered in the context of the intra-group transaction under analysis.
• Secondly, once it has been established that a service has been provided by
an enterprise to one or more related parties, it is then crucial to assess what
the charge should be, in accordance with the arm’s length principle.

12.2 Determining Whether a Service Has Been Rendered
In order to assess whether an intra-group service has been rendered one must
consider whether the activity in question provides a related party with economic
or commercial value that enhances its commercial position.
This can be tested by considering whether a third party enterprise in comparable

circumstances would have been willing to pay for the activity or would have
performed the activity in house for itself. If the answer is no, the service should not
be considered an intra-group service under the arm’s length principle.
Some intra-group services are carried out by one member of an MNE group to fulfil
an identified need and to the benefit of one or more affiliated members of the
group. In such a case, it is clear that a service has been rendered.
 Illustration 1
Company A operates a call centre to provide support to the customers of its sister
company, B, which is in another country. Clearly this is an activity that company B
would have to perform itself or acquire from a third party if company A was not
performing it, and there is a clear benefit to company B. It almost certainly
qualifies as a service for which a charge should be made under the arm’s length
principle.
 Illustration 2
Company C, is a Singaporean company that has been subcontracted by its

parent company, D, to act as investment subadvisor on an investment fund which
invests in equities from the Pacific region. Company D is the investment manager
which has launched and marketed the fund to UK investors, and if company C
was not making decisions about which equities to buy and sell, company D would
clearly need to do this itself or subcontract the work to a third party. It clearly
receives a benefit from the activities carried out by company C for it, and so they
qualify as a service and a charge is justified (indeed necessary) under the arm's
length principle.
The OECD TPG make special mention (at paragraph 7.14 under the heading
Centralised Services) of the type of services that are commonly referred to as
head office services or management services. These are services that benefit the
group as a whole and are often centralised at the regional headquarters or
parent company. Paragraph 7.14 lists many examples:
• Administrative services such as planning, coordination, budgetary control,

financial advice, accounting, auditing, legal, factoring, computer services;
• Financial services such as supervision of cash flows and solvency, capital

increases, loan contracts, management of interest and exchange rate risks,
and refinancing;
• Assistance in the fields of production, buying, distribution and marketing;
• Services and staff matters such as recruitment and training;
• Research and development and administering and protecting intangible

property for all or part of the group.
The OECD TPG say that these kinds of activities ordinarily will be considered intra-
group services because they are the type of activities that independent

enterprises would normally have been willing to pay for or to perform for
themselves. Most companies do need most or all of the above functions in order
to operate, and so if these functions are performed centrally by another group
company this will normally represent a service for which a charge should be
made.
However, depending on the service, identifying whether there is a service might

not be as simple and requires understanding of how the service benefits the
related parties. The OECD TPG highlight several situations that may not be a
service to a group company because there is no benefit to that company:
• Shareholder activities;
• Duplicative activities;
• Incidental benefits.
The OECD TPG also discuss whether on call activities are a service. All of these
situations are discussed below.
The OECD TPG also state that evidence that a payment has been made for an
alleged service or the existence of service agreements are not in themselves
sufficient to demonstrate that a service has been rendered and that it created a
tangible benefit (direct or indirect) for the related parties involved in the
transaction. Equally, the absence of payments or contractual arrangements does
not automatically lead to the conclusion that no services have been rendered.
Shareholder Activities
Some types of activity are performed by a company because of its ownership

interest in other companies, ie in its capacity as shareholder. Ordinarily such
activities do not provide an economically relevant value to the other group
members. They do not need the activity and would not be willing to pay for it if
they were independent. Such activities are referred to as shareholder activities.
The OECD TPG provide the following examples of shareholder activities:
1. Costs of activities relating to the juridical structure of the parent company

itself, such as meetings of shareholders of the parent, issuing of shares in the
parent company, stock exchange listing of the parent company and costs of
the supervisory board;
2. Costs relating to reporting requirements (including financial reporting and

audit) of the parent company including the consolidation of reports. Costs
relating to the parent company’s audit of the subsidiary’s accounts carried
out exclusively in the interest of the parent company and costs related to the
preparation of the consolidated financial statements of the MNE (however in
practice costs incurred locally by the subsidiaries may not need to be passed
on to the parent or holding company where it is disproportionately onerous to
identify and isolate those costs);
3. Costs of raising funds for the acquisition of its participations and costs relating
to the parent company’s investor relations such as communications strategy
with shareholders of the parent company, financial analysts, funds and other
stakeholders in the parent company;
4. Costs relating to compliance of the parent company with the relevant tax
laws; and

5. Costs which are ancillary to the corporate governance of the MNE as a whole.
Examples 1-3 were expanded on and examples 4 and 5 were added to the OECD
TPG on adoption of the final report on Action Points 8 to 10 of the BEPS project (see
paragraph 7.10 of Chapter VII).
The OECD TPG mention another type of activity that potentially falls within the
definition of ‘shareholder activity’, namely the costs of managerial and control
(monitoring) activities related to the management and protection of the
investment in the subsidiaries.
In order to determine whether these activities can be categorised as intra-group

services, we should carefully analyse whether they create any benefit. If the
activities are ones that the subsidiary would be likely to carry out itself if they were
not being done for it by its parent, they should probably be counted as a service,
notwithstanding that the parent is carrying out the activity partly because a
shareholder naturally wishes to manage and protect its investment.
It is recognised that in some cases activities similar to shareholder activities may be

performed by a group company other than because of a shareholder interest, in
which case it may be possible that a service is being provided subject to the tests
set down in Chapter VII (see paragraph 7.10).
The concept of shareholder activities was introduced in an OECD report published

in 1984, entitled Transfer Pricing and Multinational Enterprises – Three Taxation
Issues. This superseded an earlier term, stewardship activities, which was referred to
in the 1979 OECD report, Transfer Pricing and Multinational Enterprises, which was a
forerunner of the OECD TPG. The term stewardship activity was broader in nature
and included activities such as detailed planning services for particular operations,
emergency management and technical advice (troubleshooting), and in some
cases assistance in day-to-day management. The OECD TPG make it clear that
these activities are not within the definition of shareholder activities.
 Illustration 3
Company E is a listed Spanish company with a subsidiary in Argentina, company F.

The head office in Spain carries out a number of services for the benefit of itself
and its subsidiary. A transfer pricing analysis is carried out and the following
services are identified:
• accounting, including management accounts, local statutory accounts of

both companies, and consolidation of the Argentinian results into the
published accounts of company E;
• the head of sales in Argentina reports to the head of sales in Spain, who
monitors the performance of the Argentinian sales team;
• the group CEO splits his time between the two companies and a charge is
made for his time spent regarding Argentina.
The accounting services to company F are generally of a nature that creates a

benefit for company F and should therefore be treated as a chargeable service.
However, preparing the published accounts of the parent company is an activity
that is solely for the benefit of company E, even though some of it may relate to
the results of the subsidiary. The element of the head office accounting work that
relates to the published accounts of the parent should be split from the work on
the Argentinian statutory and management accounts and should not be charged
for. This is provided that splitting all such cost is not too onerous as noted above.

The group head of sales is carrying out managerial and control activities which, in
part, are intended to ensure that the Argentinian sales operation is effectively
managed. However, detailed interviews identify that the managerial and control
activities are the same as are performed in relation to the sales managers in Spain,
and so the group head of sales is effectively acting as part of the Argentinian
management, albeit that she is normally located in Spain. Without this senior
managerial input, the Argentinian business would have to hire a more senior local
sales executive. Accordingly, this is not a shareholder service and a charge should
be made.
Detailed interviews show that the group CEO splits his time primarily on the basis of
one week per month spent visiting Argentina to manage and coach local senior
managers and review performance and meet key local clients. This is not a
shareholder activity. However, during these visits the group CEO often meets with
an Argentinian company which is a 25% shareholder of company E, to report to
them as shareholders. This is a shareholder activity and so it is not a service and no
charge should be made for the time spent on this activity.
Duplicative Activities
The OECD TPG also state that there will generally be no intra-group service arising
from activities undertaken by one group member for another group member that
merely duplicate an activity that the other group member is performing for itself, or
that is being performed for such other group member by a third party. This
principle is clearly correct, but in practice it is rare for activities to be duplicative,
because most multinationals go to some effort to ensure that their activities are
planned and controlled in a holistic, coherent manner in order to achieve
maximum efficiency and effectiveness. If the same activity is truly being
duplicated, the multinational is being wasteful, so apparent duplication should be
treated with some scepticism. The mere fact that more than one entity carries out
an activity that is labelled with the same description does not mean that there is
necessarily duplication. Multinationals often split activities between group
companies.
A duplicative activity can sometimes constitute a valid service if it is only

temporary, for instance when the group is reorganising to centralise its
management functions and there is temporary overlap. Another exception would
be where the duplication is deliberate and there is valid business justification, such
as obtaining a second legal opinion on a particular issue.
The guidance makes clear that any consideration of potential duplicative

activities should analyse the nature of services in detail. Paragraph 7.11 (of
Chapter VII) gives as an example, the fact that a company performs marketing
services in-house and also is charged for marketing services from a group
company would not of itself determine duplication, since marketing is a broad
term covering many levels of activity.

 Illustration 4
Company G manufactures a certain product and sells it in its home market of the
USA. It is the parent company of companies H and I, which are responsible for
sales and distribution in Hungary and Italy, respectively, of the product
manufactured by company G. Company G carries out certain marketing activities
and makes a charge for marketing assistance to its two subsidiaries.
A transfer pricing analysis is carried out and it is identified that the two subsidiaries
each have their own marketing teams. However, on discussion with the head of
marketing it becomes clear that there is strict delineation of marketing
responsibilities between the local marketing teams and the head office. The local
marketing teams report back to head office on a regular basis, so any wasteful
duplication would be spotted and eliminated.
Accordingly, it is confirmed that there is in fact no duplication and the marketing

assistance is a valid service for which a charge should be made.
 Illustration 5
Company J is a law firm which has recently been acquired by company K, a large
multinational law firm with its head office in France. Company J has developed a
knowledge management system consisting of a searchable database of its know-
how and previous work carried out.
Company K has its own knowledge management system which operates using
different, incompatible software. Its policy is that all group companies must use the
group knowledge management system, in order to maximise the sharing of know-
how.
The head of knowledge management in company J considers that the group

system is inferior and therefore decides to continue operating the existing system,
setting up an arrangement where the content of the local system is automatically
added to the group system in order to satisfy group policy.
Because of the partnership, decentralised ethos of law firms, he is able to resist

efforts to persuade him to save money by closing down the local system.
A charge is made to company J for its share of the group system, on the grounds
that the group system is certainly being made available to company J and that if
no charge were to be made this would encourage the wastefulness to continue.
Although company J does use the group system it duplicates almost all of the
capabilities offered by its own system and therefore use of the group system is not
a service for transfer pricing purposes because this would be duplicative.
A charge would not be appropriate for transfer pricing purposes.
Incidental Benefits
There are some cases where an intra-group service performed by a group

member relates only to some of the group members but incidentally provides
benefits to other group members. When a MNE is looking to reorganise the group,
acquire a new company, or to terminate a division, these activities could possibly
constitute an intra-group service to the particular members of the group involved.
For instance, a member of the group might be the appropriate entity to acquire a

new business in the same country and might therefore be charged for the costs of
acquiring the new business.
Members of the MNE group involved in these activities are going to receive a
service from a coordinating related party which, in a comparable non-related
situation and circumstance, an independent party would have been willing to pay
for.
These activities may also produce economic benefits for other group members not
involved as parties in the transactions, by increasing efficiencies, economies of
scale or other synergies. The commercial position of the other group members
could be more valuable after the transaction has been entered into, but the
OECD TPG take the view that the incidental benefit would not cause the other
group member to receive an intra-group service, because an independent
enterprise would not be willing to pay for it.
This is a conclusion that is easy to agree with in cases where these benefits are truly
incidental. However, in some cases the primary purpose of the transaction is to
create efficiencies, economies of scale, or other synergies for group members,
even though they may not be directly involved. These situations are one example
of where the arm's length principle can be extremely difficult to apply, because
the transaction is one that would never arise for a company if it were an
independent enterprise, and yet the transaction makes economic sense for the
group and is carried out for the benefit of the company in question.
As noted in Chapter VII, each case has to be looked at individually and the
section in Chapter I of the OECD TPG on group synergies will also need to be
considered.
 Illustration 6
A multinational group has operations around the world and manufactures a

certain product line in three factories in Poland, Slovenia and the UK.
The group has significant overcapacity in Europe in relation to this product line
and, after a review, it is decided that one of the factories should be closed down,
because this will allow the other two factories to operate at full capacity and the
group will boost profits through saving the costs of operating the factory that is
closed down and spreading the fixed costs of the other two factories over a much
higher volume of production.
It is decided that as the UK factory has the lowest utilisation it is the one that should
be closed down.
The group companies in Poland and Slovenia are not directly involved in the UK
company or the UK business, but they will clearly benefit from increased
profitability due to having additional volume of production and thus lower unit
costs of manufacturing their own products.
On an arm's length basis, they would be unlikely to be willing to pay a competitor

to close down its factory, indeed this would probably be illegal under competition
law. Nor would they be reassured that they would pick up all of the production
previously carried out by that competitor.
Nevertheless, the boost in profitability of the two surviving factories is the whole
purpose of the transaction and so is arguably not an incidental benefit that should
be disregarded.

The costs of closing down the UK factory could be argued to be a valid service to
the Polish and Slovenian companies. However, this approach could be
controversial with tax authorities. (Indeed, the approach could be controversial if it
is decided that this is not a service for which a charge should be made.)
Another type of incidental benefit that may not constitute a service is where a
group company obtains incidental benefits attributable solely to its being part of a
larger concern, and not to any specific activity being performed. The OECD TPG
(paragraph 7.13) give, as an example, a situation where, as a result of being part
of a wider group, a company has a credit rating higher than it would have if it
were not part of the group. The OECD TPG draw a distinction, however, between
benefits arising from passive association as opposed to active promotion that
positively enhances the profit-making potential of particular members of the
group. Again, each case is looked at individually and with consideration of
Section D.8 of Chapter I of the OECD TPG.
Thus, if the higher credit rating is the result of an explicit guarantee given by
another group member, the benefit of the higher credit rating has not arisen from
passive association.
This distinction was scrutinised by the Canadian courts in a transfer pricing case
involving GE Capital, full name General Electric Canada v The Queen (2010 FCA
344)
The case concerned a guarantee provided by GE Capital in the US to its

Canadian subsidiary, which had the result that the Canadian subsidiary was able
to borrow on the market at an interest rate considerably lower than if it had been
a stand-alone company.
A guarantee fee was charged, equal to 1% of the borrowings of the Canadian

subsidiary.
The Canadian tax authority argued that although an explicit guarantee had been
given, the arm’s length guarantee fee would have been nil, because the
Canadian subsidiary could have derived the same benefits from mere passive
association with its US parent. Independent lenders would have perceived an
implicit guarantee, because the US parent would not have been willing to allow its
Canadian subsidiary to default on its liabilities.
This argument is widely considered to be an attempt to widen the application of

the passive association concept.
GE Capital argued that any implicit guarantee arising from passive association
arises only from the shareholding relationship and the arm’s length test requires us
to disregard anything that arises from the shareholding relationship. This defence
arguably attempted to narrow the application of the passive association concept
or even overturn it. In answer to this argument the Court stated that TP rules can
apply where a benefit arises from a non-arm’s length relationship. As a result the
factor of ‘implicit support’ is relevant in an arm’s length analysis under the TP rules.
The Canadian High Court, subsequently supported by the appeal court (case
reference 2010 FCA 344), did not agree with either side. It found that an implicit
guarantee would have existed and that this benefit from passive association
should not be disregarded, so in principle the arm’s length test would not allow a
charge for a benefit that would have arisen from the implicit guarantee.
Based on the evidence presented to it, the Court decided that if GE Capital
Canada had only benefited from an implicit guarantee, it would have paid

interest rates more than 1% higher than the interest rates it paid as a result of the
explicit guarantee. The Court concluded that a 1% guarantee fee was therefore
justifiable under the arm’s length test.
There are many wider implications arising from this case. They include:
• In the assessment of the credit risk of the subsidiary, the parent-subsidiary

affiliation must be considered.
• The assessment of implicit support is whether or not the debt market

participants would, by reason of the parent-subsidiary affiliation, charge the
subsidiary a lower rate of interest (or some other net economic benefit)
without an explicit financial guarantee.
• In applying the arm’s length principle, the ownership or shareholder

relationship between the parent and subsidiary must be ignored so that the
parties are transacting as if they were distinct and separate enterprises.
• The selection and application of the most appropriate transfer pricing method
to determine an arm’s length guarantee fee for a financial or loan guarantee
must consider the economic benefit to the subsidiary (in other words, its
interest rate savings as a result of the explicit financial guarantee).
• It is both difficult and important to determine the credit risk rating of a

subsidiary. This determination is a critical first step in ascertaining the arm’s
length guarantee fee.
• It is not only necessary to determine the stand-alone credit risk of the

subsidiary, but to also consider the impact of the parent-subsidiary affiliation
(the implicit support) on the stand-alone credit risk.
On Call Services
Another important issue that arises when dealing with intra-group services is in
relation to ‘on call services’.
An ‘on call service’ is a service provided by a parent company or a group service

centre that ensures the complete availability of a service for members of an MNE
group.
The question is whether the availability of the service has to be considered an

intra-group service itself (in addition to any services that are actually performed)
and therefore should be charged at arm’s length.
In the OECD TPG it is stated that in order to determine the existence of an intra-
group service we should expect an independent enterprise in comparable
circumstances to incur standby charges to ensure the availability of the service
when the need for them arises. However, it is unlikely that an independent
enterprise would incur standby charges where the potential need for the service
was remote, or where the advantage of having services on call was negligible, or
where the on call services could be obtained promptly and readily from other
sources without the need for standby arrangements.
In such cases, the most reasonable course to take, as the OECD TPG suggest, is
looking at the extent to which the service has been used over a period of several
years rather than solely the year in which the charge is to be made. In other words,
in determining whether a service has been rendered, we should concentrate on
the substance of this service and to what extent and measure it has affected the
group companies involved in the transaction.

12.3 Determining an Arm’s Length Charge
Having determined that a service has been rendered and that a charge should
be made, the second step is to determine the appropriate quantum of the
charge, consistent with the arm’s length principle.
The broad principle is the same as any other type of transaction: the charge
should be that which would have been made and accepted between
independent enterprises in comparable circumstances.
The main topics discussed by the OECD TPG in this respect are:
• direct charging methods versus indirect charging methods
• cost allocations
• CUP method versus cost plus method
• the appropriateness of adding a mark-up on top of costs.
Direct versus Indirect Charging Methods
In general large MNEs use a direct or an indirect method for charging for services.
The direct charge method is where each associated enterprise receiving the
service is charged directly for that service. This charge is on a clearly identified
basis; for example it may be using CUP or cost plus as appropriate.
The indirect charge method is where the charge is based on an apportionment

amongst various associated enterprises. The apportionment will involve the use
allocation keys as discussed below.
Direct charging is most likely to be possible and appropriate in cases where a

company is providing an intra-group service that it also provides to third parties,
which means it will have put in place a mechanism for determining an
appropriate charge, such as a system for tracking work done. (It is also likely to
mean that the CUP method can be applied.)
An indirect charge method may be necessary depending on the nature of the

service being provided. For example, as stated by the OECD TPG, indirect
charging may be necessary when the proportion of the value of a service
rendered to various entities is not quantifiable except on an approximate or
estimated basis. Certain centralised activities may be intended and expected to
produce simultaneous benefits for more than one group company, although the
quantum of the relative benefit cannot be measured.
Another instance where applying a non-direct method for recharging intra-group

services is appropriate is when a separate analysis of the relevant service activities
for each beneficiary would generate excessive administrative burden for the MNE
group and result in compliance cost that could be excessive in relation to the
activities themselves.
The OECD TPG make it clear that although there is often no alternative than to use
cost allocation and apportionment, which generally necessitate some degree of
estimation or approximation, this should be done in a way that gives sufficient
regard to the value of the services to the recipients. The OECD TPG do not come
out with a categorical rule, but they appear to be pushing for direct charging in
cases where the intra-group service is one that is provided to third parties as well.

In practice, indirect charging methods using cost allocation are far more common
than direct charging methods.
Allocations
The OECD TPG state that if an indirect charging method is used, it should be
sensitive to the commercial features of the individual case (eg, the allocation key
makes sense under the circumstances), contain safeguards against manipulation
and follow sound accounting principles, and be capable of producing charges or
allocations of costs that are commensurate with the actual or reasonably
expected benefits to the recipient of the service.
The OECD TPG go on to specify that the allocation method chosen must lead to a
result that is consistent with what comparable independent enterprises would
have been prepared to accept. This is to be achieved by choosing allocation keys
that are appropriate to the particular service being rendered and the benefits
that it creates.
Tax authorities are often sceptical when multinational groups bundle a whole
range of different services together and then split all of them across the group
using a single broadbrush allocation key, such as relative sales. They often argue
that this would not have been acceptable if independent enterprises were sharing
costs in this way. They prefer direct charges, but if this is truly not possible, they tend
to prefer the use of several different allocation keys chosen to give appropriate
allocations of the various services.
 Illustration 7
For the purposes of these illustrations we assume that low value-adding services
(discussed below) are not in point.
Company L is the parent of a multinational group. It carries out human resources

and IT services on a centralised basis for the whole group.
Interviews with the head of human resources and with the users of its services
indicate that, in the long run, the amount of time spent by the HR team on each
country is roughly proportional to the headcount of staff in each country, so
headcount is used as the allocation key for the costs of the HR Department.
Interviews with the head of IT and with the users of its services indicate that IT
expenditure falls into two main categories. Firstly, every employee around the
world has a desktop computer and the IT department provides support for this.
Secondly, the IT department operates an extremely sophisticated system for
planning, scheduling, controlling, and costing production, which takes place in
two factories in Canada and Thailand.
The costs of these two categories are therefore determined separately and the
costs of desktop support are allocated in proportion to the number of desktops in
each country. The costs of the production system are allocated only to the
Canadian and Thai subsidiaries and are split between them based on the ratio of
production capacity in the two factories.

 Illustration 8
Company M is a Chinese company which manufactures consumer products. It

sells them in China and, through its overseas subsidiaries, in the rest of the world.
Although the subsidiaries carry out some of their own marketing, the head office in
China includes a marketing team which carries out marketing for the benefit of
the group as a whole, including carrying out promotion at international fairs and
arranging global marketing campaigns based around sponsorship of international
sports people and teams.
The two main sponsorships are in relation to Manchester United football team and
a world champion (Chinese) table tennis player. Although Manchester United is a
UK football team, it was carefully selected by the head office marketing team
because it is extremely widely known and supported around the world. The table
tennis player is a well respected household name in China and Korea, where table
tennis is extremely popular as a spectator sport, but in most of the rest of the world
table tennis is a sport that is of interest to only a tiny minority.
The international fairs and the Manchester United sponsorship are of significant
benefit in all major markets, but the benefit cannot be objectively measured.
Direct charging is not possible, so it is necessary to allocate the costs across all of
the sales subsidiaries (and the Chinese parent).
An appropriate allocation key might be to split the costs in proportion to sales in

each market.
In contrast, it would probably be inappropriate to use the same allocation key for
the table tennis sponsorship costs, because this would not reflect the
proportionately higher benefit in China and Korea. Perhaps in this case it would be
appropriate to weight the sales revenue in proportion to the popularity of table
tennis in the different markets, assuming that an objective measure of this can be
found.
12.4 Transfer Pricing Method
OECD TPG Chapter VII makes clear that any analysis of intra-group services must
consider the perspective of both the service provider and service recipient.
Therefore, it is necessary to consider the value of the service to the recipient, the
amount that a comparable independent enterprise would have been prepared
to pay for the service in comparable circumstances, as well as the costs incurred
by the service provider. It is implicit in this that a one-sided analysis using just cost
plus or CUP is unlikely to be acceptable. (However, this does not necessarily mean
that both methods must not be used.)
Chapter VII says that the method to be used in determining the arm’s length
transfer price for intra-group services should be determined according to the
guidance in Chapters I, II and III.
Chapter VII says that the CUP and cost plus methods are often used for pricing
intra-group services, and this is borne out by practical experience. It would be rare
for any other method to be used (although the cost plus method is often applied
in a manner that could equally be described as being a TNMM using total mark-up
on costs as the profit level indicator).

CUP Method
The OECD TPG say that the CUP method is likely to be applied when the intra-
group service is either provided to third parties by the same entity or another
related entity or there are third party comparables, which can be used to price
the transaction.
There is no explicit statement that CUP is to be preferred, although arguably the

discussions regarding mark-ups (see below) implicitly mean that cost plus is
overridden in cases where it would imply a price higher than the market value of
the services (ie, a CUP).
Using the CUP method to price intra-group services can be very difficult as unless
the service is in relation to the main line of business for the group there will not be
any third party arrangements that can be used for the analysis. Furthermore,
independent enterprises are not likely to disclose the nature and the main
characteristics of services; therefore, even finding third party CUPs could prove a
very difficult exercise.
 Illustration 9
Refer back to the previous illustration involving company L.
Assume that the group is in the business of manufacturing and selling computer
hardware and providing IT services to customers.
In such a case, then it may well be possible and appropriate to use the CUP
method to charge for the intra-group IT services by treating these as if they were
being provided to an external customer and pricing them in a similar way.
It would have to be investigated whether the pricing system would produce a

reliable split between the group companies.
It is possible that the CUP method would only produce a figure for the total charge
to be made for this service and that an allocation key would still be necessary in
order to split the costs between the group companies.
As company L is not in the business of supplying human resources services

externally, it would be unlikely to be possible to use the CUP method to charge for
the HR services.
Cost Plus Method
The OECD TPG confirm that the cost plus method is, in the absence of a CUP, the
most appropriate method when the nature of activities involved, assets used and
risks assumed are comparable to those undertaken by independent enterprises. In
practice, the cost plus method is by far the most commonly used for intra-group
services.
Many tax authorities have reservations about the use of cost plus for many
services, in part because services that are being provided between unrelated
parties are rarely priced on the basis of the costs of the service provider plus a
fixed percentage mark-up. Few independent service recipients are willing to
guarantee a profit to an unrelated service provider and few service providers are
willing to restrict their potential profitability to just a (usually small) mark-up on costs.
However, the use of a transfer pricing method is not conditional on establishing

that the same method would have been used had the transaction been carried

out between arm’s length parties. Moreover, if there is no reliable CUP available
which reaches an acceptable standard of comparability with the controlled
transaction, cost plus is, like it or not, generally the only other method that can be
used.
Regardless of the method used in determining an arm’s length price for intra-
group services, it is important that the remuneration reflects the nature, functional
and risk profiles of the transaction. Typically, when cost plus is being used to set the
transfer price, this inherently means that the service provider has little or no risk. In
comparison, the pricing methods that are commonly seen between unrelated
parties will often expose the service provider to risks, for instance if they have cost
overruns or if they do not generate enough business to keep their staff fully utilised
and so fixed costs rise as a proportion of sales revenue. This can be a source of
controversy.
Issues related to risk are dealt with in depth in the OECD TPG Chapter IX. Although
this OECD TPG chapter relates to transfer pricing issues arising from business
restructuring, many of the comments relate equally to transfer pricing if no business
restructuring has taken place. We will look at OECD TPG Chapter IX in detail later,
for now we will note that it does confirm that it should be accepted that using cost
plus means that the service provider has no risk. The risk allocation should normally
be respected unless certain narrow conditions are met.
Mark-ups
A common issue that arises when using the cost plus method is whether a mark-up
should be added for the service provider, so that it makes a profit. A number of
countries tend to object to a profit being made on inbound services that are only
carried out internally within the group.
The OECD TPG do not explicitly state a clear, unambiguous position on this, but
paragraph 7.35 does observe that:
‘in an arm’s length transaction, an independent enterprise normally would

seek to charge for services in such a way as to generate profit, rather than
merely providing the services at cost’.
The paragraph then goes on to discuss circumstances in which an independent

enterprise may not realise a profit from the performance of a service, so it is
arguably implicit that a mark-up should be added unless there are special
circumstances to justify providing the services at cost or even below cost.
One example of a special circumstance in which it might not be appropriate for

there to be a profit mark-up would be where the supplier of the services wishes to
offer the service so that a customer does not turn to the supplier’s competitors to
obtain this particular service. This might give the competitor a chance to win the
contract to supply other services currently being rendered by the supplier to the
customer on a profitable basis.
Another example would be to open up a new relationship with a prospective

customer.
In practice, however, arguments that such circumstances apply to intra-group

services are often viewed sceptically by tax authorities, because it is often the
case that group companies do not have the same freedom of choice regarding
service providers as would be the case if they were independent.

 Illustration 10
Company N prints and distributes fashion magazines. It has a subsidiary, SP, which
carries out the service of sourcing and procuring small products/gifts that can be
given as free gifts to customers to boost sales of the magazine or help penetrating
certain markets. It does not take title to the goods; it charges a fee for
procurement services.
In this case, SP would not see any financial benefit by providing these services to
the parent company at cost. The financial benefit manifests itself in larger volume
of sales of the magazine, which benefits the parent company only (or potentially
the distributors of the magazine).
In this case, the service should generate a profit for SP.
 Illustration 11
Imagine in the above illustration that SP does not just provide this service to its
parent company. Rather, it provides the service to many unrelated magazine
publishers, but to date these have generally been fairly downmarket publications.
Until recently, it was an independent company and, in an attempt to move

upmarket, it entered into negotiations to supply services to company N, whose
fashion magazines are at the top end of the market and extremely prestigious.
Rather than simply become a customer, company N chose to purchase SP, seeing
a large potential to expand SP's business.
In such a circumstance, there might be an opportunity to argue that the kudos

and credibility of being a supplier of these services to the prestigious magazines
published by company N would be so attractive to SP that it might be willing to
provide its services to N at cost or even at less than cost if this were necessary to
win the contract with N.
Another example given by the OECD TPG of where it might be inappropriate to

add a profit mark-up is where the market value of the intra-group services is not
greater than the costs incurred by the service provider.
This is a good example of a situation where it is particularly hard to apply the arm’s
length principle, because the reality is that when an independent service provider
incurs costs in rendering a service that are equal to or higher than the market
value of those services, its customers will not normally be willing to pay more than
the market value, so the service provider faces the choice of discontinuing the
service or selling at cost (if this is equal to the market value) or even below cost. An
independent service provider is unlikely to be willing to continue selling at no profit
except in unusual circumstances, such as a price war with its competitors, which
the company considers it is in a position to win.
However, there may be a number of good reasons why it makes economic sense
to a multinational group for an intra-group service to continue to be performed
internally despite the fact that it costs more than the market value of the service.
The OECD TPG make it clear that if the market value of the service is known (and
so the CUP method is able to be used) it would not be appropriate to charge a
higher price than this, even if this would be necessary to ensure that the service
provider covers its costs and makes a profit.
Very careful analysis would be necessary in order to understand fully why the
group has decided not to obtain the service externally, despite this having a lower

price. Such an analysis might indicate that an ostensible CUP is not in fact properly
comparable, because the external service provider would not provide all of the
same benefits as the internal service.
 Illustration 12
Company Q is the UK parent company of a group that provides international

consultancy services. It acquires company R, a Mexican business providing local
consultancy services on a relatively small scale.
Q has, over the last decade, developed an extremely sophisticated computer

system for planning, managing, and tracking its consultancy projects and its policy
is that all group companies should use this.
Q charges out the cost of this system to all group companies, splitting the costs in
proportion to relative sales in each country, and adding a 7% mark-up. The charge
to Mexico is £107,000.
However, when a transfer pricing analysis is performed in relation to this

transaction, the Mexican CEO explains that this system is far more sophisticated
than is justified by the needs of the current Mexican business. It is tailored to the
needs of other countries, particularly the US and UK, which often need to manage
complex projects involving multiple inputs from many parts of the group and
involving many stages.
Current Mexican projects are much more straightforward and could be handled
using simpler software, which would cost £80,000 for the Mexican company.
The group has decided that it is important that all group companies use the same
software, because this allows the group to market itself as offering the same
sophisticated capabilities in every country, even though it is accepted that
Mexican customers are not at present interested in these capabilities.
It might be appropriate in this case for the parent company to provide the
software at £80,000, even though this means it makes a loss on the £100,000 cost
attributable to Mexico. (The answer might change in the future, if the Mexican
company starts to take on more complex projects and it makes use of the full
capabilities of the software.)
Ideally, the mark-up percentage should be determined on the basis of the mark-
up made on comparable uncontrolled transactions. If, as is often the case, such
transactions cannot be identified, it is in practice necessary to resort to using the
profitability of independent companies that provide comparable services under
comparable circumstances. This is typically found by way of a search of a
database of the company accounts. As discussed above, it can often be difficult
to ensure full comparability, because independent service providers do not usually
operate on a cost plus basis, so risk levels are often different.
It is often supposed that services that require highly paid employees should earn a
higher mark-up than services that do not. Highly paid employees might be an
indication that the service in question is highly valuable and this may be reflected
in the fee charged by independent companies that provide such services, but it
does not necessarily follow that this will give rise to a higher profit margin. It is
perfectly possible that if the service in question requires skills or experience that are
scarce, competition may drive up remuneration to the point where most of the
benefit of the higher fees has been passed to the employees. The profit potential
of a company should reflect the economic value that it is adding over and above
its inputs.

It is beyond the scope of this chapter to discuss specific rules or practice in relation
to services transfer pricing in particular countries. However, it is perhaps worth
noting that the US takes a particularly pragmatic approach to the question of
mark-ups on certain services.
The US publishes a list of services on which it will not require a mark-up. Its rationale
is that if the appropriate mark-up would be low, the amount of tax revenue at
stake is also low, so it is willing to waive any requirement for a mark-up on such
services, in order to make tax compliance easier. In practice, this is only of
relevance to services provided by a US company. If the services are being
provided to a US company, by a company in another country, that other country
is likely to expect a mark-up.
Cost Base
It is often the case that most attention is given to the question of whether there
should be a mark-up and if so, what the percentage should be. Experience
suggests that frequently there can be much more at stake in relation to the cost
base.
Generally speaking, the cost base should include all relevant costs, not just the
salaries of the staff performing the service. For instance, all staff benefits should be
included, as should all overheads, such as rent, power, telecommunications,
human resources support, IT support, etc. This can often make a much bigger
difference than any error in the mark-up.
 Illustration 13
Company S delivers intra-group services to all of its affiliated entities (let’s assume
we are dealing with a large multinational group) and the total cost for delivering
the services is 100 million. If the mark-up chosen is 7%, company S will generate a
profit of 7 million.
However, a transfer pricing analysis identifies that only 10% of the costs of the head
office human resources department are being included in the cost base for the
recharge, because they only spend 10% of their time providing human resources
support direct to the rest of the group.
50% of the time at the human resources department is spent dealing with human
resources matters in relation to the other head office departments that are being
charged out to the rest of the group.
Therefore, a further 50% of the costs of the human resources department should
have been loaded into the calculation of the costs of the other departments that
are part of the head office charge.
Similarly, the costs of the head office IT department in supporting the other head
office departments have not been included. It is determined that the cost base for
the recharge should have been 10 million higher, so company S is actually making
a loss of 3 million.
The charge should be increased by 10 million, plus a 7% mark-up on this.
There are other issues to consider in relation to the cost base. In deciding the
appropriate mark-up using comparables, it is important to consider comparability
of the cost base. If the comparables include in their profit and loss accounts types
of cost not included in the costs of the supplier of the intra-group services, the
mark-up percentage is not being calculated on a like-for-like basis.

A related issue is the question of whether a mark-up should be applied in cases

where the service provider is merely acting as an intermediary. On the face of it,
no mark-up should be applied to costs that are merely being passed on, and a
mark-up should only be made on the costs of performing the intermediary
services.
An example would be advertising companies, which often acquire advertising

space on behalf of their clients. Arguably, if such a service is being provided intra-
group, the amounts disbursed to the providers of the media space should not be
marked up and the mark-up should only be applied to the costs of negotiating
and arranging the purchase of such space.
However, if the mark-up percentage is being set by reference to comparable

independent companies and those companies have flow-through costs reflected
in their cost base, then their cost base should be adjusted to strip out the flow-
through costs. This is not always possible, depending on the level of disclosure in
the accounts of the comparable independent companies, and so the only option
might be to include flow-through costs in the cost base of the in-house service
providing company.
12.5 Low Value-Adding Intra-Group Services
The final report on Action Points 8 to 10 of the BEPS project resulted in a new
Section D to Chapter VII of the OECD TPG containing additional guidance with
respect to an elective, simplified transfer pricing approach for low value-adding
intra-group services (LVAS).
The simplified approach recognises that the arm’s length price for LVAS is closely
related to cost. It operates by allocating the cost of each category of LVAS to the
group companies that benefit from using them, then applies a standard mark-up.
The benefit of this approach is that it will reduce compliance effort and cost,
provide greater certainty to the MNE and targeted documentation to the tax
authorities (see paragraph 7.52).
The OECD describes the guidance as being intended to achieve a balance

between appropriate charges for LVAS and head office expenses and the need
to protect the tax base of payor countries. Groups may adopt the simplified
approach, on a group-wide basis only, if they choose (by election). The usual
principles for charging intra-group services on an arm’s length basis will apply if a
group does not so choose. The guidance points out that where a tax authority has
not adopted the simplified approach and as a consequence a MNE has to
comply with local requirements, the MNE is not debarred from using the simplified
approach in other jurisdictions where it is available.
When a MNE elects to use the simplified approach for LVAS in accordance with
the guidance then the charges so made are determined to be in accordance
with the arm’s length principle (see paragraph 7.53).
The guidance on LVAS is to be implemented in two steps; step one was that a
large group of countries agreed to implement the elective rules by 2018; step two
involves further work by the OECD on the potential threshold above which the
mechanism will not apply (see below), this work will help persuade other countries
that the elective approach will not lead to base erosion. At the time of writing the
OECD had not published any further information on an agreed threshold,
however, countries have started to adopt the simplified approach. For example,

New Zealand announced in February 2018 that it would introduce it with a

threshold of NZ$1m.
There are four important areas covered in the simplified approach in Section D:
1. A Standard Definition of Low Value-Adding Intra-Group Services
LVAS are defined as those which are supportive in nature, are not part of the
core business of the group, do not use or create unique and valuable
intangibles, and do not involve the assumption, control or creation of
significant risk. There are examples of services that would qualify (such as IT,
HR, accounting) and those that would not.
Examples of activities which would not be considered as qualifying for the

simplified approach are: services constituting the core business of the MNE
group: research and development services; manufacturing and production
services; sales, marketing and distribution activities; financial transactions;
extraction, exploration, or processing of natural resources; insurance and
reinsurance; services of corporate senior management and purchasing and
procurement relating to raw materials used in the production process (see
paragraph 7.45).
Examples of what may be included include: accounting and auditing, matters

related to health and safety, matters related to human resources, processing
of orders, managing accounts payable and receivable and general services
of an administrative nature (see paragraph 7.47).
The OECD TPG emphasise via examples that it will depend on the nature of
the core business whether a service is low value. For example, credit risk
analysis for a shoe distributor is likely to be low value, but not for an investment
bank (see paragraph 7.50). When we are looking at the question of the core
business we look at it from the point of view of the group as a whole, not the
service company.
Where services that would normally qualify for the simplified approach are
provided to unrelated companies the simplified approach cannot be used. In
these cases, it is expected that there is a reliable external CUP that can be
used for arriving at the arm’s length price.
2. A Simplified Approach to Determining the Arm’s Length Charge
The starting point for a group that elects to apply the simplified approach will
be to identify, on an annual basis, a pool of costs associated with categories
of LVAS. These costs will be the direct and indirect cost and, where relevant,
an appropriate part of operating expenses. The cost pool should exclude costs
attributable to activities that are ‘shareholder activities’, as defined earlier in
the chapter, and those that benefit only one other group company, which will
be charged directly. Any ‘pass-through costs’ (costs where the service
provider is merely acting as an agent and not adding any value eg rent)
should not be included.
The costs in the pool are then to be allocated among group members using
an ‘allocation key’ that is relevant to the nature of the category of services.
For example, IT costs might be allocated on the basis of the share of the
number of group users, and HR costs on the share of group headcount. A
consistent approach is expected, such that the same allocation keys should
be used for all recipient companies, and also from year to year, unless there is
a valid reason to change. Groups will need to be able to coordinate all costs

and allocation keys centrally, documenting the process and results

accordingly in order to ensure there is global consistency.
A Simplified Benefit Test Relating to Low Value-Adding Intra-Group Services
While LVAS may provide benefits to all recipients of those services, it may be
difficult to demonstrate whether the recipient would have been willing to pay
for the individual service from an independent party or perform it themselves.
The taxpayer need only demonstrate that assistance was provided rather than
being required to specify individual acts undertaken that give rise to the costs
charged. Tax authorities are to refrain from reviewing or challenging the
benefits test where the taxpayer has been fully compliant with the reporting
and documentation requirements of the simplified approach. This is because
the reporting and documentation requirements should provide all the
evidence needed for the simplified benefits test.
The simplified approach states that benefits should be considered by category

of services and that a single annual invoice describing a category of services
should be sufficient to support the charge, rather than evidence of individual
services performed.
There will be cases where LVAS have benefited just one member of the group.
In this case it should be possible to provide separate documentation to
evidence the benefit to that group member.
Determination of Appropriate Mark-Ups for Low Value-Adding Intra-Group

Services
The same mark-up should apply to all categories of LVAS, and should be 5% of
the relevant cost (see paragraph 7.61). This is lower than the range in the EU
Joint Transfer Pricing Forum’s guidance for similar low-value added services as
noted below.
The 5% mark-up should not be applied to any ‘pass-through costs’.
Importantly this level of mark-up can only be applied to the category of

services that fall within the ‘low value-adding’ service definition and thus
cannot be used as a benchmark for other categories of services not falling
within the definitional scope.
The guidance provides that tax administrations adopting the simplified

approach may include an appropriate threshold to enable them to review the
simplified approach in cases where the threshold is exceeded. This might, for
instance, be based on fixed financial ratios of the recipient party or be
determined by reference to a group-wide ratio of total service costs to
turnover of the MNE group or some other appropriate measure.
3. A Description of Documentation that Should be Prepared by a Multinational

Enterprise (MNE) Group to Qualify for the Simplified Approach
Documentation for the simplified approach should be prepared on a group-

wide basis and include: reasons why the services meet the definition for being
low value-adding, and expected benefits from each category of services;
description and justification of the choice of allocation keys; confirmation of
the mark-up applied; and calculations showing the determination of the cost
pool and the application of the allocation keys (see paragraph 7.64).

4. WHT and Low Value-Adding Intra-Group Services
The OECD TPG encourage countries only to apply withholding taxes (WHT) to
the profit element or mark-up of the charge for LVAS.
The OECD observes that the levying of WHT on these categories of services
can prevent the service provider from recovering the totality of the costs
incurred for rendering the services.
Alison Lobb from Deloitte writing in Tax Journal in November 2014 (when the
revisions were still in proposal stage) commented that: ‘The proposal is for a global
approach, and therefore in practice this simplification will only be of benefit where
it is adopted widely by countries. For countries with OECD-based double tax
treaties, amendments to the Transfer Pricing Guidelines will be likely to be sufficient
to implement the proposal for simplification; but it will remain to be seen whether
other countries, particularly those outside the OECD, choose to adopt it.’
She also observed that the simplified approach will be helpful for some groups for
whom it currently proves too difficult or too costly to provide sufficient evidence to
support small amounts of individual charges across a wide number of jurisdictions,
resulting in double taxation. In addition, there will be benefits for tax authorities
with limited resources in considering the appropriate mark-up and benefit.
Section D of Chapter VII has some similarities and some differences to the EUJTPF
report discussed below.
12.6 EUJTPF Report on Low Value-Adding Intra-Group Services
The European Union Joint Transfer Pricing Forum (‘EUJTPF’) published a report on
low value-adding services in July 2010. The report looks at a certain type of
services, those that are described as ‘the glue that holds the corporate structure
together to support its main function’; services that are routine in nature and do
not add high value to either the provider or the recipient.
One of the aims of the report is to limit the compliance burden in relation to such
services. The report suggests that this can be achieved by having a ‘narrative’ to
provide sufficient correlative evidence that the service has been rendered and an
arm’s length price charged.
In relation to the form of the narrative the report states that:
‘a dedicated written narrative could be provided in some cases. Some of the

information, if appropriate, may be given verbally. It might also be the case
that the examination of written contracts will provide an insight to the wider
context and will provide most of the information in any narrative. Each of the
approaches or some combination of them is valid. The important point is that
the outcome is an understanding of how any service provision system works.’
Once the narrative has been received the tax authorities can decide whether any
further information or explanation is required.
Turning to methodologies, the report acknowledges a cost plus method as the

most commonly used. The report states that for low value services only a moderate
mark-up will be required.

The report states that:
‘in cases where it is appropriate to use a mark up, this will normally be modest
and experience shows that typically agreed mark ups fall within a range of 3-
10%, often around 5%. However that statement is subject to the facts and
circumstances that may support a different mark up’.
For such low value-adding services it is possible that there will not be written
documentation due to the nature of the service. The lack of such documentation
is (say the EUJTPF) not to be a justification for assuming that the arm's length
principle has not been applied.
The report suggests that a useful and a proportionate documentation pack may
contain:
• A narrative;
• Written agreements;
• Details of the cost pool;
• Justification of OECD methodology applied;
• Verification of arm's length price applied;
• Invoicing system and invoices.
12.7 Cost Contribution Arrangements (CCAs)
We will look in more detail at CCAs in a later chapter. Here we will just note that
CCAs can be used within a group to share the cost of services.
A CCA would be used as a way of avoiding duplication and achieving economies

of scale.
12.8 The UN Manual
The UN Manual includes a separate chapter on services. As well as expanding on

the approach of the OECD TPG it includes detailed illustrations of how the
appropriate methodology should be applied to services.
The chapter looks at two types of safe harbour for services: low value-adding and
‘minor expenses’. For low value-adding there is a description of the types of
services it might apply to and possible parameters. It is not an examination of the
OECD approach, rather a generic explanation of the type of approach that could
be taken, as such safe harbours were in place prior to the OECD introducing their
guidelines. It was noted above that New Zealand have adopted the OECD
simplified approach; prior to this adoption they did have an existing safe harbour
for low value-adding services.
A minor expenses safe harbour is described as one where the tax authority will not
make transfer pricing adjustments if the total cost of either receiving or providing
intra-group services is below a stated threshold based on cost and a fixed mark-up
is used. Examples are given of the types of parameters that could be used for such
a safe harbour (see section B.5.5 of the UN Manual).

The description of the low value-adding services safe harbour is in line with the
OECD approach although no indication of the percentage mark-up is given or a
reduced benefits test.
Now would be a good time to have a look at B.5.5 of the UN Manual and work
through the illustrative examples.
12.9 Scoping of the Future Revision of Chapter VII
In 2018 the OECD invited comments on the scope of potential future revisions to
Chapter VII on services as part of the process of Working Party No. 6 considering
beginning a project to revise Chapter VII. The invitation for comments stated that
the concern regarding the guidance in Chapter VII relates to its practical
application, rather than potential theoretical disagreements on the underlying
principles. The OECD highlighted the following areas that require further analysis:
• Demonstrating that a service has been rendered and/or that the service
rendered provides benefits to the recipient;
• Drawing a distinction between: (i) activities which do or do not benefit the

local / affiliates; (ii) benefits that purely arise from group membership and
those that arise from a deliberate concerted action; and (iii) shareholder
activities and stewardship activities;
• Identifying in practice duplicated activities;
• Finding an appropriate allocation key for charging intra-group services;
• Determining the costs that should or should not be included in the cost base of
the remuneration for the provision of services between associated enterprises;
and
• Assessing the arm’s length conditions for services provided in connection with
the use of intangibles; services that are highly integrated with the value
creation of the MNE group; and/or involve significant risks.
The OECD have also commented that a revision of Chapter VII would also focus
on aligning it with Chapters I, VI and VIII, as well as considering whether and how
to incorporate the revised guidance on the use of profit split methods and the
ongoing work on financial transactions. Comments were received in June 2018;
nothing further has been published on this exercise at the time of writing.

CHAPTER 13
SPECIFIC TRANSACTIONS: LOANS AND OTHER FINANCIAL

TRANSACTIONS
In this chapter we are going to look at transfer pricing for financial transactions, in
particular:
– loans
– thin capitalisation
– guarantee fees
– cash pooling
– captive insurance
– hedging
– BEPS Action Point 4
13.1 Introduction
Intra-group transactions are generally regarded as falling into one of four

categories: sales of tangible goods, provision of services, licences to use intangible
property, and financial transactions.
There are specific chapters in the OECD TPG relating to services and intangible
property, and although there is no specific chapter on tangible goods, much of
the first three chapters of the OECD TPG tends to be written in the context of
setting a price for tangible goods. The post BEPS update to the OECD TPG in July
2017 contained very little on financial transactions, however (as noted in an earlier
chapter) the OECD released its final report on financial transactions in February
2020. The final report adds guidance on risk-free and risk-adjusted returns to
Chapter I of the OECD TPG and adds Chapter X on financial transactions (see
later chapter).
In this chapter the focus will be on the principles behind financial transactions,
looking at how interest rates and levels of borrowing are determined in the market
place as well as looking at the use of cash pools, captive insurance companies
and hedging.
13.2 Loans
Association Tests for Loans
This chapter will not focus on the question of determining whether a loan is subject
to transfer pricing rules, because (as we saw in an earlier chapter) each country
sets its own rules about determining exactly how closely connected two
enterprises must be before they are required to meet the arm’s length test in
relation to transactions between them.
In most cases, the rules will be the same for loan transactions as for any other kind
of transaction. However, it is worth briefly noting that some countries do have
special rules which apply transfer pricing principles to loans in cases where the
level of connection between the lender and borrower would not be sufficient to
apply transfer pricing principles to other types of transaction between them.

An example would be the UK, which has special rules introduced in 2005 with the
intention of ensuring that transfer pricing principles apply to loans made to finance
private equity type investments. Typically, private equity acquisitions of businesses
are funded by high levels of debt advanced by syndicates of lenders who are also
shareholders of the business that is acquired. It is often the case that each
shareholder has a relatively small interest in the target business and would not
normally be considered to be associated with the borrower under the normal tests
of association (which, in the UK, boil down to 50% control or 40% control in cases
where there is another party with 40% control).
As there is scope for these syndicates of shareholders to ‘act together’ because

they jointly control the target business, these special rules provide that a financing
arrangement (a loan) made by a person, P, to another person, B, will be subject to
the UK transfer pricing rules if certain conditions are met regarding ‘acting
together’.
How Interests Rate are Determined
A loan transaction involves a lender lending money to a borrower in return for the
borrower paying interest and, at some point, repaying the money lent. The interest
is a percentage of the amount lent (the ‘loan principal’).
The interest is the consideration paid in return for the use of the money and it is
therefore the relevant transfer price.
Interest rates always have two components:
• First, there should be a component of interest to reflect the use of money.

Even if there is no risk that the borrower might default on the loan, the lender is
still making the money available to the borrower and so there is a minimum
price for the use of the money. This is usually referred to as the base rate.
• Second, except in cases where the borrower is risk-free, there would normally
be a margin added on top of the base rate, to reflect the additional reward
required by the lender to compensate for the risk that the borrower might
default.
A commercial lender aims to charge interest rates on their portfolio of loans that
are low enough to be competitive with other lenders, yet high enough that it
receives enough interest income to cover its expenses and make a profit.
The expenses of a commercial lender will include the interest it pays on its own
debt funding, which would normally be close to a risk-free rate for a healthy bank,
although this is not always the case and there will be exceptions, eg when there is
a global financial crisis. The expenses also include the running costs of the bank
and any write-offs of irrecoverable loans.
In practice, the most difficult part of setting the interest rate is judging how much
risk premium to add, and historically this has been a key role of banks: assessing
the creditworthiness of individual borrowers and setting appropriate interest rate
margins to reflect this.
Interest rates on loans are therefore normally expressed as a base rate plus a
margin (eg three-month LIBOR (see below) plus 2%). Sometimes, the rate is a fixed
rate (eg 4%).

However, even when the interest rate does not mention a base rate, it will have
been determined by reference to the relevant risk-free rate and a margin to
reflect the risk that the lender might default.
The Base Rate
The base rate that is appropriate for a particular loan will vary according to a
number of factors, the key ones being:
• The currency in which the loan is made
• The term of the loan
• Whether the loan has a fixed rate or a floating rate
• The date the loan was made
Currency
Base rates vary according to the currency of the loan. This is because there is
normally a central bank which controls the base rate for a particular currency and
each central bank sets the rates according to priorities that are usually set for it by
the government that issues the currency and in reaction to the macroeconomic
circumstances of that particular currency.
For instance, interest rates on US dollar loans are affected by interest rate decisions
made by the US Federal Reserve, interest rates on British Pounds are affected by
the lending rate at which the Bank of England is willing to lend to UK banks and
interest rates on euros are effectively determined by decisions by the European
Central Bank about its lending rate to banks.
The country of the borrower or of the lender is not necessarily the same thing as
the currency of the loan. What matters is the currency.
 Illustration 1
If an Argentinian company borrows a loan denominated in US dollars, the relevant

base rate is the base rate for US dollars. If there is a greater risk of default by an
Argentinian borrower than an otherwise equivalent US borrower, this should be
taken into account in the margin.
In practice, there are a number of alternatives that can be used as the base rate
for a currency. One option is the central bank rate, which is the rate at which the
central bank announces it is willing to lend to banks. This is normally what is referred
to as the repo rate.
Another option is the interbank rate, which is the rate at which banks are lending
to one another. This is usually determined using daily surveys of banks, which result
in rates such as LIBOR (the London Inter-Bank Offer Rate) which is the average of
rates reported by a panel of London banks.
It should be noted that there are LIBOR rates for many different currencies, not just
British Pounds. For instance, there is a Euro LIBOR rate, which is normally almost
exactly the same as EURIBOR, which is the equivalent for European banks lending
in Euros. As noted below the LIBOR is expected to be replaced by the end of 2021.

Term of the Loan
It is generally the case that base rates vary depending on the term of the loan: the
length of time before the loan is due for repayment.
Ordinarily, base rates tend to be higher the longer the term of the loan, because
there is a premium to the lender for committing to make the money available for a
longer period.
However, this general tendency can at times be overridden by other factors, such
as expectations about how short-term interest rates will change in the long term. If
short-term interest rates are currently high, but there is an expectation that they will
fall, then the interest rate for a long-term loan might be lower than for a short-term
loan.
LIBOR rates are available for a variety of loan terms up to 12 months. For loans
longer than this, it is common to use interbank swap rates as the relevant base
rate.
The LIBOR and EURIBOR are to be replaced by the end of 2021 with transaction-
based alternative reference rates (ARRs). Each major currency will have a different
ARR.
Fixed/Floating Rate
Commercial loans will either have a floating rate of interest or a fixed rate of
interest.
A floating rate usually means that there is a fixed margin, but there will be a
floating reference rate which reflects changes in the rate in question.
The alternative is a fixed rate, which means that the interest paid will remain the
same over the course of the loan.
Borrowers often have the choice of whether to borrow at floating or fixed rates, so
they decide which they prefer depending on their views about whether floating
rates are likely to rise or fall and depending also on their willingness to take a risk.
It is therefore important to be absolutely clear whether the intra-group loan has a

fixed or floating rate and this should be matched when selecting the relevant
base rate.
The Date the Loan Was Made
All three of the factors discussed above (currency, term, and fixed/floating) vary
over time. Therefore, in order to determine the appropriate base rate percentage,
it is necessary to match the currency and term as at the date that the loan was
made. If the loan is floating rate, then it is necessary to continue to do this
matching at each date that the interest rate is reset.
Sources of this information include information providers such as Bloomberg and

the Financial Times and official websites such as the website of the relevant central
bank, finance ministry or tax authority.
The Interest Rate Margin
It is usually relatively straightforward to determine the base rate. Determining the

interest rate margin is often more challenging, because it is more subjective.

The interest rate margin should, in broad terms, reflect two key factors: the
creditworthiness of the borrower and the macroeconomic conditions affecting
credit spreads.
Creditworthiness
There is a very wide range of factors that potentially affect the creditworthiness of
a borrower, including the following:
• Collateral given by the borrower
• Security given by the borrower
• Asset backing for the loan. Companies with large amounts of assets that could
be sold in order to repay the loan if necessary, such as those with large
property portfolios, will generally be considered more creditworthy than
companies with proportionally fewer such assets, such as most companies in
the business of providing services.
• The level of other loans taken out by the borrower. The higher the total amount
of debt of the borrower, the greater the risk that the borrower might have
insufficient cash flow to service all the debt and, in due course, repay the debt
at maturity.
• The ranking of the debt. ‘Senior’ debt is usually considered to be less risky,
because the loan agreement entitles it to be repaid first, in preference to
other debt, if the borrower is unable to repay all of its debt. Debt that ranks
behind senior debt is referred to as junior debt, or mezzanine debt, or
subordinated debt.
• Gearing/leverage. Gearing (referred to in the US as leverage) is a measure of

the proportionate level of total debt relative to the equity capital of the
business. Equity reduces the risk of the lender, because if the borrower makes
a loss this reduces equity first of all, and there would only be any default on the
debt once the equity had been reduced to zero. Therefore, the higher the
proportion of equity, the lower the lending risk.
• Interest cover. Interest cover is a measure of the size of the interest burden
(and sometimes the repayments) as a proportion of the profits out of which
the interest will be paid. The more that the profits exceed the interest burden,
the lower the risk that if profits decline they will be insufficient to continue
servicing the loan.
• Cashflow. Interest is paid with cash, not with profits, so, strictly speaking, it is
better to measure interest cover using cash flow, but in practice EBIT (earnings
before interest and tax) is often used as a proxy. Sometimes, EBITDA (earnings
before interest, tax, depreciation and amortisation) is used as a halfway house
between profit and cash flow. Cash flow forecasts are sometimes used to test
whether the borrower is likely to be able to service the loan.
• Covenants. Commercial loan agreements often include covenants:

commitments by the borrower which help to protect the lender. For instance,
it is common to have covenants that the borrower will ensure that gearing
does not exceed a certain level or that interest cover will not fall below a
certain level. The loan agreement usually provides remedies for breaching the
covenants. For instance, the lender might have the right to demand early
repayment of the loan or to prevent the borrower from paying dividends or
taking out additional debt or incurring expenditure above certain levels.

• Guarantees. If a loan has been guaranteed by a guarantor, the guarantor is

promising to remedy any default by the borrower. This therefore reduces the
risk for the lender and the degree of reduction depends on the
creditworthiness of the guarantor.
• Business risks and volatility. Some businesses are inherently highly risky and
therefore lending to them is highly risky.
• Comfort letters can be used by lenders to judge whether a borrower can

meet payment obligations on a loan. A comfort letter will be issued by a
parent company to support a subsidiary’s application for a loan. The letter
doesn’t imply that the parent guarantees the loan. The letter just gives
reassurance that the parent is aware of the loan and supports it.
• Track record. Commercial lenders will often take into account whether a
borrower has shown a successful track record of servicing and repaying earlier
loans.
• Purpose of the debt and business plan. Commercial lenders will often want to
assess the chances of success of the purpose for which the debt is being
borrowed.
• Industry prospects. Certain industries will be perceived by commercial lenders

as being highly profitable and/or likely to experience considerable future
growth, whereas other industries might be perceived as being characterised
by thin margins and decline.
Sometimes the terms of a commercial loan will include a Cure Period. This allows a
debtor to catch up on a technical default on a loan. It is normally a fixed time
period of 30 to 90 days which allows the payment without there being a default on
the loan.
Macroeconomic Conditions
The macroeconomic conditions that should be considered include the following:
• Market sentiment. Economic cycles mean that commercial lenders do

change their views about the level of interest rate margin that they
expect/require for a given level of creditworthiness.
• Supply and demand for credit. This is closely linked with market sentiment,
although there are other factors that affect the supply and demand for credit,
including government intervention, such as the quantitative easing policies
used by many governments when required. If supply rises, interest rates for a
given level of creditworthiness will fall, and vice versa.
• Country/region. The preceding two factors vary from country to country at

any one time. It is therefore important to consider the country in question.
13.3 Thin Capitalisation
Thin capitalisation (US spelling, capitalization) is a phrase used to describe a

situation where a borrower has an excessive amount of debt capital relative to its
equity capital.
The phrase is not just used within an international tax context. Any commercial
lender will want to determine whether a prospective borrower is thinly capitalised,
because this might indicate that it would be excessively risky to make further loans
to the prospective borrower.

The term ‘thin capitalisation’ is not included in the Glossary of the OECD TPG. It is
however, in the Glossary of the UN Manual where it is defined as a high proportion
of debt capital relative to equity capital.
Paragraph 4.103 of the OECD TPG mentions thin capitalisation without discussion.
In the UN Manual paragraph B1.7.8 there is an explanation of thin capitalisation
and how financing can be more advantageous by way of debt from a tax point
of view.
Thin Capitalisation for Independent Companies
Let's first consider how levels of debt capital and equity capital are determined for
an independent company. Most independent companies will need capital to
finance the net assets of the business and they have two broad options. They can
raise debt capital by borrowing money and paying interest on this. Or, they can
use equity capital, by issuing shares or by retaining profits in reserves.
Most independent companies choose a combination of both types of capital. The

difference is that debt ranks ahead of equity, so the company must give priority to
paying interest on the debt and to making repayments. Dividends on equity can
only be paid out of profits after deducting the interest expense. Equity capital is
not normally repaid, but it cannot be repaid if this would leave the company
unable to repay its debts. Equity therefore acts as a buffer reducing the risk for the
lenders, because any downturn in profitability will simply reduce the return for the
shareholders unless the downturn is so large that the company is unable to pay the
interest.
As a result, providing debt capital is less risky than providing equity capital, so the
rate of return required by a lender is lower than the rate of return required by a
shareholder. A further benefit is that the interest paid on debt capital is generally
tax deductible, whereas dividends paid on share capital are not tax-deductible,
so the net rate of return that the company must pay to the lender (the cost of
debt) is further reduced in comparison with the rate of return that the company
must pay to its shareholders (the cost of equity).
Most independent companies can therefore reduce their overall cost of capital by
including some debt capital. If the company is able to use the debt capital to
generate a return that is higher than the net cost of the debt, it generates an
incremental profit which therefore boosts the return to shareholders, which is the
mission of the company. However, as the proportion of debt capital increases, the
level of risk being taken by the lenders increases, because the buffer provided by
the equity capital is proportionately lower. Therefore, the rate of interest
demanded by the lender increases. In addition, the level of risk being taken by the
shareholders also increases, because any downturn in profit will be spread over a
smaller amount of equity capital and will therefore have a bigger proportionate
effect. Therefore, the rate of return expected by the shareholders increases and so
the share price falls.
Accordingly, for an independent company there are limitations to the proportion

of debt capital that it can sustain. There may be a point where the interest rate
that would have to be paid on an incremental amount of debt would be so high
that it has an adverse effect on the overall cost of capital or it would be
unacceptably risky because a downturn in profitability might mean that the
company is unable to meet the interest burden and therefore might go bust. This
would be a concern to both the potential lender and to the borrower.
Each lender and borrower must make its own subjective decision about whether
the proportion of debt has reached the point of being excessive, in which case

they will consider the borrower to be thinly capitalised. If they are excessively
cautious, they might be giving up profits that they could have made. If they are
excessively incautious, they might suffer losses or even endanger their businesses.
It is important to note that whether an independent borrower is thinly capitalised is

a subjective judgement. One lender might consider a company to be thinly
capitalised and therefore refuse to lend to it, whereas another lender might be
willing to make a loan.
Thin Capitalisation on Intra-Group Funding
The reason that many countries have thin capitalisation rules within their tax
legislation is that the same considerations might not apply to intra-group funding. If
a parent company already owns 100% shareholding in a subsidiary and the
subsidiary needs additional capital, the parent company has an incentive to
choose to provide the capital by way of an intra-group loan, because the interest
will be tax deductible for the subsidiary. If the subsidiary suffers a decline in
profitability and it is unable to pay the interest or perhaps even to repay the loan,
the parent company is no worse off than if it had injected the capital as equity.
Either way, the parent will suffer the full loss, so there isn't the same limitation on the
proportion of debt capital that there would be if the lender was not also the
shareholder (or associated with the shareholder).
It is therefore common for countries to have specific rules intended to ensure that
multinationals cannot reduce the tax payable by a group company by having
excessive debt levels in that company. There is a variety of ways in which these
rules might work, but they are, generically, all known as thin capitalisation rules.
Most countries choose to use a clear, fixed debt:equity threshold. Typically, the
threshold is that the ratio of debt:equity should not exceed 3:1 or, sometimes, 2:1.
For instance, Canada has a maximum debt:equity ratio of 1.5:1.
In some countries, this is a firm limit and no deduction is allowed for interest on any
debt in excess of this ratio. In other countries, it is just a safe harbour, so no interest
deduction will be denied for reasons of thin capitalisation if the company is below
the threshold, but if the company exceeds the threshold it may still obtain a full
interest deduction if it can justify that its debt level is not greater than the arm’s
length amount. An example of the safe harbour approach would be China.
The other main approach is to restrict the amount of tax deductible interest to a
set proportion of the profits of the company, an approach often referred to as
earnings stripping rules. This approach is used, for instance, by Germany which
restricts interest deductions to 30% of EBITDA. In some cases, any excess interest
can be carried forward and potentially offset in future years. This is the approach
used by Norway, which restricts interest deductions to 25% of EBITDA, but allows
carry forward of the excess for 10 years.
It is a grey area whether earnings stripping rules and thin capitalisation rules
involving fixed debt:equity ratios are transfer pricing rules. Arguably, fixed limits are
not consistent with the arm’s length principle, because the arm’s length level of
debt will vary for different independent borrowers. Therefore, some countries use a
different approach, which is to use the arm’s length principle as an explicit test.
That is, they require the transfer pricing analysis to consider not only whether the
interest rate on intra-group loans meets the arm’s length test, but also whether the
quantum of debt exceeds the arm’s length level of debt. This is the approach
used, for instance, by the UK, which deals with thin capitalisation within its transfer
pricing rules.

The UK has introduced Advance Thin Capitalisation Agreements, which are

unilateral Advance Pricing Agreements (a subject we look at in a later chapter)
specifically to give certainty to taxpayers in relation to thin capitalisation. These
are usually expressed by setting out limits on the level of debt, below which the UK
tax authorities accept that the taxpayer is not thinly capitalised. The limits are
usually expressed in terms of a limit on debt as a proportion of capital and there
may also be other limits in relation to interest cover and sometimes a ratio such as
debt/EBITDA.
Determining an Arm’s Length Debt Level is a subject that we will return to in the
next chapter when the OECD TPG on financial transactions are considered in
detail.
13.4 Guarantee Fees
Another issue that often arises in relation to loans is guarantee fees. If a loan to a
company is guaranteed by another company in the same group, this can reduce
the lending risk for the lender, because the guarantor is agreeing that it will meet
the liabilities of the borrower if the borrower defaults. Therefore, the lender will only
suffer a loss on the loan if the borrower defaults and the guarantor also defaults.
The borrower is effectively ‘piggybacking’ on the credit rating of the guarantor. As
this is a clear benefit for the borrower, it would normally be expected that it should
pay a guarantee fee to the guarantor.
The guarantee fee should not exceed the benefit gained. In practice, the
guarantee fee is normally set to be lower than the benefit, so that both the
guarantor and the borrower benefit from the transaction. It is often difficult to
determine exactly how the benefit should be split, but the decision should be
based on the relative bargaining power of the two parties.
In order to determine the benefit gained by the borrower from the guarantee, it is
first necessary to understand the nature of the benefit. In some cases, the
guarantee simply has the result that the interest rate is lower than it would have
been without the guarantee, because the lender has lower lending risk. If so, then
the benefit is the interest rate differential.
The impact of guarantees on creditworthiness in a group situation is a complex

subject that will be discussed in more detail in the next chapter.
13.5 Cash Pooling
As mentioned in an earlier chapter MNEs often use cash pooling to help reduce
financing cost. Cash pools can be physical or notional.
When a physical cash pool is used there will be a physical movement of cash
balances into and out of one central bank account. The pool may be used to
ensure that all individual accounts are maintained at zero at the end of each day
or at a targeted amount.
If a notional pool is used this will offset positive balances against negative for all
the accounts within the pool without any funds moving between accounts. This
means that the notional pool treats all accounts as if they were part of a zero
balancing physical pool, although there is no movement of cash between
accounts.
There can be costs associated with setting up and running either type of cash pool
and these need to be balanced against the interest, FOREX, and banking cost

savings that can be achieved. Savings can also be made if less staff are needed
to monitor the treasury functions across the group.
There are many issues that need to be considered including:
• The currencies to be used in the cash pool;
• Capital contributions/increase in share capital: In some countries, such as

Germany, there have been concerns as to whether a valid contribution has
been made to a company's share capital where that company participates in
a cash pooling arrangement. The issue is that even though the funds
representing that contribution may initially have been received by the
company itself, those funds are then swept into a centrally managed
account, so that the company concerned loses control over the funds. There
may then be a risk that the company's capital may be regarded as not having
been validly increased, which can affect several shareholder issues (and also
could have knock-on effects for debt: equity ratios).
• Capital maintenance: The transfer of cash balances by a participating

company to a central account may be regarded as a return of funds to that
company’s shareholder, and for the purposes of the local laws on capital
maintenance, the relevant company’s claim for the return of an equivalent
amount may be disregarded. This may mean that the company no longer
maintains the required level of capital: this may involve a criminal offence or
may in some jurisdictions be treated as a dividend for tax purposes and could
also mean that a number of the transactions should be re-assessed – including
for transfer pricing purposes – on the basis of the company having reduced
equity funding.
• Insolvency risks/directors duties: Where participating companies have a

positive cash balance which is swept into a central cash pooling account,
they are effectively put in the position of unsecured creditors of the company
running the cash pool in respect of the relevant account balances (although it
may be theoretically possible for collateral to be provided, in practice this
may make the cash pooling arrangements unworkable). For the directors of
the participating companies to comply with their duties under the applicable
laws, they may therefore be required to consider the risk of a default by the
cash pool leader if the funds are required by the participating company.
Failure to do so may potentially expose the relevant directors to ‘wrongful
trading’ type claims. In other words, they may be held to be personally liable if
the cash pool leader were to go into insolvent liquidation, and if it were found
that they had previously allowed the company to continue to trade when
they should have known that there was no reasonable prospect of avoiding
insolvent liquidation.
• Will the authorities accept the transaction as one that independent parties
would enter on similar terms? The fact that independent parties wouldn’t enter
into a cash pooling arrangement doesn’t of itself mean the arrangement is not
arm’s length. However, each party would need to consider the benefit it
would receive and any alternative options.
There is no easy answer to how the benefits should be shared. A full functional
analysis will be required together with consideration of the contractual terms and
the ability to control risk.
The OECD provides specific guidance on cash pools in Chapter X of the OECD
TPG (see later chapter).

13.6 Captive Insurance
A number of multinational groups have implemented self-insurance arrangements

under which the group decides that it will no longer obtain external insurance for
certain risks.
Insurance companies work on the basis that although it is difficult to forecast

whether any individual insured party will suffer a loss, statistical analysis allows the
insurance company to anticipate the average loss of a portfolio of similar risks with
much greater certainty. The insurance company therefore sets an insurance
premium that reflects the average likely loss per insured party, plus running costs,
plus a profit margin for the insurance company. The benefit for the insured party is
that they are exchanging the risk of the full loss for the certainty of paying a much
smaller amount as an insurance premium. However, if a (non-insurance) group has
a wide range of group members which each have the same risks, then the group
can also benefit from this portfolio effect.
For instance, an individual company with a single factory probably could not
afford to take the risk that, say, a fire might damage the factory and prevent
production, because the losses could be proportionately very high. However, if the
group has many such factories around the world, the group as a whole might
have sufficient resources to be able to bear the costs of such a loss, because it
would be low in proportion to the group. So it might make no sense to obtain
external insurance for each factory, because this will mean that on average the
group is paying a profit margin to the insurance companies on top of the statistical
average cost of losses.
In order to prevent the results of any individual group company being distorted by
a loss, it is common to have a group company which will act as the internal insurer
for the group, writing insurance policies for the other group members and
collecting premiums from them. These are known as captive insurance
companies. In some cases, the captive insurer will enter into reinsurance contracts
with other group members, under which the group member self insures, but part of
the risk is re-insured. Having centralised the risk in the captive insurer, the group
may sometimes decide to reinsure some of the risk externally, just as an
independent insurance company might choose to do (note this is different to
fronting reinsurance which is explained in the next chapter). The captive insurer will
often be staffed with suitable staff to make these judgements.
Although the specific facts surrounding this intra-group service are different from
other kinds of services, the transfer pricing issues are generically similar to the issues
for other services. One must do the functional analysis, then look for comparability
data; this will be considered further in the next chapter that looks at the OECD
guidance on financial transactions.
13.7 Hedging
Hedging is the mechanism by which a company will mitigate certain risks and
exposures which, broadly, have the potential to impact either:
• the company’s cash flow; or
• the fair value of the company's (among other things) assets and liabilities.
A hedge is essentially a reverse bet by the company. It is specifically designed to

counter any movement in the value of a particular asset or liability – the hedged
item. A company will generally enter into one or more derivative instruments (ie a
hedging instrument) in order to achieve the desired hedge. A company may

trade in derivative instruments (eg large oil companies), but they are more usually
held as either investments or as instruments intended to mitigate, ie hedge a risk or
potential liability exposure of the company (eg changes in cash flow or the fair
value of assets). They can, therefore, be used to smooth volatility (eg in interest
rates or currency) and/or to give certainty for the company, especially in relation
to future liabilities. The most common examples of risks that a company will
mitigate with hedging arrangements include exposures to:
• movements in currency exchange rates;
• movements in the price of commodities; and
• changes in interest rates.
The perfect economic hedge will be one where the changes in the fair value of a
hedged item are exactly matched by corresponding changes (in the opposite
direction) in the fair value of the hedging instrument. In reality, very few hedge
arrangements will be perfect. That means that at any given point one or other
counterpart to the arrangements will be moving in, or out, of the money.
Derivative instruments (sometimes referred to as derivative contracts) include

futures, options, forward contracts, interest rate and currency swaps, interest rate
caps, collars and floors, forward interest agreements, commitments to purchase
shares or bonds and letters of credit.
In some MNEs hedging will be undertaken at individual company level, in other

cases hedging will be centralised. Centralisation may be in the form of a separate
finance or treasury company that has hedging as one of its group functions or it
may be a division within the parent company or a group service company.
When the hedging function is centralised transfer pricing needs to be considered;

this will be looked at in the next chapter.
13.8 BEPS Action Plan and Interest Deductions
We have seen in earlier chapters that the OECD TPG were amended following
adoption of the final reports on Action Points 8 to 10 of the BEPS Action Plan. The
executive summary to the final report on Action Points 8 to 10 makes reference to
the holistic nature of the BEPS Action Plan and the fact that work on other Action
Points will impact on transfer pricing outcomes as countries begin to implement
legislation and Double Tax Treaties are updated.
In particular one of the aims of the work on Action Points 8 to 10 was to ensure that
cash rich entities (‘cash boxes’) will not be entitled to excess profits. The profits that
the ‘cash box’ will be able to retain will be no more than a risk free return. If the
return qualifies as interest or an economically equivalent payment then those
profits should be targeted by the interest deductibility rules (see below) that
countries will adopt as a result of the final reports on Action Point 4. In addition,
legislative changes as a result of Action Point 6 and Action Point 3 could mean
additional withholding taxes and that the CFC rules are in point.
Action Point 4 (Limiting Base Erosion Involving Interest Deductions and Other
Financial Payments) was aimed at preventing excessive interest deductions, in
relation to both outbound and inbound investments, that shift profits to low tax
jurisdictions and give an unfair advantage to multinational corporations over
domestic groups.

The final report on Action Point 4 recommended that countries implement a ‘fixed
ratio’ rule that will limit net interest deductions claimed by an entity (or a group of
entities operating in the same country) to a fixed percentage of EBITDA. The fixed
percentage should be between 10-30% of EBITDA. The rule is applied to each
entity within a group, and compares its net tax deductible interest with its tax
based EBITDA. The report acknowledges that, in some circumstances, EBIT (or even
asset values) may be used instead of EBITDA. If the interest: EBITDA ratio exceeds
the fixed ratio, then the excess interest is not deductible.
In addition, countries should also adopt a ‘group ratio’ rule to supplement the
fixed ratio rule, and to provide additional flexibility for highly-leveraged groups or
industry sectors. The role of the group ratio rule is to alleviate the impact of the
core ‘fixed ratio rule’ in appropriate circumstances. There is also the option to
apply an uplift of up to 10% to the group’s net third party interest expense to help
prevent double taxation. The final report gives the option to allow carry back and
carry forward of excess interest within limits.
There are also optional elements for a de minimis floor below which all interest
could be allowed plus carry forward and carry back provisions. There are a
number of ways a de minimis rule might be defined. One would be to allow all
interest deductions up to a fixed value of (say) €1m, regardless of the core
restriction imposed by the ‘fixed ratio rule’.
The final report suggested that other limitations on interest expense, such as those
arising as a result a country’s application of the arm’s length principle or thin
capitalisation rules, also should be applied first.
In December 2016 an updated report was issued. The updated report does not
change any of the conclusions agreed in 2015, but provides a further layer of
technical detail to assist countries in implementing the group ratio rule in line with
the common approach. This emphasises the importance of a consistent approach
in providing protection for countries and reducing compliance costs for groups,
while including some flexibility for a country to take into account particular
features of its tax law and policy.
The OECD recognise that the common approach may not be suitable to deal with
risks posed by entities in the banking and insurance sectors. The OECD’s
recommendation is that countries should seek to identify specific risks in the
banking and insurance sectors. If no material BEPS risks are identified, banks and
insurers should be exempted from the ‘fixed ratio’ and ‘group ratio’ rules.
However, if risks are identified, specific rules should be enacted taking into
account the regulatory and tax regimes applicable to such sectors.
There is no agreed minimum standard for the group interest restriction under
Action Point 4. Countries are encouraged to implement similar rules.
As many of the EU members are also members of the OECD, the EU has also been
looking at limitation of interest deductions as part of its Anti-Tax Avoidance
Directive (ATAD). The ATAD was adopted in July 2016. Member States were
required to adopt and publish ATAD-compliant provisions by 31 December 2018 at
the latest (exceptions were provided), with the provisions applying from 1 January
2019. As you will see later, provisions in relation to exit taxation apply from 1
January 2020.
The EU ATAD restricts net borrowing costs to 30% of the taxpayer's EBITDA,
optionally with a €3m threshold (for exceeding borrowing costs) before the
restriction will apply. Stand-alone entities may be excluded from the scope.

Within consolidated groups, Member States may allow full or partial deduction of
exceeding borrowing costs under ‘group ratio’ conditions. Member States may
exclude loans concluded before 17 June 2016, loans used to fund long-term EU
public infrastructure projects, and financial undertakings. Carry forward of non-
deductible exceeding borrowing costs may be allowed without time limit (with an
option also to include carry-back for up to 3 years or carry-forward of unused
interest capacity for up to 5 years). There is a grandfathering clause that will end
at the latest on 1 January 2024 for national targeted rules which are ‘as effective
as the fixed ratio rules’ to be applied for a full fiscal year following the publication
date of an OECD agreement on a minimum standard.

CHAPTER 14
TRANSFER PRICING FOR FINANCIAL TRANSACTIONS

– the OECD final report on financial transactions
– Chapter X to the OECD TPG which provides guidance on financial transactions
14.1 Introduction
In the previous chapter we looked at the principles of financial transactions

including how interest rates and levels of borrowing are determined and the use of
cash pools, hedging and captive insurance companies to save costs. In this
chapter we are going to look at financial transactions with associated enterprises;
that is to say intra-group transactions.
Not all countries have the same approach to the pricing of intra-group financial
transactions and the impact of being a member of a group when looking at credit
ratings and the cost of borrowing. In this chapter we will consider how the OECD
TPG apply to intra-group transactions, look at some case law on implicit support
and look at some examples of domestic law.
14.2 The OECD Final Report on Financial Transactions
The OECD final report on financial transactions was published in February 2020. It
contains the text of Chapter X of the OECD TPG. The final report is the end product
of a work stream that came from Action Points 4 and 8 to 10 of the BEPS Action
Plan. As a consequence, it is a final report from the inclusive framework (137
members) and as such may be said to be a consensus document of the inclusive
framework.
The final report amends Chapter I of the OECD TPG and adds Chapter X. At the
time of writing, the OECD had not stated the procedure for adding Chapter X to
the OECD TPG or the date that it would be effective from. The addition to Chapter
I of the OECD TPG on risk-free and risk-adjusted returns has been looked at in the
earlier chapter of the manual covering Chapter I of the OECD TPG.
Chapter X provides guidance on applying the arm’s length principle to financial

transactions between associated enterprises. It covers the interaction of Chapter X
with Chapter I of the OECD TPG, treasury functions, guarantees, and captive
insurance companies. The main points of each of these sections is summarised
below.
The introduction to Chapter X states that:
‘The conditions of financial transactions between independent enterprises will

be the result of various commercial considerations. In contrast, an MNE group
has the discretion to decide upon those conditions within the MNE group.
Thus, in an intra-group situation, other considerations such as tax
consequences may also be present’.
As indicated by the above quote, intra-group financial transactions may not take
place at the arm’s length price due to the influence of other factors such as levels

of taxation around the world. This is not a new idea, in fact it is a principle that
applies to all intra-group transactions as noted in earlier chapters.
14.3 Interaction with Chapter I of the OECD TPG
You will recall from earlier chapters that Chapter I of the OECD TPG is where the
statement of the arm’s length principle is set down and contains guidance on the
importance of accurately delineating transactions. A key part of the guidance is
found in section D.1. Section B of Chapter X begins with a reminder that Article 9 of
the OECD Model DTC (see earlier chapter) not only deals with the amount of
interest on an intra-group loan, it also looks at the level. It then goes on to state
that the accurate delineation of a financial transaction will include consideration
of whether a “purported loan” is in fact equity.
It should be noted that in some countries a classification as equity may be

considered to give rise to a deemed distribution, on the grounds that the ‘interest’
is in substance a distribution of profits. The consequence is that the excess interest
is not deductible against profits and/or it may be subject to withholding tax, unless
withholding tax on dividends is restricted by the relevant double tax agreement or
restricted for some other reason such as the European Union Parent Subsidiary
Directive.
The guidance in section B of Chapter X states that, in line with Chapter I of the
OECD TPG, the arrangements in their totality need to be considered to determine
if the arrangements differ from those which would have been adopted by
independent enterprises behaving in a commercially rational manner in
comparable circumstances. The reference to commercial rationality is something
that was looked at in an earlier chapter of the manual when section D.2 of
Chapter I of the OECD TPG was covered, which includes the OECD view on non-
recognition of transactions. In the case of financial transactions, this could result in
loans being accurately delineated as equity.
Section B of Chapter X does acknowledge that different countries may take

different approaches to debt equity levels and does not seek to mandate a
standard approach (see paragraphs 10.8-10.11).
Experience in the UK, for example, suggests that there is no such thing as a
standard answer. Some industries tend to have high levels of debt, and some do
not. For one company, a 9:1 debt equity ratio might comply with the arm’s length
test, whereas for another company it might be difficult to justify debt of 1:1.
For instance, property investment companies tend to need high levels of debt to
finance their acquisitions and banks are willing to lend at high levels because the
property is available as security for the loan. In contrast, many businesses that
provide services have few assets that could be sold to repay a loan, so they tend
to have lower levels of debt. However, they may also have high levels of
profitability and so they may have strong cash flow to use to service loan interest
and repayments.
If we take the UK to illustrate a particular country approach, we can note that

historically it was known that the UK tax authority had an internal ‘rule of thumb’
under which it was relatively unlikely that they would challenge the debt levels of
UK subsidiaries if they had lower than 1:1 debt equity ratio. Debt levels above this
were more likely to receive scrutiny.
A second indicator was that scrutiny was more likely if interest cover (profits
divided by interest expense) was lower than three. However, although these

figures were never intended to be anything more than a broad guideline, they
were sometimes taken by taxpayers and even some tax inspectors to be safe
harbours, so they have now been fully repudiated.
The UK tax authorities insist that it is necessary to consider not only the question of
how much the company could have borrowed from independent lenders, but
also how much the company would have been likely to wish to borrow (which
might be lower than the amount they could have borrowed).
Whether applying a fixed threshold or the arm’s length test, it is important to use an
appropriate definition of relevant financial figures, such as debt, equity and
interest cover. Some countries set out specific definitions in their thin capitalisation
legislation or in rulings/guidance, whereas others rely on general principles.
Often, it is obvious what counts as debt and equity, but there are some grey areas,
such as preference shares, which have some characteristics of debt and some
characteristics of equity. Another grey area is debt-like instruments such as finance
leases. And should the debt be net of cash deposits?
Similarly, it is important to be clear about the definition of interest cover. It is usually

based around dividing profits or cash flow by interest, but should the profits be EBIT
or EBITDA? Or should it be free cash flow? And should the interest be the gross
interest expense or should it be net of interest income?
A functional analysis is a key part of the accurate delineation of transactions. This is

as true for financial transactions as it is for any type of intra-group transaction. The
functional analysis may need to be adapted slightly for financial transactions to
ensure that it covers factors such as credit worthiness and macroeconomic
circumstances as discussed in the previous chapter. If we take the example of an
intra-group loan, the functional analysis should examine or include some or all of
the following:
• Intra-group loan agreements
• Business plans of the borrower
• Forecast financial statements
• Financial modelling of loan servicing
• External loan agreements (of any group companies)
• Reports to external lenders
• Board papers
• Prospectuses issued by the group
• Interviews with the group Treasurer, CFO, operational managers
Carrying out a functional analysis in relation to an intra-group loan requires

different knowledge and expertise than might be relevant for other transfer pricing
analyses, and this should be borne in mind in determining who will carry out the
functional analysis. The same is true for other intra-group financial transactions.
Section B.2 of Chapter X points out that the guidance in Chapters I to III will apply
when determining the arm’s length conditions of financial transactions. This means
that a functional analysis is required; the 5 comparability factors and the

comparability analysis (all topics we have looked at in earlier chapters) will be in

point.
Paragraph 10.12 provides a list of factors that can be relevant in the accurate
delineation of financial transactions. They include:
• the presence or absence of a fixed repayment date;
• the obligation to pay interest;
• the existence of financial covenants and security;
• the source of interest payments;
• the ability of the recipient of the funds to obtain loans from unrelated lending
institutions (see paragraph 10.12 for the full list).
Paragraph 10.13 provides an example of how a loan may be accurately

delineated. It includes the fact that the position of both the lender and the
borrower must be looked at. This idea of looking at both parties to the transaction
is expanded on in the section looking at a treasury function. Anything above the
amount that would be lent/borrowed by an independent lender/borrower having
looked at all the facts and circumstances would be instead be delineated as
equity and thus would not be taken into account when determining the arm’s
length amount of interest. The possibility of no loan being made, or no borrowing
being undertaken must also be considered.
Reference is made to the financial services industry and the impact of regulations
and standards which may need to be taken into account in understanding the
commercial and financial relationships. This indicates that the guidance applies to
the financial services industry (see paragraph 10.15).
The functional analysis will indicate whether the lender is exercising control over
the risks associated with a loan. If the lender doesn’t have control over risk then,
applying Chapter I of the OECD TPG, it follows that the lender should not receive
the full reward, they should only receive a risk-free return to compensate for the
time value of money. The reward will go to the company that is exercising the
control over risk. The functional analysis will also look at the position of the borrower
to ensure they have the capability to meet the interest payments and any security
or collateral required.
As with any transfer pricing analysis, it will normally be necessary to obtain

comparability data to indicate what the interest rate margin would have been, on
an arm’s length basis.
Any transfer pricing analysis of intra-group funding will need to consider if the
impact of implicit support is relevant. This is an area that has also been addressed
in recent case law. The GE Capital case was looked at in an earlier chapter
looking at services; the case law on implicit support will be discussed further later in
this chapter.
It will often be the case that the lender in an intra-group transaction is not a bank
and is not in the business of making loans; it is generally accepted that this should
be disregarded in determining the arm’s length interest rate margin. It is difficult to
justify having a higher interest rate margin on the grounds that the lender is not in
business to make loans and would therefore require an extra reward in order to
make it worthwhile to go to the bother of making a loan to an unrelated party.

Guidance is provided on how the five comparability factors will apply. In the
section on business strategies there is an example of a company having a ten-year
loan from an associated enterprise to cover its short-term working capital needs.
On the facts it is determined that the group’s policy and practice is to use a one-
year revolving loan to manage short-term working capital. The accurate
delineation process may find that a lender would not grant a ten-year loan in
these circumstances hence the correct delineation would be that the loan be
characterised as a one-year revolving loan rather than a ten-year loan (see
paragraph 10.37).
In some cases, the group might have loans to or from unrelated parties. Whether
or not these loans are comparable with the intra-group loan that is under
examination will depend on comparison of the factors described above.
However, it is relatively rare to be able to use this approach.
Another possible approach is to use commercial databases of loan agreements to

try to find comparable loans between unrelated parties. In some cases, there may
be surveys available which contain information (often anonymised) about interest
rates, for instance within a certain industry.
An approach that is sometimes used is to ask a bank to indicate what interest rate
it would have charged if it had been the lender. However, tax authorities are often
sceptical about this, because the bank may not necessarily have carried out
enough analysis in order to give a reliable indication of the right interest rate and
the bank might also have been influenced by the answer that the multinational
group is hoping to hear. For this reason, this approach is sometimes taken a step
further by asking the bank to carry out a formal loan review and issue a loan offer
approved by the bank's credit committee. Even in this case, it can be difficult to
satisfy a tax authority that the loan offer was genuine and is reliable evidence (see
below for the OECD guidance on bank letters).
A further possible approach is to express the creditworthiness of the borrower as a

credit rating. It would be extremely expensive to pay one of the credit rating
agencies to carry out a full credit rating, but a more cost-effective approach is to
use proprietary software packages made available by the credit rating agencies
which will provide an estimate of the credit rating. This estimate is based on an
algorithm which reflects the correlation between various financial figures and
ratios and actual credit ratings given by the agency. Various information sources,
such as Bloomberg, can then be used to determine the interest rate margin that
corresponds with a given credit rating.
For an intra-group loan the creditworthiness of the borrower is of course irrelevant,

because it is unlikely that the borrower would be allowed by the group to default.
However, under the arm’s length test it is necessary to disregard this and consider
how the lender would have viewed the borrower if they were unrelated. (In the
section below on treasury functions we will consider the OECD guidance on
creditworthiness and comparability.)
14.4 Treasury Function
Section C of Chapter X of the OECD TPG looks at treasury function. The guidance
acknowledges that finance is a complex and important part of a business and
that the degree of centralisation of financial activities will vary. The treasury
function (‘treasury’) may be carried out within a division or in a separate
company. Treasury may be responsible for raising debt and equity and managing
relationships with banks and credit rating agencies (see paragraph 10.43).

A key concern of the treasury function will be liquidity. Paragraph 10.42 states that:
‘By identifying and taking action to address financial risk, treasury can help to
optimise the cost of capital to the advantage of the users of the MNE group’s
treasury services’
You will recall from an earlier chapter that Chapter I of the OECD TPG has a
section on group synergies (see paragraphs 1.157 to 1.173 (1.177 to 1.193)). The
section includes two examples of group synergies in relation to financial
transactions and expands on the mention given in Chapter VII at paragraph 7.13
which considers the impact on credit ratings of being in a group. This is reiterated
by paragraph 10.45 where treasury is stated to be a service and as such Chapter
VII (see earlier chapter) of the OECD TPG will be in point.
Where treasury acts as a central contact point for external borrowing then
comments on group synergies in Chapter I of the OECD TPG will be relevant
regarding the sharing of the benefits from such centralisation. In other cases, the
function of treasury will be more complex and the reward will reflect this.
Section C of Chapter X then goes on to look at some of the key functions that
treasury may be responsible for being intra-group loans, cash pooling and
hedging.
(i) Intra group loans
Here the guidance echoes that of Chapter I stating that all the options realistically
available need to be considered including the option of not taking out the loan.
As noted above a two-sided approach is taken. Paragraph 10.51 states that both
the lender’s and borrower’s perspective need to be considered.
The six step analysis at paragraph 1.60 of the OECD TPG will need to be applied to
determine who the risk lies with.
The considerations of an independent lender need to be taken into account even

where they may not apply in the controlled situation as the parent has control of
the subsidiary, for example the need for collateral or security. In this regard
paragraph 10.56 states:
‘If the assets of the business are not already pledged as security elsewhere, it
will be appropriate to consider under Chapter I analysis whether those assets
are available to act as collateral for the otherwise unsecured loan and the
consequential impact upon the pricing of the loan.’
The guidance also highlights the need of both parties to have the possibility of
renegotiating terms if the economic conditions change (paragraph 10.60).
The use of credit ratings is also discussed. They are viewed as a useful tool for a
lender in assessing creditworthiness. Reference is made to the fact that credit
ratings require both quantitative and qualitative data. Special circumstances such
as the presence of start-up entities in a group will need to be taken account of as
this will impact on credit ratings. It is stated at paragraph 10.65 that:
‘Financing transactions undertaken by the borrowing MNE or another entity in

the MNE group, for example the MNE group parent, will be reliable
comparables only where the differences between the controlled and
uncontrolled transactions do not materially affect the interest rate or
reasonably accurate adjustments can be made’

The credit rating of a specific debt issuance may provide a good comparable in
some cases (see paragraph 10.70). Publicly available credit ratings will only be
available for the MNE group as a whole and so may not apply to an individual
member of the group. Publicly available tools or methodologies need to be
treated with care as they are not based on direct comparables (see paragraphs
10.72 to 10.75).
The effect of group membership is said to be important for two reasons. Firstly,
group policy will impact on the type of transactions that would be entered into.
Secondly, there may be group support to assist in meeting obligations. Again,
reference is made to Chapter I of the OECD TPG and group synergies (see
paragraph 10.76). Implicit support will impact on credit ratings. It is stated at 10.78
that:
‘An MNE group member with stronger links, that is integral to the group's
identity or important to its future strategy, typically operating in the group's
core business, would ordinarily be more likely to be supported by other MNE
group members and consequently have a credit rating more closely linked to
that of the MNE group.’
The guidance goes on to note that for less important members the rating will be
somewhat different. It is concluded that the impact of implicit support will be
subjective (see paragraph 10.80).
Although covenants are less likely to be in place between associated enterprises

as there will be less information asymmetry the guidance highlights that an analysis
must be undertaken to determine ‘whether there is, in practice, the equivalent of
a maintenance covenant between the parties and the consequential impact
upon the pricing of the loan’ (see paragraph 10.86).
Turning to the methodologies, the availability of CUPs is discussed (see paragraphs

10.89 to 10.95). Paragraph 10.93 suggests that comparables for an intra-group loan
could include bond issuances, loans which are uncontrolled transactions, deposits,
convertible debentures and commercial papers. It is stated that internal CUPs
should not be ignored, however the guidance then goes on to say:
‘it is unlikely that an MNE group’s average interest rate paid on its external
debt meets the comparability requirements to be considered as an internal
CUP it may be possible to identify potential comparable loans within the
borrower’s or its MNE group’s financing with an independent lender as the
counter-party. As with external CUPs, it may be necessary to make
appropriate adjustments to improve comparability’ (see paragraph 10.95).
Loan fees and charges should also be taken account of if they would be seen in
transactions between independent parties.
Alternatives to CUP are also considered:
• A cost of funds approach which adds a profit margin to the cost of providing
the funds pointing out that the borrower wouldn’t accept an interest rate on
such an approach if funds could be obtained more cheaply elsewhere;
• Credit default swaps as a comparable however these can be subject to high

volatility;
• Economic modelling where reliable comparables don’t exist.

Bank letter/opinions are not to be regarded as evidence of an arm’s length price

as the funds have not been provided.
Case law on intra-group loans
KHO:2010:73
This case was an appeal to the Finnish Supreme Administrative Court in 2010.
A Finnish company replaced its external borrowings, on which it was paying

interest of a little over 3%, with an internal loan from a Swedish member of the
same group, on which the interest rate was 9.5%, reflecting the cost of external
funding of the group.
The court confirmed that the price of external financing for the group was not a
relevant basis for determining the interest rate that should be paid by the Finnish
company, when, on a stand-alone basis, the borrower would have received
significantly better terms given its own credit rating and other circumstances. The
borrower’s financial position had not deteriorated, and the Swedish lender was not
providing any additional services that would have justified a higher rate.
Interest-free Loans
An issue that arises from time to time is that a group wishes to make an interest-free
intra-group loan. For instance, the group might prefer not to label an injection of
capital as being equity, perhaps due to regulatory restrictions or exchange
controls. Chapter X of the OECD TPG doesn’t mention interest-free loans.
On the face of it, an interest-free loan would not be acceptable under the arm’s
length principle, because an independent lender would not normally be willing to
lend at an interest rate of 0%. However, in certain limited circumstances it may be
possible to justify an interest-free loan, on the grounds that the loan is in substance
fulfilling an equity function and therefore it is not appropriate to require there to be
interest. This is, in effect, a reverse application of the principle underlying thin
capitalisation rules.
This is, for instance, an argument that, in principle, is accepted by the UK tax
authorities, provided it can be shown that the loan is, in substance, equity. It is
helpful to be able to show that the borrower could not have obtained loan
finance from independent lenders if it were an independent company. The UK tax
authorities are usually only willing to accept this argument where there is clear
evidence that the loan is intended to remain in place in the long term, because
equity is rarely used for short-term funding.
(ii) Cash Pooling
Cash pooling was looked at in the previous chapter. Chapter X of the OECD TPG
starts with a summary of the reason for a cash pool and an explanation of physical
and notional pools. It then goes on to look at the correct delineation of the
transactions in the cash pool. Reference is made again to Chapter I of the OECD
TPG: it is noted that as cash pooling would rarely be undertaken by unconnected
parties the arm’s length principle can be difficult to apply due to the lack of
independent evidence. The options available need to be considered when
looking at the individual members of the cash pool; they wouldn’t enter into the
transaction if it would make them worse off.
Paragraph 10.118 recognises that the benefits from the cash pool can go beyond
the impact of interest rates, for example increased liquidity. The benefits that arise

may be synergistic in nature as set down in section D.8 of Chapter I of the OECD
TPG.
To determine if the synergies arise from deliberate actions and how they should be
shared it is necessary to determine (see paragraph 10.120):
i. the nature of the advantage or disadvantage;
ii. the amount of the benefit or detriment provided; and
iii. how that benefit or detriment should be divided among members of the MNE
group.
The guidance considers how long a balance should be treated as part of a cash
pool before being treated as a long-term loan as cash pools should be short term
arrangements driven by the need for liquidity (see paragraph 10.123).
It is stated that documentation is important and information on the cash pool

should be in the master file under the three-tier approach to documentation (see
later chapter).
The six step risk analysis set down in paragraph 1.60 of the OECD TPG will need to
be applied in order to work out which member of the group has the control of risk
and as a consequence work out the arm’s length reward to the various group
members.
The specific risk will include:
• Liquidity risk; this arises from the mismatch between the maturity of the credit
and debit balances of the cash pool members.
• Credit risk; this is the risk of loss resulting from the inability of cash pool members
with debit positions to repay their cash withdrawals.
Paragraph 10.130 states that in general the role of the cash pool leader is one of
coordination with limited risk and hence limited return. It is acknowledged that this
will not always be the case. Examples of possible scenarios are provided at
paragraphs 10.133 to 10.142.
Remuneration of the cash pool members will take place after rewarding the cash
pool leader. The guidance states it should be based on the arm’s length rate of
interest for debit and credit balances and as a result synergies will be allocated. It
is recognised that this will be a difficult exercise due to the lack of comparables
and again notes that interest rates may not be the only reason for entry into the
cash pool.
Paragraphs 10.147 and 10.148 look at cross guarantees that may need to be given
as part of the cash pooling arrangement. It is concluded that:
‘that the formal guarantee may represent nothing more than an

acknowledgement that it would be detrimental to the interests of the MNE
group not to support the performance of the cash pool leader and so, by
extension, the borrower.’
The borrower as a result may not have a benefit beyond the implicit support that
exists leading to the conclusion that if this is the case then no guarantee fee would
be payable.

(iii) Hedging
Hedging was looked at in the previous chapter of this manual. Chapter X outlines
the risk management role of hedging then goes on to look at it in a group context.
Hedging can be centralised into a treasury function; in this case the entity carrying
out the treasury function will enter into the hedges. If a hedge is arranged for in the
name of individual operating entities the guidance states that this will be the
provision of a service by the treasury entity.
Difficult transfer pricing issues arise where a treasury company or other group entity
enters into the transaction but not in the name of the operating entities. In this
case the positions are not matched within the same entity, although the group
position is matched. Likewise, issues will arise where there are natural hedges within
the group. These issues and their resolution are not discussed at length; the
guidance merely states that correct delineation will require a full analysis in line
with Chapter I of the OECD TPG. This analysis would involve considering all the
options available.
In summary the starting point will be to ensure that the transactions are correctly
delineated (as noted above), A functional analysis will be required to understand
which party has the economically important functions, assets and risks. The aim is
to determine what would happen if a third party were used to carry out the
hedging function. The six step risk analysis set down paragraph 1.60 of the OECD
TPG will be important as will consideration of all the options available including not
entering into the hedge.
14.5 Financial Guarantees
Section D of Chapter X begins by setting down the definition of a guarantee:
‘In the context of this section, a guarantee is a legally binding commitment on

the part of the guarantor to assume a specified obligation of the guaranteed
debtor if the debtor defaults on that obligation’ (see paragraph 10.155).
In other words, it is looking at explicit guarantees.
Accurate delineation will require identification of the economic benefits of the

guarantee that are beyond those that result from implicit support.
The guidance states that the guarantee may allow the borrower to borrow on
terms based on the credit rating of the guarantor. Where this is the case then the
comments on the arm’s length pricing of loans in section B of Chapter X would be
relevant. The benefit to the borrower may be calculated by comparing its cost of
borrowing without the guarantee to its cost with the guarantee, taking account of
transaction costs and removing the impact of implicit support. The impact may go
beyond price to the terms and conditions; as a result each case needs to be
looked at on its own facts (see paragraph 10.159).
The comments on credit ratings and implicit support above apply equally to
guarantees.
Where the analysis shows no economic benefit beyond any implicit guarantee
then no fee would be paid; the guarantor would merely be providing a service
and Chapter VII of the OECD TPG will be in point (see paragraph 10.169).
Paragraph 10.170 looks at the position when a guarantee leads to a larger amount
of debt being available. It states that in this case the correct delineation may be

that part of the borrowing is a capital contribution and only the part that is classed
as a loan is relevant for calculating the amount of any guarantee fee.
The section on group membership once again states that anything less than an
explicit legally binding guarantee will not justify a fee. Letters of comfort are not
legally binding. Paragraph 10.164 gives examples where an explicit guarantee
would not be paid for.
As stated in the section on cash pooling, the guidance indicates that a fee would
not be paid for cross guarantees and that any support in the event of a default
would be a capital contribution (see paragraph 10.165).
Accurate delineation will also require a consideration of the financial capacity of

the guarantor to meets its obligations under the guarantee. The business
correlation between the guarantor and the borrower also needs to be considered.
The section on pricing includes a reminder that Chapter II of the OECD TPG should
be applied including a full functional analysis. It describes a number of pricing
approaches for those circumstances where a guarantee is found to be
appropriate:
• CUP is potentially the most reliable method. However, there may be difficulty
in finding internal and external comparables.
• Yield approach looks at the spread of the interest rate with and without the
guarantee. As noted above it will be necessary to take account of implicit
support and transaction cost. The result of this analysis sets a maximum fee for
the guarantee that the borrower would be prepared to pay which isn’t
necessarily the arm’s length amount that would come from negotiation.
• Cost approach looks at the cost that the guarantor would expect if the
borrower defaults or alternatively the additional capital that the guarantor
would require for the risk assumed. It then looks at some financial models that
can be used. The guidance concludes:
‘Whatever valuation model is used, the evaluation of cost method sets a

minimum fee for the guarantee (the minimum amount that the provider of the
guarantee will be willing to accept) and does not of itself necessarily reflect
the outcome of a bargain made at arm’s length.’ (See paragraph 10.180).
• Valuation of expected loss approach looks at the expected default and

recovery rates in the event of default that may be available in the market
place. The price is then calculated using financial modelling such as the
capital asset pricing model (‘CAPM’). This is a theoretical model that
compares risk and return. CAPM is used to calculate an appropriate discount
or capitalisation rate, which is then used in the capitalised cash flow or
discounted cash flow valuation methods.
• Capital Support Method calculates the additional capital that would be

needed to bring the borrower’s capital up to that of the guarantor, then
calculates the expected return on that amount of additional capital.
Examples of the pricing methods are given in paragraphs 10.184 to 10.187.

14.6 Case Law and Guarantees
As noted above and in earlier chapters, group membership often gives rise to
synergies and implicit guarantees. As a result when trying to find the arm’s length
price for an explicit guarantee it is necessary to take account of the benefit that
would have arisen from an implicit guarantee. This is because of a legal decision
involving the Canadian subsidiary of GE Capital (GE Capital [2010 FCA 344]). It is
clearly stated in the OECD TPG that where a benefit is derived by a group
company from mere passive association with the rest of the group, this does not
justify any fee being charged for this benefit. The OECD TPG give an example
where it is perceived that there is an implicit guarantee by the group and
therefore a group company is able to borrow at lower interest rates than it might if
it were a stand-alone company (as noted above). They make clear, however, that
where the benefit is brought about by specific action by another group company,
such as giving an explicit guarantee, this is a service for which a fee should be
expected.
The GE Capital judgement takes this principle a step further by saying that even
where an explicit legally-binding guarantee has been given it is still necessary to
determine the benefit gained from this guarantee by comparing the actual
interest rate paid by the borrower with the rate that it would have paid if there
had been no explicit guarantee, but the borrower was still a member of the group
and therefore would potentially still benefit from an implicit guarantee.
It would be a good idea to refer to the earlier chapter on services to refresh your
understanding of the main issues and implications from this important case; you will
also find more detail in the case law appendix.
In Chevron Australia Holdings Pty Ltd v Commissioner of Taxation (No. 4) [2017]

FCAFC 62, an Australian transfer pricing ruling provided guidance on what
constitutes 'fair loan terms' among related parties — in this instance, a USD $2.5bn
credit facility from a US entity to an Australian entity. The arrangement permitted
the US entity to raise funds at a low rate, due in part to an explicit guarantee from
the ultimate holding company of both the US and Australian entities, and lend to
the Australian entity at a higher rate of interest. The company that borrowed the
funds at the lower rate, Chevron Texaco Funding, made a profit from the loan on
to Chevron Australia Holding PTY (CAH). This profit was later paid as a dividend to
CAH and was tax free in Australia.
The taxpayer wanted the borrowing entity to be looked at on a stand-alone basis.

However, the Australian tax authority (ATO) contended that the comparison
should be to a company in the same situation, that is to say a member of a group
with implicit support.
The judge considered whether an independent lender would have taken into
account the ‘implicit support’ of a parent to a subsidiary, with the following points
being noted from expert witnesses: In the absence of a legally binding parental
guarantee, implicit credit support was found to have ‘little, if any’ impact on
pricing by a lender in the real world. One of the key reasons that agency ratings
are not solely relied upon by banks when determining credit ratings is precisely
because they may improperly give allowance for implicit support. This contrasts to
the GE Capital case where more importance was given to the impact of implicit
support.
The issues contested included that the arm’s length rate of interest for the loan was
lower than that charged by the Lender and that an arm’s length loan agreement
would have contained other terms which reduced the amount payable by the

Borrower (such as that the loan would have been in USD and would have
contained financial covenants).
Another key issue was whether the features of the hypothetical transaction should
include the provision of security or other covenants in favour of the lender as the
inclusion of this feature can affect the conclusion as to the arm's length interest
rate for the purposes of applying the transfer pricing rules.
In April 2017 Chevron lost its appeal to Australia’s Full Federal Court (FFC). The three
Federal Court judges agreed with the initial trial judge. Matters considered by the
Court included independence assumptions required by the Australian rules – the
Court concluded these include an assumption that the parties to the loan
arrangement are independent. The FFC concluded that the ‘independence’
hypothesis does not necessarily require the detachment of the taxpayer, as one of
the independent parties, from the group which it is part of, or the elimination of all
the commercial and financial attributes of the taxpayer. The FFC also agreed that
the parent would have sought to reduce the cost of borrowing. The evidence
revealed that the borrower was part of a group that had a policy to borrow
externally at the lowest cost and the parent would generally provide a third party
guarantee for a subsidiary borrowing externally.
To summarise the above, the Court found that the loan was not a genuine ‘arm’s
length’ transaction and the taxpayer had breached transfer pricing provisions of
tax legislation. The FFC found that the Australian subsidiary would not have been
able to borrow the amount it did without the backing of the ultimate parent; that
account must be taken of the group policy to provide guarantees and that as a
result the borrowing cost of the Australian subsidiary would have been lower than
that under the intra-group agreement. The broader implications of this case
include:
• That an Australian subsidiary of a multinational group is not to be treated as if it

were an ‘orphan’ when undertaking transfer pricing arm’s length calculations.
In fact, some level of parental support may need to be assumed to exist
depending on the facts/situation.
• If a borrower is part of a group that has a policy to borrow externally at the

lowest cost and that has a policy that the parent will generally provide a third
party guarantee for a subsidiary that is borrowing externally, then in the case
of related party debt, the interest rate payable by an Australian subsidiary
should probably be set at a level that assumes a parent company guarantee
has been given.
• The appropriate interest rate on internal debt will likely be closer to the
parent’s global cost of funds for the relevant currency.
In August 2017 Chevron announced that it would not appeal. The details of the
agreement reached by Chevron with the ATO were not revealed. However, It is
understood that the ATO achieved a reduction in the coupon rate for the loan.
The Chevron case is important. You will find more detail on this case in the case
law appendix.
However, in some cases, a guarantee might also have induced the lender to lend
a higher amount than it would have lent to the borrower on a standalone basis.
Therefore, thin capitalisation issues might also arise, although this will depend on
the wording of the relevant thin capitalisation legislation. In the UK, thin
capitalisation is dealt with under the transfer pricing rules and guaranteed loans
are specifically caught. The UK approach is that the borrower should only be

entitled to an interest deduction for the interest on the portion of the loan that it
could have borrowed on a standalone basis. Accordingly, it would only make
sense for the guarantee fee to reflect the benefit from the reduced interest rate.
The guarantor might be entitled to a deduction for the rest of the interest, if it is a
UK company, as the UK rules allow for a guarantor to be treated as if were the
borrower in respect of disallowed interest.
An alternative approach would be that the borrower should be entitled to a

deduction for the interest on the extra debt that it was only able to borrow
because of the guarantee, provided it can be shown that a similar guarantee
would have been available on an arm’s length basis. If so, then determining the
benefit to the borrower would be more complicated, because part of the benefit
is being able to borrow more than it could otherwise have borrowed.
The OECD view as stated above is that there will in effect be an injection of capital
equal to the extra amount that can be borrowed.
14.7 Captive Insurance
Captive insurance was looked at in the previous chapter. Section E of Chapter X

looks at captive insurance and reinsurance captives (fronting)(not covered in the
previous chapter).
The guidance states that:
‘captive insurance is intended to refer to an insurance undertaking or entity

substantially all of whose insurance business is to provide insurance policies for
risks of entities of the MNE group to which it belongs’ (see paragraph 10.189).
In the context of the guidance, reinsurance refers to a reinsurance undertaking or

entity the purpose of which is to provide reinsurance policies for risks of unrelated
parties that are in the first instance insured by entities of the MNE group to which it
belongs.
It is recognised that regulatory requirements may impact and state that these may
be lighter for captives.
The motivation for using a captive insurer includes cost savings, stabilisation of
premiums and tax and regulatory arbitrage. The guidance states that where the
captive is used because some risks cannot be insured with a third party, then the
possibility of non-recognition as set down in section D2 of Chapter I of the OECD
TPG, needs to be considered (see paragraph 10.194).
The section on accurate delineation mainly concentrates on captive insurance

companies. The guidance points out that the captive is not taking on the risk of
the insured party, it is merely taking on the risk of insuring that risk. The insurer can
take action to minimise that risk such as diversifying its portfolio of risk or re-
insurance (see paragraph 10.197).
The guidance then goes on to look at the indicators of whether a captive

insurance company is actually insuring a risk. In summary these are:
• Diversification and pooling of risk
• There is a real economic impact for the MNE group as a whole
• Both the captive insurance and any reinsurer are regulated entities

• The insured risk would otherwise be insurable outside the MNE group
• The captive insurance has the requisite skills, including investment skills, and
experience at its disposal (see paragraph 10.213)
• The captive insurance has a real possibility of suffering losses
See paragraph 10.199 for the full list.
The guidance goes on to look in detail at the assumption of risk and risk diversion. It
sets out various ways that risk can be diversified and states that as risk
diversification is core to an insurance business, a lack of it would indicate that a
captive was not providing insurance. A lack of reserves to meet the additional risks
represented by the relatively less diversified portfolio of the MNE group may also
indicate that it is not carrying on an insurance business (see paragraph 10.208).
The economically relevant risks associated with issuing insurance policies, include
inter alia, insurance risk, commercial risk or investment risk. To accurately delineate
the insurance transaction these risks must be identified with specificity. In looking at
these, the guidance refers to part IV of the 2010 OECD Report on the Attribution of
Profits to Permanent Establishments (some aspects of this report will be looked at in
a later chapter on attribution of profits to a permanent establishments) which
describes some of the activities connected with an insurance business. The control
over these activities will need to be determined in accordance with paragraph
1.65 and 1.66 of Chapter I of the OECD TPG. You will recall from an earlier chapter
that this involves asking whether there are people to perform the decision making
relating to the mitigation of risk and the capability to carry out the decision. There
needs to be some substance in the insurer, not a mere formal agreement to
contracts. In some cases the risk of the insurance may need to be allocated to
another member of the MNE (see paragraph 10.212).
Outsourcing of underwriting is allowed in some instances. The guidance states that

a captive insurance that outsources all aspects of the underwriting process
without performing control functions would not assume the insurance risk under
Chapter I analysis.
The guidance includes two paragraphs on reinsurance (fronting), where it is noted

that this is a complex area.
Certain insurance risks must be placed with regulated insurers as a legal

requirement. This may lead to the use of a fronting arrangement in which the first
contract of insurance is between the insured member of an MNE group and an
unrelated insurer (the fronter). The fronter then reinsures with the captive insurance
most or all of the risk of the first contract.
The fronter retains a commission to cover its costs and to compensate for any
portion of the insured risk which it retains. The majority of the fronter’s premium
passes to the captive insurance as part of the reinsurance contract.
The main transfer pricing issues are said to relate to whether the transactions
involved amount to genuine insurance or reinsurance and, if there is genuine
insurance, whether the premiums payable (ultimately to the reinsurance captive)
are on arm’s length terms. (See paragraphs 10.214 and 10.215.)

Turning to pricing of captive and reinsurance the guidance puts forward several
approaches pointing out that each case will need to be looked at individually.
The suggestions set out in section E.3 of Chapter X are:
• Pricing of premiums. One suggestion is to look for an internal or external CUP. It

is acknowledged that there may be practical difficulties due to functional or
capital volume differences. The alternative would be to use actuarial analysis;
it is noted that the outcome of this analysis would not represent actual
transactions between unconnected parties and thus comparability
adjustments would be needed.
• Combined ratio and return on capital. This approach has two stages to it.
Stage 1 starts by looking at the ratio of claims plus expenses to premiums; this is
then bench marked. This gives the underwriting profit which is then added to
an investment return found in Stage 2. (See paragraph 10.220 for the full
detail.)
• Group synergy. Where the captive insurance insures the risk and reinsures it in
the open market, it should receive an appropriate reward for the basic
services it provides. The remaining group synergy benefit should be allocated
among the insured participants by means of discounted premiums. The return
to the captive will depend on the role it plays. The amounts payable to group
entities could be on the ratio to premiums paid.
• Agency Sales. This covers the situation where the insurer has access to
customers through an agency that is a group member as in the 2009 UK case
of DSG Retail and Others v HMRC (see below). The guidance concludes that
the arm's length remuneration for the insurer would be in line with the bench
marked return for insurers insuring similar risks and the balance of the profit
should be allocated to the associated company (see paragraph 10.226).
It can be noted that tax authorities are sometimes sceptical about a CUP
approach under which the captive insurer charges insurance premiums in the
same way that an independent insurance company might do. They sometimes
argue that the captive insurer should not be viewed as taking the same risks as an
independent insurance company and a small cost plus-type reward on the
captive insurer’s own running costs would be more appropriate. The correct
position will depend on a very careful analysis of the facts.
An example of this sort of argument is the 2009 UK case, DSG Retail Ltd & Others v
HMRC (this case was looked at in detail in an earlier chapter). This related to a
captive insurer in the Isle of Man which insured (in some years, reinsured) extended
warranties sold to customers in a chain of electrical retail shops in the UK. DSG
based the premiums on what it considered to be comparable uncontrolled
transactions, being the premiums charged by independent companies that
provide extended warranties.
The Special Commissioners (the name then given to the lowest level of court for
tax cases) decided that the premiums charged by independent extended
warranty providers were not comparable, because they found that the bargaining
power was different. They took the view that if the UK retailer had been
negotiating with an independent extended warranty provider the UK retailer
would have had most of the bargaining power because the best opportunity to
sell an extended warranty to someone who has just bought, say, a television is to
sell the warranty whilst the customer is standing at the cash till, paying for the
television. They accepted evidence that the extended warranty providers
generally sold their extended warranties via the product manufacturer and they
considered that the balance of bargaining power would be different in this

circumstance. (This point appears to be crucial in the decision, but the case report
does not explain in any detail whether this distinction was just an assertion that was
accepted or there was hard evidence that there is indeed a difference in
bargaining power.) They also found as a matter of fact that the risks involved in
writing large numbers of extended warranties are low, because the claims cost
does not fluctuate much from year to year.
They concluded that on an arm’s length basis the UK retailer would have
negotiated a deal under which the insurer received just a small return on its capital
and the remainder of any profit from the extended warranties would be made by
the UK retailer. This effectively meant that it was held that the arm’s length level of
premium was far lower than the actual premiums paid.
14.8 UN Manual and Financial Transactions
You will recall from earlier chapters that the UN Manual aims to provide a practical
approach to the application of applying transfer pricing principles. The UN
released an update to the manual in April 2021. The update includes a new
section B.9 that covers financial transactions.
Part B Chapter 9 of the UN Manual looks at financial transactions. In summary, the

UN Manual covers similar areas to the OECD TPG, including the process of actual
delineation and relevant characteristics of financial transactions, credit rating,
potential transfer pricing methods, financial guarantees, and cash pools. There is
also a section on simplification measures and safe harbours. It does not cover
hedging.
Although there is not a discussion on thin capitalisation, there is a discussion on the

approach that may be taken to characterise transactions as either debt or equity.
The manual suggests that credit rating could be used as an indication of debt
capacity. There is a useful section on credit rating including illustrative examples in
section B9.8
B9.4 includes an overview of common intra-group financial transactions and a

classification of different type of treasury entities/departments. Paragraph 9.4.4
suggests that the most appropriate method for treasury departments/entities
operating as service centres would be CUP or cost plus or TNMM based on cost.
Section 9.10 looks at methodologies for loans stating that CUP is the most used. For
treasury services it is suggested that cost plus to cost based TNMM may be best,
noting that cost plus may not be appropriate if the treasury function takes on risk.
Profit split is stated as not normally being appropriate for intra-group financial
transactions.
As noted above, safe harbours for loans and simplification measures are discussed
(see B9.11) and caution is advised. B9.12.2.8 describes a ‘sixth method’ for loans
using a published rate, eg the US Treasury Bond rate as a proxy for rates between
uncontrolled parties. As with the commodities ‘sixth method’ (see earlier chapter)
the use of a published rate would eliminate the need to search for comparables.
Credit rate default swaps and economic modelling are also put forward as options
for pricing intra-group loans
B9.12.3 looks at the interplay between loans and other intra-group transactions. It
contains a couple of useful worked examples.
B9.13 looks at guarantees; it covers implicit, explicit, upstream guarantees (where

a group company issues a guarantee to external creditors for the benefit of its

parent company where the latter enters into agreements with the external
creditors) and downstream guarantees (where a parent company issues a
guarantee to external creditors for the benefit of one of its subsidiaries). Letters of
intent /comfort letters and keep-well agreements (a contract between a parent
company and its subsidiary to maintain solvency and financial backing throughout
the term set in the agreement; these agreements are sometimes referred to as
comfort letters) are not treated as guarantees. B9.13.2.4 and B9.13.2.5 contain a
useful summary of when a payment for a guarantee will and will not be allowed.
The methods put forward for the pricing of a guarantee fee are in line with those in
the OECD TPG. It is noted that in order to determine an arm’s length guarantee
fee, or an arm’s length range of guarantee fees, the application of more than one
approach is not a requirement (B9.13.2.9).
This would be a good time to have a look at chapter B9 in the UN Manual.

CHAPTER 15
SPECIFIC TRANSACTIONS: INTANGIBLE PROPERTY
In this chapter we will look at transfer pricing in relation to Intangible Property, in particular:
– the life cycle of intangibles
– development of intangibles
– exploiting intangibles: Principal Structure v Licensing Out
– valuation of intangibles
– case law on valuation of intangibles
15.1 Introduction
The simplest definition of an intangible is that it is something that cannot be

physically touched. The UN Manual defines it as property with no physical
existence but whose value depends on the legal rights of the owner.
Intangibles have been at the centre of several debates and court cases in recent
years. The increasing attention of tax authorities on intangibles is mainly due to
intangibles gaining more and more value as part of large multinationals’ asset
base.
Globalisation and increasing competition have led large MNEs to work harder on
differentiating themselves from the competition and investing more in intangibles
to achieve the required competitive advantage.
As a result, ensuring that intra-group transactions involving intangibles are

thoroughly planned and priced is key in minimising the risk of tax adjustments and
penalties in case of non-arm’s length results.
Among the transfer pricing transactions of MNE Groups, transfer prices for
intangibles are the most significant and susceptible to manipulation. This is a result
of the high value and mobility of intangibles and the complexity of intangible-
related issues. Intangibles carry high value because they often produce or have
the potential to boost profitability as they provide the MNE with a competitive
advantage.
Given that an intangible asset does not have a physical presence, it is easily
transferable from one country to another. Intangible-related financial issues exist in
commercial practices, valuation, and accounting as well as in attribution of
income for tax purposes.
For example, MNE Groups often attribute research and development (R&D)
expenses to higher-tax countries which provide immediate expensing of these R&D
costs. However, in reality, the R&D costs of producing intangibles may be widely
dispersed among related entities. Subsequent transfer prices charged through
royalty fees to affiliate MNEs often fail to adequately adjust for the real risk
premium assumed for the original development of the intangibles. In other
situations, to increase deductions in a higher-tax country, the MNE Group might
impose higher transfer prices on a related MNE operating in a lower-tax country.
This shift is made possible by service charges, royalties paid to the owner or licensor
of the intangibles, or through cost-sharing arrangements.

As international business has become more complex and integrated, intangibles

have become immensely important. MNE Groups are sometimes able to generate
substantial royalties from an external party through a license of intangibles. The
value of patents and other intangibles can be as high as 70% of the value of the
average MNE Group. Intangibles subject to transfer pricing are specifically defined
as we will see below.
15.2 The Life Cycle of Intangibles
When it comes to tangible assets it is easy to see their life cycle for a company.
Normally it begins when they are bought, then they are used, accounted for and
depreciated then normally the final step will be their sale. Intangibles also have a
life cycle. Generally we can identify similar stages for an intangible as a tangible
asset. To begin with they will either be bought or developed, they will then need to
be recorded and a valuation arrived at. The intangible will then be used in the
business and where appropriate amortised. Finally it will expire or be sold.
15.3 Development of Intangibles
As noted above an intangible is a moveable asset that provides opportunities for

tax planning right from the start. There are two main ways that intangibles can be
developed – via contract R&D arrangements or via cost sharing arrangements.
Contract R&D
The typical scenario will be that the MNE will set up a separate entity to provide
the contract R&D services. The MNE will often concentrate R&D into a few
locations as this helps to increase efficiency and keep down cost. However, where
the plan is that the benefits of the R&D will remain with the principal (normally
located in a low tax territory) the tax authorities will look carefully at the
arrangement. A third party contract R&D provider would have a contract and
hence it is important that there is a contract to support the in-house contract R&D
arrangement. Ownership of an intangible is a complex legal subject. In some
jurisdictions it will belong to the person who develops it – so this is a key aspect of
the contract with the contract R&D provider. Control is another area that the tax
authorities will look at closely. Many tax authorities would expect the principal to
retain some control of the development of the R&D for there to be a true R&D
contract relationship.
← Payment – royalty
PRINCIPAL
Licence
↑ ↓
Payment for
Contract R&D Intangible
contract
services Users
R&D
↓
CONTRACT
R&D PROVIDER

Cost Sharing/Cost Contribution Arrangements
Cost sharing/cost contribution arrangements are another way to centralise

development of intangibles. Again such arrangements offer commercial
advantages of efficiency and cost saving; equally they can offer tax benefits as
well. The participants in a cost contribution arrangement will agree to share the
cost of development and the costs contributed by each participant are normally
proportionate to the benefits they expect to receive in the future.
R&D
↑ Costs
Benefits ↓
COMPANY A COMPANY B COMPANY C
We will look at the OECD TPG and Cost Contribution Arrangements in the next
chapter.
15.4 Exploiting Intangibles: Principal Structure V Licensing Out
A typical principal structure will involve setting up a subsidiary in a low tax territory.
The principal will be the company that owns the intangible thus the principal will
earn the profits from the intangible. In the context of the intangible, the principal
company acts like a distributor as it receives income from the licenses as the other
companies in the group operate as the intangible developers. The agreement
between the companies and the allocation of risk will be important for ensuring
that the structure is tax efficient with income allocated to the low tax jurisdiction
conforming with the arm’s length principle.
Licensing of intangibles within a group can lead to questions as to the arm’s length
principle for the license to use the intangible. If the group company is a fully-
fledged manufacturer then a separate license fee would be required as a third
party would need to pay such a fee. Without a central policy on intangibles it
could be that the fee ends up in a high tax territory. A structure that results in
centralisation of assets including intangibles would mean that the manufacturer
would be set up as a contract manufacturer and a separate licence fee is not
required.
This structure can give more scope for keeping the receipt of licence fees in a low
tax territory. However it may be that the manufacturer is in a high tax territory in
which case a separate license fee may be the preferred option.

15.5 OECD TPG and Intangibles
Chapter VI contains 4 sections as follows:
A. Identifying intangibles.
B. Ownership of intangibles and transactions involving the development,

enhancement, maintenance, protection and exploitation (DEMPE) of
intangibles.
C. Transactions involving the use or transfer of intangibles.
D. Supplemental guidance for determining arm’s length conditions in cases

involving intangibles.
There is also an Appendix with 29 examples.
Paragraph 6.6 gives the definition of intangibles as:
‘something which is not a physical asset or a financial asset which is capable

of being owned or controlled for use in commercial activities, and whose use
or transfer would be compensated had it occurred in a transaction between
independent parties in comparable circumstances.......’
It is clear from this that it is not just a case of ownership that will lead to the
recognition of an intangible.
Paragraph 6.8 states clearly that it is not necessary for an intangible to be capable
of being transferred separately for it to be characterised as an intangible in
transfer pricing:
‘.....while some intangibles may be identified separately and transferred on a

segregated basis, other intangibles may be transferred only in combination
with other business assets. Therefore, separate transferability is not a necessary
condition for an item to be characterised as an intangible for transfer pricing
purposes.’
Examples are given in paragraph 6.27 and include goodwill as the value of future
economic benefits associated with business assets that are not separately
identified and recognised. Future trade from existing customers is also mentioned.
Martin Zetter writing in Tax Journal in October 2014 points out that: ‘the absence or
presence of goodwill does not determine whether an intangible exists or not.
Where an intangible is not owned or controlled, it is necessary to consider whether
the use or transfer would be compensated by independent parties. For example,
when goodwill is reputational value and is transferred to or shared with an
associated party, then it should be taken into account in determining arm's length
compensation.’
The glossary to the OECD TPG includes a definition of the term ‘marketing
intangible’:
‘An intangible (within the meaning of paragraph 6.6) that relates to marketing
activities, aids in the commercial exploitation of a product or service, and/or
has an important promotional value for the product concerned. Depending
on the context, marketing intangibles may include, for example, trademarks,
trade names, customer lists, customer relationships, and proprietary market

and customer data that is used or aids in marketing and selling goods or
services to customers.’
Paragraph 6.17 states the definition of ‘unique and valuable’ intangibles as:
‘those intangibles (i) that are not comparable to intangibles used by or

available to parties to potentially comparable transactions, and (ii) whose use
in business operations (eg manufacturing, provision of services, marketing,
sales or administration) is expected to yield greater future economic benefits
than would be expected in the absence of the intangible.’
Chapter VI provides illustrations of intangible assets. It states that the illustrations

are intended to clarify the provisions relating to identification of intangibles; it is not
a complete listing and is subject to local legal and regulatory requirements. It goes
on to state that this listing should not be used as a substitute for a detailed analysis.
The illustrations are as follows:
• Patents
• Know-how and trade secrets
• Trademarks, trade names and brands
• Rights under contracts and government licences
• Licenses and similar rights
• Goodwill and ongoing concern value
The following were rejected as intangibles:
• Group synergies
• Market specific characteristics
The functional analysis will indicate if the legal owner of the intangible is
undertaking all aspects of the DEMPE.
It is not a requirement that the legal owner should perform all aspects of the
DEMPE. However, if some aspects of the DEMPE are outsourced it follows that
some of the economic returns from exploiting the intangible will go to the entities
that perform the outsourced functions, the reward for each party being based on
the functions performed, assets used and risks undertaken (see paragraphs 6.50 to
6.58).
One of the assets used may be funding; this may be provided by one group
member. Other group members may undertake the functions and risks associated
with the DEMPE. If the funder merely provides funding and doesn’t perform other
functions this will impact on the returns it receives. It will be important to analyse risk
when considering the return to the entity providing the funds. In discussing this (see
paragraphs 6.59 to 6.61) reference is made to Chapter I and the importance of
the six-step process at paragraph 1.60. You will recall that Chapter I also includes a
section on risk-free and risk-adjusted returns. Paragraph 1.117 refers to paragraph
6.61 making it clear that in the case of funding for intangibles, in some cases the
funding company will only be entitled to a risk-free return if it doesn’t take on any
of the risk associated with the DEMPE of the intangible.
When considering the reward to the funder, it is important to distinguish between

the financial risk of funding and the operational risk of developing the intangible;

the funder doesn’t always take on this risk. This is discussed further below including
the tests set out in paragraph 6.71.
Applying the arm’s length principle to intangibles is particularly difficult. Paragraph

6.33 notes some of the factors that make applying Chapters I to III of the OECD
TPG challenging. In summary they are:
• A lack of comparable transactions.
• Intangibles may be owned and/or used by different members of the MNE.
• It is difficult to isolate the impact of any particular intangible on income.
• Various members of a MNE may perform different aspects of the DEMPE in

relation to an intangible.
• Contributions to the value of intangibles can be made over several years,

which may not be the same as the years in which returns are realised.
• Contractual terms for ownership, funding and risk may differ from what would
be found in independent enterprises.
To try to deal with these difficulties there is a six step analysis to analyse
transactions involving intangibles between associated enterprises. This framework is
comparable to that introduced in Chapter I of the OECD TPG for analysing risks,
consisting of the following six steps (see paragraph 6.34):
1. Identify the intangibles used or transferred in the transaction with specificity;
2. Identify the full contractual arrangements with special emphasis on

determining the legal ownership of intangibles;
3. Identify parties performing functions, using assets, and assuming risks related to
development, enhancement, maintenance, protection, and exploitation of
the intangibles through a functional analysis;
4. Confirm the consistency between the terms of the contractual arrangements

and the actual conduct of the parties;
5. Delineate actual controlled transactions in light of the legal ownership, other

relevant contractual relations, and the actual conduct of the parties; and
6. Where possible, determine arm’s length prices consistent with each party’s
contribution of functions performed, assets used, and risks assumed.
Have a look at paragraph 6.42 which points out that returns do not always go to
the legal owner of the intangible.
The key factor will be the functional analysis.
Paragraph 6.71 contains guidance on which entity or entities within a multinational

group should share in the economic returns from exploiting intangibles. It sets out
the following tests to be met for a company to be entitled to all of the anticipated,
ex ante, returns derived from the MNE group’s exploitation of the intangible.

The company should:
• Perform and control all of the functions (including the important functions
described in paragraph 6.56; these include control over strategic decisions
and being responsible for determining the course of ‘blue-sky’ thinking)
related to the development, enhancement, maintenance, protection and
exploitation of the intangible;
• Provide all assets, including funding, necessary to the development,

enhancement, maintenance, protection, and exploitation of the intangibles;
and
• Bear and control all of the risks related to the development, enhancement,
maintenance, protection, and exploitation of the intangible.
The guidance distinguishes between ex ante (anticipated) and ex post (actual)

returns, clarifying that the compensation that must be paid to members of the
MNE group that contribute to the development, enhancement, maintenance,
protection and exploitation of intangibles is generally determined on an ex ante
basis. The allocation of actual ex post profit or loss will depend on the facts and
circumstances of the case.
Paragraph 6.81 gives guidance on payment for use of a company name as

follows:
‘As a general rule, no payment should be recognised for transfer pricing

purposes for simple recognition of group membership or the use of the group
name merely to reflect the fact of group membership.’
Section B discusses marketing intangibles and the entitlement to a share of the

return from those intangibles. The test is ‘what an independent distributor would
obtain in comparable circumstances’ (see paragraphs 6.76 to 6.78).
Section C sets downs the two main types of transactions considered relevant for
purposes of identification and characterisation of specific transactions involving
intangibles:
i. transactions involving transfers of intangibles or rights in intangibles, and
ii. transactions involving the use of intangibles in connection with the sale of
goods or the provision of services.
Section D provides further guidance on the pricing of intangibles which will

supplement that given in Chapters I to III. It is broken down into 5 sections as
follows:
1. General principles applicable in transactions involving intangibles;
2. Supplemental guidance regarding transfers of intangibles or rights in

intangibles;
3. Arm’s length pricing when valuation is highly uncertain at the time of the
transaction;
4. Hard to Value Intangibles (HTVI); and
5. Supplemental guidance for transactions involving the use of intangibles in

connection with the sale of goods or the provision of services.

Section D2 sets down some specific comparability factors that have to be

considered:
a. Exclusivity
b. Extent and duration of legal protection
c. Geographic scope
d. Useful life
e. Stage of development
f. Rights to enhancements, revisions, and updates
g. Expectation of future benefit
Looking at risk in relation to comparability the six step analysis that is found in
paragraph 1.60 of the OECD TPG must be applied. The risks that need to be
considered for intangibles include:
• risk related to future development
• risk related to product obsolescence
• risk related to infringement of rights to the intangibles
• risk relating to product liability
In respect of transfer pricing methods, Section D2 refers back to paragraph 2.10 of

the OECD TPG on not using rule of thumb and goes on to state:
‘The transfer pricing methods most likely to prove useful in matters involving
transfers of one or more intangibles are the CUP method and the
transactional profit split method. Valuation techniques can be useful tools’
(See paragraph 6.145).
Section D5 of Chapter VI looks at the choice of appropriate methodology for

transactions involving the use of intangibles in connection with the sale of goods or
provision of a service. Guidance is then given on the application of CUP where
comparables exist, profit split methods and the use of valuation techniques.
Examples looking at the use of profit split methods and valuation techniques are
included in the Annex to Chapter VI.
15.6 Valuation of Intangibles
The guidance covers the use of valuation techniques. The OECD TPG recognise
that they can be very useful in helping estimate the arm’s length price when a
CUP is not available (see paragraph 6.153).
The valuation of intangibles is difficult to carry out without an element of

subjectivity and is often subject to scrutiny by the auditors and tax authorities. The
economic value of intangibles is primarily determined by the economic and legal
environment in which the intangibles are created and exploited, the market
demand for the intangibles and the existence or absence of close substitutes.
Valuation experts usually identify assumptions in establishing value such as

expected future earnings estimates, the rate of the average cost of capital, and

other factors including the discount rate. The valuation of intangibles is also
affected by its tax treatment.
The value of intangibles can often fluctuate depending on the market,

competitors, etc. The fluctuation occurs not only over time, but at any one time
depending upon the key assumptions of the inherent risks associated with the
intangibles. These risks can include liability concerns or the possibility that
competitors will create new and better products.
Thus, the OECD TPG recognise that it is often difficult to attribute a distinct value to
each intangible on an ongoing basis.
The valuation of intangibles poses difficulties for transfer pricing decision making
and government oversight for three major reasons:
• Comparables for such assets seldom exist. Patents are rarely traded on
external markets. Usually MNEs are unwilling to sell their patents, but might
license out some of the rights to use the intangible asset.
• Intangible rights are often transferred in combination with tangible assets or

services, known as ‘embedded intangibles’. Buyers may want to acquire a
product that relies on a combination of intangibles and other assets.
• Intangibles other than patents are particularly difficult to detect because they
are not reported in financial statements.
As mentioned above Section D3 of Chapter VI looks at how the arm’s length price
can be found where the transaction involves intangibles where the value is highly
uncertain at the time of the transaction.
Paragraph 6.182 of the OECD TPG suggests that a possibility for dealing with
uncertainty is to use an anticipated benefits approach. This approach takes into
account all relevant economic circumstances at the outset to establish the price
at this point. The OECD TPG suggest that it may be that independent enterprises
would agree that subsequent outcomes are sufficiently predictable to fix the
price.
If this approach is not possible the OECD TPG go on to suggest that short term
agreements be entered into with price adjustment clauses (see paragraph 6.183).
The final possibility put forward (in paragraph 6.184) is that independent enterprises
might be willing to take some risk at the outset with a joint understanding that
there will be a renegotiation of the pricing agreement if required later. The
paragraph includes an illustration of when this might happen.
There are a number of generally accepted approaches to ascertaining the fair

market value of intangibles. These approaches are accepted by most tax
administrations and are consistent with generally accepted accounting principles
in most jurisdictions:
• Income based approach;
• Market based approach; and
• Cost based approach.

Income Based Approach
An income based approach seeks to generate a single present value from the
quantum, duration and risk associated with expected future economic benefits of
the business asset. An appropriate discount rate must be applied reflecting a rate
of return on investments appropriate to the asset being valued and the relevant
market conditions. It is generally appropriate to look at the rates an acquirer
would receive on similar investments.
The discounted cash flow (DCF) method in particular involves a rigorous review of
projected performance and is often preferred as a valuation method where
credible financial data is available.
A ‘multiples’ income method is often applied in estimating future sustainable

economic benefits (eg a profits multiplier). This method has the advantage of
permitting the asset to be compared to other internal or external valuations or
transactions for consistency. However, in isolation, this valuation method will often
be challenged under tax or audit principles as being rudimentary and subjective.
The OECD TPG recognise the value of income flow techniques at paragraph 6.157.
In addition three of the worked examples (27 to 29) in the Appendix to Chapter VI
illustrate its use. However, the OECD TPG also point out concerns with the use of
discounted cash flow projections (see paragraphs 6.158 to 6.180).
Market Based Approach
A market based approach seeks to identify comparable transactions and isolate

common components or measures which drive value. These components can
then be adjusted for any differences in circumstances and can be applied to
arrive at a valuation based on market transactions.
Note that market information will often be critical in informing income based
approaches; by way of example, market information on discount rates, multipliers
and forecasts can help increase the credibility of assumptions applied in income
based methods.
Cost Based Approach
Historic cost, in isolation, is less useful as a business valuation tool. It can provide
some background or useful context to value.
Replacement cost is often more relevant; a potential buyer can always consider
the cost of generating a business asset as opposed to purchasing one.
The OECD TPG at paragraph 6.142 state that valuation of intangibles based on
cost is generally discouraged. Paragraph 6.143 points out when it may be
appropriate to use cost based valuations.
Selection of Appropriate Methodology
When the MNE management is able to provide sufficient financial information to

perform a robust DCF valuation method, this method is more widely accepted by
tax administrations and under generally accepted accounting principles to be an
appropriate valuation methodology for a going-concern operating business.
The DCF method of valuation is based on projecting cash flows into the future,
which are then discounted to a present value.

It is often useful to run two methods in parallel to ensure that the results are
comparable and to increase accuracy.
Due to the complexity in valuing intangibles and determining where an intangible

is created (or where it should generate profits) tax authorities have been focusing
more on assessing the transfer pricing for large MNEs holding valuable intangibles.
The OECD TPG look at the tax authorities approach to valuation of intangibles in
paragraph 6.182 onwards. They state that the tax authority needs to take the
approach that would be taken by an independent enterprise. This is illustrated by
giving the example of an independent enterprise fixing a price based on a
particular projection. In this case the tax authority could enquire as to whether
sufficient investigations were made regarding the chosen projection.
As timing considerations are so important to the pricing of intangibles, paragraph

6.185 of the OECD TPG suggests the changes that the tax authority may be
permitted to make depending on the circumstances.
15.7 Case Law on Valuation of Intangibles
The Glaxo group settled a transfer pricing dispute in the US for $3.4 billion (Glaxo
SmithKline Holdings America Inc v Commission 2006). The magnitude of this
settlement helps illustrate the scope of the problem in valuing intangibles and
exploiting them correctly without triggering potential tax avoidance. Glaxo is
headquartered in the UK and holds several subsidiaries in the US. Glaxo’s primary
business is the development and manufacturing of pharmaceutical drugs. Cross-
border transactions of valuable pharmaceutical drugs generating large profit
margins have attracted the attention of revenue authorities.
In 2000, when the predecessor of Glaxo (GlaxoWellcome) merged with SmithKline

Beecham to form Glaxo, the merger triggered a transfer pricing audit in the US.
Glaxo also faced transfer pricing audit adjustments in Canada and Japan.
Glaxo’s sales of drugs in the United States generated almost $30 billion in revenues
from 1989 to 1999. During this period, Glaxo paid about $1.3 billion in US taxes.
Glaxo claimed that the UK had already taxed the MNE Group’s profits under
dispute with the IRS, arguing that any reallocation by the US would result in double
taxation of Glaxo.
Approximately 75% of Glaxo’s income in the US was attributable to Zantac.
The drug had been patented in the UK and hence the US subsidiary was acting as
distributor for the US market. However, the IRS argued that the US subsidiary of
Glaxo overpaid its UK parent for the patent it held. The IRS also argued that
marketing efforts by the US subsidiary were the determining factor in the success of
Zantac. Also, as the US was the largest market for the drug, which was also
manufactured in the US, the economic ownership of the intangible was
challenged. The IRS demanded about $8 billion in tax adjustments and penalties.
Glaxo tried to reach settlement with the IRS by referring the dispute to a
competent authority under the MAP procedure. The governmental discussions did
not reach common ground and the IRS took Glaxo to court to preserve evidence
in preparation for the anticipated trial.
In settling the Glaxo case, IRS Commissioner Mark Everson stated that transfer
pricing issues ‘are one of the most significant challenges’ tax agencies face.

The success and profits of Glaxo’s ‘number two’ drug in the US were primarily
based on successful marketing and sales in the US market, rather than the patents
that led to the new drug. Glaxo also was not able to prove clear ownership of the
intangible especially in relation to research activities within Group (economic
ownership versus legal ownership).
There is an argument that just as the law on intangibles merged with international
trade law to form international law on intangibles, it is now time to consider
merging transfer pricing regulation for intangibles with international law on
intangibles to create more uniform and sophisticated international transfer pricing
regulation. While a tax policy goal is to acquire a fair share of taxes and prevent
abusive tax avoidance, the goal in international law on intangibles is to promote
the development of intangibles, particularly with respect to patents for new
inventions.
Some would say international transfer pricing regulation should consider all of
these policy goals. It is possible for both the international legal system in respect of
intangibles and governmental tax regimes to adopt these fundamental goals
while simultaneously creating a more effective legal system regulating the transfer
pricing of intangibles.
The fast growth in transfer pricing legislation and regulations represents cross-
border expansion in the law. This expansion of transfer pricing regimes arises mostly
from the legitimate concern that if a country does not adopt detailed transfer
pricing regulation and penalties, MNE groups will favour attributing income to a
related MNE located in a second country that has transfer pricing laws and
regulations in place.
Through transfer pricing regulations governments are attempting to limit tax

avoidance by MNE groups engaged in transfer pricing manipulation. However,
effective and fair transfer pricing regulations must allow MNEs to use valuation
approaches as appropriate transfer pricing methods for intangibles. Clearer rules
on valuation of intangibles will translate in fewer ‘grey areas’, which can translate
in challenges by the tax authorities and large tax adjustments.
Hard to Value Intangibles (HTVI)
The update to the OECD TPG following the final report on Action Points 8 to 10
included the addition of a new section relating to Hard to Value Intangibles (HTVI).
HTVI are defined in paragraph 6.189 as intangibles or rights in intangibles for which,
at the time of their transfer between associated enterprises:
i. no reliable comparables exist; and
ii. at the time the transaction was entered into, the projections of future cash
flows or income expected to be derived from the transferred intangible, or the
assumptions used in valuing the intangible are highly uncertain, making it
difficult to predict the level of ultimate success of the intangible at the time of
the transfer.
It is important that you do not misuse the term HTVI in the exam. To score well you
need to ensure that when this term is used you are referring to intangibles that fall
within the above definition.
The guidance recognises that tax authorities do not have access to the same
information as MNEs and that the time delay between the valuation by the MNE

and the scrutiny by the tax authority also makes it difficult for tax authorities to
assess if an arm’s length price has been used.
To assist the tax authorities, the OECD TPG now provides that ex post outcomes
can be used in some cases to help assist in assessing the ex ante pricing. The
OECD TPG say that ex post evidence provides presumptive evidence as to the
existence of uncertainties at the time of the transaction, whether the taxpayer
appropriately took into account reasonably foreseeable developments or events
at the time of the transaction, and the reliability of the information used ex ante in
determining the transfer price for the transfer of such intangibles or rights in
intangibles. In other words, tax administrations may use such ex post evidence to
determine the pricing arrangements that would have been made at the time of
the transaction between independent enterprises, including any contingent
arrangements (such as milestones payments or renegotiation clauses) that might
have been agreed (see paragraph 6.194).
Such presumptive evidence may be subject to rebuttal if it can be demonstrated

that it does not affect the accurate determination of the arm’s length price (see
paragraph 6.192).
The circumstances in which an ex post approach cannot be used include: (see

paragraph 6.193)
• Where the taxpayer can demonstrate that ex ante projections used at the
time of the transfer to determine the pricing arrangements were reliable,
taking into account risks and reasonably foreseeable events that might have
affected the outcomes. In addition the taxpayer also needs to provide reliable
evidence that any significant difference between the projections and actual
outcomes is due either to unforeseeable developments, or to the playing out
of foreseeable outcomes whose probabilities were originally reasonably
estimated.
• The transfer of an HTVI is covered by a bi- or multilateral advance pricing

agreement.
• The difference between financial projections and actual outcomes does not
reduce or increase compensation arising from the HTVI by more than 20% of
the compensation determined at the time the transaction was entered into.
• A commercialisation period of five years has passed and the difference

between financial projections and actual outcome in this period has not been
more than 20%.
Further guidance on HTVI was issued in June 2018. The aim is to help tax authorities
and ensure consistency when applying the HTVI approach. The guidance includes
examples and a section dealing with the interaction with MAP. It will be added to
the OECD TPG as an annex to Chapter VI.
The guidance makes it clear that the focus is on the avoidance of double taxation
and ensuring that the approach is applied in a timely manner.
It is recognised that difficulties may arise as the actual outcomes may have long
gestation periods and don’t necessarily tie in with audit cycles. Wherever possible
tax authorities should identify HTVI, and apply the appropriate audit approach to
flag up any issues, as soon as possible. It is noted that this may not always be
practical and states that the guidance doesn’t seek to amend time limits which
are a matter for national sovereignty for the countries concerned.

It is stated in the guidance that tax authorities may make appropriate adjustments,
including an alternative pricing structure that reflects one which would have been
made by independent enterprises in comparable circumstances, to take account
of the valuation uncertainty in the pricing of the transaction (for example,
milestone payments, or price adjustment clauses, or a combination of these
characteristics). The nature of HTVI means that the tax authorities cannot be
expected to justify the changes by reference to comparable uncontrolled
transactions for similar intangibles.
As part of your study we recommend that you look at the examples in this
guidance – you can find it at http://www.oecd.org/tax/transfer-pricing/guidance-
for-tax-administrations-on-the-application-of-the-approach-to-hard-to-value-
intangibles-beps-action-8.htm
With regard to the interaction with the MAP (which we will look at in a later
chapter), the guidance on HTVI makes specific reference to the fact that Article
25 of the OECD Model DTC makes provision for the taxpayer to set the MAP in
motion without having to wait until the taxation considered by them to be not in
accordance with the convention has been charged against or notified to them.
The taxpayer only has to establish that the actions of one or both of the
Contracting States will result in taxation not in accordance with the DTC, and that
the risk of this is not merely possible but probable (see paragraph 14 of the
Commentary to Article 25).
The guidance also makes reference to the fact that one of the best practices
recommended in the BEPS Report on Action Point 14 is that, subject to the
requirements of paragraph 1 of Article 25, countries implement appropriate
procedures to permit, in certain cases and after an initial tax assessment, taxpayer
requests for the multiyear resolution through the MAP of recurring issues with
respect to filed tax years, where the relevant facts and circumstances are the
same and subject to the verification of such facts and circumstances on audit.
In December 2020 the OECD published the results of a questionnaire on HTVI. 40

countries responded to the questionnaire on the implementation of the HTVI
approach. Most of the countries that responded had adopted the HTVI approach
either directly via legislation or via the adoption of the OECD TPG to interpret the
legislation. The response from most countries was positive. Some OECD members
have not adopted the HTVI approach eg France.
15.8 UN Manual and Intangibles
The chapter on intangibles in the UN ManuaI is in line with the OECD approach. It
points out that group synergies are not an intangible although they do need to be
taken into account when looking at intangibles.
The UN Manual points to the importance of particular factors when analysing

intangibles. These are set out as Development, Acquisition, Enhancement,
Maintenance, Protection and Exploitation of intangibles (DAEMPE). This is slightly
different to the chapter in the OECD TPG that refers to DEMPE, the difference
being the Acquisition (A) that is added in the UN Manual.
The UN Manual is at pains to point out that there is no intention to diverge from the
OECD guidance contained in the Final Report on BEPS Action Points 8 to 10, but
rather a wish to clarify that intangibles can be acquired by an MNE group either
through development activities or by an acquisition from a third party.

There are useful sections on the application of the methodologies to intangibles

and the use of discounted cash flow for valuing intangibles.
The UN Manual also concludes that the methodologies most likely to apply where
there are one or more intangibles are CUP or profit split (see B6.5.4.25).

CHAPTER 16
COST CONTRIBUTION ARRANGEMENTS
In this chapter we are going to look at cost contribution arrangements, in particular:

– OECD TPG and cost contribution arrangements
– Structuring and documenting a cost contribution arrangement
– Case law on cost contribution arrangements
16.1 Introduction
Cost contribution arrangements (sometimes referred to as cost sharing

arrangements (CSA)) were looked at briefly in the last chapter. For ease of
reference the diagram showing a typical structure is reproduced here.
R&D
↑ Costs
Benefits ↓
COMPANY A COMPANY B COMPANY C
The arrangements can take many forms, the above diagram is just one example.
In essence a cost contribution arrangement (CCA) is a contractual arrangement.
As noted in the previous chapter the aim is to achieve advantages in terms of
efficiency and cost savings. CCAs can also result in tax being saved and so may
attract the attention of revenue authorities.
CCAs are not just used for development of intangibles. They can also be used for
any joint funding and cost sharing for acquiring property or services.
16.2 OECD Guidelines and Cost Contribution Arrangements
Chapter VIII of the OECD TPG sets down the OECD view on CCAs. The OECD
recognises that CCAs can be used in other contexts not just for the development
of intangibles.
Chapter VIII of the OECD TPG adopts the same principles as that contained in
Chapter I of the OECD TPG – that is to say they require correct delineation of the
transaction and the risks involved. They also take account of the principles in
Chapter VI for intangibles. The requirements relating to substance and control and
the restrictions to using cost as a basis for valuation of contributions, both pre-
existing and current, are likely to have an impact for participants in CCAs.
The main aim is to ensure that the contributions made under a CCA are
commensurate with the benefits received.

The definition of the concept of a CCA can be found in paragraph 8.3:
‘A CCA is a contractual arrangement among business enterprises to share the

contributions and risks involved in the joint development, production or the
obtaining of intangibles, tangible assets or services with the understanding
that such intangibles, tangible assets or services are expected to create direct
benefits for the businesses of each of the participants. A CCA is a contractual
arrangement rather than necessarily a distinct juridical entity or fixed place of
business of all the participants.’
There is also a requirement that participants in a CCA must have the capability
and authority to control the risks associated with the CCA activities (see
paragraph 8.15). This means they must be capable of making the decision to take
on the initial financial risk of participation in the CCA, and must have the ongoing
decision-making capacity to decide on whether or how to respond to the risks
associated with the CCA. It is recognised in paragraph 8.17 that it is not necessary
for all CCA activities to be performed by the personnel of the CCA participants.
Thus, participants in a CCA may outsource certain functions to a separate entity
outside the scope of the CCA. In this case, the relevant CCA participants must
exercise the requisite control over the specific risks they assume under the CCA.
The OECD TPG recognise that there are many types of CCAs. Independent
enterprises may come together to share costs. There are two key concepts to the
chapter on CCAs; firstly that there will be mutual benefit from sharing costs and
secondly that each enterprise will contribute according to the amount of benefit
they expect to receive.
Chapter VIII contains a distinction between two commonly encountered types of

CCAs: ‘development CCA’ and ‘services CCA’. It is stated that these two types of
CCAs will necessarily have different implications from a transfer pricing
perspective; development CCAs would typically create ‘ongoing, future benefits’
for participants, whereas services CCAs would typically result in ‘current benefits’
(see paragraph 8.10).
A CCA will meet the arm’s length principle if the participants’ share of the cost is
commensurate with their share of the benefits (see paragraph 8.13).
In deciding whether the arrangements are arm’s length, the earlier chapters of the
OECD TPG will be in point – that is to say factors such as the contractual terms,
economic circumstances and how risks are shared.
In deciding whether the arm’s length principle is being met, account will be taken
of any balancing payments or receipts. The documentation drawn up for the CCA
should include clauses for balancing payments where the amount of a
participant’s benefit is greater than that anticipated at the outset. If one party is
receiving a greater part of the benefits than that anticipated at the outset, it
follows that another is receiving less so they should receive a balancing amount to
reduce the proportionate amount of their contributions.
Balancing payments need to be distinguished from buy-in and buy-out payments

which are made when participants join or leave after a CCA has been entered
into. Buy-in and buy-out payments must also meet the arm’s length principle and
these are discussed further below.
It is important to take account of all contributions made not just cash, so account
must be taken of services and property. The guidance makes reference to
contributions of pre-existing value and current contributions. For example, the
contribution of patented technology to be used in the development of an

intangible would reflect the contribution of pre-existing value, whereas the

contribution of current R&D would constitute a current contribution. The value of
the pre-existing intangible (eg, patented technology) should be determined under
the arm’s length principle using guidance in Chapters I-II and Chapter VI, including
the valuation techniques as set out in Chapter VI (see paragraph 8.26).
The OECD TPG state (at paragraph 8.28) that the value of each participant’s
contribution should be determined in line with the value that would be placed on
it by independent enterprises in comparable circumstances. While contributions
should be measured based on value, the OECD recognises that it may be more
practical for taxpayers to compensate current contributions at cost. However, this
approach may not be appropriate where the contributions of different
participants differ in nature (for instance, where some participants contribute
services and others provide intangibles or other assets). Examples are also
provided in the Annex to Chapter VIII to illustrate this.
The OECD TPG also recognise the problem of estimating benefits that will arise in
the future and suggests this might, as noted above, be dealt with via a balancing
payment at a later date (see Section C5 of Chapter VIII).
If it is found that the arrangements are not in keeping with the arm’s length
principle, then adjustments will need to be made. Again the adjustment required
will be dependent on the facts but will likely lead to the need for a balancing
payment from one party to the CCA to another. The OECD TPG recognise the right
of the tax authority to make adjustments where a participant’s contribution has
not been correctly calculated, for example as a result of miscalculation of the
value of property or services contributed or as a result of unexpected benefits. The
OECD TPG do counsel against minor amendments; they also state that
adjustments should not be made in respect of a single fiscal year if overall the
arrangements are being carried out in good faith (see paragraphs 8.36 and 8.37).
Section C6 of Chapter VIII OECD TPG deals with the issue of disregarding part or all
of the terms of a CCA. Where arrangements viewed in their totality lack
commercial rationality in accordance with Section D2 of Chapter I, the CCA can
be disregarded in its entirety (see paragraph 8.40).
Section D of Chapter VIII looks at buy-in and buy-out payments plus termination of
a CCA. Changes in the membership of a CCA would generally trigger
reassessment of the shares of contributions and the expected benefits for each
participant. When a new participant joins then effectively a share of the existing
benefit is moved to the new participant and this should be compensated using
the arm’s length principle. This would be a buy-in payment. Similarly, if a
participant leaves, a buy-out payment would be expected using the arm’s length
principle to compensate them for the interest they are giving up. The guidance set
down in earlier chapters of the OECD TPG will apply including that in Chapter VI
on intangibles.
On the termination of a CCA either the participants will retain a share in the results
of the CCA activity (if there are any) or they should be compensated for
transferring their share to the other participants (see paragraph 8.49 OECD TPG).

16.3 Structuring and Documenting a CCA
The structure of a CCA should conform to the arm’s length principle. Anyone who
will benefit from the CCA should contribute to the CCA. Those who will not benefit
should not be participants. The nature and extent of each participant’s beneficial
interest should be clearly documented. The payments made should only be those
required under the arrangement including any balancing payments.
As already noted, the proportionate share contributed by each party needs to be

calculated in an appropriate manner. The OECD TPG suggest the use of allocation
keys. There should be provision for balancing payments and buy-in and buy-out
payments.
The participants should have full access to all information in relation to the
activities and the projections used to estimate contributions and benefits under
the CCA.
All of this plus any other information of interest to a tax authority will need to be
documented. The OECD TPG give a list of the information that may be relevant
(see below) but note that this is neither a minimum compliance standard nor an
exhaustive list (see paragraph 8.52):
a. A list of participants;
b. A list of any other associated enterprises that will be involved with the CCA
activity or that are expected to exploit or use the results of the subject activity;
c. The scope of the activities and specific projects covered by the CCA and how
the CCA activities are managed and controlled;
d. The duration of the arrangement;
e. The manner in which participants’ proportionate shares of expected benefits

are measured, and any projections used in this determination;
f. The manner in which any future benefits (such as intangibles) are to be

exploited;
g. The form and value of each participant’s initial contributions, and a detailed
description of how the value of initial and ongoing contributions is determined
(including any budgeted versus actual adjustments) and how accounting
principles are applied consistently to all participants in determining
expenditures and the value of contributions;
h. The anticipated allocation of responsibilities and tasks and the mechanisms for
managing and controlling those responsibilities and tasks, in particular those
relating to the development, enhancement, maintenance, protection or
exploitation of intangibles or tangibles used in the CCA activity;
i. The procedures for and consequences of a participant entering or

withdrawing from the CCA and the termination of the CCA; and
j. Any provisions for balancing payments or for adjusting the terms of the
arrangement to reflect changes in economic circumstances.
In addition records will need to be kept over the life of the CCA of any changes to
the arrangements, comparisons of actual outcomes to projections, details of
actual cost and actual contributions made (see paragraph 8.53).

Paragraph 8.51 brings in the importance of the documentation complying with the
requirements in Chapter V of the OECD TPG. When you come to look at the three-
tier approach to documentation in a later chapter, you will see that the master file
will include details of important service arrangements and agreements related to
intangibles, including CCAs. In addition, the local file will contain transactional
information, including a description of the transactions, the amounts of payments
and receipts, identification of the associated enterprises involved, copies of
material intercompany agreements, and pricing information, including a
description of reasons for concluding that the transactions were priced on an
arm’s length basis. As a result, it would be expected that in order to comply with
these documentation requirements, the participants in a CCA will maintain
materials describing the arm’s length nature of the CCA. You will note that
paragraph 8.51 states that the level of detail of the materials prepared should be
commensurate with the complexity and importance of the CCA to the taxpayer.
16.4 Case Law on Cost Contribution Arrangements
VERITAS Software Corp., 133 TC No. 14, Dec. 58,016 (Dec. 10, 2009)
This was a US case looking at ‘buy-in’ costs for a cost contribution arrangement.
The IRS argued that what had taken place was akin to a sale or spin-off of Veritas
operations hence the sum to be paid should be valued on this basis.
Veritas Software, which is in the business of developing, manufacturing, marketing,

and selling software products, went through several corporate changes; mostly
notably, it was purchased by Symantec Corp. on July 2, 2005. Prior to that, on
November 3, 1999, Veritas Software assigned all its existing sales agreements with
its European-based sales subsidiaries to a new corporation – Veritas Ireland. In
addition, on the same date, Veritas Software and Veritas Ireland entered into a
research and development agreement, as well as a technology license
agreement.
Based on the licensing agreement, Veritas Software granted Veritas Ireland the
right to use certain ‘covered intangibles,’ as well as the right to use Veritas
Software’s trademarks, trade names, and service marks. In exchange for the rights
granted by the licensing agreement, Veritas Ireland agreed to pay royalties, as
well as a ‘prepayment amount’.
In 2000 Veritas Ireland made a $166 million ‘lump sum buy-in payment’ to Veritas
Software. This amount was later adjusted downward to $118 million. At issue, from
a tax perspective, was whether the buy-in payment was ‘arm's length’.
The court rejected the IRS approach agreeing with Veritas that the amount to be
paid should be based on comparable uncontrolled royalties payable over the life
of the agreement. Further the court said that the IRS determination was arbitrary,
capricious, and altogether unreasonable.
Veritas used agreements between Veritas Software and certain original

equipment manufacturers (OEMs) as comparables. The IRS contended that the
OEM agreements involve substantially different intangibles. But the court
disagreed: it concluded that, collectively, the more than 90 “unbundled” OEM
agreements the parties stipulated were sufficiently comparable to the controlled
transaction.

In noting the comparability, the court also pointed out the following:
(1) Veritas Ireland and the OEMs undertook similar activities and employed similar
resources in conjunction with such activities, (2) there were no significant
differences in contractual terms, (3) the parties to the controlled and uncontrolled
transactions bore similar market risks and other risks, and (4) there were no
significant differences in property or services provided.
Therefore, the court was happy that the unbundled OEM agreements were
sufficiently comparable to the transaction they were looking at thus giving the
result that comparable uncontrolled transaction method (CUT) (as set down in the
US regulations) was the best method to determine the appropriate buy-in price.
The buy-in payment charged met the arm's length standard and the IRS's
contention was rejected.
Amazon.com, Inc. v Commissioner, 148 TC No. 8, (2017)
In 2017, the US Tax Court issued its long-awaited opinion in a transfer pricing
dispute involving Amazon's cost sharing arrangement (‘CSA’) with its Luxembourg
subsidiary.
Amazon contributed software and other pre-existing intangibles to a CSA, under

which a Luxembourg subsidiary (‘Luxco’) agreed to share the costs of developing
future intangibles for use in Amazon’s European operations. Consistent with the
pre-2009 transfer pricing regulations’ requirement that the parties make buy-in
payments for use of intangibles contributed to CSAs, Luxco paid $255 million to
Amazon.
The IRS challenged Amazon’s valuation method, claiming that Luxco’s buy-in
payment was not arm's length. The IRS asserted that the contributed intangibles
had an indefinite useful life (because it served as the basis for future intangibles
development) and used a discounted cash flow valuation method. This approach
increased the buy-in payment to approximately $3.5 billion.
The IRS was making essentially the same arguments that the court had
emphatically rejected in the Veritas case above. In Amazon, the court noted that
the IRS formulation of ‘indefinite’ useful life was in substance identical to the
‘perpetual’ life it had argued in Veritas. In both cases, the IRS valuation method
included the present value of income projected to be derived from future
intangibles not yet developed. This violated the regulations’ prescription that a
buy-in payment equals the value of only the existing intangibles contributed to the
CSA. The IRS arguments can be seen as an attempt – soundly rejected – to impose
the more stringent valuation methods of the new, post-2009 CSA regulations on
pre-2009 cases.
The chapter on CCAs in the UN Manual is predominantly in line with the OECD
approach. It contains some worked examples, now would be a good time to have
a look at them.

CHAPTER 17
SPECIFIC TRANSACTIONS: BUSINESS RESTRUCTURING
In this chapter we are going to look at the transfer pricing implications of business
restructuring, in particular:
– the rationale for restructuring and the role of tax
– typical models applied during restructuring
– the OECD approach including location savings
– UN Manual approach
– tax authority response to business restructuring
17.1 Introduction
Chapter IX of the OECD TPG on the transfer pricing aspects of business

restructuring attempts to deal with the growing trend of large groups undergoing
structural changes.
In an economic environment where globalisation has become the norm and

stronger competition has forced many big players out of business, looking at how
a business can function more efficiently, reducing cost and maximising profitability
potential are key to the survival of the business itself. In many cases, when a group
undergoes a structural change it often leads to centralisation of functions and risks.
Through selection of location of these centralised activities, many multinational
groups seek to maximise the tax benefits of restructuring as well.
Business restructuring is often a necessity; however, when looking at restructuring

groups also try to maximise profitability potential by looking at cost and tax
efficiencies. This is a concern for tax authorities, who are concerned that the
restructuring may be wholly or predominantly for tax purposes, and who are
therefore keen to protect their tax base from erosion through abusive planning.
Following business restructuring activities, operating companies will typically earn

lower profits than pre-restructuring, with related party transactions with a central
entity in a low-tax jurisdiction being the mechanism by which those profits are
reduced. Therefore, understanding the transfer pricing implications of business
restructuring is critical.
17.2 The Rationale for Restructuring and the Role of Tax
There have been many headlines accusing large groups of trying to evade tax by
setting up principal entities in low tax jurisdictions with the ‘excuse’ of restructuring
and making the group more efficient. Whilst it would not be fair to comment on
individual cases, it is reasonable to say that companies have adopted a wide
range of strategies to business restructuring, with some moving thousands of
employees to different locations and others simply changing the form of
transactions.

Although it is true that some groups have viewed business restructuring as a means
of driving down the group effective tax rate, there are a wide range of
operational reasons why companies seek to restructure. These include:
• Business control – It is often the case that in growing companies, management

will seek to exercise greater control of the business through a centralised
structure. This will allow greater control over key value drivers, such as global
customer relationships or intellectual property, and enable faster and more
effective execution of global strategy.
• Cost management – In the tough economic climate experienced by most

industries and countries in recent years, there is increased pressure on
companies to manage costs to maintain profitability. Business restructuring
allows multinational companies to take advantage of economies of scale,
either through purchasing efficiencies or restructuring the supply chain to
make optimal use of the existing manufacturing footprint.
• Business integration – Companies will often seek to restructure either to enable

better integration of a recently acquired business, or to position themselves
better for future merger and acquisition activity. It is much easier to ‘bolt on’ a
newly acquired business to a centralised structure than a decentralised one.
Clearly, an efficient operational and organisational structure is not only a way to

increase revenue, but also a necessity for a group to be able to manage all
operations efficiently and stay competitive. Therefore, although there are cases
where restructuring projects have been driven by tax, in most cases tax is looked
at to ensure that the cost efficiencies are not eaten away by tax implications.
It should be noted that true business restructuring is a commercially-led activity

often involving changes to key operating processes and movement of personnel.
In most organisations it would be very difficult for tax to drive decision making
around such a restructure. They tend to be disruptive and potentially put a
business at risk for the sake of a lower tax rate (as much as all tax practitioners
would love to be the ones driving, it is normally the case that business and
commercial considerations come before tax).
It should also be noted that tax efficiency is in fact the responsibility of a company,
rather than a negative characteristic. Companies have a duty to shareholders to
optimise the value of a business, albeit in a responsible manner. Tax costs are
simply one element of cost to a business, and need to be factored into any major
business decision. As the OECD TPG note:
‘MNEs are free to organise their business operations as they see fit. Tax
administrations do not have the right to dictate to an MNE how to design its
structure or where to locate its business operations. They are free to act in their
own best commercial and economic interests in this regard. In making this
decision, tax considerations may be a factor.’ (See paragraph 9.34).
Structural changes will almost invariably lead to changes in the intra-group pricing,
financing and allocation of risk and functions; therefore, transfer pricing is a key
subject and a planning tool when setting up the new structure.

17.3 Typical Models Applied During Restructuring
Before considering the OECD approach to business restructuring, it is necessary to

understand what it means in practice in terms of the models used and related
transactions. The theory behind business restructuring, as targeted by the OECD
TPG, is relatively straightforward. It typically involves the movement of economic
activity from high-tax jurisdictions to low-tax jurisdictions, with an accompanying
shift in transfer pricing model to reflect the new balance of activity.
The first step is to establish the entity that will undertake increased activities. This
entity is commonly known as the Principal. Location will often be determined by a
number of factors, including amongst other things:
• Proximity to the markets the Principal will be responsible for
• Cost and standard of living (since often senior personnel will need to be
located there)
• Availability of skilled resource to support the Principal’s activity
• Ease of doing business
• Tax profile
In many cases, the location chosen will have a low tax rate. This could be for a
number of reasons, including a low underlying corporate tax rate, the availability
of significant tax losses to offset future profits, or tax incentives offered by the local
government to encourage the relocation of certain qualifying activities to that
territory.
Having established the Principal, some or all of the business value chain is
reorganised to move value to the Principal. This may involve:
• Manufacturing – whilst core manufacturing operations are unlikely to move to

the Principal, key decision-making processes might.
The Principal may undertake the planning and scheduling process, take
responsibility for capacity and inventory planning, and insulate the
manufacturing entities from key risks beyond the delivery of core
manufacturing processes.
Centralised structures typically involve conversion of manufacturing operations

to contract manufacturers, whereby the Principal determines what the local
manufacturing operation should produce and purchases all the output.
Alternatively, some structures adopt a toll manufacturing model, whereby the

Principal owns the raw materials through the manufacturing process, and the
local operations simply provide a conversion service.
• Purchasing – the centralisation of procurement can often have obvious

advantages in generating economies of scale through simple aggregation of
demand. However, further cost advantages can be achieved through taking
a more strategic approach to purchasing, including supplier management,
redefining product requirements and timing of purchases.
Centralised purchasing operations may either purchase the materials directly

from suppliers and sell on to related party operations (earning a margin on the
resale price) or else facilitate global purchasing arrangements for operations

to purchase directly from suppliers, but pay the central procurement entity a
fee.
• Development and maintenance of intangibles (IP) – a fundamental feature of

many business restructurings is the movement or development of IP within the
Principal.
For new IP, this will involve either direct development within the Principal entity,
or engaging related parties (or third parties) to undertake development work
on a contract R&D basis. This would require sufficient management and
control of the R&D process from the Principal, which would need people with
sufficient technical capability to understand the R&D process, as well as
responsibility for budgets and key decisions through various development
tollgates.
For existing IP, this would need to be transferred to the Principal in some form
(discussed in more detail below), and would require active management to
protect and maintain its value.
Reward for owning valuable IP would be either through royalty charges to

related parties using and benefitting from that IP, or else would be embedded
in the price of products sold by the Principal to related parties.
• Selling – business restructuring exercises will also often involve centralisation of

functions around the selling process, with distributors being converted to
‘limited risk distributors’ (LRDs).
The term LRD is a catch-all term, and can actually involve a range of functions
being undertaken.
Typically, as the Principal entity takes greater responsibility for demand

planning, key account management, pricing and portfolio management, and
key risks such as inventory or credit risk, the profit earned by the LRD is
reduced. This is achieved through an increased sales price, often with
purchases being direct from the Principal entity.
In some cases, responsibility for customer contracting is removed from the

local sales operation, which would be converted to some form of sales agent
earning a commission.
The above are just some examples of the shift in functionality seen in business
restructuring. Some industries, such as consumer goods, have seen many
companies adopt all aspects of these within their value chain. For others, it is more
common to see only some aspects.
Many companies find it difficult to transition to a full centralised model in one go,
either because of system constraints, lack of resources to manage the transition or
the scale of disruption that it would entail. Therefore, these Principal structures may
initially involve only one aspect (such as procurement) but develop into a full
Principal over time.
It should also be noted that often the Principal may not be part of the title chain.
This may again be due to system constraints, or else complexities arising from
where the manufacturing and sales activities take place in the same country. In
those cases, the Principal may be rewarded through some form of service fee that
is sufficiently high to reflect the value that it contributes.

17.4 The OECD Approach
As mentioned in Chapter 1 of this manual, the OECD TPG were reissued in 2017.
Many of the amendments had already been seen when the OECD TPG were
updated in May 2016 following endorsement of the BEPS reports. Chapter IX was
not directly affected by the BEPS Action Plan, however the conforming changes
that were needed to ensure that Chapter IX reflected the changes to earlier
chapters of the OECD TPG means that the chapter is quite different in
appearance to the 2010 version.
Chapter IX of the OECD TPG contains two key areas:
• Arm’s length compensation for the restructuring itself; and
• Remuneration of post-restructuring controlled transactions.
The chapter begins by stating that there is no legal or universal definition of

business restructuring (see paragraph 9.1).
4 examples of types of business restructuring are set out in paragraph 9.2. In

summary these are:
• Conversion of a fully-fledged distributor into a limited risk distributor;
• Conversion of a fully-fledged manufacturer into a contract or toll

manufacturer;
• Transfers of intangibles or rights in intangibles to a central entity within the

group; and
• The concentration of functions into a regional or central entity with a

corresponding reduction in scope or scale in other group entities.
The focus of the chapter is the application of Article 9 of the OECD Model DTC in
the context of business restructuring; it does not cover the attribution of profits
under Article 7 (see later chapter).
The principles that are established earlier in the OECD TPG regarding the
treatment of related party transactions apply equally to business restructuring:
‘This chapter starts from the premise that the arm’s length principle and these
Guidelines do not and should not apply differently to restructurings or post-
restructuring transactions than to transactions that were structured as such
from the beginning.’ (See paragraph 9.9)
Nevertheless, it is a feature of business restructuring that they often include

structures and constructs not typically seen between independent parties,
creating substantial complexity. Furthermore, the amounts of tax at stake are often
substantial. These factors therefore explain the specific consideration given to
business restructuring within the OECD TPG.
Arm’s Length Consideration for the Restructuring Itself
The OECD TPG recognise that the process of business restructuring itself may give
rise to a cross-border transfer of something of value. A payment to reflect such a
transfer is commonly known as an ‘exit charge’. In some cases, this may be
obvious, such as tangible or intangible assets. In those cases, the same principles
apply as would be the case if those assets were being sold in normal
circumstances.

Complexities arise where there is a business restructuring resulting in a significant

shift in the profit profile of entities restructured but no obvious transfer of assets.
Consideration needs to be given as to how the arm’s length principle would apply,
by challenging whether the restructured entity operating on a stand-alone basis
would be prepared to accept the restructuring without need for additional
payment.
The first step of this process is to understand the nature of the restructuring.
Specifically see OECD TPG paragraphs 9.13–9.18; it is important to identify the
difference in functional and risk profile pre- and post-restructuring, and the
economic nature of what has been shifted.
Contracts will be an important starting point (see paragraph 9.17). As with other
transactions, the contractual allocation of risk should not be immediately
discounted. Nevertheless, tax authorities are entitled to consider whether the
behaviour of the parties accords with the written contracts. Where there are no
written contracts or the behaviour of the parties differs from the written agreement
then the facts will need to be used to deduce the actual transaction. Again the
principles in Chapter I of the OECD TPG need to be applied, in particular Section
D1.1 of Chapter I.
The functional analysis will have a key role in accurately delineating the
transactions and recourse needs to be had to Section D1.2 of Chapter I (see
paragraph 9.18).
As noted above, the guidance on the analysis of risk in the context of business
restructuring is found in Part I of Chapter IX paragraphs 9.19 to 9.23. The guidance
stresses the importance of identifying economically significant risk. It is only
economically significant risk that will have an impact on profit. When analysing risk
the six step process set down in Section D1.2.1 of Chapter I of the OECD TPG must
be applied.
When looking at the transfer of risk, it is important to look beyond contractual

assumption of risk to ensure that the party claiming to be responsible for risk
actually has the capability to control risk as discussed in Chapter I of the OECD
TPG. A risk cannot be moved from one associated company to another if it was
not controlled by the transferee to begin with. Likewise, it is important to establish
that once the contractual assumption of risk is changed, the transferee has the
capability to control the risks (see paragraph 9.21).
A key consideration is the economic importance of the risk; the OECD TPG point
out (see paragraph 9.22) that only where economically important risk is transferred
would we expect to see a significant reallocation of profit potential.
Chapter IX emphasises the importance of Chapter I in analysing risk; it also

includes a useful example of how an analysis may be undertaken in the context of
inventory risk at paragraph 9.23.
Another important area is understanding the business rationale for restructuring.

Section B.2 of Part I gives guidance on the role of group synergies. As stated
above, the business restructuring may be motivated by a desire to streamline,
centralise and make cost savings. If this is the case then the OECD TPG point out
that they would expect the anticipated synergies to be recorded. Chapter IX
conforms with Chapter I by recognising that where concerted action is taken to
achieve synergies, then those who contribute should be rewarded. It follows that
those who don’t contribute but merely facilitate the savings, such as a central
procurement function, would merely be rewarded by a fee for the function they
perform. It should be born in mind that in some cases the synergies don’t

materialise and in other cases they don’t necessarily lead to additional profit; they
may just allow an MNE to remain competitive.
To fully understand the restructuring, the OECD TPG state that consideration should
be given to the options realistically available to the restructured party (see
paragraphs 9.27 to 9.30). By evaluating the restructured entity as if it were
operating independently from the rest of the group, it should be established
whether it would be prepared to accept the restructuring or whether it would
have had more profitable options available to it. This is not to say that the mere
reduction in its future profits should give rise to compensation. Nevertheless, if it
can be established that the restructured entity had material profit potential that it
has given up, then at arm’s length, this would be rewarded.
When addressing this question, the rights of the party need to be considered. Take
the case of a distributor with long-term contractual rights (either written or implied
by behaviour) to distribute a product. In the event of a restructuring, it would need
to be questioned whether the distributor would accept a lower, more stable return
for its activities. If margins pre-restructuring were volatile, or there are declining
margins in the industry, it may be possible to make the case that no compensation
would be required as it would be rational to accept the terms of the restructure.
However, if the party would expect to maintain higher, stable profits, and would
ordinarily have the contractual rights to be able to do so, then some
compensation must be given to recognise this profit potential foregone.
As mentioned above, when looking at synergies documentation will be important.

As pointed out in paragraphs 9.32 and 9.33, a business restructure should be
mentioned in the master file and the applicable local files (which we will look at in
a later chapter).
Section C of Part I looks at recognition of the transactions within the business

restructure. Here the OECD TPG reinforce the point that taxpayers are free to
organise their business operations as they see fit. Nevertheless, they also recognise
that tax authorities have the right to determine the tax consequences of the
structure in place (see paragraph 9.34).
In general, tax authorities should only disregard the structure of the transaction in
exceptional circumstances as set down in paragraph 1.122 (1.142) of the OECD
TPG. In summary this states that an accurately delineated transaction can only be
disregarded and replaced with an alternative transaction where the transaction
viewed as a whole differs from the one that would have been entered into by
independent enterprises in comparable circumstances behaving in a
commercially rational manner taking into account the options available to them
(ie. the same circumstances that apply to any transaction, not just business
restructuring).
Even if the tax authority were to disregard the transactions as structured by the
taxpayer, the alternative transaction used for taxation would nevertheless need to
recognise certain commercial realities. For example, if the restructuring involved
the closing of a factory, any substituted transaction would need to recognise that
the factory is no longer available (see paragraph 9.35).
The discussion in Chapter IX on the reallocation of profit potential following a

business restructure emphasises the importance of the functional analysis in
understanding the changes that have taken place (see paragraph 9.42). If we
take the case of a distributor with local marketing intangibles being converted into
a limited risk distributor we would need to be clear on what is going to happen to
the local intangibles – will they also move across? If they do then it will need to be
for an arm’s length remuneration. If they stay this will impact on the functional

analysis after the restructure and the reward to the restructured distributor going
forward.
Section E of Part I deals with transfers of something of value; this includes tangibles,
intangibles and transfer of a going concern. As with other sections of Chapter IX
there is reference back to earlier chapters in the OECD TPG, in particular Chapter
VI when looking at intangibles and the importance of considering who is
responsible for the development, enhancement, maintenance, protection and
exploitation (DEMPE) of intangibles and Chapter I when looking at the impact for
the arm’s length price when looking at transfer of a going concern that includes
the transfer of an assembled workforce. When there is a transfer of a going
concern then valuation techniques may need to be considered as the price will
not necessarily be the sum of the value of the component parts.
There is recognition that a restructure may include the transfer of loss making
activities; as with profitable activities the options available need to be considered.
Rather than transfer a loss making activity it can be closed down, however it may
be commercially more profitable to transfer it (see paragraphs 9.71 to 9.73).
The OECD TPG also address whether it is necessary for the restructured party to be
indemnified against restructuring costs (such as plant closure and redundancy
costs). (See paragraphs 9.75 to 9.97.) In doing so, consideration is given to the
terms of the agreement(s) between the parties, both written and what is implied
by the behaviour of the parties pre-restructuring. This should be evaluated in the
context of local commercial law. The overriding principle is whether at arm’s
length another party would be willing to indemnify the restructured entity.
Although the answer is heavily dependent on the specific facts and
circumstances, it could be the case that the Principal, the parent company or a
new entity benefitting from additional business (or a combination of all three)
could be willing to pay.
Post Restructuring Transfer Pricing
As a guiding principle, the determination of arm’s length transfer prices following a

business restructuring should be no different to any other related party transactions
(see paragraph 9.98). Nevertheless, the OECD TPG acknowledge that there are
certain features of a business restructuring that create specific challenges.
One issue is that comparability analysis may be harder to apply (see OECD TPG
Part III Section A2 Chapter IX). There are already inherent difficulties in identifying
comparable data from independent parties to test related party transactions
given there are often fundamental differences in the way that multinational
groups and independent firms operate. This is often placed under further stress
following business restructuring where transactions are frequently structured in a
way that is not seen between unrelated parties, with substantial differences in the
division of responsibility and risk. Such a fact pattern does not necessarily mean
that a controlled transaction is not arm’s length, and it is necessary to find a
reasonable solution. This places increased importance on a thorough functional
analysis to identify the key economic drivers of the transaction.
When we are looking for comparables there is the added complication that it may
not be appropriate to compare a restructured entity with a new entity. Paragraph
9.102 cites the case of a previously successful fully-fledged distributor that has
been converted to a limited risk distributor and points out although it is new to
being a limited risk distributor it may not be comparable to a newly set up limited
risk distributor as the company that has been restructured will still be operating
from a base of many years of successful operation as a distributor. In addition, it
will not need to go through a market penetration exercise in the same way as a

newly set up company. The application of comparability analysis as set down in

Chapter III of the OECD TPG will be very important.
A further difference for the application of transfer pricing models to business

restructuring is the interaction with the restructuring itself (see paragraphs 9.114 to
9.116). Although it is discussed at length whether payment is required to
compensate for foregone profit potential, it is important to consider whether the
same outcome can be achieved through post-restructuring pricing. It may be the
case that parties would agree to forego an upfront payment in return for a more
beneficial transfer price. As such, the arrangements would need to be considered
holistically to determine whether they comply with the arm’s length standard.
Location Savings
You will recall from an earlier chapter looking at comparables, that location
savings are the net cost savings from relocating a business or part of its functions to
a new geographical location. Location savings may be part of the business
strategy in a restructuring. In this case the OECD TPG point out that section D1.5 of
the OECD TPG needs to be considered.
Chapter IX expands further on location savings (see paragraphs 9.126 to 9.131). In

many cases, business restructurings result in the shift of labour-intensive activity
from high-cost countries to low-cost ones to create efficiencies within the business.
However, it has become an increasing trend for tax authorities in those low-cost
territories to assert that a proportion of those cost savings should be shared with
the new operations creating the savings.
The OECD TPG do not directly rule out the case for sharing the savings. However,
they note that in the case of routine activities operating in a competitive market, it
is likely that the Principal would have the option realistically available to use third
parties in that territory. As such, at arm’s length, very little would be attributed to
the routine entity, and standard benchmarking could be used to determine the
appropriate return. However, in the case where the new entity performs more
specialised services, there may be a case for attributing greater returns. However,
it is arguable that these additional returns relate more to the nature of the services
being provided than to a share of location savings.
17.5 UN Manual
Location Savings
The UN Manual’s view on location savings and location specific advantages (LSA)
was also covered in the earlier chapter on comparability. To recap, the UN
Manual expands on the OECD definition, discussing location rents and how they
should be allocated in the absence of comparables or benchmarking, stating that
relative bargaining power reflecting that of independent enterprises would be a
key metric.
Location savings, plus the specific benefits of a location, such as highly skilled
manpower and knowledge, proximity to a growing local/regional market (see
B.3.4.5.9 of the UN Manual for more examples) are added together to give LSA.
The additional profit that arises as a result of LSA is called a location rent.
Location rent is equal to or less than LSA.

The allocation of location rent depends on a number of factors. The first

consideration is the level of competition in the marketplace. The next is the realistic
alternatives available to the associated enterprises. Another consideration is that
the competitors may not have access to the LSA. This can arise as the MNE has
production intangibles that lead to lower costs or because the MNE is the first to
operate in the location. When that is the case they will be in a good position to
extract rents until competitors arrive. The UN Manual observes that:
• There may be no location rent in a competitive market as they may all be

passed on to the customer
• The MNE may still gain a greater market share even if LSA are all passed on to
the customer
• If an MNE has exclusive access to LSA then this gives a competitive advantage
• Competition may dissipate LSA over time
The question of how the rents should be allocated between associated enterprises
is dealt with briefly. If CUPs are available then they would be first choice.
Benchmarking is another possibility. In the absence of CUPs and benchmarking, an
allocation based on the relative bargaining power of the associated enterprises
which reflects the arm’s length position is given as a key metric to allocate the
rents.
The section finishes with some guidance on other matters that need to be
considered:
• Government rules eg exchange controls, subsidies for certain sectors. These

are stated to be a market condition. The comparable should be to companies
that are subject to the same restrictions or benefits.
• Market level of the company; is it operating at wholesale or retail level?
• Other market conditions that can influence the transfer price eg availability of
substitutes or level of demand.
Section D of the UN Manual provides country examples.
China and India Contributions to the UN Manual: Location Savings
China defines location savings as the net cost savings derived by a MNE when it
sets up its operations in a low cost jurisdiction. Net cost savings are commonly
realised through lower expenditure on items such as raw materials, labour, rent,
transportation and infrastructure even though additional expenses (‘dis‐savings’)
may be incurred due to the relocation, such as increased training costs in return
for hiring less skilled labour.
The approach taken by China is to identify if there are LSA that give rise to
additional profits then identify the transfer pricing method that will be used to
allocate the profits.
At section D2.20.16 it is stated that China has noted that many MNEs have set up
sales subsidiaries which have been involved in heavy marketing and sales activity
to build up brand image amongst Chinese customers leading to additional profits
for the MNE. The Chinese tax authorities take the view that these additional profits
should be taxed in China if they derive from the unique characteristics of the
Chinese market. In D2.20.16 it is also stated that where Chinese customers have
chosen to go abroad to secure a lower price for a luxury brand then a proportion

of that sales revenue should be attributed to the marketing contribution made by

Chinese subsidiaries.
India has also contributed to the UN Manual on location savings (see section D3.7).
The Indian transfer pricing administration states that the concept of ‘location
savings’ is one of the major items to be reviewed when carrying out comparability
analysis during transfer pricing audits. Location savings is interpreted as any cost
advantage. The manual states that India provides operational advantages to
groups such as low-cost labour or skilled employees, lower cost raw material, rent,
training, and tax incentives.
In addition, India also provides the following LSA in addition to location savings:
highly specialised skilled manpower and knowledge, access and proximity to
growing local/regional market, large customer base with increased spending
capacity, superior information network, superior distribution network, incentives
and market premium.
Again, the incremental profit from LSA is known as ‘location rents’. The main issue
in transfer pricing is the quantification and allocation of location savings and
location rents among a group.
The Indian transfer pricing administration believes it is possible to use the profit split
method to determine arm’s length allocation of location savings and rents in
cases where comparable uncontrolled transactions are not available taking into
account the bargaining power of the parties.
Business Restructuring
This chapter in the UN Manual is surprisingly short given that it begins by noting that
business restructuring increasingly affects developing countries. In recent years a
number of large MNEs have either (i) transferred their manufacturing facilities into
low-cost countries, eg where the cost of labour of a skilled workforce is lower
and/or (ii) similarly moved certain distribution functions and/or (iii) similarly moved
valuable intangible property out of the jurisdiction where they were acquired,
developed or exploited.
The chapter sets down a process for setting or testing the arm’s length principle in
business restructuring operations and gives a worked example.
Similar to the OECD TPG chapter on business restructuring, it outlines that

restructuring can involve the conversion of fully-fledged manufacturers to toll
manufacturers and fully-fledged distribution companies to limited risk. It also
mentions the transfer of either trade or marketing intangibles to foreign intellectual
property holding companies.
The chapter does include a section on operational considerations of the

restructure. You may find it useful to look at Chapter B.8 of the UN Manual at this
point.

17.6 Tax Authority Response to Business Restructuring
The issue of business restructuring is a major concern for many tax authorities, given
the significant erosion of the tax base that it can create. Tax authorities have
sought to address the issue in a number of ways:
• Identifying and targeting restructuring
• Imposing exit charges
• Additional tax challenges
These are discussed in more detail below.
Identifying and Targeting Restructuring
The most obvious step taken by tax authorities has been to focus efforts on
identifying where business restructuring has taken place. In some instances, this
involves formal disclosures – in Australia, the International Dealings Schedule
requires taxpayers to disclose any business restructuring as part of the tax return,
whilst in the UK, any taxpayer hoping to maintain a low risk status with HMRC would
be expected to discuss the restructuring with the Customer Relationship Manager
at an early stage in the process.
In other cases, tax authorities are looking for the signs of potential restructuring. This
might be in the form of significant changes to the profit profile of the taxpayer, as
disclosed through the tax return, or through news and media releases about
organisational changes. Tax authorities will often look at public disclosures about
business restructuring (including the purpose and expected benefits) to determine
whether they accord with tax position being taken, and challenge any
discrepancies.
Imposing Exit Charges
As noted in the OECD TPG, there needs to be consideration of whether there

should be a payment to reflect the restructuring itself. In practice, tax authorities
take a range of approaches to this. In part, it depends on the scope of local tax
legislation. For a lot of countries, exit charges are restricted to Capital Gains Tax
applied where a tangible or intangible asset has been transferred. If it is not
possible to identify such an asset that has been transferred, then no exit charge
can be applied, even with a significant reduction in local profit (although the
natural consequence of this is increased focus on post-restructuring transfer
prices).
At the other end of the spectrum, some countries, such as Germany, will seek to
apply an exit charge based on the net present value of profits transferred out of
the jurisdiction, with considerable efforts required to demonstrate that such a
charge is not payable.
The European Union Anti-Tax Avoidance Directive (ATAD) which was approved by
the ECOFIN in July 2016 (as we saw in an earlier chapter) includes a requirement
for an exit charge (taxing market value minus tax value for assets) when a
taxpayer moves a permanent establishment or residence of a company out of a
Member State’s taxing jurisdiction. A five-year deferral is possible subject to certain
requirements being met. The Directive applies from 1 January 2020 for exit charges
(the other provisions of the ATAD apply from 1 January 2019).

Additional Tax Challenges
Although there is considerable focus on transfer pricing, there are a range of other
avenues that tax authorities will consider when challenging business restructuring:
• Permanent Establishment (PE) (A subject we will look at in detail in later

chapters) – Tax authorities will often seek to establish whether the Principal
created through the business restructuring has a taxable presence in the local
country. Many aspects of a centralised model can give rise to a PE risk,
including (but not limited to) ownership of stock in the country, time spent in
the country delivering local country support, the need to register the Principal
for local Goods and Services Tax (GST)/Value Added Tax (VAT)/customs
purposes, and local sales operations legally binding the Principal in its
negotiations with customers.
If a PE is established by the tax authority, profits would need to be attributed

to the PE in the same way that would apply to a branch, and as a result, a
significant portion of the profits transferred to the Principal may once again
become taxable in the local country. Furthermore, additional penalties may
apply. In determining whether a PE exists, consideration needs to be given to
both the underlying tax legislation in a country, and the relief provided by any
relevant double tax agreements (DTAs). The variation that exists in both
legislation and DTAs means that a structure rolled out identically in a number
of countries may have a different PE analysis in each.
• Withholding tax (WHT) – WHT may well apply to certain payments made by
local operations to a Principal. Although this is not relevant to a basic buy-sell
Principal structure where the only transactions involve the flow of physical
goods, it is relevant to more complex models where payments to the Principal
may be in the form of variable royalties or value-added service fees. Under
such circumstances, the local tax authorities may not accept the transaction
and seek to substitute it with other transactions in such a way that carries the
largest WHT burden.
• Controlled Foreign Company (CFC) rules – Where the Principal entity is not the
parent company within a group, consideration needs to be given to CFC
legislation in the parent jurisdiction (and any jurisdictions for holding
companies between the parent and the Principal). CFC legislation is complex
and requires separate analysis, but broadly speaking, depending on the tax
rate in the Principal, the nature of the income it earns and the extent of the
activities it undertakes, the tax authority might seek to deem the Principal to
be a CFC of the parent and tax the profits it earns. Mitigation against this risk
depends on the specific rules of the jurisdiction in question but generally
requires an appropriate level of substance in the Principal through undertaking
sufficient economic activities.
The EU Anti-Tax Avoidance Directive (ATAD) mentioned above also includes a

requirement for Member States to have CFC rules. The ATAD sets down that
there should be taxation of certain non-distributed income of CFCs. Entities of
which the voting/capital/profit rights are directly or indirectly owned for more
than 50% by residents and PEs that are not subject to tax are CFCs if their
actual corporate tax paid is lower than the difference between the corporate
tax that would have been charged in the Member State of the taxpayer and
the actual corporate tax paid by the entity or PE. The rules do allow the
Member States to include certain ‘de minimis’ carve outs and there are
certain exceptions for cases of substantive economic activity and for genuine
or non-tax driven arrangements. Member States were required to have

conforming provisions by 31st December 2018 and the CFC rules are operative
from 1 January 2019.
• Indirect taxes – Changes to the transaction model will have a knock-on effect
for indirect taxes, which can be an area that tax authorities will seek to
challenge. Following conversion to a typical Principal structure, it is often the
case that prices for goods sold into distribution territories will increase. In some
cases this will be a dramatic increase. Whilst this is of benefit to customs
authorities charging duty on an ad valorem basis, this nevertheless brings with
it the challenge of why prices have changed. It can be difficult to justify to
customs officials why prices have significantly shifted when the underlying
product entering the country has not changed at all. The challenge to defend
against is that historic pricing has been incorrect, and business restructurings
can often lead to customs audits for periods prior to the restructuring.
17.7 Conclusion
Disputes may occur between a parent company and tax authorities in relation to
whether business decisions are commercial and not purely tax driven. Furthermore,
tax authorities are likely to be concerned if valuable intangibles are transferred
from existing manufacturers without adequate compensation.
From a business perspective, restructuring seen as a whole may constitute a

commercially sound business decision, focusing on optimisation, removing
duplication and reducing costs. However, tax authorities may see it as a taxable
transfer of intellectual property rights as well as a significant part of the business.
Documentation, proof of sound commercial rationale and risk analysis are key to
supporting the business decision. The fact that transfer pricing is high on the to-do
list of most tax authorities is a clear warning. It is also important to understand that
although most countries comply with the OECD TPG on business restructuring there
are still differences in approach and methodology amongst different jurisdictions,
so it is very important to look at the overall picture when considering a
transformation project, map the tax effects in each jurisdiction and consider any
transfer pricing issues that may arise. Any cost or tax benefit should be checked
against any tax risk or exit charge, which might be triggered by the restructuring
process. Last, but not least, robust documentation, clear intra-group agreement
and strong proof of commercial rationale driving the transformation are essential
in reducing the risk of potential tax audits and consequent adjustments.

CHAPTER 18
NON-RECOGNITION ISSUES
In this chapter we are going to examine the circumstances in which a tax authority may
seek to disregard or not recognise a transaction between associated enterprises.
18.1 Introduction
Non-recognition refers to the extent to which a tax authority may, for tax purposes,
set aside the contractual terms of a controlled transaction. The tax authority may
take one of two approaches:
• disregard the actual transaction, or
• disregard the actual transaction and, substitute another, notional, transaction

which the tax authority asserts is closer to the substance that might be
expected if the taxpayer had been dealing at arm’s length with an
independent party.
These approaches are illustrated in paragraphs 1.127 and 1.128 (1.147 - 1.148) of
the OECD TPG (see below).
The arm’s length principle certainly governs the prices and other conditions of the
controlled transaction. However, there are cases where it is not just the prices or
other conditions associated with the controlled transaction which are being
challenged by the tax authorities, but also the nature of the transaction.
There is a wide variety of domestic tax law anti-avoidance approaches which may
permit non-recognition of transactions other than through transfer pricing
measures per se. These approaches include:
• ‘Substance over form’ and ‘abuse of law’ doctrines: these are, respectively,
common law and civil law concepts which require that the purpose of the
legislator prevails over the actual form of a transaction if that form is not
specifically contemplated by the law and the same economic results could
have been obtained in another manner.
• ‘Sham’ doctrine: legal form of the transaction does not cover the reality
intended by the parties.
• General anti-avoidance rules.
• Targeted anti-avoidance rules, such as the US economic substance doctrine

which may disallow tax benefits if there is no purpose or effect to a transaction
other than tax minimisation.
In addition, as we saw in an earlier chapter, the EU have brought in an Anti-Tax

Avoidance Directive (ATAD) which contains a general anti-abuse rule (GAAR)
allowing tax authorities to ignore non-genuine arrangements where the main or
one of the main purposes is to obtain a tax advantage that defeats the object or
purpose of the tax provision. Arrangements will fall within the GAAR to the extent

they are not put into place for valid commercial reasons which reflect economic
reality.
However, the main focus of this chapter is on the OECD approach to non-
recognition in a transfer pricing context. The content of the OECD TPG in this
context is important because in many countries there is an explicit or implicit
requirement to respect the OECD TPG in domestic tax law.
18.2 The OECD Transfer Pricing Guidelines (TPG)
Following adoption of the final reports on Action Points 8 to 10 of the BEPS Action
Plan, Section D2 of Chapter I of the OECD TPG looks at when a transaction can be
disregarded. It is still the case that transactions can only be disregarded in the
exceptional circumstances as laid down in the OECD TPG (see paragraphs 1.122
to 1.125 (1.142 - 1.145)).
Section D2 recognises that the non-recognition of transactions is controversial and

can lead to double taxation (see paragraph 1.123 (1.143)). Paragraph 1.121
(1.141) states that:
‘every effort should be made to determine the pricing for accurately

delineated transactions under the arm’s length principle.’
The OECD TPG recognise that associated enterprises may enter into a much wider
variety of arrangements than non-related enterprises. Therefore the OECD TPG
state that when tax authorities are considering non-recognition:
‘the key question is whether the actual transaction has the commercial
rationality of arrangements that would be agreed between unrelated parties
under comparable economic circumstances, not whether the same
transaction can be observed between independent parties’ (see paragraph
1.122 (1.142)).
The emphasis in the OECD TPG is on commercial rationality as per the above
quote. Further the guidance goes on to say:
‘the mere fact that the transaction may not be seen between independent
parties does not mean that it should not be recognised.’(see paragraph 1.122
(1.142)).
Non-recognition will be allowed where:
‘the arrangements made in relation to the transaction, viewed in their totality,

differ from those which would have been adopted by independent
enterprises behaving in a commercially rational manner in comparable
circumstances’(see paragraph 1.122 (1.142)).
The OECD TPG point out in the same paragraph that it is also relevant to look at
whether the multinational group as a whole is left worse off on a pre-tax basis as a
result of the transaction.
When non-recognition takes place then if a substitute transaction is appropriate

the transaction will be replaced by a structure that closely corresponds with the
facts of the actual transaction and achieves a commercially rational expected
result arriving at an acceptable price for both parties (see paragraph 1.124
(1.144)).

Section D2 finishes with some worked examples. The first looks at an internal
insurance arrangement; here the transaction is not recognised (see paragraph
1.127 (1.147)). The second looks at the sale for a lump sum payment of all the
future outcome of research and development activities. In this case the
transaction is replaced by an alternative structure (see paragraph 1.128 (1.148)).
 Illustration 1
As an illustration, differing tax authorities have different approaches to when a

loan should be treated as debt and when it is to be treated as equity. Consider
the position for the Zeta group of companies:
Zeta Holdings Ltd (resident in

Cayman Islands)
Zeta (UK) Limited (resident in UK) Zeta Romania SRL (resident in

Romania)
The UK and Romanian subsidiaries are established with nominal equity capital of
£1/€1 respectively. They are both financed by Zeta Holdings Ltd by way of loans of
£1 million each at an interest rate of 6%. They have both incurred substantial losses
and have few assets.
In the UK, it is likely that transfer pricing rules would operate so as to deny a tax
deduction for the interest charged on the basis that taking into account all
factors, at arm’s length, Zeta (UK) Limited could not have borrowed £1 million
However, that is the full extent of the transfer pricing impact – the UK rules would
not then provide for the loans to be recognised as share capital and the interest
payable to be substituted for a payment of dividends.
In Romania, the thin capitalisation rule of 3:1 debt:equity applies. Because this is
breached, a tax deduction will be denied to Zeta Romania SRL for the interest
charged. Moreover, the interest is classified as a dividend and withholding tax is
applied accordingly.
Factors to be Considered
Against what criteria can it be assessed whether or not ‘arrangements made in

relation to the transaction, viewed in their totality, differ from those which would
have been adopted by independent enterprises behaving in a commercially
rational manner’?
Options Realistically Available
The OECD TPG note, at paragraphs 1.38 and 1.122 (1.142), that an independent
enterprise will compare a proposed transaction with the other ‘options realistically
available’ to it. This is in the context of comparability analysis rather than non-
recognition, but if faced with an “irrational” controlled transaction it would appear
instructive to ask the question ‘Is there another option realistically available which
independent parties acting at arm’s length might have chosen?’ If so, that option
may provide a means of substituting the actual transaction undertaken.
This concept is clearly not without difficulty. Practical issues include: How are the
options identified? What is ‘realistic’? When is that judgement to be made? The

OECD TPG do not seek to answer such questions. Paragraphs 9.27 to 9.31 of
Chapter IX of the OECD TPG do expand on the question; however the discussion
focuses on the fact that the question has to be looked at taking into account the
wider economic circumstances and that the requirement to consider other
options, including the option not to enter into a transaction, is not to be taken to
be a requirement to document every possible hypothetical option.
In the absence of objective tests, there is clearly scope for disagreement between
taxpayers and tax authorities and a risk that tax authorities will use hindsight to
argue that the taxpayer could at arm’s length have chosen a ‘clearly more
attractive’ option.
Risk
When looking at the question of non-recognition the allocation of risk between the
parties of a transaction is important. You will recall from earlier chapters that the
allocation of risk is an important part of the functional analysis and that the
contractual allocation may not reflect the actual position. The OECD TPG pay
special attention to the allocation of risks underlying the contractual terms as well
as the capabilities and responsibilities to manage those. Section D.1.2.1 sets out a
six step analytical framework for identifying risk and who has control over risk.
Chapter IX on business restructurings at paragraphs 9.19 to 9.42 also looks at risk.

Emphasis is placed on whether the parties conform to the purported allocation of
risks and whether the party to which the greater risk is allocated has the capacity
to control it. It is made clear that the assessment of who has the capacity to
control risk is no different to that set down in Chapter I of the OECD TPG. The
discussion in Chapter IX emphasises the importance of applying the six step
analysis set down in Section D1.2.1 of Chapter I. This analysis should be used first to
confirm which company has control of the risk before the business restructure and
whether any change has taken place as a result of the restructure. Applying the
guidelines in Chapter I of the OECD TPG, it cannot simply be assumed that the
party that has contractually assumed the risk controls the risk.
Business Restructuring Aspects
As we have seen in earlier chapters, in recent years a large number of MNEs have
embarked on complex value chain restructuring projects leading to the formation
of principal entities in lower tax jurisdictions, which are meant to take on the more
‘non-routine’ functions and the major business risks (eg stock, customer, product,
etc.) and therefore, attract a large portion of the group profits.
Centralisation is often pursued by large MNEs not as way to achieve tax

advantages, but in the pursuit of cost efficiencies, better control and as a way to
stand against the competition.
However, when centralisation also generates a tax advantage it is crucial for the
contractual arrangements to match the economic substance in each of the
parties.
For example, the common use of limited risk distributors or sales agents, which act
on behalf of a main super distributor, is often challenged by tax authorities. If
successful, the transaction as drawn up would not be recognised; instead it would
be replaced with a transaction between a local entity which is taking on more risk
and functions than the main distributor.
Another classic example is in relation to contract R&D arrangements, where an

enterprise pays a related party for the development of IP, which is then exploited

by the paying enterprise. Tax authorities often argue that the R&D company might
be acting on its own behalf and the IP being created resides locally and does not
belong to the enterprise paying for the R&D expenses. It is important to ensure that
not just the pricing reflects the nature of the transaction (eg using net cost plus to
remunerate the R&D service provider), but also the risk and functional profile (eg a
clear direction on the R&D has to be provided, all losses that might arise from
unsuccessful launch or use of the IP should be covered by the paying enterprise
and not by the R&D service provider, etc.).
Commercial evidence is also very important when assessing the nature of a

transaction or facing a non-recognition challenge by a tax authority.
The existence of official documents (eg board papers) clearly showing the
commercial goal to be achieved when setting up a contractual arrangement
between related parties does not prevent challenges from tax authorities, but it
provides evidence that the reason for entering into the contract was driven by
commercial needs (eg cost reduction, market penetration, volume discounts,
etc.).
Paragraphs 9.34 to 9.38 in Chapter IX of the OECD TPG address the issue of non-
recognition from a business restructuring perspective. This confirms the following:
• The taxpayer has the freedom to decide whether and to what level they
perform the functions and take on the risks, and what resources they employ.
The entrepreneurial freedom of disposition also includes that the taxpayer can
freely decide if the functions are performed by the taxpayer themselves or by
another company within the group, are allocated to several companies or are
assigned to a subcontractor.
• Tax authorities do have the right to determine the tax consequences of the
structures put in place by MNE subject to their obligations under DTC; in
particular Article 9 of the OECD Model DTC and their domestic law.
• Only in exceptional circumstances will non-recognition be appropriate.
• MNE groups implement business models that may be rarely, if ever, found at
arm’s length. That does not automatically make them irrational.
• It is not appropriate to expect the members of a MNE group to behave as if

they were independent of each other – what matters is the outcome.
• In evaluating ‘options realistically available’ a wide range of factors needs to

be assessed including all the conditions of the restructuring, rights and assets of
the parties, compensation for the restructuring itself and post-restructuring
remuneration.
• The restructuring must make commercial sense for the individual members of
the MNE group, as well as the group as a whole.
• A notional transaction to be substituted for the actual transaction must

respect as closely as possible the facts of the case.
Paragraphs 9.122-9.125 give an example of restructuring transactions where non-

recognition may be appropriate. This is replicated, in slightly modified and
abbreviated form, below.

 Illustration 2
Pre-reorganisation
Company A in Country A: Head

Office and valuable brand owner
Contract manufacturing Distribution company in Country

company in Country B C
Post-reorganisation
Company A in
Country A: Head
Office
Transfer of
brand
↓
Contract Distribution company Company Z in Country

manufacturing in Country C Z: valuable brand
company in Country owner – No staff
B
No risk-bearing
capacity
The MNE group headed by Company A manufactures and distributes branded

goods. The group derives most of its revenues and profitability from its valuable
brand which is owned by Company A and maintained and developed by 125
staff in Country A. Company Z is formed in Country Z. The brand names are
transferred from Company A to Company Z in exchange for a lump sum.
Thereafter, Company A is remunerated on a cost plus basis by Company Z (and
Companies B and C) for the services it performs, but the excess profits after
remunerating Companies A, B and C for their “routine” functions now accrue to
Company Z.
• No reliable evidence can be found of independent enterprises allocating the

valuable brand, and attached risks, as Companies A and Z have done;
• Company Z lacks substance: it has no staff to control risks associated with the
brand development. Those functions in fact continue to be performed in
Company A whose senior management team visit Company Z once a year to
formally validate strategic decisions already taken in Country A.
There is a disconnect between the legal ownership of the brand on one hand and
the economic substance and continuing beneficial ownership on the other hand.
A tax authority may well be expected to seek to set aside the brand transfer. In
practice, one would not expect a properly advised taxpayer to enter into a cross-
border reorganisation so blatantly lacking in substance. The correct delineation of
the transaction following a full investigation and the application of the principles
set down in Chapter I of the OECD TPG may lead to the conclusion that this

transaction is in substance a funding agreement between Company A and

Company Z.
18.3 Country Examples of Case Law and Other Guidance on Non-

recognition
There are only a limited number of cases in which national courts have agreed to
set aside for tax purposes contractual arrangements entered into between related
parties, or substitute different notional arrangements. That reflects the fact that
most national tax systems will, unless there is a huge variation between substance
and form, respect the actual transactions undertaken and instead challenge the
transfer pricing. Each proposed transaction nevertheless needs to be evaluated
against the landscape of the national tax systems of each of the associated
enterprises. Relevant developments in Canada, the US and Australia are
considered below.
Canada
There have been two important cases in Canada that have looked at non
recognition: GSK Canada 2010 [FCA 201] and Cameco Corporation [2020 FCA
112].
GSK Canada 2010 [FCA 201]
This case illustrates how controversial and difficult non-recognition and substitutions
can be especially when in relation to intangible property. (This is a case we look at
in other parts of this manual as it covers many issues.)
This case highlights two main issues:
• How the OECD TPG are applied within the local legislation; and
• Whether a transaction can be taken in isolation when applying the arm’s

length principle.
The court case revolved around the fixing of the price paid by a Canadian
subsidiary (Glaxo Canada) of a pharmaceutical company to a related non-
resident company for ranitidine (the main ingredient used for manufacturing a
branded prescription drug).
Glaxo Canada was paying a price over five times higher to buy the ranitidine from
the Glaxo Group than it would have paid to buy the ranitidine from generic
manufacturers.
Glaxo Canada paid a royalty to its UK parent company (and IP owner) to

manufacture and sell the branded drug Zantac in the Canadian market. Glaxo
Canada’s rights under the intra-group agreement allowed the Canadian entity to
manufacture, use and sell various Glaxo Group products (including Zantac), make
use of other trademarks owned by the Glaxo Group, gain access to new Glaxo
Group products and receive technical support.
However, Glaxo Canada was also obliged to acquire the main ingredient for the
drug (ie ranitidine) from a Glaxo approved source (ie Adechsa S.A., (Swissco)
subsidiary of the GSK Group).
The Canadian Revenue Authority (CRA) reassessed Glaxo Canada by increasing

its income on the basis that the amount it had paid Swissco for the purchase of

ranitidine was ‘not reasonable in the circumstances’ within the meaning of the
Glaxo Canada’s position was that the price paid to Swissco was reasonable in the
circumstances when viewed in consideration with the license agreement and its
business circumstances which were not the same as the generic purchasers; most
notably it had the right to sell Zantac.
Glaxo Canada appealed the CRA’s reassessment to the Tax Court of Canada
(TCC), which affirmed the CRA’s adjustment of the transfer price on the basis of
the prices generic drug companies were charged for ranitidine. The TCC
supported the CRA’s position that, in determining the reasonableness of the
amount paid, the license agreement was an irrelevant consideration because
‘one must look at the transaction in issue and not the surrounding circumstances,
other transactions or other realities’.
The CRA’s position was that the appropriate analysis is what is the arm’s length
price for the active ingredient and any other circumstances should be
disregarded. According to the CRA, it is not important whether the buyer wanted
to acquire the ranitidine for the generic market or the premium brand market.
The CRA viewed the transaction as a separate item and not in the context of the
larger picture. The lack of clear guidance in the legislation left room for
interpretation.
The Supreme Court of Canada upheld Glaxo Canada’s appeal that the licence
agreement must be taken into account in examining the purchase price for
ranitidine; the case was remitted back to the TCC to determine pricing. The
Supreme Court decision was also interesting in holding that OECD TPG are not
binding.
In addition to the points noted above it can be said that the wider implications of
the Supreme Court decision include:
• That as a result of statements made by the Supreme Court, an arm’s length

price can fall within an acceptable range of prices
• The guidance given by the Supreme Court suggests that Canadian courts
must keep an eye on the bigger picture in making their transfer pricing
determinations, and the determination of arm’s length pricing is quite distinct
from the narrower concept of fair market value
• The decision lead to changes being made to the detail of Canadian

legislation eg removing ‘in the circumstances’ from the reasonableness test
This case shows how difficult and controversial the application of transfer pricing
principles can be. The taxpayer should carefully consider all the implications when
making decisions on contractual arrangements for intra-group purposes when the
transactions are particularly complex.
Cameco Corporation [2020 FCA 112]
This case looked at the transfer price between Canadian affiliates that produced
uranium and its distribution affiliates.
Cameco’s Canadian mines produced a significant amount of uranium at only $10

per pound, which was the lowest cost of production globally. In 1999, Cameco
established the Swiss affiliate Cameco Europe AG (CEL) to distribute the uranium

produced by these mines and established Cameco Europe SA (CESA) to purchase

the uranium that Russia had formerly used in its nuclear arsenal. CEL later assumed
the role of CESA thereby selling Russia's inventory of uranium and Cameco
production of uranium.
This litigation involved the period from 1999 to 2006. During the early years of this
intercompany arrangement, uranium prices fell below $10 per pound as Russia
inventory sales dominated the market. In 2003 and 2004, uranium prices rose with
the average price over the 1999 to 2004 period being $11 per barrel. Uranium
prices continued to rise going from an average of $27.93 per pound to $47.68 per
pound over the period considered. As such, CEL received substantial profits under
the intercompany pricing policy while the Canadian affiliates earned very little
profit. The CRA challenged this intercompany pricing arrangement using several
approaches to argue that all of these Swiss based profits should represent
Canadian taxable income.
The CRA presented essentially three different positions. Its legal position was that
the restructuring was a sham and all of CEL's profits should be taxable in Canada.
Its second position was based on the claim that CEL was a limited function
distributor so its gross margin should be very modest. Its third position was the
recharacterisation rule (now non-recognition).
The first two approaches were later abandoned. The CRA sought to rely on the
recharacterisation rule. The CRA reassessed Cameco’s 2003, 2005 and 2006
taxation years, reallocating approximately $480 million of the Swiss subsidiary’s
profits under this rule.
For recharacterisation to apply under the Canadian law, the CRA needed to show
that the transaction or series of transactions (i) would not have been entered into
between persons dealing at arm’s length and (ii) can reasonably be considered
not to have been entered into primarily for bona fide purposes other than to
obtain a tax benefit (section 247(2)(b)(i) and (ii) of the Act). The CRA sought to
interpret the first condition very widely.
In June 2020, the Federal Court of Appeal (FCA) released its decision upholding
the Tax Court’s decision in favour of the taxpayer and confirmed a restrictive
interpretation of the recharacterisation provisions contained in sections 247(2)(b)(i)
and (b)(ii). The FCA confirmed that the test in section 247(2)(b)(i) imposes an
objective test (ie. whether any arm’s length persons could enter the transactions
on any terms), and not a subjective test, taking into consideration the situation of
the particular taxpayer.
The FCA also ruled that if the recharacterisation rule had applied, the
consequence should not have been drastic transfer pricing adjustments, such as
the total reallocation of profits to Cameco Canada, because this would amount
to disregarding the existence of the other parties to the transactions (CEL and
CESA).
An appeal was lodged in October 2020 but was dismissed in February 2021.
The implications from this case include:
I. That in Canada the recharacterisation provisions will not apply where the
taxpayer’s arrangements are commercially rational and where nothing
prevents the determination of a transfer price, even if there is a tax-oriented
purpose to the particular structure utilised.

II. That where recharacterisation does apply, the CRA is required to substitute
arm’s length terms and conditions for the transaction it has determined would
have been entered into between arm’s length persons, rather than
substituting the terms and conditions with nothing on the presumption that the
no transaction would have occurred.
III. Given the Tax Court’s willingness to ascribe commercial rationality to any
transaction or series that could be transfer priced under sections 247(2)(a) and
(c), it remains difficult to predict the types of transaction that would be so
commercially irrational that the recharacterisation provisions in Canada would
apply.
United States
US transfer pricing regulations permit the IRS to not recognise transactions that lack
economic substance and amend to a form which more closely equates to the
economic substance. (Treasury Regulation s.1.482–1(d)(3)(ii)(B))
There are a number of leading cases where the IRS has failed to persuade the
Courts that non-recognition is appropriate. For example, in Eli Lilly v Commissioner
856 F.2d 855, a US corporation transferred patents and know-how to a Puerto
Rican manufacturing subsidiary. The IRS asserted that this transfer should be
disregarded on the basis that the US company could have retained the revenue
streams from the intellectual property transferred. That was rejected by the Tax
Court and Court of Appeals.
In a 1992 decision of the Tax Court, Kwiat v Commissioner 64 TCM (CCH) 327, a
purported lease with reciprocal put/call options was substituted for a secured
loan. This was not a transfer pricing case as such because the parties to the
transaction were not associated enterprises. However, it serves as a contextual
reminder of the need to consider all possible legal tools at the disposal of the tax
authority which might ultimately result in the disregard or non-recognition of a
transaction.
In the Kwiat case, the appellant taxpayers leased shelving equipment to another
party. There was a put option permitting the taxpayers to sell the equipment at a
projected profit to the taxpayers.
The Tax Court held that the rights and responsibilities of ownership of the shelving
had passed to the purported lessee: the lease was in substance a sale and the
taxpayer was denied tax depreciation in respect of the assets in question.
In June 2016 the US Tax Court ruled in favour of the taxpayer in Medtronic Inc. &
Consolidated Subsidiaries v. Commissioner (T.C. Memo. 2016-112).
The IRS contended that the royalties payable from a Puerto Rican affiliate to
Medtronic, Inc should be increased by $US1.4 billion. The Tax Court rejected the
IRS’s proposed transfer pricing method as arbitrary, capricious, and unreasonable.
The court instead accepted Medtronic’s use of the Comparable Uncontrolled
Transaction (CUT method) but made adjustments to take account of differences
between the licenses and the comparable transaction relied on by Medtronic.
The IRS argued that the Puerto Rican company was merely an assembler which
only had a minor role in design manufacturing and development. The court
disagreed; they found that the company in Puerto Rico was involved in every step
of the manufacturing process and that the functions it performed were critical to
overall profitability of the group. It was found that product quality was a critical
factor. An interesting point is that the court found that a license could add to the

assets of the licensee. The rights under the license can become an intangible asset
for the licensee.
Australia
Glencore [2020 FCAFC 187]
This is an Australian case that looks at an intra-group agreement for the purchase
of copper concentrate by Glencore International AG (GIAG), a Swiss company,
from its subsidiary, Cobar Management Pty Ltd (CMPL), an Australian company
which owned and operated a mine in Australia.
Pricing arrangements under the contract for the sale of the copper concentrate
were changed in 2007; the taxpayer stated that change was necessary as there
was uncertainty in the price of copper and an increase in costs associated with
the mine.
Before 2007, the intercompany contract between GIAG and CMPL only
compensated GIAG for its very modest marketing functions as CMPL was
responsible for paying for the actual cost of smelting. From 2007 to 2009, GIAG
paid third-party smelters and charged CMPL 23% of revenue in a price
participation arrangement. The taxpayer contended that the new arrangements
would give CMPL a better risk position. The Australian Tax authority (ATO) thought
the change was excessive and that it shifted too much profit to GIAG.
The ATO wanted to amend the pricing mechanism to what they thought
independent enterprises would have entered into. In addition to changing the
pricing mechanism the ATO sought to add additional clauses to the contract.
Amended assessments were issued to reflect the mechanism previously used and
increased the consideration paid under the contract by AUD241 million over the
years 2007 to 2009.
The taxpayer defended the change in arrangements by providing evidence of

similar contracts from 2003. The ATO argued that these contracts were not
comparable to a contract entered into in 2007. The court recognised that the
comparables were not perfect but stated that they still provided appropriate
reference points.
The ATO and the taxpayer presented evidence to the court from experts. The
experts did not agree on the pricing that independent parties would have agreed
on. The court took the view that a taxpayer did not have to demonstrate its pricing
is more likely than that put by the ATO or ‘to lead evidence to negate positive
claims put up by the Commissioner’, they just have to show through the evidence
that its position is within a range of ‘commercially acceptable arm’s length
outcomes’.
A key aspect of the case was how far the tax authority could go in making
changes to the intra-group agreement. In this case the transfer price was based
on a formula; the court stated that if required an alternative formula could be put
forward by the tax authorities to meet the arm’s length principle however
additional clauses that didn’t affect the calculation of the consideration may not
be added. The characteristics of the hypothetical independent parties were also
examined; it was determined that the characteristics assigned to the hypothetical
parties must only be those that would impact on the consideration payable under
the hypothetical contract only – objective attributes and features should be
assigned. Any features of the agreement that arose from the special relationship
that exists between group members shouldn’t be included. Risk was an important

part of the formulation of the consideration thus it was appropriate to take

account of how a hypothetical party would take account of the risk when
determining the mechanism for pricing. It was agreed that there could be a range
of arm’s length outcomes that could meet the requirements of the law in Australia
as what controls the range of acceptable arm’s length outcomes is the concept
of what might reasonably be expected; and that a degree of flexibility and
pragmatism is required in determining if the taxpayer has met the requirements of
the law.
The court stated that the OECD TPG are merely ‘guidelines’, stating that the
language in them is ‘highly generalised’ and ‘frustratingly opaque’. It was noted
that ‘reconstruction’ (now recharacterisation) was not limited to circumstances in
the OECD TPG.
The final decision on this case was made by the Federal court in November 2020.
The court agreed that the evidence of the taxpayer’s expert should be accepted
and that the rate of 23% used in the agreement was within an arm’s length range.
(An appeal was denied by the High court in May 2021).
Important points that arise from this case include:
• Substituting pricing methodology is permitted (as discussed above).
• Substituting terms which do not define price is not permitted (as discussed
above).
• Commercial prudence is sufficient. A taxpayer will succeed if it can

demonstrate that its position is a ‘sufficiently reliable prediction’ which is
reasonable, even where the ATO attempts to assert that it is instead
unreasonable.
• Taxpayers are not required to maximise profitability at the expense of risk

mitigation. The court recognised that business decisions occur with a range of
considerations in mind, and that there can be a trade-off between risk and
profits.
• It is not appropriate to rely on estimates or forecasts as certainties. This will be

the case even if in hindsight they turned out to be correct.
• Comparables that are not identical are still important as they can provide a
valid reference point.
• In Australia the OECD TPG do not replace the text of the statutes. The OECD
TPG are only a guide as to how a revenue authority or a taxpayer might apply
the ‘arm’s length principle’. They do not replace the text of the legislation in
Australia.
• Prior contracts are not determinative of whether the contract in dispute is

arm’s length. It was held that the ultimate issue for determination is whether
the contract in dispute was priced at arm’s length, not whether an arm’s
length party would have agreed to amendments to a prior agreement.
• The Australian courts will take a rigorous approach to evidence (this is in line
with the Chevron decision – see earlier chapter), thus taxpayers should ensure
they have good documentary evidence of the analysis that was undertaken
to arrive at pricing.

CHAPTER 19
PERMANENT ESTABLISHMENTS
The concept of permanent establishments (PEs) are covered in this chapter. We will
consider:
– basic features – fixed place of business PEs, construction sites and dependent agency
PEs
– the application of the concept to specific activities such as offshore activities
– the auxiliary and preparatory activities exclusion
– the multilateral instrument
– case law
– double taxation
19.1 Introduction to Permanent Establishments
Understanding the rules relating to permanent establishments (PEs) has two steps
to it. The first step is to understand when a PE exists. The second step is to look at
how profits are attributed to that PE (this is done in conjunction with considering
the tax implications). We will look at the second step in the next chapter.
Broadly an overseas resident has a substantive presence in a state if he meets the

threshold of having a PE (Article 5 OECD Model Treaty, see below).
At the same time, if that person is carrying on business through the PE then tax may
be due in the state in which the PE is established as well as in the state of
residence (Article 7 OECD Model Treaty, see later chapter).
Therefore, on its own, without a business activity through it, a PE may not of itself
give rise to a tax liability.
In its simplest form a PE exists where a company is resident in one country (referred
to as the head office) but also has a business conducted from a fixed base in
another country (the branch). The income attributable to the other fixed base
usually attracts a tax liability in the second country. In many businesses the branch
will have its own management structure maintaining separate accounts. It will not
have a separate legal persona as it is just part of the company.
The fundamental rationale behind the PE concept is to allow, within certain limits,
the taxation of non-resident enterprises in respect of their activities (having regard
to assets used and risks assumed) in the source jurisdiction.
A practical example of the use of a branch is where a bank or regulated financial

business with capital requirements starts a business in a foreign country. It will use a
branch so it can meet its local regulatory capital requirements by relying on the
capital of the whole entity.
In its more complex form of deemed PE, it is a tax ‘fiction’ enabling tax authorities
to impose corporate taxes on the deemed branch. A third type of PE is again a
tax ‘fiction’ where there is a deemed branch providing services.

19.2 Physical Presence PE
Article 5 (1) states:
‘the term ‘permanent establishment’ means a fixed place of business through

which the business of an enterprise is wholly or partly carried on.’
Paragraph 12 of the 2017 commentary to the OECD Model Treaty tells us that:
‘Whilst no formal legal right to use a particular place is required for that place
to constitute a permanent establishment, the mere presence of an enterprise
at a particular location does not necessarily mean that location is at the
disposal of that enterprise.’
The commentary then goes on to provide examples by way of illustration.
The first example is that of a salesman who regularly visits a major customer to take
orders and meets the purchasing director in his office to do so. The commentary
states:
‘In that case, the customer’s premises are not at the disposal of the enterprise
for which the salesman is working and therefore do not constitute a fixed
place of business through which the business of that enterprise is carried on
(depending on the circumstances, however, paragraph 5 could apply to
deem a permanent establishment to exist).’
Dr. Juris Arvid Aage Skaar (academic and author of the seminal 1995 book
‘Permanent Establishment: Erosion of a Tax Treaty Principle’ Kluwer Law
International) makes the following comments in relation to this example:
‘it is not convincing if the Commentary suggests that a salesman who visits
customers to take orders is an example of a person who is merely present in
the office of the purchasing director. Is the salesman’s activity (taking orders
from the customer) not a business activity? The premises are factually used by
the salesman in connection with sales work, and this is the enterprise’s core
business. This can hardly be considered to be “mere presence”. However, the
essence is that the OECD countries have agreed that a salesman’s regular
and lasting use of the facilities of the customer in such a way shall not
constitute a PE.
The example shows that if the requirement of a legal right of use test for PE is
abandoned, a PE may be constituted on the basis of facts that the OECD
countries otherwise agree should be kept outside source-state taxation.
Because of this, it is necessary regarding the ‘examples’ to explain situations
that will not be considered to be a PE anyway.’
The second example (at paragraph 15 of the 2017 commentary) is that of an

employee of a company who, for a long period of time, is allowed to use an office
in the headquarters of another company (eg a newly acquired subsidiary) in order
to ensure that the latter company complies with its obligations under contracts
concluded with the former company.

The commentary states:
‘In that case, the employee is carrying on activities related to the business of
the former company and the office that is at disposal at the headquarters of
the other company will constitute a permanent establishment of his employer,
provided that the office is at his disposal for a sufficiently long period of time
so as to constitute a ‘fixed place of business’ (see paragraphs 28 to 34) and
that the activities that are performed there go beyond the activities referred
to in paragraph 4 of the Article.’
Skaar comments generally that:
‘The use of examples by the 2003 OECD Commentary [which remain in the
2017 version] is confusing because no distinction between factual use and
implied legal right of use to a place of business is made. It is unclear to what
extent and under what circumstances factual use is sufficient for a PE to exist.’
The fact that a PE is a fixed place of business suggests that there is some
permanency and that there is a space of some kind taken up by the business. If it
is possible to demonstrate that there is a temporary nature to an operation, it may
not be a PE. However, if the intent at the outset is one of permanence, then even
in circumstances of a sudden curtailing of the operation in unforeseen
circumstances (such as a death) then there is more likely to be a PE. Frequent use
of an office/premises for short bursts of time is suggestive of permanence.
A general note with regard to the length of time needed to indicate permanence
can be found at paragraph 28 in the 2017 Commentary, where it is noted that a
PE is normally not found to exist in situations where a business has been carried on
in a country in a place that was maintained for less than six months. This is
considered a loose ‘rule of thumb’ with regards to the time required to establish a
PE, and all facts and circumstances, including the nature of the business itself,
would need to be examined. For example, recurrent activities require
consideration of the period of time the business is operating in one place along
with consideration of the number of times a year during which that place is used,
which then may extend over a period of years (paragraph 29 of the 2017
Commentary).
Broadly, use of different premises in a locality may also constitute a PE. Even the
use of moveable (but substantially fixed and solid) premises may constitute a PE.
The latter is illustrated by the fact that a drilling rig used in the exploration of oil
may move from place to place and still be considered a PE.
Offshore activities raise important questions as to what constitutes a PE. Often in

practice there are separate articles dealing with offshore activities. ‘Extraction’
activities are covered in Article 5(2)(f). Arguably, exploration should be considered
preparatory in nature (ie within Article 5(3)) but drilling rigs do give rise to a PE in
many situations.
The OECD commentary makes reference (paragraph 41 in the 2017 version) to

automated vending machines as constituting a PE if there are persons setting up
and operating the machines – if not, it may not be a PE. This concept is similar to
the question of whether a server constitutes a PE. This and related issues are
discussed in more detail later in this manual.
Equipment leased to a lessee in a state should not create a PE of the lessor, even
where staff operate the equipment, provided that the staff are there to look after
the operation or maintenance of such equipment only, and they are under the
direction and control of the lessee (see paragraph 36 of the 2017 commentary).

Article 5(2) of the OECD Model Treaty tells us that the term PE includes:
• a place of management;
• a branch;
• an office;
• a factory;
• a workshop, and
• a mine, an oil or gas well, a quarry or any other place of extraction of natural
resources.
Note this is not an exhaustive list.
19.3 Construction Site PE
A building site or installation project constitutes a PE only if it lasts more than 12

months (Article 5(3)). The commentary tells us that the 12-month test applies to
each individual site or project. The time period we look at is to include preparatory
work in the country where the construction project is to be carried on. Difficulties
can arise with the identification of separate building projects.
When it comes to construction sites under Article 5(3) there are a number of
practical questions that are raised, including whether different parts of a locality
can constitute different construction sites – these issues depend on facts and
circumstances such as whether there is a commercial and geographic coherence
to a project. The kind of sites and activity included for this purpose include that of
the installation or construction of buildings, roads, bridges, canals, pipelines,
excavating and dredging. Demolition and clearance of sites fall under this
category as well. Even assembling movable property may be included.
The final BEPS Report also considered certain concerns with regard to the
possibility of related parties splitting up contracts into several parts – each covering
a period of less than 12 months and attributed to a different company in the
group, in order to prevent the creation of a PE under this paragraph. It was
considered that the principal purpose test (PPT), suggested as part of the work
undertaken on Action Point 6 on preventing treaty abuse, would be sufficient to
deal with this type of problem. An example of how the operation of the PPT could
counteract such actions has now been included in the commentary on the use of
the PPT rule (see paragraph 182 to commentary on Article 29 Example J). Where
the PPT was not present in a particular treaty, an alternative specific rule is also
included in the OECD commentary to Article 5(3) (see paragraph 52) which could
be used by a contracting state when drafting their treaties which applies
automatic aggregation under certain conditions.
This alternative aggregation provision is limited to scenarios where ‘closely related’

enterprises perform ‘connected activities’. A de minimis exception applies, relating
to activities carried on for not more than 30 days (which would not be aggregated
with other periods).
The 2019 case Kompleksowa obsluga boduwnictwa sp z oo v Swedish Tax Agency

HFD 2019 ref 36 (22 ITLR 419) provides a good illustration of the 12-month
construction site PE rule. As noted by the Editor in the ITLR, under the Poland-
Sweden double taxation convention, a construction site PE only existed if the work
continued for 12 months or more. In this case the Polish company conducted work

on the construction of an ethane tank in Sweden (the size of such tanks shows they
are a considerable engineering task, like a power station). The construction of the
bottom plate of the tank began on 27 August 2014 and ended on 8 October 2014
when the company staff left the site and the equipment was removed. A
completely different company then entered the site and from 9 October 2014 to
15 February 2015 that other company constructed the walls of the tank. At the end
of October 2014 the Polish company was asked to quote for work on the roof
dome of the tank. It was successful, and it began work on the roof from 16
February and ended on 17 October 2015. Clearly, if the entire period from 27
August 2014 to 17 October 2015 was taken into account, then the construction site
lasted for more than 12 months; if the period when the other company was
engaged in constructing the walls of the tank was ignored, then the Polish
company was not onsite for a total of 12 months.
The Board for Advance Tax Rulings was asked to give a ruling, and concluded that
there was not a PE because the interruption could not have been avoided and
was not merely temporary. This was challenged before the Supreme Administrative
Court. The Court noted that the treaty was based upon the OECD Model. The
treaty had been concluded in 2005 but the Court considered that it was
appropriate to refer to the relevant commentary to the 2017 Model which, even
though it contained further examples and was more comprehensive, did not
extend the scope of application of the treaty article. The OECD Model
commentary for Article 5 stated that the entire period during which work was
carried out on the site should be included, but not periods when work was
interrupted unless those interruptions were merely temporary. The OECD
commentary gave examples of temporary interruptions – those due to poor
weather, lack of materials, or strikes. As the interruption here did not arise from any
of these temporary causes, it was not a mere temporary interruption but rather a
period that should not be included in computing the 12 month duration.
Consequently, the Polish company was not engaged on the site for 12 months,
and no PE existed.
This case is also a very good example of treaty interpretation. It is notable in that
firstly the Court determined whether it was appropriate to use the version of the
commentaries in force in 2005 (when the treaty was concluded) or the most
recent version, and then, having determined that it was appropriate to refer to the
2017 version, it applied the commentary’s explanation of a temporary interruption.
The Supreme Administrative Court effectively then treated the commentary’s
explanation of a ‘temporary interruption’ as an exhaustive definition of
circumstances that should be regarded as merely temporary. You should revisit this
case when you review the later chapter on treaty interpretation.
19.4 Exclusions from Definition of a PE
Certain circumstances are deemed not to give rise to a PE (per the 2017 version)
(Article 5(4)):
a. facilities used solely for storage, display or delivery of goods or merchandise

belonging to the enterprise;
b. maintenance of a stock of goods or merchandise belonging to the enterprise

solely for the purpose of storage, display or delivery;
c. maintenance of a stock of goods or merchandise belonging to the enterprise

solely for the purposes of processing by another enterprise;

d. the maintenance of a fixed place of business solely for the purpose of

purchasing goods or merchandise or of collecting information for the
enterprise;
e. the maintenance of a fixed place of business solely for the purpose of carrying
on, for the enterprise, any other activity; or
f. the maintenance of a fixed place of business solely for any combination of

activities mentioned in subparagraphs (a) to (e).
provided that such activity, or in the case of sub paragraph (f) the overall activity
of the fixed place of business is of a preparatory or auxiliary character.
The key element of Article 5(4) is the concept that activities which are preparatory
or auxiliary to the business in nature are not sufficient to create a PE.
Amendments have been made to this paragraph of Article 5, to ensure that these
exclusions are up to date with the modern business environment. A problem with
the previous version (2014 OECD Model Treaty) is that it did not deal sufficiently
with modern, sophisticated and highly integrated supply chains. The provision of
locally held stock, accessible to the customer at short notice, may in itself create
significant value for a business, and this should be taken into account.
The qualification added to Article 5(4) (as part of the 2017 update following the
BEPS Report) such that each activity identified in the paragraph (eg. using facilities
for storage, or maintaining stock for the purposes of storage, display or delivery,
etc), regarded alone or collectively, is itself limited to being preparatory or auxiliary
in character, or the overall activity of the fixed place of business is of a preparatory
or auxiliary nature, is there to combat the opportunities for abuse of the
exceptions.
Additional guidance included in the commentary to clarify the meaning of

‘preparatory or auxiliary’ can be found at paragraph 59 of the updated 2017
commentary. Further examples for each of the sub-paragraphs are included, and
it is provided that as a general rule an activity has a preparatory nature if it is
carried on in contemplation of the carrying on of what constitutes the essential
and significant part of the activities of the enterprise as a whole. An activity has an
auxiliary nature if it is carried on to support, without being part of, the essential and
significant part of the activity of the enterprise as a whole. In addition, the
changes clarify that the activities listed at sub-paragraphs (a) to (e) refer to
activities that the enterprise carries on solely for itself, and it must be able to
demonstrate this.
At the heart of these changes is the desire to ensure that the profits from the core
activities performed in a country can be taxed in that country. When an enterprise
used a fixed place of business not only for the activities listed in paragraph 4, but
also for other activities which are not preparatory or auxiliary in nature, then the
exclusions listed in this paragraph will not apply and a PE will be created (the
profits of which, in respect of all its activities, will be attributable to such PE).
An anti-fragmentation rule was included as sub-paragraph 4.1 of Article 5 as part

of the 2017 update. This rule was included to counteract the use of the specific
activity exclusions in paragraph 4 to artificially avoid PE status – this might be done
by fragmenting one cohesive business into smaller operations such that each part
is seen as only engaged in some form of preparatory or auxiliary activity.
Under the anti-fragmentation rule the specific activity and preparatory or auxiliary
exceptions will not be available where there is either an existing PE in the local

state, or where the collective activities carried on by all the group companies in
the local territory (by way of presence in that state, or by residence in that state)
are not of a preparatory or auxiliary nature. The 2017 commentary notes at
paragraph 79 that in order for this rule to apply at least one of the places where
the activities are carried out must constitute a PE, or where this is not the case, the
overall activity resulting from the combination of the relevant activities must go
beyond what is just preparatory or auxiliary.
The anti-fragmentation rule as described applies only to the preparatory or

auxiliary/specific activity exemptions. This means it does not apply where there are
no premises in the local territory ‘at the disposal’ of the non-resident company. In
addition, the commentary notes that if a state considers that the amendments to
the sub-paragraphs of Article 5(4) (stating that each must be subject to the
preparatory or auxiliary nature requirement) are not necessary, then such state
might alternatively just use the anti-fragmentation rule instead.
19.5 Agency PE
The ‘agency PE’ concept set down in Article 5(5) was updated in 2017 such that
an enterprise is treated as having a PE where a person habitually concludes
contracts on behalf of an enterprise, or where the intermediary ‘habitually plays
the principal role leading to the conclusion of contracts that are routinely
concluded without material modification’ by the overseas company. In addition
the 2017 version includes the activity of an intermediary that results in contracts for
the transfer of ownership, or the granting of the right to use, property owned by
the non-resident enterprise, or for the provision of services by the non-resident.
As noted in the 2017 commentary, this means that an agency PE will now exist
where a non-resident company’s goods or services are being sold in a local
market by a sales entity that is doing all the work to secure the sales and establish
contractual terms, with the limited involvement of the non-resident – so even
where the contracts do not legally bind the enterprise to third parties. This covers
commissionaire arrangements where the commissionaire does not enter into
contracts with third parties that legally bind the non-resident, but the non-
resident’s property is nevertheless transferred to the third party. A key
consideration would then be whether the commissionaire had played a ‘principal
role’ leading to the conclusion of the contracts (see the updated commentary at
paragraph 94).
The 2017 commentary does note (at paragraph 96) that arrangements whereby a
person concludes contracts on their own behalf, and in order to fulfil this obligation
obtains the goods and services from the non-resident, does not create a PE. Thus
the actions of a local distributor, including a so-called ‘low-risk’ distributor, should
not cause the creation of a PE of the non-resident enterprise, where title to the
property passes from the non-resident enterprise to the distributor and then to the
customer.
In addition no PE will be created where there is merely a local marketing activity in

a country, which is designed to identify sales opportunities and contractual terms
are actively determined and decided by the non-resident enterprise itself. In
addition, no PE is created where the activities are limited to the ‘preparatory or
auxiliary’ activities per Article 5(4) (see above).
At this point it is worth having a brief look at how Article 5(5) was in the 2014 version
of the OECD Model Treaty as this will help you understand why Action Point 7 of
the BEPS Report lead to the changes that resulted in the 2017 version.

In the 2014 version of the OECD Model Treaty, Article 5(5) stated that even if the
enterprise does not fall under paragraphs 1 and 2 there may still be a PE where:
‘a person – other than an agent of an independent status to whom

paragraph 6 applies – is acting on behalf of an enterprise and has, and
habitually exercises, in a Contracting State an authority to conclude contracts
in the name of the enterprise,’
The exception being if:
‘the activities of such person are limited to those mentioned in paragraph 4

which, if exercised through a fixed place of business, would not make this
fixed place of business a permanent establishment under the provisions of that
paragraph.’
The position with regard to commissionaire arrangements under the 2014

commentary to the treaty is not entirely clear, which was a major concern and led
to the changes as noted above. Two cases highlighted this – in both a PE was held
not to have been established under commissionaire arrangements.
In December 2011 a judgement by the Norwegian Supreme Court, in the Dell

Products Ireland case (Dell Products v. Staten v/Skatt øst, Case HR-2011-02245-A)
found that the Norwegian company (Dell AS) acting as commissionaire did not
create a PE for the non-resident principal company (Dell Ireland). Per the facts, the
commissionaire arrangements entered into between Dell AS in its own name (with
clients) did not legally bind the principal. Therefore Dell AS could not be said to
have the authority to conclude contracts in the name of the Irish company.
The Norwegian court referred to the 2010 French Supreme Court case, Zimmer
(France V Zimmer Ltd CE 31/3/2010 304715). Under this case the French Court had
held that ‘something additional’ had to be present in the arrangement between
the commissionaire and the principal, beyond an ordinary commissionaire
agreement, for the principal to be bound legally towards a third party. This was
held to be the case in Zimmer regardless of the fact the commissionaire is
dependent on the principal.
The discussion draft of amendments to the commentary on Article 5 picks up on

these issues, however the final BEPS Report also considered these problems and as
noted above Article 5(5) of the OECD Model Treaty has now been amended to try
to deal with PE avoidance via commissionaire structures. Indeed the OECD
consider the 2017 changes to Article 5(5) have now dealt with this and such
commissionaires are now caught as PEs.
Independent Agents
Article 5(6) states that an enterprise is not deemed to have a PE where it carries on
business in a state through an independent agent provided that such persons are
acting in the ordinary course of their business.
The requirements for an agent to be considered ‘independent’ are much tighter in

the 2017 version of the OECD Model Treaty than in the 2014 version. The 2017
version notes that when ‘a person acts exclusively or almost exclusively on behalf
of one or more enterprises to which it is closely related’ then that person cannot
be considered an independent agent. This can generally be taken to mean
where more than 90% of an agent’s sales are to related parties (see paragraph
112 of the commentary to Article 5). Enterprises are ‘closely related’ when, based
on all the facts and circumstances, one enterprise has control of the other or both
are under the control of the same person/ enterprise.

The commentary to Article 5 also notes that a person acting almost exclusively for
one enterprise or related enterprises over an extended period of time is unlikely to
fall within the definition of an independent agent (see paragraph 111 of the
commentary to Article 5).
19.6 Position of Subsidiary
Article 5(7) acknowledges that where one company in one state controls another
in the other state, the fact that there is control shall not of itself give rise to a PE in
either state.
Thus a parent company may have a PE in the state of a subsidiary, by virtue of the
fact that space owned by a subsidiary is at the disposal of the parent company
and is used by it to carry on part of its own business in that state. A subsidiary can
also act as a dependent agent (under Article 5(5)) of a parent company. Similarly,
the same comments can be made generally in relation to related companies in a
multinational group.
19.7 Related Party
Paragraph 8 was added to Article 5 as part of the 2017 update to define a closely
related person (this term was used when we looked at Article 5(3) and Article 5(6)
above). The commentary to Article 5(8) notes that it is not the same concept as
associated enterprises that we will see in a later chapter when we look at transfer
pricing and Article 9, although it does state that there will be some overlap.
The definition of related party is based on control, looking at beneficial ownership

of more than 50% (see paragraph 119 to 121 of the 2017 commentary).
19.8 Services PE
As seen above, the 2017 (sometimes referred to as post BEPS) wording in respect of
the agency PE allows for the possibility of a PE being created where a person (the
agent) has been involved with regard to the arrangements/ contracts for the
provision of services by the non-resident company. This type of PE would be an
agency PE, as it requires a person to act as agent/ play a principal role, with
regard to an agreement for the provision of services. This differs from the type of PE
envisaged below, and only found in the commentary to Article 5, which can be
created when a person (such as an employee) is actually providing services in the
other state – the provision of those services itself being the nexus to create a
taxable PE presence (regardless of where the contract for such services was
signed or negotiated).

The commentary since 2008 has provided an alternative services PE. This addition
does not amend the definition of a PE but allows states to include alternative
wording into the PE article should states wish to during negotiations. The example
given at paragraph 144 in the 2017 commentary is as follows:
‘Notwithstanding the provisions of paragraphs 1, 2 and 3, where an enterprise

of a Contracting State performs services in the other Contracting State:
a) through an individual who is present in that other State for a period or

periods exceeding in the aggregate 183 days in any twelve-month
period, and more than 50 per cent of the gross revenues attributable
to active business activities of the enterprise during this period or
periods are derived from the services performed in that other State
through that individual, or
b) for a period or periods exceeding in the aggregate 183 days in any

twelve-month period, and these services are performed for the same
project or for connected projects through one or more individuals who
are present and performing such services in that other State.
the activities carried on in that other State in performing these services shall be
deemed to be carried on through a permanent establishment of the
enterprise situated in that other State, unless these services are limited to those
mentioned in paragraph 4 which, if performed through a fixed place of
business, would not make this fixed place of business a permanent
establishment under the provisions of that paragraph. For the purposes of this
paragraph, services performed by an individual on behalf of one enterprise
shall not be considered to be performed by another enterprise through that
individual unless that other enterprise supervises, directs or controls the manner
in which these services are performed by the individual.’
The introduction of this alternative has given rise to much comment. The
alternative has been criticised for leading away from a consensus of opinion
approach. Some think that it will lead to conflicts with the definition of construction
sites which can also fall under services. There are existing treaties with such
provisions in them and some comment that the OECD could have drawn more
heavily from these. There is also of course the comparable provision in the UN
Model Treaty.
Of course, the articles in the OECD Model Treaty are not or will not always be
adopted in full in all international tax treaties and therefore when looking at
specific cases specific country treaties have to be examined together with any
specific country legislation and international tax cases.
The UN Model Treaty is designed to aid developing states to tax a larger part of
the overseas investor’s income. In particular, the UN Model Treaty recognises
services PEs without a fixed base (discussed later).
The 2016 US Model Treaty in general reflects the OECD Model Treaty articles
relevant to PEs.

19.9 E-commerce Issues
As discussed in a later chapter, with regard to e-commerce matters the OECD

Commentary to Article 5 does deal with various e-commerce issues, at paragraph
122 onwards of the 2017 commentary. It is most interesting to note in this respect
that a server can create a PE of the company which owns the server, so this may
be an Internet Service Provider, however, it will not create a PE for the companies
which merely have websites hosted on such a server. A website is considered an
intangible entity which does not have a place of business nor a facility such as
premises, etc
However, there are various national observations and reservations with regard to
the commentary. For example, a controversial Spanish case, the 2012 Dell case
(ruling of Mar.15.2012, Central Tax Court), reflected the reservation noted by Spain
in previous versions of the commentary. In line with Spain’s position on software
driven PEs, the Spanish Court ruled that an Irish Company had a Spanish PE, as the
online store (the website through which Spanish sales were made) could qualify as
an ‘online PE’, even though the server was situated outside Spain. Whilst the server
was not located in Spain, and there were no employees of the Irish company in
Spain, the Spanish subsidiary company did employ people to translate the
website, review the contents and administer the site, which was considered a
factor. In addition, the Court noted the observations on the commentary made by
Spain in the 2003 and 2005 versions, which stated that Spain had a number of
reservations on the OECD approach.
However, it is notable that on appeal, the Spanish National Appellate Court

(decision of June 2015) and the Supreme Court (in June 2016) decided the matter
with regard to the fixed place of business PE and the agency PE concept (Dell
Spain acted as commissionaire and Dell Ireland was bound by the contracts under
Spanish law), without referring to the creation of a ‘virtual PE’. It is interesting that
the Appellate Court did comment that according to the OECD commentary to
the Model Treaty, an online website does not in itself have a location that can
constitute a PE, although the place where the server is located could constitute
one. The Supreme Court confirmed the point as to the commissionaire
arrangement, which is the issue of most importance in this case. In so concluding
the Spanish Court differs from courts in France and Norway which have concluded
that commissionaire arrangements do not give rise to PEs. However, such
conclusions have been overtaken by the changes to the Agency PE definition in
the treaty as discussed above.
19.10 The Multilateral Instrument (MLI)
The changes to the OECD Model Treaty as a result of Action Point 7 are to be
implemented for many countries under the multilateral instrument (MLI) (pursuant
to Action Point 15 of the BEPS Project). This puts into place the various changes
arising from the BEPS Project on a multilateral basis, without the need for countries
to renegotiate all their existing double tax treaties.
Supplementary guidance on how to determine the profits to attribute to a PE was

the subject of consultation at the end of 2016, see a later chapter in this regard.
The provisions of Action Point 7, on PEs, are not considered a minimum standard.
Therefore, under the MLI contracting states will have the flexibility to opt out of the
provisions entirely (or in part) and therefore out of the BEPS changes. This is
accomplished through the mechanism of reservations, which are specifically
defined for each article of the MLI. Part IV of the MLI contains the provisions with
regard to Action Point 7 and the artificial avoidance of PE status, in its Articles 12 to

15. Where changes are to be implemented to existing treaties, the parties need to
notify the depositary (the OECD acts in this capacity) of their tax agreements
which contain a relevant provision which will be subject to change.
The UK, for example, (when it signed the MLI in June 2017) has opted into the
provisions with regard to anti-fragmentation (with regard to preparatory and
auxiliary activities), however, it has not opted into the other provisions with regard
to changes to the definitions of a PE. There are reports and databases available
from the various accountancy firms which indicate which countries have opted
into/out of which BEPS provisions. In addition the OECD maintains a matrix on its
website of which provisions countries have opted into.
19.11 Relevant Case Law
In addition to the case law analysed above, the issue of whether a PE exists or not
has, over the years, given rise to many disputes, which have been litigated. Some
notable cases are highlighted below. Although these cases pre-date the 2017
update they are still relevant.
Roche Vitamins [2012] STS 201/2012
In a further Spanish case, the subsidiary of a non-Spanish company was

considered to be the overseas company’s PE.
In brief, the facts were as follows:
• Up to 1999 the Spanish subsidiary Roche Vitamins SA performed the functions

of manufacturing, import and distribution of goods (as a fully-fledged
distributor).
• In 1999 the group undertook a restructuring and the Spanish activity was
changed.
Roche Vitamins SA concluded two contracts with its related party Roche Vitamins
Europe Ltd, a Swiss company:
i. A manufacturing contract, whereby Roche Vitamins SA committed to

produce and package, in their facilities, the products ordered by Roche
Vitamins Europe. Such products would be sold at cost-plus with a mark-up of
3.3%.
ii. An ‘agency contract’, where Roche Vitamins SA was designated by Roche

Vitamins Europe as its ‘Spanish agent’ to promote the sale of particular
products. The Spanish subsidiary committed to “represent, protect and
promote” the other party’s interests, receiving in consideration an amount
equal to 2% of the total sales promoted in Spain.
Therefore, after the restructuring operation, the Spanish subsidiary became, at

least from a contractual point of view, a contract manufacturer and sales
commission agent, whose profits are considerably lower than those received by a
fully-fledged distributor.
The Supreme Court confirmed, on appeal, the existence of a Spanish PE of Roche

Vitamins Europe, as a result of the activities carried on by the Spanish subsidiary. In
particular, the Court considered that the Spanish subsidiary operated as a
dependent agent of the Swiss entity, as it carried on, under the two contracts, the
activity which could have been done directly through a fixed place of business
(being the sale and distribution of the goods produced).

The fact that Roche Vitamins SA had no capacity to contract or negotiate on

behalf of Roche Vitamins Europe did not exclude the application of the
dependent agency clause of the double tax treaty, as interpreted under the
commentaries to the OECD Model Treaty (1997 version). In particular, under the
‘Agency Contract’ the Spanish subsidiary was obliged to promote the goods that
were sold by Roche Vitamins Europe, which function was considered by the Court
as a greater involvement in the Spanish market. It was a key element leading to
the conclusion that Roche Vitamins SA was not merely processing purchase orders
issued by the Swiss company.
The Supreme Court also considered the general PE clause found at Article 5(1) of
the treaty and held that the Spanish Subsidiary (Roche Vitamins SA) constituted a
‘fixed place of business’ of Roche Vitamins Europe. The Supreme Court upheld the
position of the tax authorities in that the Subsidiary was held to constitute a fixed
place of business of the foreign related company on the basis that all the activity
of the Spanish company was directed, organised and managed by the Swiss
Company (Roche Vitamins Europe). Some commentators consider the decision in
this case to be controversial.
Director of Income Tax v e-Funds IT Solutions [2017] SLP No 27494 of 2017
In this case before the Indian High Court, the result of which was confirmed by the
Supreme Court (2017), it was held that an (indirect) Indian subsidiary of a US
company was not a PE, notwithstanding the close association between the US
company and the Indian company, the division of functions, the assets used and
the risks assumed. The High Court held that a consideration of these factors was
not the appropriate test to determine whether a PE was present. The Court
considered the PE definition in the US-Indian double tax treaty and focused on the
conditions necessary to fulfil these PE tests. As the Indian Revenue was not able to
prove the ‘right to use’ test and having space at ‘the disposal’ of the company
test were satisfied, no fixed PE was found to exist. Similarly in the absence of any
evidence that employees of the Indian subsidiary were under the control or
supervision of the US parent and providing services on behalf of the US company,
or that the Indian employees were participating in negotiations with customers, no
service PE or agency PE were found to exist either.
The Supreme Court also noted that even though there was a mutual agreement
between the states that referred to PEs in India, this did not establish a binding
decision on the matter.
As noted above, care needs to be taken with regard to employees, in particular

with regard to a subsidiary’s employees that might be taken (on the facts) to be
economic employees of the parent company or group. If they are, this may result
in the parent group having a PE in the foreign jurisdiction, perhaps through an
agency PE (if they fulfil the criteria for entering into contracts on behalf of the
parent entity) or as a services PE (where the treaty in question has included the
services PE wording) or perhaps through a physical PE if the premises where the
employees work falls within the requirements needed to create such a PE for the
parent group.
GE Energy Parts Inc v ADIT [2017] ITA No. 671/Del/2011
It is notable that the availability of information such as that found on social media
sites may have an impact. In 2014 the ITAT (Indian tax tribunal) held that the
LinkedIn profiles of the employees of the GE group submitted by the tax authorities
were admissible as evidence in determining the existence of a PE in India. The
tribunal’s view was that the LinkedIn profiles were not hearsay, but were akin to
admissions made by a person; and they had considerable bearing on the subject

matter of this appeal. However, the taxpayer is free to rebut the information
contained in their LinkedIn profile by bringing on record contrary facts to dislodge
the claims made. The tribunal merely commented on the evidential issue and did
not comment on the existence of a PE at that stage. In later appeals in this matter
the tribunal determined that there was an agency PE of GE in India, however they
did not agree with the profits attributed to it by the tax authorities (27 January 2017
— Appellate Tribunal).
It is therefore possible that the tax authorities in India may use social media
platforms to collect factual evidence for use in assessment/appellate
proceedings, not only for pending assessments, but potentially also to reopen
completed assessments. The case is very similar to Reuters’ special report on
Google’s taxable presence in the UK, published in 2013, with research that
included ‘CVs and endorsements on networking website LinkedIn’. The case shows
the importance for firms of ensuring that anything in the public domain, including
all the various social networking websites such as Facebook, LinkedIn and Twitter, is
properly reviewed and accurate.
AB LLC and BD Holdings LLC v SARS [2015] ZATC 2
The South African Tax Court here had to determine whether a US company,
providing strategic and financial advisory services for the South African airline
industry had created a PE in South Africa. The Court reviewed Article 5 of the US-
South Africa double tax treaty, (which differs from the OECD Model Treaty) in
particular the second paragraph of the article (paragraph (5)(2)(k) in this case),
which ‘includes especially’ as a PE ‘the furnishing of services including consultancy
services’.
The Court held that a PE had been created, on two counts. Firstly, on the basis the
US company employees had been in South Africa for over 183 days. The Court
considered that the wording of the second paragraph of Article 5 in the treaty,
which is preceded by the words ‘includes especially’, creates a standalone
definition of a PE by way of an extension to the first paragraph of the article.
Secondly, it found there was in any event a fixed place of business created by the
use of the boardroom (it had exclusive use during relevant working hours). The US
company was liable for tax on the profits attributable to the services rendered. In
addition the Court held that the penalties imposed by the tax authorities (100% of
the tax) were not disproportionately punitive and that the US company had been
negligent in not considering the tax consequences of the agreement it had
entered into. The company had argued that it did not deliberately ignore South
African law, but just misunderstood it. Interestingly the Court commented that the
enterprise had a global reach and that it was ‘not a novice in the area of tax
liability’, and therefore should accept responsibility for its own error.
Al Hayat Publishing Co Ltd [2014] No 360890
This French case relates to the collection of information and whether it is core to
the business, or merely preparatory and auxiliary. The French Conseil d’Etat
considered the application of Article 5(4) in the France-UK treaty. Al Hayat UK
publishes the Arabic language daily, Dar Al Hayat. Its Paris bureau researches
French current affairs and writes articles for publication in the newspaper. The
Court held that the bureau’s activities were not core to Al Hayat’s business and so
did not constitute a taxable PE in France. The core activity of the business was the
editing and commercialisation of the business, rather than the writing of articles.
The bureau’s activities were confined to the collection of news and its transmission
to the UK head office. The decision is in line with paragraph 22 of the OECD
commentary on Article 5.

Formula One World Championship Ltd v Commissioner of Income Tax, International

Taxation 3, Delhi & Anr [2017] Appeal 3850 and 3851
The case related to the UK company holding the commercial rights to stage, host
and promote the Indian Grand Prix in 2011 through to 2013 at a location owned by
an Indian company. The existence of a PE was upheld as the Court found that the
racing circuit constituted a fixed place from which a business/economic activity
was conducted. In addition the various agreements between the Indian company
and the UK company (together with its affiliates) could not be looked at in
isolation. In the Court’s view, the facts pointed to the conclusion that the UK
company had made its earnings in India through the racing circuit over which it
had complete control during the event. The duration of the event and the number
of days for which the UK company’s personnel had access to the circuit would not
make any difference.
The Supreme Court ([2017] 80 taxmann.com 347 (SC)) upheld the judgment of the
High Court, and notably referred to the OECD commentary and also to Klaus
Vogel on Double Tax Conventions when discussing the permanence requirement
for a PE and the duration of the activities in the case.
Aska Gmbh [2017] SKM2017.213.SR
In this case, a Scandinavian sales manager was required under the terms of his
contract to work from home in Denmark. Although he spent the majority of his
employment duties travelling and visiting clients, the Danish tax board ruled that
the manager’s use of a home office for administrative work (for which he was not
reimbursed) constituted a PE of his German employer. It was held to be irrelevant
whether the home office was owned or rented or in any other way made
available to the foreign corporation. As long as the business of the non-resident
entity was carried out effectively and habitually (and these activities were not
preparatory), then a PE was in existence. It was also emphasised that there must
be recurring work from the home office not just sporadic or occasional. The Danish
tax board determined that the administrative work was directly related to the
main business of the company and could not be said to be non-core or
preparatory.
Wolf v R [2018] TCC 84, 21 ITLR 506
This Canadian case illustrates the interpretation and application of a ‘services PE’
provision similar to that found in the OECD commentary to Article 5, at paragraph
144. In order for that provision to operate, an individual must be present in the
state for 183 days in a 12 month period, and more than 50% of the ‘gross revenues
attributable to active business activities of the enterprise during this period’ (in the
OECD drafting) must be derived from services performed in the other states
through that individual. This case raises issues concerning the meaning of
‘enterprise’, ‘business’, and ‘gross revenues attributable to active business
activities’ which are relevant to the application of this provision.
As noted by the Editor of the ITLR, the background facts are not particularly
complex. Mr Wolf was an aeronautical engineer resident in the US, who provided
services to Bombardier Inc in Canada. In 1994 he had invented a fuel line system
for aircraft which had been patented, and the patent had been transferred to
him. He had then entered into a Manufacturing and Licence Agreement with a US
company to exploit that patent, and he was entitled to receive from that
company profits from the sale of fuel lines and profits from the sub-licensing of his
patent. He had assigned the right to receive those payments to a US LLC,
Wolfbend, in which his brother and other members of the company participated.

During 2012, Mr Wolf supplied services to Bombardier Inc as an independent

contractor hired by an agency, spending 188 days in Canada in a 12 month
period, and earning from those consultancy services a little over Can$26,000. The
only issue was the taxation of those earnings under the Canada-US double tax
treaty, considering Article 5(9) deemed services to be provided through a PE if the
183 days of presence and the 50% of gross active business revenue tests were both
met. He was over on the days and Mr Wolf also failed to prove that more than 50%
of the gross active business revenues were not derived from services performed in
Canada.
As noted in the judgment, whilst an enterprise whose commercial activities were

licensing or sub-licensing was different from an enterprise that provided
engineering services, in this instance all three commercial activities were directly
related to the design of fuel lines for aircraft. The degree of interlacing and
interdependence between the three sources of income was such that both the
Canadian and US commercial activities were part of the same enterprise, which
was thus a single business. Mr Wolf’s revenues from all three sources therefore
qualified as gross revenues, and all three sources qualified as ‘active business’
revenues. On the evidence, Mr Wolf’s enterprise was sufficiently active in earning
the licensing and sub-licensing income. With respect to the licensing activities, Mr
Wolf’s responsibilities included designing, in accordance with the patent, the fuel
lines ordered and their customization for the client’s particular aircraft. Revenues
to be included in the calculation of ‘gross active business revenues’ were only
those generated during the 188-day period that the taxpayer had been in
Canada. The onus had been on the taxpayer to prove that the revenues earned
in Canada had not represented more than 50% of the gross active business
revenues of the enterprise. He had failed to prove this and so the Can$26,000
earned in Canada was deemed to be attributable to a PE under Article 5(9) and
was taxable in Canada. The judgment was confirmed on appeal.
Centrica India Offshore Ovt Ltd v CIT [2014] WP(C) No. 6807/2012
In this April 2014 case heard at the Delhi High Court, employees seconded to India
from group companies in the UK and Canada were deemed to constitute PEs of
those companies in India.
A wholly owned subsidiary of the UK company, Centrica PLC, resident in India,

provided back office services to overseas group companies, charging cost plus
15% for those services. To help support the establishment of the Indian back office
services in the first year of their operation, some overseas employees with
managerial and technical expertise were seconded to India. The Court decided
that under the terms of the India-UK and India-Canada tax treaties, these
secondees constituted a PE in India that was providing technical services.
The decision arose because of the way the secondments themselves were
structured. The Court distinguished between the concepts of economic employer
and legal employer. It also referred to the OECD commentary which says that the
foreign company will not constitute a service PE if the seconded employees work
exclusively for the Indian enterprise and are released for the period in question by
the foreign enterprise – that was not the case here. The Court further observed
that there was no purported employment relationship between the Indian
company and the secondees and, notably, that the company had no right to
terminate the employment contract and the employees had no right to sue the
company for non-payment of salary.
This case highlights the need to ensure that secondment arrangements are very
carefully managed to avoid the risk of PE exposure when employees are
seconded to India, but could also have wider implications.

It is important that you are familiar with the commentary to Article 5 as this is an
important area of the syllabus.
19.12 Discussion Drafts and the BEPS Project
Prior to BEPS, the OECD had launched a consultation (in November 2011) with
regard to updating the commentary on Article 5. In October 2012 the OECD issued
a revised discussion draft based on the results of the consultation, with a working
party meeting in February 2013 to further consider the latest draft. The OECD
Model Treaty and commentary were updated in 2014 but the proposed changes
to the PE article were not included.
As noted above, changes to the PE article per the final BEPS Report have been
implemented through the MLI and in the 2017 update to the OECD Model Treaty –
and thus some of the proposals outlined below (per the 2012 discussion draft) have
been covered. However, the 2012 proposals discussed some 25 issues, and so
certain issues have not been dealt with.
Some of the key areas highlighted in the 2012 discussion draft proposed
amendments to the commentary in the following areas: the meaning of the
phrase ‘at disposal of’ with regard to a physical presence; amendments to clarify
the meaning/use of a home office, with a focus on the employer’s ‘right to use’
the premises; amendments to introduce exceptions to the general practice of
taking 6 months as a signifier of permanency, in particular with regard to activities
of a ‘recurrent nature’; further clarification of the position of secondees in the host
country; the creation of a deemed PE where a main contractor uses a
subcontractor in the host state (time spent by such subcontractor counts towards
the 12 months); non-inclusion of time spent on ‘snagging’ and other repairs in the
guarantee period; and, clarification of the meaning of ‘to conclude contracts in
name of the enterprise’ with regard to agency and commissionaire arrangements.
The update to the 2017 OECD Model Treaty has been discussed above, and
clearly addresses some of these issues where a state opts in. However, only
amendments to Article 5(4), (5) and (6) have been made. Whilst this does deal
with the problems surrounding commissionaire arrangements, the preparatory or
auxiliary exemption, and some construction site issues, many of the other matters
raised in 2012 remain. It is unclear whether we will see any additional changes so
soon after the 2017 update, however, further amendments at some point in the
future are possible.
19.13 Double Taxation Relief
Double taxation relief is a connected issue that arises from a company having a
PE. In most of the examples the companies would incur double taxation in country
1 and 2. Therefore the OECD Model Treaty under Article 23 grants double taxation
relief ie either the amount of tax paid in country where the PE is situated to be
offset against the tax payable in the country of residence of the legal entity or the
income is exempted in country of residence of the legal entity. Of course, if the
company is located in a tax haven then there is the potential of double taxation
as this relief is rarely available. We will look at double taxation in more detail later in
the manual.

CHAPTER 20
ATTRIBUTION OF PROFITS TO PEs
In this chapter we will look at the attribution of profits to a PE in particular:

– rejection of force of attraction principle
– business profits (Article 7 OECD Model Treaty)
– 2010 and 2017 updates to the Model Treaty
– practical application of the transfer pricing process
– attribution of profit in excess of the total profit of the enterprise
– comparison of the Article 7 OECD approach to Article 9
20.1 Introduction
Having established the existence of a PE the second and probably more difficult
issue is how to attribute profits to the PE.
Again the starting point of the analysis is OECD Model DTC. The OECD Model DTC
determines in several of the articles the countries rights to tax income dependent
on residence or source. The two articles that are most directly relevant to transfer
pricing are Articles 9 and 7.
In these articles the OECD Model DTC distinguishes between the attribution of
profits to a PE and transfer pricing between separate entities by including different
articles for each of these situations:
• Article 7 – attribution of business profits between the parts of a single entity

using the separate entity principle;
• Article 9 – transfer pricing between two separate associated enterprises using

the arm’s length principle.
We will concentrate on Article 7 as it is relevant to attribution of profits to PEs

although of course reference is made to Article 9 Associated Enterprises. We have
looked at Article 9 in an earlier chapter – you may find it useful to review Article 9
at this point.
20.2 Rejection of Force of Attraction Principle
The first overriding principle of double taxation treaties is that a company resident
in one country will not be taxed on its business income in the other State unless it
carries on that business in the other country through a PE situated in that country.
The second principle is that the taxation right of the State where the PE is situated
does not extend to income that is not attributable to the PE.
The interpretation of these principles has differed from country to country. Some
countries have pursued a principle of general force of attraction, which means
that all income such as other business profits, dividends, interest and royalties
arising from sources in their territory was fully taxable in that country if the
beneficiary had a PE there, even though such income was clearly not attributable
to that PE. The approach has been rejected by the OECD.

20.3 Business Profits (Article 7 OECD Model Treaty)
The business profits article in a tax treaty sets out the limits of the source state’s
taxing rights in relation to business profits. This was rewritten and updated in the
2010 version. The 2014 and 2017 updates did not amend Article 7. However, the
2017 update did make some minor changes to the commentary. The comments
below refer to this latest version.
If an enterprise resident in a contracting state has a PE in the other state through

which it carries on business it can be taxable in that other state, as well as the state
of residence (subject to double taxation relief). The tax is limited in the source state
to no more than the profits attributable to the PE. If, however, the enterprise does
not have a PE in the other state it can only be taxed in the state of residence
(Article 7(1) OECD Model Treaty).
With reference to establishing the profits of a PE, where there is one, and for the
purposes of the double tax relief per Articles 23A and 23B, the assumption to be
made is that the PE is a separate and independent enterprise from the main
enterprise of which it is a part. The main thrust of the approach is that the arm’s
length principle should apply.
The PE is to be treated as carrying on the same or similar activities under the same
conditions under which the main enterprise operates (Article 7(2)). The transfer
pricing aspects of the ‘separate and independent enterprise’ concept has been
under substantial review within the OECD. Guidance was issued in 2010 in the
Report on the Attribution of Profits to Permanent Establishments which provided
details of how to apply transfer pricing concepts to PEs. However, an updated
Report which clarifies some aspects of this approach has been issued, post BEPS,
as outlined below.
The separate and independent enterprise fiction found at paragraph 2 is restricted

to the determination of profits that are attributable to a PE. With reference to
paragraph 28 of the commentary, such fiction does not extend to create notional
income for the enterprise taxable by a contracting state under its domestic law
under paragraph 4 Article 7 (this paragraph allows taxation of income
notwithstanding paragraph 1 Article 7, however as the commentary makes it clear
you cannot ‘create’ income under the arm’s length principle which is not there!).
Paragraph 2 does not specifically deal with the issue of whether expenses are
deductible when computing the taxable income of the enterprise in either state.
Paragraph 30 of the commentary states the conditions for deductibility are a
matter to be determined by domestic law, subject to the provisions of the
convention (in particular paragraph 3 of Article 24, on non-discrimination). Thus PEs
must be accorded the same right as resident enterprises to deduct the trading
expenses that are, in general, authorised by that state’s tax law to be deducted –
without any restrictions other than those also imposed on resident enterprises.
Generally the arm’s length concept present in paragraph 2 is thought sufficient for
the determination as to whether any particular expense should be deductible. See
paragraphs 38 to 40 of the commentary.
Paragraph 3 of Article 7 deals with reciprocal adjustments to the profits calculated

in respect of the PE. The combination of Article 7 and Articles 23A and 23B should
ensure there is no unrelieved double taxation of profits that are properly
attributable to the PE. However, this may require that the two states resolve
differences based on different interpretations of paragraph 2 – it is key that there
are mechanisms available to allow such resolution (to the extent necessary to

eliminate double taxation). Paragraph 3 allows for these corresponding

adjustments to avoid double taxation, on consultation of the two states involved.
Generally, interest, royalty and similar payments made by the PE to its head office
are not allowable on the basis this relates to ‘internal’ borrowings. However
deductions are allowed in certain circumstances, and where third party
borrowings are involved. See below for a discussion of the 2010 Report on the
Attribution of Profits to Permanent Establishments, which considers interest
expense. As highlighted the provisions of Article 7 do not go into detail with regard
to any applicable deductions. The premise of paragraph 2 Article 7 is that the
arm’s length principle should apply and therefore the OECD 2010 Report on the
Attribution of Profits to Permanent Establishments should be used, along with the
principles set out in the OECD Transfer Pricing Guidelines.
In practice, for example in the US, certain excess interest payments to third parties
made by a PE may be disallowed on the basis that the underlying capital of the PE
cannot support the excess payments (ie this is a form of ‘thin capitalisation’). This
calculation looks at the assets of the branch compared to the worldwide assets of
the enterprise to test what level of proportional liabilities should be attributed to
the PE.
Finally, importantly, Article 7(4) of the OECD Model Treaty cedes priority to another
specific income article of a treaty, if an item of income (forming a part of the
profits of an enterprise) is dealt with under that article. This is the same as the
previous Article 7(7).
20.4 2010 and 2017 Updates to the Model Convention
In 2010 an update was made to the Treaty, to the Report on the Attribution of
Profits to Permanent Establishments (2010 PE Profits Report), and to the OECD
Transfer Pricing Guidelines – the intention being that these could now be read
consistently with each other. The updated 2010 PE Profits Report takes into
account the 2010 version of Article 7 of the Treaty and the commentary thereto, so
as to provide maximum certainty as to how to attribute profits per the Treaty.
However, it should be noted that an update to the 2010 PE Profits Report was
issued in March 2018, to take account of the changes to Article 5 (definition of a
PE) in the 2017 OECD Model Treaty (post BEPS).
Considerable changes were made in 2010 to Article 7 and the commentary. The
paragraphs of this Article 7 are discussed above, however, we have outlined
below further comments with regard to the methodology the OECD now
envisages should apply when attributing profits to a PE. It is notable that
paragraph 9 of the commentary to Article 7 states that this version of Article 7
reflects the approach developed in the 2010 PE Profits Report and must be
interpreted in light of the guidance contained in it – so Article 7 and the 2010 PE
Profits Report need to be read in conjunction with each other.
The OECD’s principal aims are to:
i. develop a methodology of attributing profits to a PE so as to minimise the

incidence of double taxation and
ii. achieve a degree of harmonisation on the taxation of PEs worldwide.
The key concept underpinning the 2010 PE Profits Report is that a PE should be
considered as if it is a separate (and independent) legal entity from the enterprise
of which it is a PE, which is reflected in paragraph 2 of Article 7. The 2018 additional

guidance also reinforces this as a high level basic approach, regardless of whether
the state uses the approaches set out in the 2010 PE Profits Report.
The effect of this analysis is that the PE will be attributed with the profits which it
would have earned if it had been a separate legal enterprise carrying on a
separate business on the assumption that all transactions between the PE and the
rest of the enterprise to which it ‘belongs’ were conducted on arm’s length terms.
This approach under Article 7, combined with the application of the 2010 PE Profits
Report, and the application of the updated Transfer Pricing Guidelines (see
below), to transactions between the PE and the entity of which it is a part, is called
the Authorised OECD Approach (AOA).
Analysis of the PE as a separate entity means that the economic ownership of

assets, the need for capital and the assumption of risk will be apportioned
between the PE and the rest of the enterprise.
Paragraph 20 of the treaty commentary contains a two step approach to the

attribution of profits. Firstly the activities carried on through the PE need to be
identified. It is stated that this needs to be done through a functional and factual
analysis and that the guidance found in the Transfer Pricing Guidelines will be
relevant for that purpose.
Per paragraph 21 the steps undertaken will lead to:
• the attribution to the PE, as appropriate, of the rights and obligations arising
out of transactions between the enterprise of which the PE is a part and
separate enterprises;
• the identification of significant people functions relevant to the attribution of

economic ownership of assets, and the attribution of economic ownership of
assets to the PE;
• the identification of significant people functions relevant to the assumption of

risks, and the attribution of risks to the PE;
• the identification of other functions of the PE;
• the recognition and determination of the nature of those dealings between

the PE and other parts of the same enterprise that can appropriately be
recognised; and
• the attribution of capital based on the assets and risks attributed to the PE.
The reference to ‘significant people functions’ above can be found in the 2010 PE
Profits Report. Part I paragraph 15 explains that such functions are significant
functions relevant to the ownership of assets or to the assumption and
management of risks of the PE which are performed by people in the PE. The
actual functions which are relevant will vary from business sector to business
sector.
Then under the second step, such dealings will be priced as determined again by
applying the principles developed for the application of the arm’s length principle
between associated enterprises (per the Transfer Pricing Guidelines) by reference
to the functions performed, assets used and risk assumed by the enterprise through
the PE and through the rest of the enterprise – see paragraph 22 of the
commentary.

With reference to the 2010 PE Profits Report, in the case of dependent agent PEs
(DAPEs), the analysis described above will be applied to determine the functions
undertaken by the dependent agent both on its own account and on behalf of
the non-resident enterprise. On the one hand the dependent agent will be
rewarded for the service it provides to the non-resident enterprise (taking into
account its assets and its risks (if any)). On the other hand, the dependent agent
PE will be attributed the assets and risks of the non-resident enterprise ‘relating to
the functions performed by the dependent agent enterprise on behalf of the non-
resident, together with sufficient capital to support those assets and risks’. The
dependent agent PE is attributed with the profits on the basis of those assets, risks
and capital.
The OECD also considered how internally developed intangible assets should be
treated. The OECD determined that the PE should be attributed with any profits
arising in connection with the internally developed intangible asset to the extent
that the decisions/activity in connection with that asset were undertaken by staff
who took on the active management of the asset development programme. The
2010 PE Profits Report cross refers to Chapter VI and VII of the Transfer Pricing
Guidelines (which have been updated post BEPS – see the 2017 version of the
Guidelines).
Consideration would also be given to where the economic ownership of the asset
lies. On this point the OECD considered that an intention to use the asset is not as
important as the nature of the role the intended user performed in the
development of the asset: that is, who made the decisions in relation to the asset?
If these people ‘belong’ to the PE, it will likely follow that the PE will have economic
ownership of the asset and will then be attributed with any profits arising in relation
to that asset.
As noted above in respect of Article 7(2), the commentary makes it clear that
when looking at deductible expenses, domestic law will be in point subject of
course to Article 24 on non-discrimination.
The question of deductions for interest is looked at in more detail in the 2010 PE
Profits Report. The OECD has retained its stance that in most scenarios there should
be no deduction for interest on ‘internal’ loans from head office. However, where
there are ‘treasury dealings’ between the two then an interest deduction may be
considered. In addition, the position where third party lending is involved is more
complex, and where external debt is on-lent to the PE then a ‘notional’ interest
deduction can be allocated to the PE. This will be based on the amount of ‘free’
capital which shall be allocated to the PE, and thus the amount of debt capital
(see paragraphs 150 to 171 of the 2010 PE Profits Report).
When looking at a possible interest charge to be allowed the 2010 PE Profits Report
talks about the idea of a PE being allocated a certain amount of “free capital”
which will be used to determine its borrowing capacity (paragraph 115 et seq, Part
I). It states that the free capital will be the amount required to support its functions,
assets and risks.
To deal with the different measures of free capital that may apply between
different states the 2010 PE Profits Report comments (paragraph 162 Part I) that the
rules of the state where the PE is located should prevail provided certain
conditions are met and subject to Article 7.
The 2017 case Re BNP Paribas (21 ITLR 142) illustrates one of the problems arising
from the AOA (under Article 7 requirements) where the PE is treated as a separate
enterprise but is not treated as a resident of the state where it is situated. As
discussed by the Editor of the ITLR, in this instance the BNP Paribas head office in

France made ‘loans’ to its branches in China, India, the Philippines and Thailand.
‘Interest’ was paid on those loans (‘loans’ and ‘interest’ in parentheses because a
head office cannot make loans to its branches, which are part of the same legal
entity; nor can it receive interest from part of itself). The four countries concerned
all imposed flat-rate withholding taxes on that ‘interest’. Arguably, they should not
have done so, but clearly those countries saw this as an interest payment having a
source in its territory. The bank sought to claim a credit for the tax withheld against
its taxation in France. The French Revenue discovered this on audit and refused
the claim based on the relevant tax treaties.
The difficulty, of course, was that the PEs were not residents of the source states.
That was a requirement to claim the credit. This situation is not dealt with by the
provisions in Article 23 of a treaty, and there was no relief.
Whilst it may be that it is only banks and insurance companies that are likely to
operate through branches, it is notable that this type of situation is not covered,
and perhaps there is an argument that it should be. This serves as an illustration
that the AOA does have certain issues with regard to its application in practice.
Another case which also looks at the loan position between a bank and its
overseas branch is Irish Bank Resolution Corporation and Irish Nationwide v HMRC
[2019] UKUT 277. In this case the Upper Tribunal found that HMRC had been correct
to deny deductions for interest paid by PEs to the companies they were part of.
The 1976 Ireland-UK treaty in point here contains the earlier version of the business
income article (which is Article 8 in this treaty) and the distinct and separate
enterprise concept we are familiar with is present (in Article 8(2)).
Both Irish Bank and Irish Nationwide (which were tax resident in Ireland) traded in
the UK through PEs, so that the profits attributable to these establishments were
chargeable to UK corporation tax. Both appellants’ claims for deductions in
relation to interest paid by the PEs to the Irish companies had been denied. HMRC
had attributed to the PEs a notional amount of equity and debt capital (including
an amount of free capital) based on the legislation, which was calculated as if the
PEs were separate and independent enterprises, and denied the interest
deductions on this basis. HMRC considered that the returns submitted by the Irish
companies for the PEs had understated the amount of equity capital each PE was
deemed to hold, and so overstated the amount of loan capital and the
associated interest charges.
Irish Bank and Irish Nationwide contended that the treatment of their tax returns so
as to disallow the deduction for interest paid was precluded by the double tax
treaty. They argued that the adjustment implemented by HMRC under the
legislation was contrary to the treaty, which obliged HMRC to compute a PE’s
profit by reference to that establishment’s books of accounts only. Their argument
was based on the fact the treaty was based on the 1963 version of the OECD
Model, and that the approach using ‘notional free capital’ was only introduced in
the 2010 treaty update.
The Upper Tribunal observed that the starting point was indeed the actual records,
including (for example) the capital actually attributed to the PEs. However, under
Article 8(2) of the Convention, where there was a mismatch between the PE’s
books of account and the position as it would have been, if the PE had been ‘a
distinct and separate enterprise engaged in the same or similar activities and
dealing at arm’s length with the enterprise of which it is a permanent
establishment’, then the latter prevails and should be applied. The UK legislation,
which did apply the separate enterprise principle, was correct in that it allowed
such possible distortions to be remedied, by ensuring that the PE’s books of
account were adjusted to reflect the distinct and separate enterprise approach,

by way of the capital attribution adjustment undertaken by HMRC. Therefore the

denial of the interest deductions was entirely compliant with the approach taken
in the treaty. This decision was upheld by the Court of Appeal in August 2020.
The 2014 update to the Model Treaty made only minor amendments to the
commentary to Article 7, replacing the word ‘income’ with ‘profits’ in certain
paragraphs, and including a new paragraph on trading emission permits and
credits. Similarly the 2017 update to the Model Treaty only made minor changes to
the commentary. Some consequential changes were made to deal with the
amendments to Article 1 and Article 5. A new paragraph 59.1 was added to
confirm the operation of the corresponding adjustment mechanism at Article 7(3).
In addition an updated paragraph 62 on the same issue allows states to add an
additional paragraph to the Article which sets a time limit for possible changes to
the PE profits under Article 7(2) which would need to be adjusted for by the other
state.
March 2018 Additional Guidance on Attribution of Profits to a PE
Additional guidance was needed to cover the changes made to Article 5 in the
2017 update, as part of the BEPS Project. After consultation a final version of the
Additional Guidance was issued in March 2018. The focus is the dependent agent
concept (and how profits are attributed to commissionaires and other
intermediaries) and also the anti-fragmentation aspect of the changes to the
preparatory and auxiliary exclusion. A key element of the guidance is that the
separate and independent enterprise concept is maintained as an appropriate
high level principle to be applied by all states, whether or not the relevant treaty
uses the AOA under Article 7. This guidance is in your KVR book.
The report does note the interrelationship between Article 9 and Article 7, and
comments that the order these articles are applied should not impact the profits
taxable in the host state under Article 7, so long as states are consistent in their
approach and double taxation is not allowed to arise in the host state. The report
also notes that the ‘risk control functions’ seen under Article 9 (and the 2017
updated Transfer Pricing Guidelines) are not exactly aligned with the ‘significant
people functions’ concept seen under Article 7, and so should not be used
interchangeably. Risks that have been attributed under Article 9 cannot also be
allocated again under Article 7 (as there would be a possibility of double
taxation).
The report does clearly indicate that ultimately, even considering the arm’s length
payment to the intermediary, the profits of the PE might still be positive, nil or even
negative (so a loss). This has widely been accepted as a welcome comment.
The most significant part of the report is set aside for examples, which should assist
with analysis of the 2017 changes made to Article 5.
The first example examines the anti-fragmentation rule at Article 5(4.1). It looks at
an internet based sales business which operates in the host state by way of a
warehouse and also a geographically separate merchandising and information
gathering office in the same host state (with 15 staff). The analysis comments at
paragraph 14 that provided the business activities carried on at the warehouse
and at the office constitute complementary functions that are part of a cohesive
business operation, the warehouse and the office constitute two PEs in host state
under Article 5(1) of the treaty, as each of these locations is a fixed place of
business through which the business of the overseas entity is partly carried on, and
the overall activity resulting from the combination of the activities carried on in the
host state is not of a preparatory or auxiliary character.

With regard to the attribution of profits to each of these PEs, the analysis proceeds
by applying the AOA.
For the warehouse: Under step one, the functional and factual analysis shows that
the head office staff deal with the purchasing of goods from unrelated suppliers
and the sale of goods to third party customers. However, the analysis also shows
that the warehouse personnel are responsible for leasing the warehouse and hiring
the independent service providers for delivery in the host state. Consequently, the
warehouse PE is hypothesised to have the overseas company’s rights and
obligations arising from these transactions. In addition the significant people
functions relevant to the management of the warehouse are performed by the
personnel in the host state. Accordingly, the PE is hypothesised to be the
economic owner of the warehouse. With regard to the recognition of an internal
dealing between the PE and the head office, this is hypothesised as the provision
of storage and delivery services by the warehouse PE to the head office. Under
step two, the Transfer Pricing Guidelines are applied by analogy to determine the
arm’s length pricing of the internal dealing between the PE and the head office. In
this case, that would equal the amount the overseas company would have had to
pay if it had obtained the storage and delivery services from an independent
warehouse provider in the host state.
For the merchandising and information gathering office: Under step one, the
functional and factual analysis shows that the office personnel are responsible for
the merchandising and collection of information activities in the host state. Thus
this PE is attributed the rights and obligations arising from any dealings with
unrelated parties in the process of providing these services. In addition the
significant people functions relevant to the management of the office are also
performed by the personnel there, so the PE is hypothesised to be the economic
owner of the office. The internal dealing between the PE and the head office is
therefore hypothesised as the provision of merchandising and collection of
information services. Under step two of the AOA, the Transfer Pricing Guidelines are
applied by analogy to determine the arm’s length remuneration of the internal
dealings. In this case, the amount the overseas company would have had to pay
if it had obtained the same merchandising and collection of information services
from an independent enterprise in the host state (on the basis of the same
functions, assets and risks profile).
The other examples in the guidance deal with marketing and sales intermediaries,
and also a procurement agent intermediary that creates a DAPE. Some
commentators have indicated that whilst the guidance does initially indicate that
the AOA does not have to be followed by all states in their treaties, it is notable
that all the examples provided follow the strict two step approach of the AOA.
Thus it does not really deal with the problem of how the AOA might co-exist
alongside alternative approaches.
Notably the report does not contain any country positions or comments. In
addition the report is relatively brief and arguably provides only limited general
guidance in the application of the complex AOA principles to the new updated
concepts of a PE in the post BEPS 2017 version of Article 5. However, the examples
are very good and provide an examination of both the PE status and the
attribution of profits in the scenarios provided. It is suggested you read through all
the examples and note how the profits are attributed to the parties involved, in
particular in the dependent agency scenarios (under the updated 2017 version of
Article 5(5)).

20.5 Practical Application of the Transfer Pricing Process
As noted above, firstly, the functional analysis is used to determine the pricing
methodology. This is done by selecting the functional profiles that will be linked to
the pricing methodology. This will be determined in most cases by the location of
the significant people functions.
By analogy with the principles in the OECD TPG it is necessary to determine the
least complex part of the organisation (PE or Head Office (HO)) in order to test for
compliance with the arm’s length principle.
This is done by using the information obtained from the results of the functional
analysis. The tested entity is usually the entity where the pricing method can be
applied to give a reliable result and where comparable data can be located.
Normally this is the least complex entity (see paragraph 2.59).
So what is the least complex entity? It is the entity generally performing the routine
functions, owning limited intangible assets and incurring the least risks. This can
either be the PE or the HO.
As we saw in an earlier chapter when we looked at the choice of transfer pricing

methodology, there are a number of recognised labels used in determining the
profile of an entity which can be used to link to the choice of the transfer pricing
method. For ease of reference these are reproduced here.

Examples of functional profiles and links to pricing methodologies
Functional profile Description Pricing method

Group This is the entity containing Residual profit after
entrepreneur/intangible the decision makers, taking rewarding the other
owner the investment risks (eg entities in the supply
research, new markets and chain for their
innovation). The group functions.
entrepreneur can take
several forms. For example it
can be a manufacturer, the
group researcher or the
group product designer.
Contract manufacturer A contract manufacturer CUP/Cost plus
produces goods under the method/Transactional
direction and using the net margin method
technology of the group
principal (usually by
reference to a contract). Its
retain the group
manufacturing contract. In
its most limited risk form it will
be a toll manufacturer with
the principal supplying and
retaining ownership of all
materials.
Service provider A service provider supplies CUP/Cost plus
services to other group method/Transactional
companies usually by net margin method
reference to a contract. Its
provide contracted services
at budgeted costs.
Distributor A group distributor distributes CUP/Resale minus
goods supplied by its method/Transactional
principal. Its risk profile can net margin method
vary dependent on the
structure of the operation.
Of course entities exist that do not completely fall directly into these categories as
they may be performing multi-functions and more than one pricing method has to
be applied.
As noted in the earlier chapter, the functional analysis provides the information
required for performing the comparability studies (also referred to as economic
analysis or benchmarking) ie the information obtained from the functional analysis
will be used to select comparables using the available comparability factors
contained in the OECD TPG (which we have looked at in earlier chapters).
The type of comparability analysis will be determined by the choice of transfer

pricing testing method.

20.6 Attribution of Profit in Excess of the Total Profit of the Enterprise
As noted above, another issue is whether the profits of a PE can be higher than the
profits of the enterprise as a whole. This is a question of interpretation of Article 7.
Paragraph 1 talks about the attribution of profits as we saw above.
The OECD Commentary states that Article 7 paragraph 1 should be read in

conjunction with paragraph 2, which states what profits should be attributed to a
PE ie a PE should be attributed the profits which it might be expected to make if it
were a distinct and separate enterprise engaged in the same or similar activities
under the same or similar conditions and acting wholly independently.
Therefore the view is that paragraph 1 does not restrict the amount of profits that
can be attributed to a PE to the amount of profits of the enterprise as a whole.
(OECD Model Treaty 2010 Commentary Paragraph 17).
Profits may therefore be attributed to a PE even though the enterprise as a whole

has never made profits. The converse can also apply ie the application of Article 7
may result in no profits being attributed to a PE even though the enterprise as a
whole has made profits.
20.7 Comparison of the Article 7 OECD Approach to Article 9
As we have seen above, the reason why we have the AOA for the attribution of
profits to a PE is that we cannot apply Article 9 to ‘dealings’ – one of the key
reasons for this is that a PE cannot legally contract with its HO so an approach that
beings with looking at contractual relationships and comparing them to the
functional analysis cannot be used.

CHAPTER 21
PE IN THE UN MODEL DOUBLE TAX CONVENTION
In this chapter you will cover the following:

– The definition of a PE in Article 5 of the UN Model DTC
– The rules relating to attribution of profits as set down in Article 7 of the UN Model DTC
21.1 Introduction
The layout of the UN Model DTC is very similar to the OECD Model DTC. Permanent
Establishments (PEs) are defined in Article 5 and the allocation of business profits to
PEs is dealt with in Article 7.
21.2 Definition of a PE
If you have a copy of the permitted text Kees van Raad Materials on International,
TP and EU Tax law, you will see that there is a copy of the UN Model DTC in there.
Article 5 looks very similar to Article 5 in the OECD Model DTC although there are
some differences.
The first important difference comes in paragraph 3 where the definition of a

building site, construction or installation project etc is 6 months rather than 12
months.
Paragraph 3 also includes an additional paragraph relating to services:
‘The furnishing of services, including consultancy services, by an enterprise

through employees or other personnel engaged by the enterprise for such
purpose, but only if activities of that nature continue (for the same or a
connected project) within a Contracting State for a period or periods
aggregating more than 183 days in any 12-month period commencing or
ending in the fiscal year concerned.’
You will recall from an earlier chapter that the OECD Model DTC includes an
option to insert a paragraph relating to services in the Commentary.
In the 2017 update to the UN Model DTC, paragraph 4 was aligned with the OECD
Model DTC Article 5 paragraph 4 and anti-fragmentation provisions identical to
those in the OECD Model DTC were added as paragraph 4.1.
In the list of what is deemed not to constitute a PE in paragraph 5, ‘delivery’ is not

mentioned in the UN Model DTC, but is mentioned in the OECD Model DTC.
Therefore a delivery activity might result in a PE under the UN Model DTC, without
doing so under the OECD Model DTC.
Paragraph 5, dealing with the definition of a dependent agent, has an additional

subparagraph specifying that there will be a dependent agent if there is no
authority to conclude contracts but the person habitually maintains in the first-
mentioned State a stock of goods or merchandise from which they regularly
deliver goods or merchandise on behalf of the enterprise.

Paragraph 6, dealing specifically with insurance companies, does not appear in

any form in the OECD Model DTC. Furthermore, as noted in an earlier chapter, the
OECD, as part of its work on Action Point 7 of the BEPS Action Plan, determined
that a separate rule was not required for this sector (see earlier chapter on
Permanent Establishments). Paragraph 6 states:
‘Notwithstanding the preceding provisions of this Article, an insurance

enterprise of a Contracting State shall, except in regard to re-insurance, be
deemed to have a permanent establishment in the other Contracting State if
it collects premiums in the territory of that other State or insures risks situated
therein through a person other than an agent of an independent status to
whom paragraph 7 applies.’
Paragraph 7 equates to paragraph 6 in the OECD Model DTC.
Paragraph 8 of the UN Model DTC equates to paragraph 7 in the OECD Model

DTC and deals with subsidiaries.
In summary, we can say that the definition of a PE in the UN Model DTC is wider in
scope than the OECD Model DTC. This remains the case following amendments to
both Model DTCs in 2017 as a result of the final report on Action Point 7 of the BEPS
Action Plan as noted in the previous chapter, although the gap is narrowing.
21.3 Article 7 Business Profits
We see a greater divergence between the two models when we look at Article 7.
Paragraph 1 of the UN Model DTC is essentially the same as paragraph 1 of the

OECD Model DTC before the 2010 update that we looked at in an earlier chapter
(this was in fact the 2008 version and will be referred to as such in this chapter) plus
the addition of a limited force of attraction rule. This allows the country in which
the PE is located to tax not only the profits attributable to that PE but also other
profits of the enterprise derived in that country to the extent allowed under the
Article. It is noted that the force of attraction rule is limited to business profits
covered by Article 7 and does not extend to income from capital (dividends,
interest and royalties) covered by other treaty provisions.
Paragraph 2 is the same as paragraph 2 of the 2008 OECD Model DTC; that is to
say it includes the distinct and separate enterprise approach. The Commentary to
the UN Model DTC stresses the importance of the arm’s length approach and
states that normally the starting point will be the records kept by the enterprise of
the amounts paid.
Paragraph 3 is fairly prescriptive in terms of the amounts that can be deducted.

The Commentary to Article 7 notes that members from developing countries felt
that it would be helpful to include all the necessary definitions and clarifications in
the text, with a view, in particular, to assisting developing countries. Some of those
members also felt that provisions prohibiting the deduction of certain expenses
should be included in the text of a bilateral tax treaty to make sure that taxpayers
were fully informed about their fiscal obligations.
As in the case of the OECD Model DTC, the rules on deductions are for the
purposes of establishing the profits attributable to the PE. They do not impact on
domestic law which will have the final say on which deductions are allowed for tax
purposes.

With regard to royalties, the Commentary refers to the inherent difficulty of trying to
allocate ownership of IP when there is only one legal entity.
In the case of services, a distinction is drawn between the situation where services
are provided by an entity where this is its main business and services that are just
part of the general administration of a business. A profit is only appropriate in the
first case with the proviso that a benefit is also received by the recipient.
The rules relating to the deduction of interest often cause debate. It is noted that
at paragraph 41 the Commentary states:
• From the legal standpoint, the transfer of capital against payment of interest
and an undertaking to repay in full at the due date is really a formal act
incompatible with the true legal nature of a permanent establishment.
• From the economic standpoint, internal debts and receivables may prove to
be non existent, since if an enterprise is solely or predominantly equity funded
it ought not to be allowed to deduct interest charges that it has manifestly not
had to pay. Whilst, admittedly, symmetrical charges and returns will not distort
the enterprise’s overall profits, partial results may well be arbitrarily changed.
In the case of external borrowing it is necessary to calculate how much can be

allocated to the PE.
Paragraph 4 & 5 mirror the 2008 OECD Model DTC allowing apportionment of
profits in some cases and requiring consistency in treatment year by year.
The final paragraph mirrors paragraph 4 in the current Article 7 giving priority to
other articles.
In summary we can say that the UN Article 7 is the ‘former’ OECD Article 7; that is
to say the 2008 version plus some extensions.

CHAPTER 22
COMPLIANCE ISSUES
In this chapter we are going to look at compliance issues, in particular:

– why documentation is important
– the OECD TPG on transfer pricing compliance
– domestic law approaches to transfer pricing compliance
– unilateral or multilateral documentation
– non-documentation considerations
– safe harbours
22.1 Introduction
In simple terms, compliance in the context of transfer pricing is predominantly

achieved by taxpayers transacting with related parties on an arm’s length basis. If
companies can achieve this, then the most significant costs of non-compliance
(being penalties and double taxation) are substantially mitigated. However, it is
not sufficient for taxpayers to transact on an arm’s length basis; they must also be
able to demonstrate that this is the case. This chapter explains the steps that
taxpayers must go through to demonstrate compliance with transfer pricing
regulations. It also addresses some of the broader compliance requirements of
taxpayers as a result of transacting across borders.
Compliance with transfer pricing regulations comprises a number of aspects, with

the importance of each varying across different countries. These include:
• Maintenance of primary documents and records, including (but not limited to)
accounting records, invoices, and intercompany agreements;
• Disclosures to be made to the tax authority at the time of filing the tax return;
• Analysis to be maintained or prepared for the purpose of presenting to the tax

authority upon request.
Whereas the first two tend to be factual or quantitative information, the third
aspect tends to be more qualitative in nature. It is analysis that provides the
evidence upon which taxpayers rely to demonstrate the arm’s length nature of
their pricing, and is referred to as transfer pricing documentation. In terms of
resources used and cost to the taxpayer, transfer pricing documentation tends to
be by far the most significant aspect of transfer pricing compliance.
22.2 Why is Documentation Important?
It is worth considering the role of documentation for both the taxpayer and the tax
authority. It is generally the objective of tax authorities to ensure that taxpayers
pay appropriate taxes based upon the application of the arm's length principle. In
the absence of documentation, tax authorities would have only limited
information on which to evaluate transfer prices.
They would have access to statutory accounts, tax returns and publicly available
information, but none of these are sufficient to undertake anything but a high level
assessment of whether transfer prices are arm's length. From the perspective of the
tax authority therefore, the objective of documentation is typically not to provide

an exhaustive assessment of all aspects of transfer prices, but rather to allow them
to be able to assess whether or not to pursue a transfer pricing enquiry.
Furthermore, if a tax authority does start a transfer pricing audit, documentation
allows them to be a lot more focused on the issues with the highest risk. Thus, for
tax authorities, documentation is a crucial part of the process in allowing them
optimal use of their resources in the policing of transfer pricing.
For the taxpayers, preparation of transfer pricing documentation should be more

than simply about meeting a compliance requirement. Through the preparation of
documentation, taxpayers are able to proactively manage their transfer pricing
risk.
Documentation provides a platform for the taxpayer to present its case. Clearly
any analysis needs to be factually accurate and economically sound.
Through the preparation of robust documentation, the taxpayer has the

opportunity to present the facts in the most favourable light and to a large extent,
determine the criterion through which transfer prices are evaluated. Provided the
taxpayer has made a reasonable attempt to follow OECD principles in
determining the choice of method and the means of application, it can be very
difficult for tax authorities to successfully apply a radically different framework.
The preparation of adequate transfer pricing documentation is often sufficient to

discharge the burden of proof regarding the arm’s length nature of prices (where
this rests with the taxpayer), and put the onus back on the tax authority to
demonstrate that the arm’s length standard has not been met.
Furthermore, the very process through which taxpayers prepare documentation

can help in the identification and management of transfer pricing risk.
Documentation can require the collection of considerable amounts of facts and
data regarding the nature of cross-border dealings. As a process, it can therefore
provide some discipline to tax risk management in the area of transfer pricing,
allowing the tax function within a multinational company to identify early those
countries or transactions with significant risk and devote resources accordingly.
22.3 The OECD Guidelines on Transfer Pricing Compliance
There are two key reference points for taxpayers when considering transfer pricing
compliance. Clearly, local country legislation, regulation and tax authority
guidance are crucially important. However, in many cases, such guidance is built
(either explicitly or implicitly) upon the principles set out in the OECD TPG.
Therefore, it is important to consider in the first instance what the OECD TPG have
to say about compliance and documentation.
The OECD TPG address a number of aspects of compliance. Chapter IV focuses

on administrative approaches. In addressing tax authorities, the OECD TPG have
no strict authority, and acknowledge that tax compliance procedures are a
matter of domestic sovereignty. It is within the rights of every jurisdiction to establish
its own compliance procedures. Nevertheless, they seek to offer guidance for
administrative procedures that enforce compliance. The OECD TPG encourage
restraint and reasonableness in administrative practices, consideration of the
burden of proof and the application of penalties. Chapter V deals with
documentation.
The BEPS Action Plan released in July 2013 contained proposals relating to
documentation.

Action Point 13 is as follows:
‘Develop rules regarding transfer pricing documentation to enhance

transparency for tax administration, taking into consideration the compliance
costs for business. The rules to be developed will include a requirement that
MNE’s provide all relevant governments with needed information on their
global allocation of the income, economic activity and taxes paid among
countries according to a common template.’
Thus it was proposed that there should be a common template for documentation
going forward.
Guidance on a new documentation package was issued in February 2015.
Chapter V now takes a three-tier approach and is much more prescriptive in terms
of the information to be provided than the original Chapter V. The three-tier
approach consists of:
1. A master file, containing specific information relevant for all MNE group
members. The master file is intended to provide a complete picture of the
MNE’s global operations.
2. A local file, referring specifically to material transactions of the local taxpayer;

and
3. A country-by-country (CbC) report, containing high-level data with respect to

the global allocation of the MNE’s income and taxes and the certain
measures of economic activity.
The OECD believes that use of the three-tier approach that requires MNEs to
prepare a CbC report of the master file and the local file will result in MNEs
articulating consistent transfer pricing positions. This will provide the tax authorities
with useful information to assess transfer pricing risk and to make decisions about
where resources can best be used to commence and carry out audits. This will
allow the tax authorities to be able identify more easily when MNEs are using
transfer pricing and other methods to artificially move substantial amounts of
income to low tax environments.
The master file is intended to provide a high-level overview in order to place the
MNE group’s transfer pricing practices in their global economic, legal, financial
and tax context. It is not intended to require exhaustive listings of minutiae (see
paragraph 5.18). The information will be organised into 5 categories:
1. Organisational chart;
2. Business description – this will include profit drivers, supply chains, main
geographic markets, a functional analysis and any important business
restructures;
3. Intangibles including the MNE’s overall intangibles strategy and policies, an

identification of important intangibles and who owns them, the location of
important R&D facilities and R&D management, details of important
agreements such as cost contribution arrangements, and any intangible
transfers;
4. Financial transactions including policies and important third party

arrangements, identification of any MNE member that provides a central
financial function including details of the country where it is formed and the
POEM;

5. Financial position including allocation of tax; this includes annual consolidated

financial statements and a list of any unilateral advance pricing agreements
(APAs) and relevant tax rulings.
The local file will provide more detailed information relating to MNE group
members’ specific intercompany transactions. This local file is required to contain:
• Information regarding the local entity, including organisational structure,

description of the local business, business strategy and key competitors;
• Detailed analysis of material related party transactions, including copies of

material intercompany agreements, functional and comparability analyses,
selection and application of the most appropriate transfer pricing
methodology and copies of any APAs or relevant rulings to which the local
jurisdiction is not a party but ‘are related to’ the related party transactions;
• The amounts of intra-group payments and receipts for each category of

controlled transactions involving the local entity, broken down by tax
jurisdiction of the foreign payor or recipient; and
• Local entity financial information, including local entity financial accounts for
the financial year; information on allocation schedules showing how financial
information used in applying the transfer pricing method can be tied to the
financial statements and summary schedules of data for comparability and
the source of trial data.
Paragraph 5.22 states that where a requirement of the local file can be fully
satisfied by a specific cross-reference to information contained in the master file,
such a cross-reference should suffice.
The CbC information is to be reported to tax authorities and only used at a very
high level. The OECD has issued guidance on ‘appropriate use’ of the CbC report:
• high level transfer pricing risk assessment
• assessment of other BEPS shifting related risks
• economic and statistical analysis, where appropriate.
Both the final report on Action Point 13 of the BEPS project and the bilateral and
multilateral Competent Authority Agreements (CAA) for exchange of Information
(see below) state that the CbC report should not be used by itself as a basis for
proposing changes to transfer prices using global formulary apportionment. The
CbC report can be used by tax authorities when planning a tax audit or as the
basis for making further enquiries on transfer prices and other matters during an
audit. The OECD Forum on Tax Administration has prepared a handbook to
support tax authorities in making effective use of CbC information for the purposes
of tax risk assessment. The OECD has also issued guidance on the inappropriate
use of CbC reports.
Action Point 13 and the CAA contain consequences for non-compliance with the
appropriate use of a CbC report. Non-compliance can result in the loss of the right
to receive and use the CbC report. The countries using CbC reporting have given
a commitment to monitor compliance by other jurisdictions and report failures.
Some of the reporting will be via peer reviews.

The data points that will be required to be reported for each country will be the
following:
• Revenues (from both related and unrelated party transactions);
• Profit before income tax;
• Income tax paid (cash basis);
• Current year income tax accrual;
• Stated capital;
• Accumulated earnings;
• Number of employees; and
• Tangible assets (excluding cash and equivalents).
Further guidance on each tier and a CbC report template are provided in the
annexes to Chapter V. You should be familiar with these templates. The clear
implication is that the template is designed to highlight those low-tax jurisdictions
where a significant amount of income is allocated, without some ‘proportionate’
presence of employees. This means, in practice, that there will be pressure to
ensure that profit allocations to a particular jurisdiction are supported by the
location in that state of sufficient appropriately qualified employees, who are able
to make a 'substantial contribution' to the creation and development of
intangibles.
Concerns have already been noted regarding the confidentiality of this data, as
well as the potential for adjustments by tax administrations based on a formulary
apportionment approach, leading to many more transfer pricing controversies.
The OECD has also noted that some countries (for example Brazil, China, India and
other emerging economies) would like to add further data points to the template
regarding interest, royalty and related party service fees. These data points are not
included in the current template, but the compromise is that the OECD has
agreed that it will review the implementation of this new reporting. A concern in
this context is that there may well be a tendency to expand CbC reporting,
particularly in developing countries. The emerging market economies that
implement CbC reporting will likely require the reporting of interest, royalty and
related party service fees; they will also be likely to require CbC reporting for any
company doing business in their jurisdiction, regardless of where the MNE parent is
located.
The OECD undertakes annual reviews of CbC. For each jurisdiction, the review
covers the domestic legal and administrative framework, the exchange of
information framework and measures in place to ensure the confidentiality ad
appropriate use of CbC reports.
A virtual public consultation meeting took place in May 2021. Points raised include:
making CbC reports more widely available; decreasing the threshold for CbC
reporting; that the focus should be on improving existing standards rather than
changing them; amending the content of the CbC report; technical issues relating
to how to define a group; common control standards for filing; the inclusion of
entities that are not resident anywhere; and the need for the publication of
additional information by the OECD on how the data from CbC reports is being
used. At the conclusion of the meeting it was noted that the work on CbC
reporting will be aligned with the ongoing work on the digitalization of the

economy. At the time of writing there has been no further publication on the
outcomes of the consultation.
Chapter V also includes guidance on compliance issues in Section D. It states the

need for tax authorities to take account of the costs involved in collecting
information.
With regard to timing, paragraph 5.30 suggests that the best practice is to require
that the local file be finalised no later than the due date for the filing of the tax
return for the fiscal year in question. The master file should be reviewed and, if
necessary, updated by the tax return due date for the ultimate parent of the MNE
group. Paragraph 5.31 goes on to recognise that the CbC reporting may need a
longer time frame and states that in some cases it may be extended to one year
following the last day of the fiscal year of the ultimate parent of the MNE group.
To deal with issues of materiality Chapter V encourages individual countries to

establish specific materiality thresholds for the master file and local file, in addition
to developing regimes for reducing compliance obligations for small and medium
enterprises (SMEs).
Taxpayers should not be required to keep documentation for unreasonable

amounts of time and should be allowed to keep it electronically where it can still
be made available to the tax authority in a timely manner.
Chapter V recommends that taxpayers review and update the master file, local
file and CbC report annually, although significant leverage from prior year
documentation is permitted where appropriate. Database searches for
comparable companies may be updated every three years instead of annually,
however the financial data for the comparables should be updated each year.
Language requirements should be governed by local law. In addition countries

are encouraged to permit filing of transfer pricing documentation in commonly
used languages where it will not compromise the usefulness of the documents (see
paragraph 5.39).
There is a section on penalties recognising that each country will have their own
and that care needs to be taken to ensure that penalties regimes do not result in
taxpayers favouring one jurisdiction rather than another. It is suggested that
protection from penalties could be used as a mechanism to encourage
compliance.
Confidentiality is covered with the tax authorities being told that all steps should be
taken to maintain the confidentiality of information provided for transfer pricing
purposes.
Paragraph 5.46 states that:
‘The requirement to use the most reliable information will usually, but not
always, require the use of local comparables over the use of regional
comparables where such local comparables are reasonably available.’
In June 2015 the OECD published its package of model legislation and competent
authority agreements to help tax administrations implement the requirement for
multinational companies to report transfer pricing information on a CbC basis. The
guidance on implementation has been updated several times, the latest being in
November 2019. The updates provide guidance on the treatment of minority
interests and various definitions within the model legislation such as the definition of
revenue. It is likely that these updates will continue.

There has been widespread uptake of CbC reporting. Australia, Belgium, Canada,
India, the Republic of Ireland, the UK, Italy, Poland, Netherlands, Finland, South
Korea, Turkey and the US are just a few examples of the countries introducing
legislation. However not all countries are taking on board the OECD
recommendations relating to the master file and the local file – the US is an
example of this.
The ‘Country by country reporting implementation package’ will, according to the

OECD: ‘facilitate a consistent and swift implementation of new transfer pricing
reporting standards developed under action 13 of the BEPS action plan, ensuring
that tax administrations obtain a complete understanding of the way multinational
enterprises (MNEs) structure their operations, while also ensuring that the
confidentiality of such information is safeguarded’.
The implementation package consists of model legislation requiring the ultimate

parent entity of an MNE group to file the CbC report in its jurisdiction of residence,
including backup filing requirements when that jurisdiction does not require filing.
The package also contains three Model Competent Authority Agreements
(MCAA) to facilitate the exchange of CbC reports among tax administrations. The
model agreements are based on the Multilateral Convention on Administrative
Assistance in Tax Matters, bilateral tax conventions and Tax Information Exchange
Agreements (TIA’s).
The model legislation includes an exemption for a group with consolidated total
revenue of less than €750m (or an equivalent local currency). This means that CbC
reporting will not apply to all groups.
In principle, the report should be iXBRL tagged and filed electronically by the
ultimate parent entity (UPE) of the group with its home tax authority by the filing
deadline – which can be anything up to 12 months from the end of the
accounting period. The report should then be automatically shared by the UPE’s
tax authority with the relevant countries mentioned in the report.
In cases where the UPE is not required to file a report in its home tax jurisdiction, or
the automatic exchange mechanism does not work, most countries have
implemented a backstop in the form of a secondary filing requirement. In this
case, constituent entities of the group must file a CbC report directly with their
local tax authorities. Alternatively, in certain circumstances the UPE can elect to
file the CbC report as a surrogate in another territory, thereby making use of the
tax authority automatic sharing mechanism and reducing the compliance burden
of multiple secondary filing obligations.
The OECD model legislation defines a group as:
‘a collection of enterprises related through ownership or control such that it is

either required to prepare Consolidated Financial Statements for financial
reporting purposes under applicable accounting principles or would be so
required if equity interests in any of the enterprises were traded on a public
securities exchange’.
The legislation provides further clarification and details on a number of areas

including: the secondary filing mechanism; notification of the requirement to file;
scope of reporting in situations where consolidated financial statements are not
produced; definition of a MNE group; currency conversion for calculating the
€750m revenue exemption; expectations regarding the penalty regime; and
electronic filing format.

The activation of ‘automatic exchange relationships’ is part of the implementation

of CbC reporting in accordance with the BEPS Action Point 13 minimum standard.
The full list can be found at www.oecd.org/tax/automatic-exchange/country-by-
country-exchange-relationships.htm.
Rules and Procedures
From the perspective of the taxpayer, there is an acknowledgement that some

work will be required on their part in order to demonstrate that transfer prices meet
the arm’s length standard.
Taxpayers are advised to give consideration to what transfer pricing arrangements

are appropriate before pricing is established through the application of principles
established in earlier chapters of the OECD TPG.
For example, it would be prudent for taxpayers to understand whether CUPs exist,
and whether conditions have changed from previous years to inform whether
transfer prices should change. Taxpayers should apply the same prudent
management principles that would govern other business decisions of similar
complexity, and should therefore expect to prepare and obtain certain materials
to help achieve this.
In this regard, taxpayers should accept that it may be necessary to prepare

written documents that would not otherwise be required in the absence of tax
considerations. However, it is also clear that the taxpayers should not be expected
to incur disproportionately high costs relative to the complexity. For example,
taxpayers should not have to undertake an exhaustive search for CUPs if there is a
reasonable case for believing that such CUPs do not exist. Notwithstanding this,
taxpayers should recognise that tax authorities will need to make an assessment
on the arm's length nature of transactions based on information presented by the
taxpayer, however incomplete that information is. See Section B1 of Chapter V of
the OECD TPG.
Furthermore, there should be an acknowledgement that the greater the

complexity of the issues, the more significance will be attached to the
documentation. Therefore, the taxpayer should take responsibility to ensure
adequate document retention and voluntary disclosure of information, in order to
help to improve the persuasiveness of analysis.
For their part, tax authorities are discouraged from being too onerous in their
expectations of taxpayers. They should seek from taxpayers only the minimum
documentation needed to make reasonable assessment of transfer prices. They
should request information to be prepared only if it is indispensable for verifying the
arm's length nature of transactions.
Furthermore, tax authorities should be reasonable in the type of information they

request from taxpayers in documentation. Requests for documents that became
available only after the transaction was entered into should be limited to avoid
the use of hindsight. Instead, tax authorities should have regard for what the
taxpayer would have reasonably had available at the time of the transaction.
They should take care not to ask for what is not in the possession or control of the
taxpayer. This includes acknowledgement that it may be difficult to identify data
from foreign affiliates, particularly where such information is in practice not
necessary to make a reasonable assessment of the taxpayer. See Section D of
Chapter V of the OECD TPG.

22.4 Domestic Law Approaches to Transfer Pricing Compliance
It is not possible to set out all the requirements in all countries. Nevertheless, there
are common themes of which taxpayers should be aware:
• Legislative requirement for documentation – In some countries (such as the

US), taxpayers are required to have in place documentation at the time that a
return is filed with the tax authority. Indeed in certain jurisdictions, taxpayers
meeting certain criteria are expected to submit that documentation with the
return to the tax authority.
• Administrative practices around documentation – In some territories, while

there is no legislative requirement for documentation, the administrative
practice of the tax authority renders it essential to prepare contemporaneous
documentation. Short response times (sometimes two weeks or less) to a
request for the submission of documentation ensure that taxpayers are well
advised to maintain analysis. This can be compounded by a refusal by tax
authorities to consider any evidence not submitted with the initial response to
a documentation request.
• Specific content – Taxpayers should be aware that some tax authorities

impose specific requirements on the content of documentation. This may be
as simple as requiring documentation to be maintained in local language or a
specific format. Alternatively, tax authorities may require specific information
to be included, such as transactional data or information relating to the local
business operations. In addition, the tax authority may specify the form of the
analysis, such as requiring the local entity to be the tested party irrespective of
the policy applied by the taxpayer, or requiring the use of local comparable
data rather than regional or global sets.
• Transfer pricing disclosures – A growing trend is for tax authorities, including

those of countries such as Australia, Denmark and Malaysia, to request
information relating to transfer pricing as part of the tax return. Information to
be provided typically discloses the size and type of transactions involving the
taxpayer, as well as the location of the counter parties. Taxpayers may also be
required to provide information on the transfer pricing method(s) applied and
the level of contemporaneous documentation maintained.
• Safe harbours – These are simple rules or provisions that taxpayers can follow to
have certainty over tax treatment (eg. Cost plus a defined margin for
specified services). The existence of safe harbours may provide relief from tax
compliance burdens, where taxpayers meet the defined criteria. We look at
these in more detail below.
Many countries use the OECD TPG as the starting point for establishing
documentation requirements. However, taxpayers should avoid the assumption
that preparing documentation consistent with OECD principles will be sufficient to
avoid compliance penalties. In particular, for countries where taxpayers are
aware that they have significant transfer pricing risk (typically arising from
losses/low profit or complex transactions), careful consideration should be given to
local tax authority requirements and expectations. Australia is an example of this –
the local file requirements go beyond the template set down by the OECD. It will
be important to ensure that the local file requirements as set down by the OECD
are customised if a MNE wants to be sure of compliance with Australian rules.

22.5 Unilateral or Multilateral Documentation?
The range of different environments created by tax authorities (not all will adopt
the OECD approach outlined above) provides taxpayers with a choice of how
best to approach compliance. Once mandatory compliance issues have been
dealt with, the most fundamental choice to be made is around approach to
documentation, both in terms of the level of resources to commit and the degree
of global co-ordination.
In many ways the simplest is to compile unilateral documentation in countries that

are deemed to be sufficiently at risk of an enquiry. Such analysis is prepared on a
stand-alone basis, often by the local tax team rather than the global one and is
designed to meet the specific requirements of an individual country.
The functional analysis and benchmarking are likely to be focused on the country
being documented, rather than a broader perspective of the supply chain.
Such an approach has the obvious drawback that it can be highly inefficient, as
there is likely to be significant duplication in effort where unilateral documentation
is prepared in multiple territories.
More significantly, it can result in inconsistencies in the analysis being prepared.
One of the key developments in tax authority behaviour in relation to transfer

pricing has been the increased sharing of information. Companies taking the
approach of preparing separate reports in different territories run the risk of
contradictory analysis (such as different approaches to the use of CUPs, different
benchmarking approaches) which can significantly undermine the analysis
prepared.
The alternative is to prepare some form of multi-jurisdiction documentation. This

could take the form of bilateral, regional, business unit-focused, or even global
analysis.
Such documentation presents consolidated analysis for the purposes of supporting

cross-border transactions in a number of territories in a single report. It provides the
tax authority with a more complete overview of the taxpayer's value chain.
It can be more onerous to prepare, requiring potentially substantial co-ordination

of resources by taxpayers.
Nevertheless, there are likely to be efficiencies in production and benefits in the

ability it creates to monitor and control a group's transfer pricing risks from a
central perspective.
One potential drawback from the multi-jurisdiction approach is the lack of

flexibility to meet individual country requirements. Taxpayers may produce robust
analysis that is consistent with the approach set out in the OECD TPG, yet fail to
meet country-specific compliance needs.
There have been several initiatives to address this matter. The Pacific Association
of Tax Authorities (PATA) has provided guidance on a documentation package
that, if followed by taxpayers, would be accepted as appropriate documentation
in Australia, Canada, Japan and the US.
The PATA Documentation Package has three operative principles that need to be
met to avoid the imposition of PATA member transfer pricing documentation-

related penalties. First, MNEs need to make reasonable efforts, as determined by

each PATA member’s tax administration, to establish transfer prices in compliance
with the arm’s length principle. Second, MNEs need to maintain contemporaneous
documentation of their efforts to comply with the arm’s length principle. Third,
MNEs need to produce, in a timely manner, that documentation upon request by
a PATA member tax administrator.
In summary the requirements under the PATA approach are to have

documentation showing:
• Organisational structure: showing the parties involved in the controlled

transactions, including any only indirectly involved. Details of the worldwide
group and history of shareholding changes.
• Nature of the business/industry and market conditions: including the mission

statement of the group, a description of the structure, intensity and dynamics
of the competitive environment, details of the broader regulatory and other
factors affecting the taxpayer’s business, a description (catalogue) of
intangible property and copies of annual reports and financial statements.
• Controlled transactions: including a description of the controlled transactions

that identifies the participants, the scope, timing, frequency of, type and the
value of the controlled transactions as well as the currency of the transactions;
the terms and conditions of the transactions and their relationship to the terms
and conditions of each other transaction entered into between the
participants; identification of internal data relating to the controlled
transactions and any similar transactions.
• Assumptions, strategies, policies: including details on market and business

strategies and how strategies have impacted on pricing.
• Cost contribution arrangements (CCA): full detail is required including copies

of the arrangements and information on buy-in and buy-out payments with full
information on how the contribution of each member is determined.
• Comparability, functional and risk analysis: full details on comparability

including how comparables were selected and evaluated; details of any
adjustments and any aggregation; details on the establishment of ranges of
outcomes and the extension of the analysis over a number of years with
reasons for the years chosen.
• Selection of the transfer pricing method: identification of the method selected

and the reasons why it was selected, including, for example, economic
analysis and projections relied upon; a description of the data and methods
considered and the analysis performed to determine the transfer pricing;
explanation as to why other methods were not selected (where countries
have a hierarchy of methods, it may only be necessary to explain why higher
ranking methods were rejected).
• Application of the transfer pricing method: documentation of assumptions and

judgments made in the course of determining an arm’s length outcome (refer
to the comparability, functional and risk analysis section above);
documentation of all calculations made in applying the selected method in
respect of both the taxpayer and the comparable; appropriate updates of
prior year documentation relied upon in the current year to reflect
adjustments for any material changes in the relevant facts and circumstances.

• Background documents: these are documents that provide the foundation for
or otherwise support or were referred to in developing the transfer pricing
analysis.
• Index to documents: this is a general index of documents and a description of

the record-keeping system used for cataloguing and assessing those
documents (required in the US and encouraged, but not required, by other
PATA members). The general index is not required to be prepared
contemporaneously.
Although the PATA countries are adopting legislation to bring in CbC reporting
based on the OECD rules, they still have local rules that differ from the template
laid down by the OECD. As mentioned above the Australian reporting
requirements require more detail in the local file and the proposed legislation in
the US contains additional guidance not in the OECD template.
Within Europe, the EU Joint Transfer Pricing Forum has also provided guidance on
documentation that should meet compliance requirements in Member States. The
guidance is in the form of a code of conduct issued in June 2006. The approach
supported is that of the master file concept.
Under this approach, documentation is compiled in two parts: a centralised

masterfile contains relevant standardised information for the global group, and this
is supplemented by country-specific appendices addressing local country issues.
The combination of efficiency and flexibility created by this approach means that
it is being increasingly used by taxpayers, not just to address their European
compliance needs but their global ones as well.
The use of EU transfer pricing documentation (TPD) is optional for MNEs.
The masterfile should contain the following items:
a. a general description of the business and business strategy, including changes

in the business strategy compared to the previous tax year;
b. a general description of the MNE group’s organisational, legal and

operational structure (including an organisation chart, a list of group members
and a description of the participation of the parent company in the
subsidiaries);
c. the general identification of the associated enterprises engaged in controlled

transactions involving enterprises in the EU;
d. a general description of the controlled transactions involving associated

enterprises in the EU, ie a general description of:
i. flows of transactions (tangible and intangible assets, services, financial),
ii. invoice flows, and
iii. amounts of transaction flows;
e. a general description of functions performed, risks assumed and a description

of changes in functions and risks compared to the previous tax year, eg
change from a fully-fledged distributor to a commissionaire;
f. the ownership of intangibles (patents, trademarks, brand names, know-how,

etc.) and royalties paid or received;

g. the MNE group’s inter-company transfer pricing policy or a description of the

group's transfer pricing system that explains the arm's length nature of the
company’s transfer prices;
h. a list of Cost Contribution Arrangements, Advance Pricing Agreements and

rulings covering transfer pricing aspects as far as group members in the EU are
affected; and
i. an undertaking by each domestic taxpayer to provide supplementary

information upon request and within a reasonable time frame in accordance
with national rules.
The resolution states that ‘The country-specific documentation would be available

to those tax administrations with a legitimate interest in the appropriate tax
treatment of the transactions covered by the documentation.’
Country-specific documentation should contain, in addition to the content of the

masterfile, the following items:
a. a detailed description of the business and business strategy, including

changes in the business strategy compared to the previous tax year;
b. information, ie description and explanation, on country-specific controlled

transactions, including:
i. flows of transactions (tangible and intangible assets, services, financial),
ii. invoice flows, and
iii. amounts of transaction flows;
c. a comparability analysis, ie:
i. contractual terms,
ii. functional analysis (functions performed, assets used, risks assumed),
iii. characteristics of property and services,
iv. economic circumstances, and
v. specific business strategies;
d. an explanation of the selection and application of the transfer pricing

method(s), ie why a specific transfer pricing method was selected and how it
was applied;
e. relevant information on internal and/or external comparables if available; and
f. a description of the implementation and application of the group’s inter-

company transfer pricing policy.
Any MNE that opts for EU TPD should apply the approach collectively to all
associated enterprises.
Each member of the MNE should inform its tax authorities if the documentation is
adopted.

As regards the Member States the code of conduct states:
‘Since the EU TPD is a basic set of information for the assessment of the MNE
group's transfer prices a Member State would be entitled in its domestic law to
require more and different information and documents, by specific request or
during a tax audit, than would be contained in the EU TPD.’
With regard to the documentation the code of conduct states:
‘The sort of documentation that needs to be produced by an enterprise that is

a subsidiary enterprise in a group may be different from that needed to be
produced by a parent company, ie a subsidiary company would not need to
produce information about all of the cross-border relationships and
transactions between associated enterprises within the MNE group but only
about relationships and transactions relevant to the subsidiary in question.’
It should be irrelevant for tax administrations where a taxpayer prepares and stores
its documentation as long as the documentation is sufficient and made available
in a timely manner to the tax administrations involved upon request. Taxpayers
should, therefore, be free to keep their documentation, including their EU TPD,
either in a centralised or in a decentralised manner.
The way that documentation is stored – whether on paper, in electronic form or in

any other way – should be at the discretion of the taxpayer, provided that it can
be made available to the tax administration in a reasonable way.
The country-by-country reporting proposals, including automatic exchange,

(consistent with OECD BEPS Action Point 13) were adopted in May 2016, and the
EU Directive on Administrative Cooperation (‘DAC’) was appropriately amended
to include this exchange – this is known as DAC4. This is effective as from 1 January
2017, in respect of calendar year 2016. These reporting requirements relate to
global MNEs which have a total consolidated group revenue in excess of €750
million (or local currency equivalent). Thus smaller groups are excluded.
Suggestions for public reporting of country-by-country reports have been taken on

board. A draft proposal to change the EU Accounting Directive was released in
April 2016. The global MNEs (as above) operating in the EU would be required to
publish key information such as turnover, number of employees, nature of activities
etc. for every EU Member State in which a company is active, as part of their
financial statements. In February 2017 the European Parliament published a report
with suggested changes. In June 2021 agreement was reached on the wording of
the directive. The agreed wording now needs to be endorsed by the European
Commission and Parliament. You should review any developments in this area,
and in the OECD approach.
22.6 Non-documentation Considerations
Transfer pricing compliance issues extend beyond the realms of documentation,

even if that is often the focus of consideration. Throughout the lifecycle of related
party transactions, there are a number of areas where compliance issues need to
be addressed. Even if the driver behind these issues is not corporate tax, the tax
practitioner still needs awareness of what needs to be done.

Before the Transaction
As well as analysis to ensure that transfer prices are set on arm’s length basis,
companies must also ensure that several other factors are addressed before
transactions are even entered into. Most notably, there is the issue of
intercompany agreements. From the perspective of the OECD TPG, intercompany
agreements are relevant, insofar as they can provide a starting point for
understanding the expected division of responsibility, risks and assets between the
parties. However, from an OECD perspective they are not essential, and in their
absence the terms of the legal arrangements between the parties can be
deemed from the behaviours exhibited.
However, in practice this approach is not followed by all jurisdictions. In certain

territories, it is required to have a legal agreement in place before deductions will
be given in relation to intercompany charges. Predominantly, this relates to royalty
payments for use of intellectual property, and fees paid for related party services.
Indeed, royalty payments may require pre-approval by tax or finance authorities
and be subject to strict limits.
Related to this is the interaction between transfer pricing and foreign exchange
controls. Foreign exchange controls are particularly prevalent across BRIC (Brazil,
Russia, India and China) and developing economies, and can restrict
multinational companies’ ability to remit payment for services provided. Thus, even
where it can be demonstrated that the terms of a transaction meet the arm’s
length standard, executing the transaction on the terms desired may not be
possible. For countries where foreign exchange controls are relevant, appropriate
approvals should be sought before the transaction is entered into where possible.
Executing the Transaction
Where a related party transaction is entered into, due consideration should be

given to accounting requirements. Transactions should be recorded appropriately
and records maintained to support the statutory accounts, with appropriate
remittance. In some cases, taxpayers may choose to offset transaction flows in
opposite directions. As you will recall from an earlier chapter this concept of
intentional offsetting is acceptable under OECD principles, although care needs to
be taken to evaluate the arm’s length nature of each side of the series of
transactions, and recording the basis for believing that the set off is reasonable.
Furthermore, consideration needs to be given to secondary tax implications from
offsetting transactions (such as indirect tax or withholding tax).
Appropriate invoicing is also important, particularly where transactions involve the

provision of related party services. In some countries, tax deductions will not be
given for related party service charges unless supported by an invoice giving an
adequate description of the services provided. Other countries may require the
basis for calculating the charge to be included with the invoice.
Where intercompany transactions relate to services, royalty payments or interest,

consideration needs to be given to withholding tax obligations. Taxpayers should
be aware not only of the rate that is payable, but also the timing of the
obligations. In some territories, the timing will be determined by when payment is
actually made, but in others, it will be determined with reference to when the
service is provided (or interest or royalty becomes payable), which may well be at
an earlier date. Furthermore, where withholding tax rates are reduced under a
double tax treaty, taxpayers should ensure they understand whether this reduction
is automatically applied, or whether a further process needs to be followed in
order to benefit from the treaty rate.

After the Transaction
Many companies manage their transfer pricing through the use of adjustments.
These adjustments are made either periodically through the year, or at year end,
to ensure that the group’s transfer pricing policy is met. For example, there may be
retrospective adjustments to intercompany selling prices to ensure that a
distributor earns an operating margin within a targeted range. Consideration
should be given to the acceptability of such adjustments for local tax authorities.
Furthermore, there should be awareness of the potential customs implications of ex
post adjustments to the transfer price.
Where adjustments result in a downward adjust to the price, then the taxpayer will
have overpaid customs duties. However, claiming refunds from customs authorities
can be an arduous process, and indeed is not always possible. A more significant
risk arises where the adjustment results in an increase to the transfer price. This
could result in additional customs liability that may need to be disclosed, with
potential penalties and likely increased attention from customs authorities in future.
22.7 Safe Harbours
Up to 2013, the OECD TPG concluded that transfer pricing safe harbours were not
generally advisable, and consequently the use of safe harbours was not
recommended. Despite this recommendation many member countries did have
some form of safe harbour rules.
Section E was updated in 2013 and the position on safe harbours was changed.
Paragraph 4.96 acknowledges that the original guidelines were generally negative
towards the use of safe harbours.
The OECD TPG state that safe harbours will be most appropriate when directed at
low risk transactions and/or taxpayers (see paragraph 4.96).
The OECD TPG recognise that safe harbours can be a benefit to tax
administrations as well as taxpayers.
Paragraph 4.102 of Section E contains the following definition:
‘A safe harbour in a transfer pricing regime is a provision that applies to a

defined category of taxpayers or transactions and that relieves eligible
taxpayers from certain obligations otherwise imposed by a country’s general
transfer pricing rules A safe harbour substitutes simpler obligations for those
under the general transfer pricing regime. Such a provision could, for
example, allow taxpayers to establish transfer prices in a specific way, eg by
applying a simplified transfer pricing approach provided by the tax
administration. Alternatively, a safe harbour could exempt a defined category
of taxpayers or transactions from the application of all or part of the general
transfer pricing rules. Often, eligible taxpayers complying with the safe harbour
provision will be relieved from burdensome compliance obligations, including
some or all associated transfer pricing documentation requirements.’
So we can see that a safe harbour can take many forms.
Certain forms of safe harbour are not covered by the definition in Section E; these
include administrative simplifications and exemption from certain documentation
requirements, APAs and thin capitalisation rules (see paragraph 4.103).
The OECD TPG set out a discussion on the pros and cons of safe harbours.

The advantages of safe harbours are identified as:
• Simplification of compliance and reduction of compliance cost;
• Provision of certainty;
• Allowing tax administrations to redirect resources to higher risk taxpayers and

or transactions.
(see paragraph 4.105)
The OECD TPG also state the concerns that:
• Use of safe harbours may mean that the arm’s length principle is not adhered
to;
• Unilateral adoption of safe harbours may increase the risk of double taxation
or double non-taxation;
• There could be an increase in inappropriate tax planning. Taxpayers may shift

income or change the size of transactions to ensure they come within the safe
harbour rules;
• There may be issues of equity and uniformity where apparently similar

taxpayers are not able to use the safe harbours because of the criteria.
(see paragraphs 4.110 to 4.126)
Section E concludes that:
‘However, in cases involving smaller taxpayers or less complex transactions,

the benefits of safe harbours may outweigh the problems raised by such
provisions.’ (see paragraph 4.129).
The recommendations are that where safe harbours are adopted:
• There should be a willingness to modify safe harbour outcomes in the Mutual

Agreement Procedure (MAP). (MAP will be covered in the next chapter.)
• The use of bilateral or multilateral safe harbours is best.
• There should be a clear recognition that a safe harbour, whether adopted on

a unilateral or bilateral basis, is in no way binding on or precedential for
countries which have not themselves adopted the safe harbour.
The OECD TPG go on to state that for complex and high risk areas it is unlikely that
safe harbours can offer a workable solution.
The final recommendation is that tax administrations should carefully weigh the
benefits of and concerns regarding safe harbours, making use of such provisions
where they deem it appropriate. We can see this as somewhat of a move from
the previous negative stance on safe harbours.

22.8 Scoping of Future Revision of Chapter IV of the OECD TPG
In 2018 the OECD invited comments on proposed revisions to Chapter IV of the

TPG as they were considering beginning a project to amend it.
The OECD specifically asked for comments as follows:
• What additional aspects or mechanisms to minimise the risk of transfer pricing

disputes should be included as part of the guidance on transfer pricing
compliance practices (eg co-operative compliance, risk assessment tax
examination practices)?
• Is there any additional guidance that would be useful in relation to

corresponding and/or secondary adjustments to minimise the risk of double
taxation?
• Is there any additional guidance that could be provided on advance pricing

arrangements (APA)? Are there any features of APAs or specific initiatives
related to APAs that could strengthen their role in minimising transfer pricing
disputes? What are the advantages of such initiatives and the implementation
challenges? (APAs are dealt with in the manual in a later chapter.)
• Are there any other mechanisms or issues relevant to the administration of

transfer pricing and/or to prevention and resolution of transfer pricing disputes
for which guidance should be developed as part of the revision of Chapter
IV?
Although Working Party No. 6 decided that there was no need at this stage to
revise or supplement the current guidance on safe harbours and arbitration, public
comments were also invited on those issues.
Comments were received in June 2018; nothing further has been published on this
exercise at the time of writing.
22.9 Summary
The approach taken by taxpayers to transfer pricing compliance is influenced by

many factors. At its heart is the core assertion in the OECD TPG that transfer pricing
is not an exact science. The natural corollary to this is that it is impossible to be
prescriptive in matters of compliance because so much comes down to matters of
judgement.
Ultimately taxpayers must decide how much resource they are willing and able to
commit to transfer pricing compliance, based upon their risk profile and their own
appetite for risk. Having done that, they must then use that resource to produce
analysis that best persuades tax authorities of the arm’s length nature of their
pricing.
The OECD TPG, as well as local country guidance, provides some direction as to
what might be appropriate, but not certainty. As such, the approach taken to
transfer pricing compliance is a fundamental part of the strategic approach to
global tax risk management for many companies.

CHAPTER 23
AVOIDING DOUBLE TAXATION AND DISPUTE RESOLUTION I

– transfer pricing audits
– corresponding adjustments
– secondary adjustments
– Article 25 of the OECD Model Tax Treaty (MAP)
– arbitration in double tax treaties
– BEPS Action Point 14
– the EU arbitration convention
– The EU Directive on Tax Dispute Resolution Mechanisms
– Overview of Article 25 of the UN Model Tax Treaty
23.1 Introduction
Despite the OECD TPG and the arm’s length principle being accepted by the
majority of countries, tax authorities often seek to apply the methodology in the
OECD TPG differently. These differences can often leave MNEs in the middle
seeking to avoid double taxation as a result of the differences in interpretation.
Double taxation can be either juridical or economic in nature.
Juridical double taxation occurs when tax is imposed in two (or more) territories on
the same taxpayer in respect of the same income. This may arise where, for
example, a company resident in one territory derives source income in another
country and the domestic tax legislation of both countries taxes that income. It
can also arise when more than one tax authority considers the taxpayer to be
locally tax resident.
Economic double taxation occurs when more than one tax authority includes the
same income in the tax base of different taxpayers. Transfer pricing disputes can
trigger both economic and juridical double taxation.
There are a number of ways to reduce or eliminate the impact of double taxation.
The OECD TPG Chapter IV covers avoiding and resolving dispute resolution.
It is noted however, that in some circumstances taxpayers may accept a certain

amount of economic double taxation because it is more costly to defend an audit
than accept the additional tax. Furthermore, where a tax authority introduces
administrative simplification procedures, such as safe harbours, in order to access
the safe harbour and consequently reduce the compliance burden, the taxpayer
may accept an element of double taxation.

Taxpayers therefore need to consider the various ways in which the risk of double
taxation can be reduced or eliminated and/or whether a certain level may be
considered acceptable. This is often carried out by the tax department as part of
an MNEs transfer pricing risk management strategy. In considering their strategy,
taxpayers should weigh up the advantages and disadvantages of dispute
management versus dispute avoidance. The various options are illustrated in the
following diagram:
In this chapter we will concentrate on the dispute management aspect. We will

explore from a theoretical and practical perspective, how taxpayers can minimise
double taxation as part of the audit process and we will describe the dispute
resolution frameworks available (MAP, EU Arbitration Convention). We will also
consider how and why most tax disputes are settled away from formal dispute
resolution mechanisms. In the next chapter, we consider how advance pricing
arrangements (APA) can be used in order to avoid double taxation arising in the
first place.
In considering the options following settlement, it is interesting to note that against

a backdrop of budget deficits and increasingly aggressive tax authority enquiries,
the EU Arbitration Convention is not widely and routinely used. Given that transfer
pricing audits are widely reported as a key priority of tax authorities around the
world, it appears that taxpayers are not turning to EU arbitration to resolve their
issues. It seems, whilst factors such as cost and reputational risk have their part to
play in dissuading stakeholders from pursuing such channels, other influences such
as tax authority prudence and speed of resolution (or otherwise) have an impact
here.
23.2 Transfer Pricing Audits
Transfer pricing audits are the process whereby a tax authority undertakes a
review of an enterprise’s transfer pricing affairs to ensure it is compliant with local
legislation. The OECD TPG note that examination practices will vary widely among
OECD member countries (see paragraph 4.6).
The burden of proof in determining whether pricing arrangements are arm’s length
will also differ. Although in many jurisdictions it will be with the tax administration,
(see paragraph 4.1) the UK is one of a growing group where the onus is on the
taxpayer via self-assessment.

Transfer pricing audits constitute a significant business risk to MNEs. Transfer pricing
risk management is therefore crucial for taxpayers to identify risks areas prior to a
potential audit. Pre-audit planning can include defence strategies such as the
preparation of contemporaneous transfer pricing documentation or negotiation of
a unilateral or a bilateral/multilateral APA to obtain certainty going forward.
Transfer pricing disputes will include factual enquiries. The interpretation of the
facts in a transfer pricing context rarely offer one ‘correct answer’. As a result,
disputes may arise and the transfer pricing audit can be a long, drawn out
process.
The resolution of a transfer pricing dispute rests in an area where judgement and
degrees of differences apply. By working with the tax authorities to focus their
enquiries on relevant information and providing the information in a way that
supports the reasonableness and accuracy of the taxpayer’s transfer pricing
policy, the company increases its chances of resolving transfer pricing audits
quickly. The best strategy for an early settlement involves:
• a cooperative approach;
• active involvement from the beginning;
• transparency and guidance; and
• submission of the supporting evidence such as transfer pricing documentation

file.
Joint audits are a method of monitoring compliance with transfer pricing

legislation.
There are two possible options for joint audits:
1. A joint team of individuals from more than one tax authority acting as one
team to jointly identify issues. Independence would need to be protected, as
there may be a loss of individuality for tax authorities.
2. A three-way engagement involving the taxpayer, the tax authority of Country

A and the tax authority of Country B. The advantages of this option are that
risks can be assessed, there is a reduction of compliance burdens and a
discussion of the relevant factors can be conducted at the same time. This
should allow more collaboration whereby all sides can hear the arguments put
forward, also potentially avoiding double taxation arising and the subsequent
need to take adjustments to MAP.
Joint audits, sometimes referred to as simultaneous tax examinations, often derive

their authority from Article 26 of the OECD Model DTC. In addition, as noted later in
this chapter, the Multilateral Instrument (MLI) also allows coordinated multilateral
operations.
Article 26 OECD Model Treaty in its basic terms deals with exchanges of
information between tax authorities for carrying out the provisions of the treaty
and also the enforcement of domestic tax laws, which would include for the
prevention of evasion of taxes. Such information relates to persons who are not
residents of either of the contracting party states, as well as residents. There is a
requirement of confidentiality on the tax authorities in their exchanges of
information, particularly in relation to commercially sensitive information which
may affect the trading position of a company if it were disclosed to third parties.
The intent of the exchange of information article is to increase the volume of
relevant information that would otherwise possibly not be exchanged. However,

this is within the context of the normal administrative laws or practices in existence
in the party states (with the exception of banking secrecy).
Article 26(1) allows for exchange in three different ways: on request; automatically;
or spontaneously. However, the commentary notes (at paragraph 9.1) that the
states are not limited to these methods and that contracting states may use other
techniques to obtain information which may be relevant. The commentary gives
the example of simultaneous tax examinations as a possible other technique.
Directive 2011/16/EU on Administrative Co-operation is more comprehensive in its

scope than Article 26 of the OECD Model Tax Treaty. Consultations between
Member States are encouraged by the directive to establish certain criteria for
exchanging information under the directive itself and any bilateral agreements. It
can be used itself to assist third countries outside the EU, where certain conditions
are met.
The European Union’s Joint Transfer Pricing Forum (JTPF) produced a report in 2018
on a coordinated approach to transfer pricing controls. The report discusses Joint
Audits, Simultaneous Audits and Multilateral Controls. The report encourages
Member States to implement legislation that permits the active presence of visiting
foreign officials. In addition it encourages Member States to agree to a
Memorandum of Understanding (MoU – a template is included), in case they wish
to establish sustained coordinated transfer pricing controls programmes. The report
also includes generic guidelines and a set of best practices in the context of
coordinated transfer pricing controls for organisational matters, taxpayer’s rights
and obligations, and the various phases within a joint transfer pricing audit. For
more information see the Europa.eu website.
23.3 Corresponding Adjustments
The conclusion of transfer pricing audits may result in the agreement of a transfer
pricing adjustment arising from the application of Article 9(1) of the OECD Model
Double Tax Convention (DTC). (See OECD TPG paragraph 1.6.) Where a transfer
pricing adjustment has been made in one territory and there is no corresponding
adjustment in the second territory, prima facie there is double taxation.
 Illustration 1
Economic double taxation
A Ltd resident in State A sell goods to an associated company B Ltd in State B. A

Ltd shows a profit of 20 million on sales of 160 million to B. B Ltd has made profits of
35 million in State B. The authorities of State A adjust the profits of A Ltd by 2 million
as they have conducted a transfer pricing audit and determined that the selling
price to B Ltd was too low. This means that A Ltd will pay tax in State A on 22 million
of profit rather than just 20 million. Without further adjustment the total to be taxed
would be 22 million in A Ltd and 35 million in B Ltd giving a total of 57 million.
However the total profits are actually 20 million and 35 million giving a total of 55
million. This increase to 57 million rather than 55 million arises as the profits of B Ltd
include the price paid to A Ltd for the goods. It follows that if A Ltd had charged B
Ltd the higher price then B Ltd would have made less profit. To summarise:
Pre the transfer pricing Post the transfer pricing

adjustment adjustment
Profit in A Ltd 20 22
Profit in B Ltd 35 35
Total 55 57

To prevent the 2 million adjustment being taxed in both State A and State B, there
needs to be a corresponding adjustment.
Article 9(2) of the OECD Model DTC seeks to address this by providing that where
an adjustment has been made as envisaged under Article 9(1):
‘then that other State shall make an appropriate adjustment to the amount of
the tax charged therein on those profits. In determining the adjustment, due
regard shall be had to the other provisions in this Convention and the
competent authorities of the Contracting States shall if necessary consult with
each other.’ (OECD Model DTC Article 9(2))
So we can see that corresponding adjustments mitigate double taxation in cases

where one tax authority increases a company’s taxable profits (ie by making a
primary adjustment) as a result of applying the arm’s length principle to
transactions involving a related party in a second tax jurisdiction.
The corresponding adjustment in the illustration above would constitute a

downward adjustment (a decrease in profits) to the tax liability of the related party
(B Ltd), made by the tax authority of the second jurisdiction (State B), so that the
allocation of profits between the two jurisdictions is consistent with the primary
adjustment and double taxation is avoided.
The commentary to the OECD Model DTC notes that the adjustment is not
automatic and is therefore subject to the agreement of the other tax authority.
(See commentary on OECD Model DTC Article 9, paragraph 7.)
In practice, tax authorities consider requests for corresponding adjustments under

mutual agreement procedures (MAP) included in the relevant double tax
convention (DTC), in order to address the economic double taxation caused by a
transfer pricing adjustment.
DTCs mitigate the risk of double taxation by providing agreed rules for taxing
income and capital. They also provide guidance on effective tax dispute
resolution mechanisms to be applied in cases where the competent authorities of
the contracting territories to a transaction are in disagreement.
Article 9 does not specify the method by which a corresponding adjustment

should be made and therefore the method used is left to the discretion of the
relevant tax authorities.
OECD member countries use different methods to provide relief in cases where a
primary adjustment has resulted in double taxation. Tax authorities bilaterally agree
on what method is appropriate depending on the facts and circumstances of
each case. (See commentary on OECD Model DTC Article 9, paragraph 7.)
A corresponding adjustment can be made in two ways: by recalculating the

profits subject to tax of the party to the transaction in the second territory (ie,
making the corresponding adjustment in the tax return) or by granting the
associated party in the second territory tax relief against its own tax paid for the
additional tax arising from the primary adjustment. (See commentary on OECD
Model DTC Article 9 paragraph 7.)
Once a tax authority has agreed to make a corresponding adjustment, timing

needs to be considered as the corresponding adjustment may either be made in
the year during which the original transaction took place or an alternative year
such as the year in which the primary adjustment was determined. This issue is not

addressed by the OECD Model DTC. (See commentary on OECD Model DTC
Article 9, paragraph 10)
The former approach is generally preferred as it achieves a matching of income

and expenses and more accurately reflects the economic position as it would
have been if the controlled transaction had been at arm’s length. (See
commentary on OECD Model DTC Article 9, paragraph 10.) Timing issues may also
raise a question as to whether a party to the transaction is entitled to interest on
the portion of the overpaid tax.
Article 9 does not impose specific time limits within which corresponding
adjustments should be made and therefore the provisions of the tax treaty or
domestic laws of the relevant territory apply.
Relief under Article 9 may not be available if the time limit provided by the treaty
or domestic law for making corresponding adjustments has expired. (See
commentary on OECD Model DTC Article 9 paragraph 10.)
The multilateral instrument (MLI) (see below) encourages treaties to be amended

to allow a corresponding adjustment. Article 17 of the MLI enables states to
implement Article 9(2) of the OECD Model DTC, committing them to making the
‘appropriate adjustment’. States can choose not to adopt this, if their treaties
already contain the provision; or on the basis that they will make the
corresponding adjustment, notwithstanding the treaty terms; or they will
endeavour to resolve the issue under MAP.
23.4 Secondary Adjustments
Primary adjustments and their corresponding adjustments change the allocation

of taxable profits of MNEs for tax purposes as they result in the adjustment of tax
computations, where necessary, to reflect the position that would have existed
had the related party transaction taken place at arm’s length. Unless these
adjustments in the tax computations are matched by payments between the
affected parties, the economic circumstances of the parties will be distorted. This
distortion can have a significant and continuing impact on capital structure and
more generally on profit potential, and thus on future tax liabilities. To address this
problem some territories have introduced provisions into domestic legislation that
require secondary adjustments to be made.
Secondary adjustments attempt to account for the difference between the re-
determined taxable profits and the original profits. They will treat additional profits
resulting from primary adjustments as having been transferred in some other form
such as a constructive dividend, equity contribution or loan and tax them
accordingly. (See OECD TPG paragraph 4.68). Whilst they do not themselves
restore the financial situation of the parties to what it would have been had the
transaction which gave rise to the transfer pricing adjustment been made at arm’s
length, secondary adjustments can be administered to encourage the restoration
of funds to their proper place or, failing this, allow adjustment of the tax effects of
the distortion that might otherwise arise. (See OECD TPG paragraph 4.70).
The OECD Model DTC does not address the topic of secondary adjustments.
(Commentary on OECD Model DTC Article 9 paragraph 8). Secondary
adjustments are however, discussed in the OECD TPG although many countries do
not actually require or recognise them. (See OECD TPG paragraphs 4.68–4.78).
Where countries have introduced provisions into their domestic legislation that
require secondary adjustments to be made they are usually compulsory, although

tax authorities will generally allow taxpayers to prove the exact nature of the
transaction and in some cases to repatriate funds, for example by way of a
constructive dividend, in order to avoid the secondary adjustment.
The exact form that a secondary transaction takes and the consequence of the
secondary adjustment will depend on the facts of the case and on the tax laws of
the country that asserts the secondary adjustment. (See OECD TPG paragraph
4.68). This example, taken from the OECD TPG, illustrates the point:
Related parties located in Territory A and Territory B enter into a related party
transaction. As a result of the transaction being deemed not to occur at arm’s
length, Tax authority A makes a primary adjustment that results in an increase
in taxable profits in Territory A. Tax authority A then elects to make a
secondary adjustment that treats the additional profits as being a loan from
the related party in Territory B. In this case, an obligation to repay the loan
would be deemed to arise. The loan approach therefore affects not only the
year in which the secondary transaction is made but also a number of
subsequent years until such time as the loan is considered to be repaid.
(Paragraph 4.69).
Tax Authority A could alternatively treat the additional profits as being a dividend
in which case withholding tax may apply.
Secondary adjustments are not a common occurrence. A questionnaire was

circulated to all Member States by the European Union Joint Transfer Pricing Forum
in June 2011 to discuss secondary transfer pricing adjustments. Of the 27 Member
States that responded to the questionnaire, only nine had legislation in place that
allowed for secondary adjustments. In practice, in most European countries,
secondary adjustments are rarely enforced. Within the 18 Member States not
having such legislation, no Member State planned to introduce this legislation.
 Illustration 2
The following illustrates the application of transfer pricing adjustments between the
UK and a foreign territory, in this case, the US:
UK Ltd, a subsidiary of US Inc, purchased finished goods and services from US Inc
for resale into the European market but left the amounts due outstanding on inter-
company account.
The Internal Revenue Service (IRS) challenged the accumulation of the trading
balance and contended that part of the trading debt should be re-categorised as
long term funding debt on which an interest charge should be imputed. The
pricing basis for the underlying transactions was not challenged.
After lengthy negotiations, a settlement was reached and signed between US Inc
and the IRS.
Interest income was imputed by the IRS on a deemed loan for the four calendar
years 2000 to 2003. In 2006, US Inc raised an invoice to UK Ltd for this interest, which
was recorded as a profit and loss charge in UK Ltd’s statutory accounts and paid
in 2007.
HMRC initially refused UK Ltd’s deduction claimed for the interest on the basis that
there was no legal obligation for UK Ltd to pay interest to US Inc.

HMRC also stated that a taxpayer may not make a corresponding adjustment
unilaterally and the only mechanism by which to achieve deduction is through
MAP (we look at the MAP below).
A MAP application was made on the basis that a corresponding adjustment was
being claimed in the UK for a transfer pricing adjustment made in the US in
accordance with Article 9(2).
The application was successful and UK Ltd were granted a deduction for the
interest charged from US Inc. In this case, it is noted that as the interest was
actually charged and paid, no secondary adjustment would have arisen.
The process took around nine months to complete once the application was
made to HMRC. Much of this time was waiting for the competent authorities to
discuss the matter in the first instance.
23.5 Article 25 of the OECD Model Double Tax Convention (DTC) (MAP)
Introduction
The mutual agreement procedure (MAP) is a mechanism designed for dispute

resolution matters when dealing with double taxation. Where, as a result of the
actions of one or both fiscal authorities that is party to a double taxation
agreement, a taxpayer is suffering double taxation, most treaties contain a
mechanism for arbitration.
The taxpayer states its case to the competent authority of either state (2017 OECD
Model Treaty). If the competent authority is unable to resolve the matter
unilaterally, the competent authorities of both contracting states consult to
endeavour to resolve the matter by mutual agreement.
The taxpayer must present his case within three years from the first notification of
the action resulting in taxation not in accordance with the provisions of the treaty.
There is no requirement normally for the competent authorities to come to a

formal agreement as to a resolution; they ‘shall endeavour’ to, but may not
actually, do so. In addition here there is no time limit set for resolving the dispute.
This can be problematic as it means although the MAP process may have been
initiated, the issue under dispute may never actually be resolved.
This phrasing has thus long been problematic. However the 2017 update to the
OECD Model Treaty amended the commentary, and implemented so far as
possible the aims of the minimum standards set under the BEPS Project, at Action
Point 14, which are discussed further below. A new paragraph 5.1 notes that ‘shall
endeavour’ means that the competent authorities are obliged to seek to resolve
the case in a fair and objective matter, on its merits, in accordance with the terms
of the convention. It is perhaps hoped that this phrasing will push states to actually
come to a resolution with regard to matters placed before them.
Paragraph 3 also provides for the competent authorities to consult together for the
elimination of double taxation in cases not provided for in the convention.
Following the 2017 update there is an additional interaction between this

paragraph and Article 3(2) which allows the domestic meaning of a term to be
used where the treaty has no definition. Where the contracting states use Article

25(3) to come to a mutual agreement which includes the definition for a term,
then this agreed definition should take precedent over the domestic term.
Paragraph 4 of Article 25 provides a mechanism for the competent authorities to

communicate with each other.
MAP is covered by Article 25 of the OECD Model DTC. Paragraph 5 which covers
arbitration was added to Article 25 in the 2008 update to the OECD Model DTC
meaning that taxpayers can request arbitration if an outcome is not achieved as
a result of competent authorities consulting with each other. MAP arbitration,
together with the EU Arbitration Convention and the new EU Directive on Tax
Dispute Resolution, which are separate mechanisms for EU countries, is discussed in
further detail below.
The commentary to Article 25 indicates that MAP should be used to resolve

difficulties arising from the application of the Convention in the broadest sense
(2017 commentary to Article 25, paragraph 1). In practice it is used where double
taxation has arisen in areas for which it is the specific purpose of the Convention to
avoid double taxation (2017 commentary to Article 25, paragraph 9).
Common scenarios of double taxation where MAP is required include:
• Cases of transfer pricing adjustment (and no corresponding adjustment) due

to the inclusion of associated enterprise profits by the taxing authority in one
state, under paragraphs 1 and 2 of Article 9;
• Issues relating to attribution of profits to a PE, under paragraph 2 of Article 7;
• Excess interest or royalties under the provisions of Article 9, paragraph 6 of

Article 11 or paragraph 4 of Article 12;
• Situations regarding ‘thin capitalisation’ when the state of the debtor

company has treated interest as dividends, based on Article 9 or paragraph 6
of Article 11;
• Cases of misapplication of the Convention with regard to residency

(paragraph 2, Article 4), or the existence of a PE (Article 5).
The BEPS Action Plan contained a commitment (Action Point 14) to address
obstacles that prevented countries resolving treaty related disputes under Article
25. The final report on Action Point 14 is discussed below together with the MLI that
is being used to implement the outcomes of the final report.
Transfer Pricing Adjustments and MAP
Transfer pricing adjustments may give rise to economic double taxation (OECD
Model DTC, commentary to Article 9, paragraph 5). To eliminate double taxation
that may be caused by transfer pricing adjustments, Article 9(2) of the Convention
states:
‘..that other State shall make an appropriate adjustment to the amount of the
taxes charged therein on those profits.’
However, this adjustment is not automatic and only applies when the second state
agrees to the adjustment both in terms of the quantum and principle (OECD
Model DTC, commentary to Article 9, paragraph 6)

In order to give effect to this, Article 9(2) says that:
‘… the competent authorities of the Contracting States shall if necessary

consult each other.’
Such consultation will take place by way of MAP.
To prevent economic double taxation, the taxpayer may request the competent
authority of the first country to discard or decrease the transfer pricing adjustment.
Alternatively, the taxpayer may call for the competent authority of the second
country to enforce a corresponding adjustment. The UK competent authority for
example is open to receiving such representations from taxpayers although the
decision will of course be made between the competent authorities.
Article 7 of the Convention and the OECD Report on the Attribution of Profits to
Permanent Establishments, provides a mechanism for entering MAP where a
branch or permanent establishment exists, similar to that in Article 9 of the
Convention.
Administration of MAP
MAP can be entered into either at the request of the taxpayer or the request of
the tax authorities.
i. Taxpayer begins MAP proceedings
Dealings between the taxpayer and either state – this allows the taxpayer to
apply to the competent authority of either contracting state regardless of any
remedies available under domestic law.
The competent authority is obliged to consider if the taxpayer’s case is

justified, and if so, take appropriate action. If the complaint of double taxation
is wholly or partly due to measures taken in the taxpayer’s resident state, a
quick resolution may be provided by making the necessary adjustment or
allowing appropriate relief. Resorting to MAP in this instance may not be
necessary. However, the exchange of information and opinions with the
competent authority of the other contracting state may be useful for
interpretation purposes. If the competent authority of the resident state views
the objection to taxation to be in whole or part as a result of a measure taken
in the other state, it must commence proceedings for MAP.
ii. Tax authority begins MAP proceedings
Dealings between states – when the competent authority of the resident state
considers MAP to be the appropriate route in addressing the case, an
approach is made to the competent authority of the other state.
If we use the UK as an example, we see that the UK has no set form of presentation
for cases to be dealt with through MAP. Specific treaties may state certain
information that is necessary and the UK’s treaty partner may have domestic
guidance too. UK taxpayers should specify the relevant year(s), the point of
taxation not in accordance with the treaty and the full names/addresses that the
MAP claim relates to.
Note, the UK stipulates that a taxpayer cannot pursue domestic legal remedies
and MAP simultaneously. This approach is the same as that adopted by most
countries and is discussed in the commentary to Article 25 (OECD Model DTC,
commentary to Article 25, paragraph 76).

As noted above, the competent authorities of both states are obliged to use their
best endeavours to reach mutual agreement and resolve difficulties arising
through the interpretation or application of the Convention to eliminate double
taxation. However, Article 25 of the Convention does not guarantee relief from
double taxation. Nevertheless, for MAP cases involving transfer pricing
adjustments, this can be a very effective mechanism for eliminating double
taxation.
The MAP allows competent authorities to resolve, where possible, difficulties in the
application or interpretation of the Convention. MAP may also address more
complex situations of double taxation, such as cases of a resident of a third state
having a PE in both contracting states.
It is notable the UK encourages the use of Article 26 of the Convention, regarding

exchange of information, to assist competent authorities so they have all the
necessary facts during the decision making process. Competent authorities may
communicate directly, so it is not necessary to go through diplomatic channels.
For transfer pricing MAP matters, taxpayers may have the opportunity to present
the relevant facts to the competent authority orally as well as in a written format if
invited to do so.
Stages of MAP and Illustrative Time Frames
The OECD issued an illustrative process and timeline as part of its report issued in
2007 (OECD Report: Improving the Resolution of Tax Treaty Disputes). The key steps
and illustrative time frames for MAP included in this report are detailed below. The
illustrative timeframe is generally applicable to the majority of MAPs, which are
initiated by taxpayers (as opposed to a tax authority) except where specific treaty
clauses state otherwise. The BEPS best practices should also be noted going
forward.
Stage 1 – Notification and Acceptance into MAP Process
The taxpayer initiates MAP by submission of a MAP request. A three year time limit
is provided for by the Convention, or a time period may be outlined by domestic
provisions. The taxpayer then receives confirmation of the receipt of the MAP
request from its domestic competent authority and the MAP request is forwarded
to the other competent authority.
The taxpayer or associated enterprise in the other country is also encouraged to

contact their competent authority and provide all supporting materials to both
competent authorities promptly and simultaneously.
The case is reviewed by whichever competent authority is the adjusting

competent authority and subsequently there may be requests for the taxpayer to
provide additional information within a month after initiation of MAP by the
taxpayer. The adjusting competent authority determines the eligibility for MAP and
notifies the taxpayer if the case is accepted or rejected.
If the case is accepted, a proposal is made to the relieving competent authority to

commence MAP negotiations and an opening letter is issued.
Stage 2 – Negotiation
MAP is a process of consultation rather than litigation.
MAP consultations with the other state are initiated and a position paper is issued
by the adjusting competent authority. Ideally, this occurs within four months, but

no later than six months after agreement between the competent authorities to
enter into MAP consultations.
A review of the case is conducted by the relieving competent authority, which

comprises a preliminary screening for completeness of the position paper,
notification of missing information and determination whether it can provide
unilateral relief to the taxpayer. The relieving competent authority provides a
response to the position paper within six months of receiving it.
Negotiations then occur between the competent authorities. (Face to face

meeting(s) between the competent authorities can be organised in this stage, or
in any other stages when necessary).
The taxpayer is not a formal party to the consultation, however experience has
shown that it is advantageous for taxpayers to be involved at an early stage, in
particular where the case involves transfer pricing adjustments.
Stage 3 – Implementation
The mutual agreement between the competent authorities is documented in the

form of a ‘memorandum of understanding’ and should be submitted immediately
after conclusion of the mutual agreement.
The approval process for MAP includes a one month deadline for the taxpayer
and other ‘interested parties’ eg where the administrative-territorial subdivisions or
any local tax authorities’ consents are necessary or required to respond.
The mutual agreement is then confirmed with the relevant terms and conditions
and an exchange of closing letters occurs.
After acceptance of the mutual agreement by the taxpayer (and if relevant,

other parties), it is implemented, but no later than three months after the
exchange of closing letters.
How it is implemented will depend on the domestic rules of the adjusting state.
An efficient MAP process is largely dependent on the audit process being

effective. The time scales outlined assume that the dispute under audit which has
led to an initial adjustment, has an associated analysis which is well documented
and at an accepted standard (ie in line with the OECD TPG). Where this is not the
case, it will prolong MAP proceedings. The length of each stage of MAP will vary
depending on the nature of the case. Where translation is required, this may also
add to the length of time the process takes.
The time limits applicable will depend on the specific terms of the treaty.
The taxpayer does not play a formal part in the consultation and negotiation
process of MAP. However, as a major stakeholder, the taxpayer should always be
kept informed of important milestones in its case. In addition, the taxpayer has the
right to decide to accept (or decline) the agreement between the competent
authorities. MAP cases, especially transfer pricing cases, are often complex and
highly fact specific. Therefore it may be useful for the taxpayer to present facts
and related questions in person. So the taxpayer may be asked to informally
participate in the MAP process at the discretion of the competent authorities,
despite not being part of the consultations.
MAP is not an alternative to the usual transfer pricing enquiry process. A transfer
pricing enquiry seeks to determine an arm’s length level of profits for the

entity/branch in question relating to the inter-company transaction(s) at hand.

MAP establishes how the double taxation of these profits will be relieved in
principle by the treaty partners.
It is advantageous for taxpayers to present cases early to invoke MAP, especially

for transfer pricing enquiries. For instance, if a UK treaty partner is applying an
unsuitable transfer pricing methodology during the course of an audit, the UK tax
authority may be able to help demonstrate that an alternative method is more
appropriate.
Of course there have been some instances where the MAP procedure has not
produced a solution, a key one being the Glaxo case in the US. The disagreement
between Glaxo and the IRS had been ongoing for many years (since the late
1980s). The IRS position was to challenge the transfer pricing used by Glaxo UK to
remunerate (or better ‘under remunerate’) the US subsidiaries for marketing drugs
developed in the UK.
Once Glaxo realised that the IRS would enforce the adjustment on the US profits,
the enterprise invoked its right under the DTA to force the tax authorities in the UK
and US to enter into negotiations under MAP and to arrive to a common
agreement as to the arm’s length transfer pricing.
Glaxo had possibly wished for the UK tax authorities to help convince the IRS that
they were being unreasonable and should perhaps reconsider the size of
adjustment.
The worst case scenario envisioned by Glaxo was probably that in case the IRS
had convinced the UK tax authorities that the adjustment was legitimate, on the
basis that the transfer pricing was wrong there should be a corresponding
adjustment to reduce UK taxable profits by the same amount.
However, as per previous paragraphs, the MAP does not force the two competent
authorities to come to an agreement. The competent authorities are asked to
‘endeavour’ to resolve ‘any difficulties or doubts arising as to the interpretation or
application of’ the DTA. Unluckily for Glaxo, the tax authorities did not reach an
agreement.
There are various concerns with regard to how the MAP operates, notwithstanding
the work undertaken as part of the BEPS Project. For example, the MAP itself is
complicated in application and time consuming, and, as with the previous case,
unless a state has signed up to arbitration (see below) there is no guarantee of a
successful outcome.
The taxpayer is generally not a party to the negotiations which is perhaps not
ideal, although states would argue that this allows for full and frank discussions
between the two states with regard to the matter in hand. In addition, the very
broad ranging rules and practices of different countries may well create issues
which hinder the operation of the MAP; for example, countries may have different
approaches to the start date for the three year time limit for commencing MAP
which could cause issues.
A problem which is highlighted in the OECD Model Commentary with regard to

the MAP relates to the interaction of the MAP with domestic legal remedies
available in a state. Whilst the Commentary does put forward suggestions as to
how this interaction might proceed (see paragraphs 25 to 27), there is no
consistent approach between states, and thus in some countries a taxpayer may
well be able to litigate a matter further if they did not agree with the MAP
outcome, while in others this may not be the case.

Nevertheless, it is clear the MAP has become an increasingly important tool for
taxpayers and tax authorities alike in addressing double taxation, as it allows for
competent authorities to consult with each other on the application of double
taxation treaties. A collaborative global environment has allowed the process to
become more efficient, where authorities exchange information to reach an
appropriate outcome.
The EU Arbitration Convention and the new (2017) Directive on Tax Dispute
Resolution Mechanisms also provide alternatives to MAP and tax treaty arbitration,
in dealing with the elimination of double taxation related to adjustments of
associated enterprise profits (the former), and more generally (the latter), and
these are discussed further in the sections below.
23.6 The Tax Treaty Arbitration Procedure
Article 25(5) was added in the 2008 version of the OECD Model Treaty. It makes
provision for arbitration to be used if agreement has not been reached within 2
years of the date when all information needed has been provided to both
competent authorities (per the 2017 update). Prior to 2017 this paragraph referred
to 2 years of the presentation of the case to the other competent authority. The
procedure is not based on prior authorisation by the competent authorities. This
article was updated as part of the BEPS process – see Action Point 14 below.
Arbitration under Paragraph 5 is only available for cases brought under Paragraph
1 of Article 25 – it does not extend to those brought under Paragraph 3.
A request for arbitration must, per the 2017 update to the OECD Model Treaty, be
made in writing. It will not be granted where a decision has already been
rendered by a court or administrative tribunal of either state. Arbitration can only
be requested where there are unresolved issues.
The annex to the commentary on Article 25 includes a sample form of agreement

that the competent authorities may use as a basis of mutual agreement to
implement the arbitration process. The 2017 update to the OECD Model Treaty
amended this agreement to ensure it is consistent with the various changes made
to Article 25(5) and to the commentary, and with BEPS Action Point 14.
The sample agreement includes details of the application of the arbitration

process including the timetable that should be followed. The time line included in
this sample is illustrated in the table below. It is noted that this is only an example
and the competent authorities can amend this when they conclude their bilateral
agreement. (See commentary on 2017 OECD Model Article 25 Annex paragraph
1.)
The arbitration process addresses the issues in respect of which competent

authorities were unable to reach a decision through MAP, rather than focusing on
the overall case itself. The objective of the arbitration process is to enable a
decision to be made on the overall case based on the arbitrator’s choice of
proposed resolution, or on their own determination of the unresolved issues.

An Example of the Arbitration Process Timeline
The above time line is taken from the commentary on 2017 OECD Model Treaty
Annex – Sample Mutual Agreement on Arbitration. See also below for Arbitration
under the MLI changes, on which the 2017 update was based.
23.7 BEPS Action Point 14
The aim of Action Point 14 was to ‘Develop solutions to address obstacles that
prevent countries from solving treaty-related disputes under MAP, including the
absence of arbitration provisions in most treaties and the fact that access to MAP
and arbitration may be denied in certain cases’.
The final report on Action Point 14 was published in October 2015 and contains
measures aimed at strengthening the effectiveness and efficiency of the MAP,
such as specific actions to be taken by countries, suggested changes to legislation
and administrative practices, and changes to the OECD Model Treaty and its
Commentary. The measures are being implemented by the MLI that came into
force on 1 July 2018. We look at this in more detail below.
The final report sets out three core proposals, the first of which is to be adopted by
all countries involved. This requires countries to commit to minimum standards on
the resolution of international tax disputes. Agreement is also required to a peer-
based monitoring mechanism to ensure the countries are meeting the new
standards, which is being executed through the Forum on Tax Administration’s
MAP Forum.
The minimum standards have three main objectives:
1. To ensure that countries implement Article 25 of the OECD Model Tax Treaty in
good faith. This means that states should ensure all their treaties have at least
paragraphs 1 to 3 of the OECD Model Treaty Article 25, and ensure the MAP is
available for transfer pricing disputes and anti-treaty abuse provisions.
2. To ensure that domestic administrative procedures don’t block access to the

MAP process and promote timely resolution of treaty-related disputes. The
requirements in this regard are not so clear, and include the provision of
‘adequate resources’ to the MAP process along with the publication of clear
guidance.

3. To allow taxpayers access to the MAP process when the requirements at

paragraph 1 of Article 25 for taxpayers to access the MAP process are met.
This deals with the issues surrounding the ability of the state of residence of the
taxpayer to decide whether the claim is justified. The standard requires
contracting states to either use an amended version of Article 25 such that the
taxpayer can choose which state to approach with their claim or to
implement a bilateral notification or consultation process when the home
state does not consider the request to be justified.
The second core proposal relates to ‘best practices’; the best practices
complement the agreed minimum standards. However they are voluntary and it
should be noted that only some of the OECD BEPS/G20 countries were willing to
commit to them. The final report contains 11 best practices. We have set out
below some examples of the best practices. We recommend that you read the
final report.
Best Practice 1: Countries should include paragraph 2 of Article 9 in their tax

treaties. This is required as some countries are of the opinion that without
paragraph 2, they cannot make a corresponding adjustment.
Best Practice 4: Countries should implement bilateral APA programs. Bilateral

agreements present an increased level of certainty for both governments and
taxpayers and can prevent transfer pricing disputes (as noted in the next chapter).
Best Practice 8: A country’s published MAP guidance should include an

explanation of the relationship between MAP and domestic law administrative
and judicial remedies.
Best Practice 11: Countries’ MAP guidance should provide guidance on

multilateral MAPs and APAs.
The third core proposal relates to the need for a compulsory, binding arbitration
procedure, and identified 20 countries (at that time), including the UK, the US,
Australia, Canada, France and Germany, that had agreed to implement
mandatory binding MAP arbitration in their bilateral tax treaties. According to the
OECD, the 20 countries that made this commitment were involved in more than
90% of the outstanding MAP cases at the end of 2013. Therefore, this is considered
a significant move forward and over time it is likely more countries will follow suit.
Peer reviews of the implementation of this minimum standard are an ongoing

process, being undertaken by the OECD through the MAP Forum set up for the
task. To date the various reviews (as of mid 2021 over 80 initial peer reviews have
been completed) have shown that countries are committed to improving the
dispute resolution process.
The 2016 US Model Treaty includes such a mandatory binding arbitration process
where agreement under MAP is not reached within two years. The treaty provides
for a three-person arbitration panel, with two of the arbitrators chosen by the
countries involved, and the third selected by the other two. The provisions apply a
‘last best offer’ approach in which each country submits a position paper with a
proposed resolution and the arbitration panel chooses between them.
A typical issue raised by MNEs is the amount of time and resources used in
resolving international tax disputes. Thus even the threat of a mandatory binding
arbitration process, even if it is not ultimately invoked, will provide an incentive for
the contracting states involved to look in good faith for a resolution.

In September 2020 the OECD published statistics on the MAP; these statistics
looked at the position at the end of 2019. The key points highlighted by the OECD
include:
• Transfer pricing cases take more time on average than other cases:
approximately 30 months are needed for transfer pricing cases and 22 months
for other cases.
• Over 70% of transfer pricing MAPs concluded in 2018 resolved the issue. Over
50% of transfer pricing MAP cases closed were resolved with an agreement
fully resolving the taxation not in accordance with the tax treaty and 15% of
them were granted a unilateral relief while almost 5% were resolved via
domestic remedy. Finally, only 2% were closed without an agreement being
reached.
The following diagrams are taken from the report which can be found at
http://www.oecd.org/tax/dispute/mutual-agreement-procedure-statistics.htm.
An Example of the MAP statistics
At the end of 2020 the OECD issued a consultation seeking the views of
participating states with regard to the impact and application of BEPS Action Point
14. Specifically the OECD is looking for: experiences with, and views on, the status
of dispute resolution along with suggestions for improvements, including

experiences with MAP in those jurisdictions that obtained a deferral; views on how
to strengthen the Action Point 14 minimum standard and how to strengthen the
MAP statistics reporting framework.
Alternative to MAP for CbC reporting – ICAP
As we saw in an earlier chapter, Action Point 13 of the BEPS project contained a

minimum standard that required countries to introduce Country-by-Country (CbC)
reporting rules for certain large MNEs.
The International Compliance Assurance Programme (ICAP) is a voluntary

programme for multilateral cooperative risk assessment and assurance providing
increased and earlier tax certainty for MNE groups and giving tax authorities
assurance that any tax risks have been identified. As noted in the Tax Journal (Tax
and the City Briefing, March 2021) the ICAP process enables participating MNE
groups to meet the relevant tax administrations to talk through their country-by-
country reports (CbCR) and other documentation and provide clarity to aid
understanding of their cross-border activities. This should help tax administrations to
reach an early decision about the level of transfer pricing risk, PE risk and other
specific international tax risks, if any, presented by the data contained in the
CbCR and which the relevant tax authorities agree to include. More importantly it
should reduce the possibility of disputes later arising and thus the need for a MAP.
The first ICAP pilot was launched in 2018 with eight participating tax administrations
(including the UK). A second pilot followed in 2019 with 19 participating tax
administrations. In February 2021, the OECD published the ICAP handbook for tax
administrations and MNE groups which contains information on the process for
ICAP reflecting the experience and feedback of tax administrations and MNE
groups that participated in the two pilots. The handbook includes a detailed
description of each stage of the ICAP process, the documentation and
information an MNE group participating in ICAP will provide, the level of comfort
they may achieve as a result of participation in the programme, and a
comparison of ICAP with other possible routes to greater tax certainty (such as
advance pricing agreements).
The focus of the pilots was initially on transfer pricing and PE risk but the
programme now includes any other material international issues agreed between
the MNE group and participating tax administrations such as hybrid mismatch
arrangements, withholding taxes and the application of relevant international
treaties. The PE risks within scope are both the risk of a PE not being recognised for
tax by the jurisdiction in which it is located and the risk that profits have not been
allocated correctly to the PE.
The timeline for ICAP will depend upon a number of factors, but typically an MNE
group’s ICAP risk assessment will be completed and outcome letters issued within
24-28 weeks following delivery of the main documentation package, at the start of
the risk assessment and issue resolution stage. It is hoped that the process will
reduce the amount of disputes that have to go into the MAP and it will certainly
be a quicker process than MAP.

23.8 Dispute Resolution in the Multilateral Instrument (MLI)
As already noted the MLI implements BEPS recommendations in several areas,

including dispute resolution. As explained above, Action Point 14 of the BEPS report
requires, as a minimum standard, the inclusion in tax treaties of a version of the
MAP in Article 25(1)–(3) of the OECD Model Treaty, modified to allow the taxpayer
to present its case to either state. The previous version of Article 25(1) required the
case to be presented in the state where the taxpayer is resident.
Article 16 of the MLI provides that the modified version of MAP will apply to treaties
in place of, or in the absence of, MAP provisions. States can make limited
reservations. These include not allowing claims to be made to either state, but
implementing a process whereby the state receiving the taxpayer’s notification
will notify or consult with the other state, if it does not consider the taxpayer’s case
to be justified.
We looked at Article 17 on corresponding adjustments above.
Article 19 of the MLI sets out a mechanism for mandatory binding arbitration (MBA)
where competent authorities are unable to reach agreement under MAP. It is
optional and will only apply to treaties where both states choose to apply it. States
can choose to allow three years (rather than two years) for agreement to be
reached under MAP before MBA is possible; and can choose to exclude issues
which have already been decided by a court.
States can reserve the right for the MBA provisions in the MLI not to apply to
treaties which already provide for MBA. The UK for example will reserve the right to
maintain any existing stronger MBA provisions. States can also reserve the right for
the MBA provisions not to apply to cases entering MAP before the MLI comes into
force for a given treaty, unless the parties agree to MBA applying to that specific
case.
As we have noted, when the report on Action Point 14 was published in October
2015, 20 states were committed to MBA, being Australia, Austria, Belgium, Canada,
France, Germany, Ireland, Italy, Japan, Luxembourg, the Netherlands, New
Zealand, Norway, Poland, Slovenia, Spain, Sweden, Switzerland, the UK and the US.
When the MLI was signed in June 2017 the US (as they didn’t sign the MLI), Poland
and Norway did not commit. However Andorra, Fiji, Finland, Greece, Lichtenstein,
Malta, Portugal and Singapore added their commitment making the total 25 at
that time.
The arbitration panel will consist of three independent and impartial members with
expertise or experience in international tax. Each competent authority will appoint
one member and then those two members will appoint a third member to chair
the panel, who cannot be a national or resident of either state. There are
provisions for an OECD official to appoint panel members, if states or the panel
members do not make the necessary appointments.
States can choose which type of arbitration they want to adopt for their ‘covered
agreements’. Either:
• ‘final offer’ or last best offer arbitration (‘baseball arbitration’): this is the
default option, under which each competent authority will submit to the
arbitration panel its proposed resolution; and (if it chooses) a supporting paper
and a response to the other state’s proposed resolution. The panel will decide,
by simple majority, which resolution to accept. It will not give any reasons for its
decision and the decision will have no precedent value; or

• ‘independent opinion’ arbitration: under this option, each competent

authority provides the panel with ‘any information that may be necessary for
the arbitration decision’. The panel decides the issues and provides a
reasoned decision. Again, the decision is by simple majority and it does not set
a precedent.
If one party to the treaty has chosen independent opinion arbitration and the
other has chosen final offer, independent opinion will apply to the treaty, unless
the state which has chosen final offer has chosen not to apply MBA to treaties with
states that have chosen independent opinion arbitration. In that case, MBA will not
then apply unless the parties reach agreement on the type of arbitration process
that will apply to that treaty.
The arbitration decision will be binding on both states unless the taxpayer does not
accept the decision. It will also not be binding if a court holds that the decision is
invalid or if a person directly affected by the decision pursues litigation in relation
to the issues.
As noted above, the Sample Agreement in The Annex to the commentary to

Article 25 in the OECD Model Treaty follows these lines.
The states concerned will agree how costs will be borne. In the absence of
agreement, each state will bear its own costs and those of the panel member
they appointed, and the cost of the chair and any other expenses will be shared
equally.
There is also an optional provision for the arbitration decision not to be binding and
not implemented if the states concerned agree on a different resolution of all
issues within three months of the arbitration decision. This will only apply to a
particular treaty if both states have made the reservation. The reservation can be
made in respect of all treaties or only those where independent opinion arbitration
applies.
23.9 EU Arbitration Convention
The EU Arbitration Convention, the convention on the elimination of double

taxation in connection with the adjustment of profits of associated enterprises
(90/436/EEC), was introduced in 1995 as a mechanism by which double taxation
arising from transfer pricing adjustments for transactions between two EU Member
States would be eliminated. It should apply to all transactions although some
jurisdictions, such as Bulgaria and Italy, do not accept that it covers financial
transactions.
An application for relief under the EU Arbitration Convention should be made

within three years of the notification of the adjustment that is likely to lead to
double taxation. (See (90/436/EEC), Article 6 paragraph 1).
Disputes settled under the EU Arbitration Convention should reach their conclusion
within a three-year timescale from the commencement of proceedings. (See
Revised Code of Conduct for the effective implementation of the Convention on
the elimination of double taxation in connection with the adjustment of profits of
associated enterprises (Revised Code of Conduct), paragraph 4.).
Article 4 of the EU Arbitration Convention includes similar wording to Article 9 of the

OECD Model Treaty. The conclusion reached under the EU Arbitration Convention
should therefore be in accordance with the arm’s length principle as detailed in
the OECD TPG.

The procedure is a two stage process: firstly, following a request by a taxpayer, the
tax authorities should negotiate under a mutual agreement procedure (similar to
MAP) to agree a resolution to the double taxation and then a second, arbitration
phase, if resolution is not reached, whereby the tax authorities consult
independent experts to make a binding decision.
Phase 1 – Mutual Agreement
When an application for relief under the EU Arbitration Convention is made, the
onus is on the taxpayer to initiate the application; this may be done at the same
time as a MAP application under a DTT.
There is no set form of presentation of the application. However, the code of

conduct has a list of information that should be provided with the request. See
Revised Code of Conduct, paragraph 5 which includes:
• Identification of the taxpayer that has suffered double taxation;
• Identification of the other parties to the transactions;
• Details of the facts of the case;
• Details of the periods for which the transaction(s) cover; and
• Copies of the tax assessment notices and details of any appeals, the result of
which have led to double taxation.
The Revised Code of Conduct states that the competent authority should respond
within one month.
If the competent authority believes that the enterprise has not submitted the
minimum information necessary for the initiation of a mutual agreement
procedure as stated under point 5(a), it will invite the enterprise, within two months
of receipt of the request, to provide it with the specific additional information it
needs.
Member States undertake that the competent authority will respond to the
enterprise making the request in one of the following forms:
i. if the competent authority does not believe that profits of the enterprise are
included, or are likely to be included, in the profits of an enterprise of another
Member State, it will inform the enterprise of its doubts and invite it to make
any further comments;
ii. if the request appears to the competent authority to be well-founded and it

can itself arrive at a satisfactory solution, it will inform the enterprise
accordingly and make as quickly as possible such adjustments or allow such
reliefs as are justified;
iii. if the request appears to the competent authority to be well-founded but it is

not itself able to arrive at a satisfactory solution, it will inform the enterprise that
it will endeavour to resolve the case by mutual agreement with the
competent authority of any other Member State concerned.
If a competent authority considers a case to be well-founded, it should initiate a

mutual agreement procedure by informing the competent authority(ies) of the
other Member State(s) of its decision and attach a copy of the information as
specified under point 5(a) of the Revised Code of Conduct. At the same time it will

inform the person invoking the EU Arbitration Convention that it has initiated the
mutual agreement procedure.
The competent authority initiating the mutual agreement procedure will also
inform – on the basis of information available to it – the competent authority(ies) of
the other Member State(s) and the person making the request whether the case
was presented within the time limits provided for in Article 6(1) of the EU Arbitration
Convention and of the starting point for the two-year period of Article 7(1) of the
EU Arbitration Convention. (See paragraph 6.3 f to g of the Revised Code of
Conduct.)
Where the competent authority agrees that an enterprise is suffering double

taxation, the competent authority is responsible for initiating negotiations with the
other tax authorities to agree a resolution.
This resolution may be achieved by any method deemed suitable, such as

telephone calls or meetings between the two tax authorities. (See Revised Code
of Conduct, paragraph 6.1 (c)).
The taxpayer is not a formal party to the negotiation although in practice they
may have some involvement such as presenting their case to the competent
authorities. The taxpayer should also be kept informed of any developments to the
application for relief of double taxation as the case progresses. (See Revised Code
of Conduct, paragraph 6.3 (b)).
If the authorities are able to agree a suitable resolution, then the tax authorities are
advised in the Revised Code of Conduct to inform each other appropriately and
make such adjustments or allow such relief as quickly as possible.
The outcome of the negotiations should be communicated to the taxpayer.
The taxpayer should be part of the negotiation at this stage and is permitted to
reject an agreement that has been reached between the two tax authorities if it
believes that the outcome is not consistent with the arm’s length principle. (See
Revised Code of Conduct, paragraph 6.3 (b).)
Phase 2 – Advisory Commission
The second phase of the EU Arbitration Convention is the setting up of an Advisory

Commission. Although there have been a number of cases under the mutual
agreement phase of the EU Arbitration Convention, it is rare for this second phase
(arbitration) to be used.
If the tax authorities are unable to agree a resolution within a two-year period from
the application date, then the tax authority that initiated the proceedings is
responsible for setting up an Advisory Commission to arbitrate on the matter.
The Advisory Commission should be provided with all of the information necessary
to make their judgement and make their decision within six months of being
established. Within the next six months, the tax authorities may agree an
alternative decision which differs to that of the advisory commission. If no
alternative is agreed, then the tax authorities must act in accordance with the
decision of the Advisory Commission. (See Convention on the elimination of
double taxation in connection with the adjustment of profits of associated
enterprises (90/436/EEC), Article 12.)
Once the decision has been made, the decision should be communicated to the
taxpayer and it may be published if the taxpayer agrees.

23.10 EU Directive on Tax Dispute Resolution Mechanisms
In October 2017 the EU Council formally adopted the Directive on Tax Dispute
Resolution Mechanisms (EU 2017/1852), which came out of the work of the EU as
part of its Corporate Tax Action Plan, and the work undertaken in the BEPS Project.
This directive operates alongside the Arbitration Convention (discussed above),
and provides a wider scope for the use of arbitration for intra EU cross-border
double tax disputes.
With the processes in the MLI (discussed above) and this new directive, the way
tax disputes are resolved is likely to change significantly in the near future. The
directive will apply to disputes relating to tax year commencing 1 January 2018, for
cases submitted from 1 July 2019, however states may agree to apply the
procedures to disputes from earlier tax years.
Whilst the Arbitration Convention is limited to transfer pricing disputes, this directive
is wider in scope and applies to disputes arising from the interpretation and
application of agreements and conventions entered into for the elimination of
double taxation on income and capital. The directive encourages arbitration as a
way to resolve such disputes, after a MAP procedure, however it also allows
Member States the option of choosing other methods, under its Alternative Dispute
Resolution procedure.
Notably if a dispute is covered by both the arbitration provisions in the MLI and the
directive, the directive will take precedence where the dispute has been brought
to the attention of the Member States under the provisions of the directive.
However, as an area prone to disputes, it remains likely that future arbitration
cases will still be transfer pricing disputes.
The basis of the mutual agreement concept included in the directive is the
procedure found in Article 25 of the OECD Model Treaty, but with some further
refinements. Any person who is a resident of an EU Member State for tax purposes
and whose taxation is directly affected by a matter giving rise to a dispute may,
within three years from the receipt of the first notification of the action resulting in
or that will result in the dispute, submit a complaint simultaneously to each of the
concerned EU competent authorities. If the affected person is an individual or
‘small’ undertaking then the complaint only has to be submitted to their state of
residence, and that competent authority will then inform the other states involved
within 2 months.
Per Article 3, a complaint under the directive must be accompanied by detailed

information and supporting documentation relating to the dispute as specified in
the directive (which is one area where the directive differs from the treaty MAP).
The information must be in the language of each receiving state or any other
language they may accept. The competent authorities then have three months to
request further information (to be supplied within three further months). It should be
indicated within six months whether they have accepted the complaint. There are
limited grounds for rejecting the request and the absence of a decision on this is
deemed acceptance.
Within the same six month period (from having received all the necessary
documents), any of the competent authorities involved can decide to resolve the
dispute on a unilateral basis. Otherwise, under Article 4 the Member States
concerned shall endeavour to solve the dispute by way of MAP within a period of
two years. If the Member States reach an agreement under this process within this
time, then this is binding on the competent authorities and enforceable by the
taxpayer subject to his acceptance of the agreement. If the Member States fail to

reach an agreement the competent authorities will inform the taxpayer of the
reasons for their failure.
Where the MAP fails, or where the MAP was not commenced as the complaint
was rejected by one (but not all) of the competent authorities concerned (and
there are no rights of appeal in the domestic courts), the taxpayer may request
the competent authorities of the Member States set up an Advisory Commission.
Such a Commission consists of three to five independent arbitrators and up to two
representatives of each Member State. If the Member States do not set up such a
Commission within the relevant timescale then the taxpayer may apply to their
national courts to do so – which is a significant benefit as it allows the processes to
move on and not be ‘locked’ by a Member State not complying. Although it is not
entirely clear in the wording of the directive, this Advisory Commission will proceed
to give a reasoned opinion on the dispute (there is reference to this approach in
Article 10).
However, the directive does provide Member States with the flexibility to set up an
Alternative Dispute Resolution Commission instead of the Advisory Commission
(Article 10). This Alternative Dispute Resolution Commission may take on a different
composition and form, as agreed between the parties, and can consider any
dispute resolution process or technique to solve the question in dispute in a binding
manner. It is interesting that Article 10 does indicate that as an alternative to the
independent opinion process (used by the Advisory Commission), the Alternative
Dispute Resolution Commission could use the ‘final offer’ arbitration process
(otherwise known as ‘last best offer’ arbitration) – which notably is the default
option for arbitration under the MLI, as discussed above. Arguably this latter
approach is likely to offer a quicker and more cost-effective method of dispute
resolution.
The taxpayer does have more extensive rights under the directive, as they can
provide the deciding panel with information, evidence or documents that may be
relevant for the decision. In addition, taxpayers may also appear (or be
represented) before the Commission with the consent of the competent
authorities.
The Advisory Commission, and the Alternative Dispute Resolution Commission, must
deliver their decisions within six months, with a possibility to extend by a further
three months (see Article 14). Decisions are by majority vote, with a casting vote to
the chair as required. The competent authorities are bound by the decision unless
they agree a different resolution within six months of notification of the decision
(Article 15). The taxpayer must agree to the decision in order to be bound by it. If a
Member State fails to implement the decision then the taxpayer is able to enforce
implementation of the decision by resorting to national courts in that Member
State (Article 15(4)).
If the competent authorities agree, and the taxpayer consents, then the final
decisions may be published in full, excluding any information on trade, business,
industrial or professional secrets or processes. Where full publication is not agreed
then the competent authorities must publish an abstract of the decision and
details of the arbitration method used.
The set time limits for the different processes, of MAP and arbitration, should mean
proceedings will actually be resolved, and within shorter time-frames. This fact,
combined with improved access to the process and more involvement of the
taxpayer, does indicate that this directive should be a great improvement on the
existing rules.

23.11 Overview of Article 25 of the UN Model Tax Treaty
The UN Model Tax Treaty also contains MAP procedures in Article 25. The UN Article
25 contains Alternative A and Alternative B. Paragraph 1 of both alternatives
differs from the OECD version as it only allows a case to be brought to the
competent authority where the taxpayer is resident unless it is a claim under Article
24. Alternative B includes a provision for mandatory binding arbitration although
there are differences to the OECD version:
I. Arbitration must be requested by the competent authority of one of the

contracting states, not the taxpayer;
II. The arbitration procedure may be initiated if the competent authorities are
unable to reach an agreement on a case within three years from the
presentation of the case rather than within two years;
III. The competent authorities can depart from the arbitration decision if they
agree on a different solution within six months after the decision has been
communicated to them.
The commentary to Article 25 suggests that states that do not wish to include
Alternative B may wish to include provision for voluntary arbitration.

CHAPTER 24
AVOIDING DOUBLE TAXATION AND DISPUTE RESOLUTION II
In this chapter we look at further aspects of avoiding double taxation and dispute
resolution in particular:
– Advance Pricing Arrangements (APAs)
– international perspective and trends
– APAs and the BEPS Action Plan
24.1 Introduction
An advance pricing arrangement (APA) is an administrative approach to avoiding

transfer pricing disputes arising by agreeing in advance the criteria for applying
the arm's length principle to transactions. It is a procedural arrangement between
parties that differs from classic tax ruling procedures as it is more fact specific. (See
OECD TPG Annex to Chapter IV paragraph 3).
Agreeing an APA allows taxpayers to gain certainty over their tax affairs from a
transfer pricing perspective.
The OECD TPG include commentary on APAs in Chapter IV, together with an
annex to Chapter IV, which details ‘Guidelines for conducting APAs under MAP’
(MAP APAs). The process gives tax authorities and taxpayers the opportunity to
consult over transfer pricing issues in a less adversarial way than may be the case
as part of an enquiry or during litigation. (See OECD TPG paragraph 4.143).
24.2 What is an APA?
An APA is defined in the OECD TPG as an agreement between a taxpayer, one or

more associated enterprises and one or more tax administrations, to determine in
advance an appropriate set of criteria that satisfies all parties, and can be used to
determine arm's length transfer pricing for the transactions covered by the APA
over the duration of the agreement. (See paragraph 4.123.)
An APA can be unilateral, bilateral or multilateral. Following Action Point 14

mentioned in an earlier chapter, the OECD TPG now state that best practice is a
bilateral APA. (See paragraph 4.134.)
A unilateral agreement is made between the taxpayer and their respective tax
administration. As unilateral APAs only deal with tax issues within one jurisdiction
there is still a risk that double taxation can occur (as the counter-party tax
authority may take a different stance on the matter).
When a taxpayer makes an application for a unilateral APA it is recommended

that, where a suitable treaty is in place, the tax authority informs the competent
authority of the other territory and invites them to participate in a bilateral APA.
(See paragraph 4.140.)
Some tax authorities may still agree unilateral APAs in particular circumstances, for
example where the amounts at stake are small so there is very little to gain with a
bilateral agreement and/or the majority of the transfer pricing risk lies in the
taxpayer's home country or where the other party to the transaction is resident

within a jurisdiction with which there is no treaty or the treaty partner has no formal
APA process.
There may be good reasons why unilateral APAs may be preferred not least, as
they only involve one tax authority, they tend to be easier to agree.
In bilateral or multilateral arrangements, two or more countries participate.
A multilateral APA is simply a series of complementary bilateral APAs with each of

the countries using the bilateral APA procedure.
Bilateral or multilateral APAs are often referred to as MAP APAs and will be agreed
under mutual agreement procedures with the other tax authority.
MAP is covered in Article 25(3) of the OECD Model DTC, although this article does
not expressly mention APAs. It is however considered within the OECD TPG that this
article covers MAP APAs, as the specific transfer pricing cases subject to an APA
are not otherwise provided for in the OECD Model DTC. (See paragraph 4.150.)
Article 25 does not oblige a competent authority to enter MAP and the willingness
may depend on the policy of the two countries involved. (See Annex to Chapter
IV paragraph 16.)
There is flexibility over which transactions can be the subject of an APA, although
some tax authorities prefer all issues to be covered. (See paragraph 4.147.) The
criteria that is to be agreed can include the transfer pricing method to be used to
demonstrate arm’s length pricing, the comparables and/or method of selection
and appropriate adjustments to the comparables and the critical assumptions in
relation to future events. (See paragraph 4.135.)
It is difficult for APAs to go beyond agreeing methodology and for example agree
fixed results as these would rely potentially on forecasts and budgets. For example,
it may not be reasonable in the case of a financing transaction to agree a fixed
interest rate but it may be reasonable to agree a percentage point pegged to an
index. (See paragraph 4.136.)
There are four separate steps to the APA process, which are: expression of interest
or preliminary discussions, formal submission of an application, evaluation and
agreement.
The above diagram illustrates the APA process.

Step 1 – Initial Package Submission
Prior to the expression of interest or pre-filing meeting stage some tax authorities
may require certain documents, pertaining to the APA, to be submitted for
consideration.
Step 2 – Expression of Interest
This is a feature of many domestic APA processes, which can assist in dealing with
the actual APA application more quickly. (See OECD TPG Annex II to Chapter IV
paragraph 29.) The expression of interest stage makes the process more efficient
and limits waste of resources by determining whether the APA process will be
beneficial for the parties. For MAP APAs this will also allow the relevant competent
authorities to have preliminary discussions, which will also help clarify expectations
from the process.
Step 3 – Formal Application
If informal approval is received by an enterprise to enter the APA process, a formal

written application should be submitted. This will usually be in the form required by
domestic procedure.
The same information should be provided regardless of whether it is a bilateral or

multilateral APA application. The actual contents will also depend on the facts
and circumstances but should include all information necessary for the tax
authorities to consider the application. (See Annex to Chapter IV paragraph 38).
The information provided should include the critical assumptions in respect of the
operational and economic conditions that will affect the transactions under
consideration during the course of the APA. An assumption will be critical if, where
the assumption is incorrect, this will result in the methodology not reflecting an
arm's length price.
Step 4 – Review and Evaluation
Once an application has been received, the tax authorities will evaluate its
contents and continue to liaise with the business as necessary.
Step 5 – Negotiation
The MAP APA process is a two-stage process: Stage 1 fact finding, review and
evaluation and Stage 2 the competent authority discussions. Under the first stage
all the relevant information is gathered and the taxpayer may have a high level of
involvement in this process. The second stage is a government to government
process and so may have less taxpayer involvement. (See Annex to Chapter IV
paragraph 63.)
Step 6 – Final Signing
Both a MAP APA and a unilateral APA will be normally a written document, signed
and agreed by all parties to the agreement.
The amount of time taken to complete the APA process is dependent on the
complexity of the case and the type of agreement sought.

Step 7 – Implementation and Monitoring
Tax administrations will want to monitor compliance with the APA. This can be
carried out either by way of the filing of annual reports on compliance with the
APA requirements and/or the tax authority continuing to monitor the compliance
with the APA as part of the regular audit cycle. (See paragraph 4.148.)
Where the terms of an APA have not been complied with or there has been a
change in circumstance that for example affects one or more material
assumptions, the APA will be reconsidered. In these circumstances, an APA may
be revoked from an effective date, cancelled either from an effective date or
from the start or renegotiated. (See Annex Chapter IV paragraphs 74-85.)
24.3 Duration of an APA
As noted above, an APA helps provide certainty. Thus there is motivation for the
APA to cover more than one year. However, in determining the length of an APA
account needs to be taken of the fact that circumstances can change. As a result
there is no one rule on the time frame that an APA will cover. They can be as short
as a year and do not normally go beyond 5 years. However, there can be
exceptions too. The OECD address this issue and other issues that have been
looked at above in Annex II to the OECD TPG Chapter IV. You should look at this
appendix as part of your study of APAs.
24.4 International Perspective and Trends
Many countries have an APA process in place, including the US, Australia,
Canada, France, Germany, Russia, and Japan. Many countries such as
Switzerland do not have formal APA programs, but have previously allowed
informal unilateral and bilateral APAs to be completed.
The APA programs will be at varying stages of development and will also vary as to
whether a territory will allow unilateral, bilateral or multilateral APAs, a combination
of all three or not at all.
The regulations relating to the APA process are found in the domestic legislation of
each country and therefore differ from one country to another. In order to illustrate
the different approaches to the APA process, and the various stages of
development, we will summarise the approaches to the APA process in the US,
China, India, Ireland and the UK.
US
The US was the first country to introduce a formal APA process by outlining a set of
procedures for a business to follow. In order to reduce the backlog of claims and
align process with MAP, the IRS underwent a period of reorganisation and the APA
program is now included within an advance pricing and mutual agreement
program.
The US will agree all forms of APAs including unilateral, bilateral or multilateral. The
taxpayer has to propose and present to the IRS a transfer pricing method and all
relevant facts so that a proper transfer pricing analysis can be performed. As with
the APA process in the UK, the administration evaluates all the information and
liaises closely with the taxpayer to ensure all information is disclosed so the analysis
is a fair representation of the taxpayer’s affairs.

Following agreement, the taxpayer must submit an annual report for the duration
of the agreement. The APA applies for the agreed term from the effective date of
the APA, which can be prior to the agreement being finalised. Rollbacks are also
possible with the permission of the IRS, allowing the APA to cover earlier tax years,
which can resolve existing enquiry issues.
A user fee is charged for participation in the programme. The US also has a slightly
different, simplified and more streamlined, program for taxpayers with either less
than US$200 million total income or covered transactions less than US$50 million (or
less than US$10 million for intellectual property).
China
Access to the APA program is limited to the largest taxpayers, as an applicant's

annual related-party transactions must exceed a stated limit.
Most APAs completed within China have been unilateral APAs, however bilateral
and multilateral APAs are also available. Applications for APAs should be sent to
the SAT and the municipal tax authority simultaneously. The APA process follows
the same process as described in the OECD TPG and pre-filing meetings are
encouraged. This is followed by examination, evaluation and negotiation, which
usually leads to completion of the APA. The overall processing time for most APAs
in China is less than two years, and this is likely to reduce over time.
SAT allows for APAs to be rolled back to previous years as long as the relevant
transactions are the same or similar to those covered by the APA.
In 2017 China agreed its first APA for a cost sharing agreement (CSA) with a
fortune 500 company in Guangdong Province. This agreement is further evidence
that APAs are a viable means of reducing transfer pricing risks in China for
intangibles which can be high risk.
India
To address the increasing number of transfer pricing disputes arising in India, the
Union Budget 2012 introduced APAs into the Indian transfer pricing regime. The
APA scheme formed a part of the direct taxes code (DTC) (Section 118(7) of the
direct tax code), which was proposed in 2010 but had not yet been implemented.
The basic framework has been inserted in the Finance Act 2012.
Access to the APA program is available to all taxpayers falling within the ambit of
Indian transfer pricing legislation as no threshold limit is currently prescribed.
Ireland
The Irish Revenue Commissioners introduced a formal system for APAs from 1 July
2016. Prior to this Ireland did not have a formal system but did accept requests
from treaty partners. The formal system only applies to requests for bilateral APAs; it
does not cover unilateral APAs. The Irish revenue has confirmed that they would
be open to negotiating multilateral APA, but not unilateral. The system was
introduced as a result of Action Point 14 of the BEPS Action Plan. As noted in the
previous chapter the best practices put forward in the final report include having a
bilateral APA program.
There is no fee for requesting an APA. For the application to go forward there must
be a double tax treaty in place with the other state. Multilateral APAs will be
achieved via a system of concluding bilateral APAs with the parties concerned.
The revenue authorities have stated that they will attempt to complete formal

applications within 24 months. The typical period for an APA will be 3 to 5 years,
rollback is possible in certain cases. Once an APA is agreed the company will
need to file an annual report which is due on the date for filing the annual
corporation tax return. Non-compliance with the terms of an APA could lead to
cancellation or revocation of an APA.
UK
HMRC have operated an APA programme since 1999 and so now have
considerable experience of working on and negotiating APAs. The APA program is
open to UK taxpayers, including a non-resident entity trading in the UK through a
permanent establishment. Due to limited resources allocated to the APA process,
HMRC generally only consider more complex and challenging transfer pricing
issues. This is in line with the OECD TPG as it is noted by the OECD that the APA
program will not be suitable for all taxpayers due to the expense and time taken
by the procedure. (See paragraph 4.158.)
As we saw earlier in this chapter bilateral or multilateral APAs are often referred to
as MAP APAs and will be agreed under mutual agreement procedures with the
other tax authority. MAP is covered in Article 25 of the OECD Model DTC. From a
UK perspective, access to bilateral and/or multilateral APAs will only be possible
where the relevant clause is included within the tax treaty in question.
HMRC currently aim to complete an APA application process within 18-21 months
of formal submission. Unilateral APAs are generally completed in a shorter time
frame.
The UK also has advance thin capitalisation agreements (ATCA) which are a form
of unilateral APA.
An ATCA is a process designed to help resolve financial transfer pricing issues.

(Statement of Practice 01/12 paragraph 11). It can be used to agree in advance
whether or not HMRC considers that a company is thinly capitalised and what
transfer pricing adjustments are required to affect an arm's length result.
24.5 APAs and the BEPS Action Plan
The BEPS Action Plan is likely to have a significant impact with regard to APAs. You
will recall from an earlier chapter that the 11 best practices put forward in the final
report on Action Point 14 (issued in October 2015) included the implementation of
APAs (Best Practice 4).
In addition, as we see increased global enforcement of tax claims through the

operation of the collection of taxes articles in treaties, and greater information
available to tax authorities through the BEPS CbC reporting requirements (under
Action Point 13), it is likely APAs will become more important to MNEs as part of
their global tax compliance activities.
APAs are included in the categories of information which tax authorities will be
required to exchange with each other, pursuant to the work undertaken on Action
Point 5 – with respect to countering harmful tax practices. One area targeted
under this Action Point is the need for greater transparency. This is envisaged by
implementing a framework for the compulsory spontaneous exchange of
information on certain taxpayer specific rulings. This will include details of unilateral
APAs or other cross-border unilateral rulings in respect of transfer pricing.

Action Point 5 was a minimum agreed standard and it was up to individual

countries to implement the relevant domestic legislation to allow them to identify
the rulings which needed to be exchanged, and the relevant countries to send
the information to.
The 2017 OECD Model DTC includes references to bilateral and multilateral APAs in
the Commentary to Article 25; in particular in the Commentary to paragraph 3.
The European Union has also taken up this approach with regard to countering
harmful tax practices. Under the December 2015 amendments to the Directive on
Administrative Cooperation (2011/16/EU) the scope of the article on automatic
exchange of information was extended to include the details of advance cross-
border tax rulings and APAs entered into by a Member State. The provisions apply
post 1 January 2017, with the scope being sufficiently broad to cover a wide range
of situations, including both unilateral and bilateral/multilateral advance pricing
arrangements and decisions.
Case law on APAs
These decisions concerns tax rulings which validate APAs.
Ireland/Apple State Aid [SA.38373]
This case concerned two Irish subsidiaries of Apple Inc managed and controlled in
the US, so their tax residence at the material times was not in Ireland. As Irish
corporations, they were not US residents either. Each subsidiary had a branch in
Ireland. The essential issue concerns the profits properly attributable to those
branches and whether they were under-taxed selectively, giving rise to unlawful
state aid.
The Irish Government granted two APAs to Apple. The tax treatment in Ireland
enabled Apple to avoid taxation on almost all profits generated by sales of Apple
products in the entire EU Single Market. This was due to Apple's decision to record
all sales in Ireland rather than in the countries where the products were sold.
Under the agreed method, set down in the APAs, most profits were then internally
allocated away from Ireland to a ‘head office’ within Apple Sales International.
This ‘head office’ was not based in any country and did not have any employees
or own premises. Its activities consisted solely of occasional board meetings. Only
a fraction of the profits of Apple Sales International were allocated to its Irish
branch and subject to tax in Ireland. The remaining vast majority of profits were
allocated to the ‘head office’, where they remained untaxed.
Apple Operations Europe benefitted from a similar tax arrangement over the same
period of time. The company was responsible for manufacturing certain lines of
computers for the Apple group. The majority of the profits of this company were
also allocated internally to its ‘head office’ and not taxed anywhere.
The European Commission (EC) argued that the tax rulings issued by Ireland
endorsed an artificial internal allocation of profits within Apple Sales International
and Apple Operations Europe, which had no factual or economic justification. As
head office had no physical presence or employees only the Irish branch of Apple
Operations Europe could be allocated the income from intellectual property as it
alone had the capacity to generate any income from trading, ie from the
production of certain lines of computers for the Apple group. Therefore, sales
profits of Apple Operation Europe should have been recorded with the Irish
branch and taxed there.

The EC stated that Ireland must allocate to each branch all profits from sales
previously indirectly allocated to the ‘head office’ of Apple Sales International and
Apple Operations Europe, respectively, and apply the normal corporation tax in
Ireland on these re-allocated profits. The decision did not ask for the reallocation
of any interest income of the two companies that could be associated with the
activities of the ‘head office’.
Apple was told to repay €13 billion of tax plus interest.
In July 2020 the General Court of the European Union annulled the decision of the
EC that Ireland had granted Apple €13bn in unlawful tax advantages.
The court held that the reasoning that the lack of physical presence or employees
in the head office meant that all the profits should go to the Irish branch was
based on erroneous assessments of normal taxation under Irish law. In particular,
the EC had not attempted to show that income deriving from the intellectual
property was representative of the value of the activities actually carried out by
the branches themselves. It had simply identified the functions performed by the
company as a whole and presumed that they had been performed by the
branches when they could not be allocated to the head offices.
The EC had also concluded that, even if it was wrong on the main line of
reasoning, the rulings departed from a reliable approximation of a market-based
outcome in line with the arm’s length principle as a result of inappropriate
methodological choices. Those choices conferred a selective economic
advantage on the companies. The court held that, although the rulings were
incomplete and occasionally inconsistent, the Commission had not succeeded in
demonstrating that the methodological errors to which it had referred had
resulted in a reduction in the chargeable profits of the companies in Ireland. It was
held that the EC’s decision, that a selective advantage had been granted, was
not based on the actual activities of the branches.
In summary the General Court:
• Concluded that there were errors in applying the Irish law (which essentially
was in line with OECD authorised approach to profit attribution to a branch) as
there wasn’t sufficient focus, and consequently evidence collected, on the
activities carried on through the branches. As a result the EC had wrongly
adopted the ‘exclusion approach’, attributing to the branches what it
considered was not attributable to the US head office without establishing,
with supporting evidence, that the branch managers actively managed those
functions or risks. Apple’s evidence indicated that personnel around the world
and in the US were involved in managing those functions and risks outside
Ireland and the branches essentially provided support for the R&D function.
The ‘centre of gravity’ of the Apple Group was in the US; and the court found
that the strategic decisions, in particular concerning the development of the
Apple IP, appeared to be in the US. Further, the subsidiaries only had one
director based in Ireland.
• Determined the same deficiencies in the EC’s approach were found in relation
to the allocation of procurement activities to an Irish branch.
• Rejected the EC view that choice of wrong tested party had created a tax
advantage.
• Accepted Ireland made some errors, but these were not sufficient to create a
tax advantage.

• Said the EC failed to perform evidence based comparative analysis on

functions performed by branches.
• Stated that operating costs was also an acceptable basis of PLI under OECD
approach, not just sales (as per EC).
• Stated that the EC had failed to show the approach taken did reduce tax.
This case shows that while it is possible for the EC to target tax rulings under state
aid rules, the EC needs to prove its case.
According to George Peretz QC and Tarlochan Lall of Monckton Chambers (Tax

Journal, 31 July), the decision ‘shows the difficulties the European Commission
faces in proving selective tax advantages that may constitute unlawful state aid’.
As was established in Portugal v Commission, ‘the very existence of an advantage
may be established only when compared with “normal” taxation’. The focus of this
case was state aid, however the discussion on the allocation of the profits to the
Irish branch does focus a lot on the attribution of profits to PEs.
This case also shows that the EU will look carefully at transfer pricing reports when
deciding if there has been illegal state aid, (basically state aid without permission).
APAs cannot be used as a means of attracting companies into a country by
offering a tax advantage not otherwise available.
The EC appealed the decision in September 2020.
The EC contends that: ‘The General Court’s failure to properly consider the
structure and content of the decision and the explanations in the Commission’s
written submissions on the functions performed by the head offices and the Irish
branches is a breach of procedure’. Further, the court showed ‘contradictory
reasoning’, which amounts to a failure to state reasons.
In a press statement announcing the decision, Executive Vice-President Margrethe

Vestager said that the General Court judgment raises important legal issues that
are of relevance to the Commission in its application of state aid rules to tax
planning cases, and went on to say that the EC ‘respectfully considers that in its
judgment the General Court has made a number of errors of law’. For this reason,
the Commission is bringing this matter before the European Court of Justice.
Netherlands and Starbucks [SA.38374]
The European Commission (EC) opened an investigation into the transfer pricing
arrangements that Starbucks had in the Netherlands.
The case looked at an APA between the Dutch tax authorities and Starbucks. The
APA covered royalties to a UK limited partnership for coffee-roasting know-how,
and payments to a Swiss affiliate for green coffee beans, thereby reducing the
group’s taxable profits in the Netherlands.
The EC stated that the methodological choices in the transfer pricing report
provided by the tax adviser for Starbucks, which were accepted by the Tax
Authorities in the APA, did not lead to a reliable approach to a market result and
thereby did not fulfil the arm’s length principle.
More specifically the EC were concerned by the use of TNMM when the OECD
TPG show a preference for CUP then went on to say that if TNMM was the correct
choice that it had been applied incorrectly. As a result it was argued that the
Dutch Co didn’t get any economic value for the use of the IP related to coffee

roasting hence no fee should have been paid. The price paid for the green coffee
beans was too high.
The European Court of Justice (ECJ) did not agree with the commissioners. They
ruled in favour of Starbucks as the EC had not conclusively established that
Starbucks had benefited from any clear financial and economic advantage
because of the tax deal that had been provided to it by the Netherlands.
It was stated that within EU law there was a defacto arm’s length principle
however the impact would be limited if states applied the OECD TPG.
The ECJ recognised that determining arm’s length transfer prices is not an exact
science. To this end, it ruled that the EC should verify whether the prices applied
are a reliable approximation of a market-based outcome, and whether there is
any variation that goes beyond inaccuracies inherent in the methodology used to
obtain that approximation.
It was for the EC to demonstrate that methodological errors, or real or alleged lack
of analysis, led to a reduction of the tax burden.
The fact that there was a methodological error did not in itself demonstrate State
Aid.
In other words the EC needed to prove that the transfer pricing accepted by the
Member State was ‘flawed’ and that the flaws led to reduced taxes.
The General Court carried out a comprehensive assessment of the EC’s decision
and underlying evidence, and it pointed out various findings by the EC that were
not supported by evidence.
It was stated that the EC could not easily refute the taxpayers choice of
methodology (TNMM in this case) as the OECD TPG allow the choice of the most
appropriate method. The EC had sought to amend the choice of tested party,
however the General Court didn’t see that there was more reliable data for the
alternative party.
In addition, the ECJ questioned the reference by the EC to the more recent OECD
TPG rather than those in existence at the time of the facts.
It can be said that the wider implications of this decision include that the ECJ has
given a comprehensive and clear answer to questions relating to the application
of an arm’s length principle, the allocation of the burden of proof and the status of
the OECD TPG.
24.6 Conclusion
As we discussed at the beginning of the chapters on double taxation, taxpayers

will need to consider the various ways in which the risk of double taxation can be
reduced or eliminated as part of their transfer pricing risk management strategy.
APAs are likely to become an increasingly popular strategy particularly where tax
authorities can streamline the process and make it more accessible to smaller
taxpayers. APAs are also seen as less adversarial than other forms of dispute
resolution. Although often costly and time consuming they can have less
reputational risk as the APA process takes place behind closed doors as opposed
to the open court of litigation.

In the future, it is anticipated that the number of APA programs will increase and,
as the US has recently implemented, will increasingly be closely aligned with the
MAP programs. Although there is no fee in a number of territories, like the UK where
there does not seem to be a plan to introduce one, the APA process is resource
intensive for tax authorities and so it is likely that an increasing number of them will
impose fees to access the program. As APA programs are perceived as difficult for
all but the largest taxpayers, additional fees may be a good thing where this
allows tax authorities to increase resources to allow greater access to the
program.

CHAPTER 25
E-COMMERCE AND TRANSFER PRICING
In this chapter we cover:

– the OECD Ottawa conference in 1998
– OECD technical advisory groups (TAGs)
– permanent establishments and transfer pricing for e-commerce
– the BEPS Project and Action Point 1 on the digital economy
– post BEPS developments in this area
25.1 Introduction
Simply stated e-commerce relates to commercial transactions conducted

electronically on the internet. It includes sales, marketing, banking and supply
chain management. It may be business to business or business to consumer.
Whilst this is currently a very topical issue, the taxation of profits deriving from e-
commerce has been a problem in the sphere of international tax for some time.
The traditional mechanisms used, in particular in double tax treaties (which
allocate taxing rights in various ways), do not deal well with the particular issues
arising from the flexible and dynamic business models and structures found in the
e-commerce environment.
Domestic tax legislation of states, and the work of bodies such as the OECD, have
struggled to keep up with this rapidly changing aspect of the world economy. This
in turn has, arguably, led to the erosion of the tax base of many jurisdictions as
global businesses are able to maintain a profile and sell to customers without the
creation of a traditional physical presence in a country.
This chapter examines some of the earlier work of the OECD in this area,
specifically the results of the 1998 Ottawa Conference and Report, and the
subsequent reports issued by the ‘Technology Advisory Groups’ (TAGs) on various
relevant issues, including the interaction with the PE concept. We then consider
the more recent work of the OECD under the BEPS Project, which commenced in
2013, leading to a final report in respect of Action Point 1, on the ‘Tax Challenges
of the Digital Economy’, issued in October 2015. We will also consider the post BEPS
developments in this area, including the recent further reports (and agreements)
on e-commerce and digitalisation and the unilateral actions of some states.
25.2 Earlier Reports: The OECD Advisory Groups’ Work on Tax and E-
commerce
The Ottawa OECD conference in 1998 was an important step forward in

addressing the challenges presented by e-commerce in the direct tax field; the
fear of a number of tax authorities being that their national tax bases would be
eroded by businesses selling into their jurisdictions without the need for a significant
physical presence.
A report produced by the OECD Committee on Fiscal Affairs entitled ‘Electronic

Commerce: Taxation Framework Conditions’ was formally adopted by the OECD.

Key conclusions of the report included:
1. the taxation of e-commerce should not be a barrier to its development. Tax

systems should be ‘certain and transparent’ and enable business readily to
understand the tax consequences. Efficiently administered tax systems should
also minimise compliance costs; and
2. tax authorities themselves should utilise these technological developments to

the benefit of their own administration of tax.
The OECD report also suggests that ‘neutrality’ should mean that similar
transactions in similar situations have the same tax consequences. The principles
noted in this report were referred to in the final 2015 BEPS Report on Action Point 1.
However problems arise with this approach. For example, there is a difference
between a physical transaction of actually buying a book in a bookstore from an
assistant and a ‘virtual’ transaction of downloading the contents of a book from
the internet.
The former is the purchase of a physical product as a part of a trade of selling

books.
It is not clear whether the virtual transaction is the same, or whether it is more
specifically the payment of a royalty of some form for the right to use and store the
material. The latter is possibly less likely, since the end use is probably just the same
as buying the physical book.
However, the buying of software over the internet does more obviously present
different considerations and the issue as to whether it is a royalty payment is more
immediate. Withholding taxes potentially apply on cross-border royalty payments,
whereas they do not on trading payments. Amendments to the OECD Model
Treaty Commentary attempt to address such issues, as noted below.
It was therefore agreed by the OECD at Ottawa to review a number of e-

commerce tax areas affecting the OECD Model Treaty and to delegate the task in
each area to TAGs. Four main issues arise from a direct tax standpoint (and many
remain points of discussion in the BEPS Reports):
1. the concept of a permanent establishment (this is discussed in more detail

below);
2. the characterisation of income. The TAG on Treaty Characterisation Issues

Arising from e-Commerce concluded in February 2001 that the commentary
to Article 12 of OECD Model Treaty should reflect the potential consequences
of e-commerce-related transactions.
In essence, if software is downloaded for the purpose of purely operating a

computer program then the payment represents a trading receipt in the
recipient’s hands
Conversely, if the software is itself being used to create other computer

programs or is modified for further use, then the payment is a royalty.
Limited uses of copyright, such as the ability to copy material onto the hard
disk of a computer, are not to be regarded as giving rise to a royalty.
It is notable the commentary to Article 12 has been amended to reflect these

issues.

Although the normal distinction is between royalties and trading income, some
tax treaties contain technical service fee articles, and articles dealing with the
use of scientific equipment or know-how. In the latter cases it may be that
withholding taxes still apply if the payments are classified as being within the
terms of those articles;
3. transfer pricing issues. A discussion paper produced by the Business Profits TAG
in 2001 specifically focused on the transfer pricing issues arising in respect of e-
commerce transactions through a potential PE (discussed below).
Specific trade intangibles can be problematic. For example, patents may

clearly be seen to have been created and registered by a company for the
benefit of others. However, it is not always clear, especially with market
intangibles (such as brand names and customer lists), which company
created that right or benefit. Global business in multinationals tends to be
conducted through the combined efforts of a number of companies in
different jurisdictions with divisional aims. The use of the internet integrates that
activity even more. The possibility of finding ‘comparable uncontrolled prices’
in relation to such integrated activities is frequently fraught with difficulty.
The treatment of intangibles remains an issue and is considered further in the

2015 BEPS Report on Action Point 1, and also the work undertaken on Action
Points 8 to 10.
4. tax residence. This is a further area investigated by the Business Profits TAG.
We will not look at this as our concern is with areas of the TAG relating to
transfer pricing.
25.3 E-commerce and Permanent Establishments
Well before the BEPS project, there was a series of OECD reports on the attribution
of profits to a PE which provided the basis of discussion between tax authorities
and business which led to changes to the interpretation of OECD Model Article 5,
dealing with PEs, and in 2010 to changes to Article 7 and its commentary.
The final report was published in December 2005 ‘Are the Current Treaty Rules for
Taxing Business Profits Appropriate for E-Commerce’. The underlying principles of
these reports are that:
• A PE should be treated as a notional separate entity from its head office: this
forms the basis of the ‘working hypothesis’.
• In undertaking an assessment of the functions of the PE it is important to test

and examine the ‘key entrepreneurial risks’ associated with the operation of
the PE as well as its underlying assets (including intangibles).
• An arm’s length price is then ascertained based on those functions and risks.
This is a very broad outline of the range of the OECD discussions on transfer pricing
which have taken place over a number of years (see earlier for the detail of the
2010 report on attribution of profits to a permanent establishment and the
changes made to Article 7 and its commentary which reflect these principles).
In March 2018 the OECD published a final report to give additional guidance on
attribution of profits to permanent establishments (Action Point 7). The 2018 report
sets out high level general principles for the attribution of profits to PEs, confirming
the principle that PEs are treated as separate entities for these purposes. The report

considers in particular the application of the 2017 amendments to Article 5 of the

OECD Model Treaty, and the examples focus on this.
It is notable that the Business TAG had issued a discussion paper in February 2001
on the transfer pricing aspects of permanent establishments carrying on e-
commerce activities.
Four scenarios were used in the 2001 discussion paper to illustrate the areas that
deserved further consideration in relation to e-commerce and which we also see
being discussed further in the 2014 and 2015 BEPS reports on Action Point 1. A
summary of the scenarios is as follows:
1. The first scenario involves a stand-alone server used by a foreign company in a

location and on which a website is operated for the benefit of the company.
Where the location server operations are purely retail in nature then it is
unlikely that there will be significant profits attributable to those operations,
since there are unlikely to be substantive assets bearing any significant risks.
Head office is also likely to own from abroad the hardware and intangibles
associated with the local operation. Therefore to the extent that the server
represents a PE there is unlikely to be much of a profit attribution, because
nearly all the risks (including credit risks) are with head office.
In these circumstances a ‘cost plus’ basis seems the most appropriate method
to allocate profits to the operations.
Importantly the OECD Model Treaty Article 5 commentary, as revised in 2003,

specifically refers to the possibility of a server alone representing a PE without
there necessarily being human intervention. The commentary previously
referred to business carried on through ‘automatic equipment’, such as a
vending machine, as possibly constituting a fixed place of business. A similar
analysis can be applied to a server. However, there are differences in that
mirror servers may be used to generate the required end user information and
it may not be possible to identify from which server such information
emanated.
2. The second scenario refers to the use of mirror servers for the retailing activities.
The same profit attribution principles apply as in the first scenario, but it is
admitted that it is not always possible to precisely identify the value added by
each server. Therefore the allocation of profits between server locations on a
possible ‘cost plus’ basis may require a reasoned approximation.
3. The third scenario deals with the situation where there is an existing PE having
people, providing, say, technical support and after-sales services, and the
server happens to form part of that PE.
Again the attribution of profits to the server itself is not likely to be significant.
The activities of the personnel may or may not have a significant profit
attribution, depending on the nature and risks of the operations.
In some circumstances, the OECD Model Treaty Article 5(6) (prior to the BEPS
Action Point 7 changes now in the 2017 OECD Model, see earlier chapter)
deems there to be a PE where an agent habitually exercises ‘authority to
conclude contracts in the name of the enterprise’. The OECD in its revised
commentary (in 2003) considers in the context of e-commerce whether an
internet service provider (ISP) or an independent owner of a server would
constitute a PE.

Where the ISP provides services to others and enables them to access the
internet, or provides general hosting services or designs web-sites, then it is
likely that the ISP will be regarded as an independent agent and therefore not
a PE.
However, where the ISP provides a more pro-active part in the actual conduct
of the trade or business of an enterprise then it may constitute a PE of the
enterprise.
Further, the exceptions that apply to avoid a PE where ‘preparatory or

auxiliary’ functions are carried out (see Article 5(4)) are less easy to ascertain in
the case of a server which allows digitised products/software to be
downloaded.
The OECD Commentary to Article 5 observes that, in broadly non-

active/passive circumstances, auxiliary activities in the case of e-commerce
can include provision of communication links, or relaying certain information
through a mirror server, or advertising, or gathering market data or supply of
certain information. If head office itself has as its main purpose these activities
then they cannot be considered auxiliary. As noted in an earlier chapter the
BEPS Action Point 7 changes to the preparatory and auxiliary activities
provisions at Article 5(4) that have been implemented via the MLI for existing
treaties, and the updated OECD 2017 Model Treaty, should also be noted.
4. The fourth scenario in the February 2001 OECD TAG discussion paper assumes
that the PE has developed intangible assets essential to the business and owns
hardware used in the server/web-site operation.
In these circumstances the PE is likely to have substantial profits allocated to it

for transfer pricing purposes, since the functions are probably significant to the
business as a whole and key entrepreneurial risk is associated with those
activities.
Some countries, such as the UK and Ireland, have made clear that, of itself, a
web-site and server will not give rise to a PE for domestic direct tax purposes.
Clearly, if there are other operations/activities then there may be a PE. (See UK
HMRC press releases dated 11 April 2000 and 12 February 2001, also Irish
Revenue statement in March 2001.) Some countries, such as Spain, have
maintained that a local website (so a software driven presence) is enough to
trigger a taxable presence.
Clearly, prior to the work undertaken as part of the BEPS Project, the aim of the
OECD was to “fit” e-commerce within the existing OECD Model Treaty without
making alterations to the Model Articles, but to change the commentary to the
Model where appropriate, and such approach has been effective, up to a point.
It seemed as if case law was overtaking such developments. In the 2012 Dell case
(Ruling of Mar. 15. 2012 R.G. 2107–07, Central Tax Court), the Spanish lower court
introduced the concept of an ‘online permanent establishment’ for the first time.
In line with Spain’s previous position on software driven PEs, the Spanish court ruled
that an online store could qualify as an online PE, even though the server was
situated outside Spain and no activity was performed by the Irish company itself in
Spain. However, on appeal in 2015, the Appellate Court, whilst confirming the
existence of a PE on the basis of both a physical presence and dependent
agency through the Spanish subsidiary, did not confirm the lower court’s finding of
a ‘virtual’ PE. Such issues relating to the digitised economy are highlighted in the
July 2013 BEPS Action Plan (Action Point 1) and considered in the later reports
issued on this matter (see below).

25.4 The BEPS Project
As noted above, the issues surrounding e-commerce and the digital economy
were at the forefront of discussions within the BEPS Project. Action Point 1 of the
BEPS Action Plan related to consideration of how to correctly approach the
taxation of the digital economy.
The Final Report – October 2015
The final October report initially examines the key tax policy considerations and
makes reference to the Ottawa principles which were developed at the time of
1998 OECD Report (see above), such that any proposed legislation should reflect:
neutrality; efficiency; certainty and simplicity; effectiveness and fairness; and
flexibility. It then discusses the evolution and expansion of information and
communication technology across the economy, and provides an overview of the
new business models now seen, and key aspects of such models. For example: e-
commerce such as business-to-consumer, or business-to-business, models; cloud
computing (such as providing infrastructure-as-a-service, platform-as-a-service,
content-as-a-service etc); payment services; high frequency trading, and online
advertising.
The final report reiterates the approach highlighted in earlier drafts, based on the
premise that the digital economy cannot realistically be separated and dealt with
independently from the rest of the economy – as it is, in essence, the economy.
With regard to Indirect Tax matters the report refers to the OECD’s international
VAT/ GST Guidelines, which now take into account digital issues (for example at
Guidelines 2 and 4), particularly with regard to the matters such as the remote
supply of digital goods and services to VAT exempt businesses or to a centralised
location for resupply within an MNE (not subject to VAT).
In direct tax matters three main policy concerns are noted: nexus (ability to have a
presence without being liable to tax); data (how to attribute value to the
generation of data through digital products and services and how to determine
profit share on the basis of this value); and how to properly characterise the
income within these new business models. Four key recommendations were put
forward in this respect:
• Amend the exceptions to the definition of PE found in Article 5(4) of the OECD
Model Treaty. This work has been carried out under Action Point 7 (as we saw
in an earlier chapter) and attempts to ensure each of the exclusions in that
paragraph is restricted to activities that are ‘preparatory or auxiliary’, and
introduces anti-fragmentation rules (see further the discussions in the earlier
chapter on Article 5 – permanent establishments).
• Amend the definition of a PE as applied to agency transactions, to catch

artificial arrangements relating to the sales of goods or services where
contracts are in effect concluded but currently fall out of scope of the
dependent agency definition of a PE (again see discussions in earlier chapters
on Article 5).
• Update the OECD Transfer Pricing Guidelines – this has been actioned within
the context of the work being undertaken on Action Points 8 to 10. It is notable
that digital economy companies place heavy reliance on intangibles for
creating value and generating income. Information technology improvements
mean global value chains are more prevalent, which allows MNEs to become
more integrated. Therefore there is a greater need now for value chain
analysis and more scope for use of the transactional profit split method.

• Improved Controlled Foreign Company (‘CFC’) rules (see later chapter for
more detail) – the report notes that the work undertaken with regard to the
design of effective CFC rules, and the proposed recommendations (the six key
building blocks) under Action Point 3, should help combat some of the
planning opportunities highlighted. In particular using appropriately drafted
CFC rules could assist a country in shifting profits earned through the digital
economy back to the parent entities within a group. By way of example, the
definition of CFC income could include revenues typically generated in the
digital economy such as license fees and income from sales of digital goods
and services.
Arguably such changes did not address the real issues – such as how to deal with
the issue of value creation by the users of digital services. Some of the options put
forward in the earlier drafts were not recommended in this final report, such as
creating a new nexus test (eg a significant presence test) or the imposition of a
withholding tax on digital transactions. However, the report does leave it open to
countries to introduce such measures within their territories as additional
safeguards against BEPS. In 2018 we started to see such unilateral action, as noted
below.
25.5 Implementation and Post BEPS Developments
Post the final BEPS reports from October 2015, we have already seen changes in
many of the Action Point areas, not least due to the operation of the BEPS MLI.
However, much work on the digital economy was delegated in the final 2015
report to the Task Force on the Digital Economy (TFDE) which is a subsidiary body
of the OECD’s Committee on Fiscal Affairs, and is part of the Inclusive Framework
on BEPS. The TFDE’s remit was to review and report on the detailed issues and
possible solutions in the period up to 2020.
It is notable that many of the OECD member states feel that the changes to the
2017 OECD Model Treaty, most of which are in the BEPS MLI, do not go far enough
to combat the issues raised by digital businesses. Tech giants such as Google and
Facebook have a significant worldwide presence, and yet continue to pay little
tax in jurisdictions where they have a large number of users. A global approach to
tackle this would be seen as an optimal solution.
The TFDE has issued an Interim Report in 2018, a Policy Note in January 2019 which
led to a public consultation, a further updated Programme of Work document in
May 2019, and details of proposals for a Two Pillar Approach at the end of 2019.
Updated versions of the Two Pillar Approach were issued in October 2020
(blueprints), and these have formed the basis of the eventual agreement, subject
to some further amendments, on the approach to take by over 130 countries
(taking part in the BEPS Inclusive Framework) mid 2021. The OECD released an
updated version of the agreed approach in a statement of 1 July 2021.
Interim Report of 2018
Since the original report of 2015 it had become apparent that the taxation of the
digital economy is now, and will remain for the future, one of the most
fundamental and challenging aspects of international taxation.
The post BEPS Interim Report from March 2018 (the ‘2018 Report’) covered a
multitude of areas and notably it highlighted that differences still exist between
various states over the action needed. The 2018 Report was concerned with
analysing how digitalisation affects business models which in turn affects the
international tax framework; looking at whether the BEPS measures in place were

having an impact; considering whether interim proposals were appropriate; and

also examining how tax authorities and companies can take advantage of
digitalisation to improve the tax compliance process.
The 2018 Report provided an in-depth analysis of new business models and
attempted to explain how these models give rise to value-creation, and how to
reconcile this with the allocation of income rights. The 2018 Report examined in
particular how digital platforms create ‘size without mass’, so how income or profits
(ie business) can be extracted from a territory without any physical connection in
the form of a PE, and noted the importance of user participation and data
exploitation, and also the heavy reliance on intangible assets.
One problem identified was how to equate the creation of value with the business
generated from large numbers of users in a territory. Different countries had
differing opinions on whether it could be said such users were creating value, and
if they were, how this should be captured. There was some divergence as to
whether the existing tax framework could deal with this type of problem. Could
income or revenue from a number of users, consumers or clients be treated as a
proxy to value, and therefore taxed?
Another important element identified in the 2018 Report was that exploitation of
data was a key component of this activity. Thus users create value for the platform
business they are using, which may justify taxation in the place where the users are
located. The next steps would involve valuing the input of the user, or as noted
using some form of proxy for assessment of this value.
So the 2018 Report did not provide a solution, but rather indicated that members
had very different opinions on these points and on the way forward. Further review
was needed of the ‘nexus’ and ‘profit allocation’ rules for determining taxing
rights.
TFDE Work during 2019 and 2020
The one thing that was evident from the publication of the 2018 Report was that it
did not suggest anything concrete – it was still aiming for a consensus of approach
by 2020, which for some countries was leaving it too long (as seen by the unilateral
implementation of digital services tax, noted below).
Issues that remained for many countries with regard to the allocation of taxing
rights derive from three characteristics which were frequently observed, as
previously noted: ‘scale without mass‘ which impacts the distribution of taxing
rights over time by reducing the number of jurisdictions where a taxing right can
be asserted over a business’s profits; heavy reliance on intangible assets which
strains the rules for allocating income from intangible assets among different parts
of a multinational group; and data and user participation which poses challenges
to the existing nexus and profit allocation rules, especially in situations where
highly-digitalised businesses exploiting data and user-generated content have little
or no taxable presence in the jurisdiction where users are located. However, there
were differences of opinion among OECD countries on the extent to which these
issues should result in changes to international tax rules.
In January 2019 the TFDE issued a policy note for consultation containing high level
proposals on policy design, looking in particular at profit allocation and nexus rules.
The note put forward four proposals, divided into two ‘pillars‘ of work. This was then
updated and amended in May 2019 when the programme of work was
established. Two consultation documents were published towards the end of 2019,
with an updated version issued in October 2020. This version was then updated by

the OECD in a statement pursuant to the 2021 agreement between the G7 and
then over 130 countries taking part in the BEPS Inclusive Framework.
Two-Pillar Approach – BEPS 2.0
The details of this approach were originally put forward in the 2019 consultation
documents with updated blueprints issued in October 2020. In a statement in
January 2020 the members of the OECD Inclusive Framework on BEPS had agreed
to get behind the proposed solution, based on the two-pillar approach, with the
aim of reaching a final agreement by the end of 2020. However, delays occurred
due to the pandemic, and agreement by the BEPS Inclusive Framework countries
on an updated approach was only reached mid 2021 (as noted above).
Pillar One covers new nexus and profit allocation rules to determine how taxing
rights should be allocated among countries, particularly ‘market’ countries where
value is created by large MNEs, not just digital businesses (which is a move away
from the original proposals). Pillar Two is a global anti-base erosion system (GloBE),
enabling countries to tax profits shifted to jurisdictions with low effective rates of
tax, building on BEPS principles.
The OECD did set out its preliminary assessment of the proposals, which suggests
the combined effect of pillars one and two would lead to a significant increase in
global tax revenues and redistribution of taxing rights to market jurisdictions. The
greatest effect would be felt by multinational businesses in the digitised and
intangible-intensive sectors. The February 2020 economic analysis of the two-pillar
solution put the combined effect at up to 4% of global corporate income tax
revenues annually (around $100bn), mainly due to the effects of new GloBE rules in
pillar two. The reallocation of taxing rights in pillar one would bring a small tax
revenue gain for most jurisdictions (mainly to low and middle-income economies).
The following is an overview of the latest proposals issued by the OECD in July 2021
– itself based on the earlier ‘blueprints’ for the two pillars. Your KVR book does not
contain the blueprints or later OECD document, however, it does contain the
earlier versions of the two pillars which had been proposed at the end of 2019. In
these overviews reference has been made to the Tax Journal articles: 11 June 2021
‘The G7 Tax Deal’ Rhiannon Were and Lucy Urwin; 25 June 2021 ‘Briefing –
International review for June’ Tim Sarson; 16 July 2021 ‘The OECD’s statement on
international tax reform’ Sandy Bhogal and James Chandler.
Pillar One
The aim of the pillar one proposals is to align taxing rights more closely with local
market engagement. Under pillar one, a portion of an MNE group’s residual profit
would be taxed in the jurisdiction where the revenues arise. It will apply to MNEs
with a turnover above €20bn (reviewable after seven years with a view to
reducing it to €10bn if it has been successful); and profit before tax of at least 10%
of revenue (determined by reference to accounting standards). The idea is that
this approach will capture large MNEs (including the digital/ tech companies) and
avoids the need to identify specific digital businesses.
‘Market’ jurisdictions where consumers are based are then to be allocated a share
of residual profits (referred to as ‘amount A’) based on a formula, and then a fixed
return for routine marketing and distribution activities (‘amount B’). The proposals
then have dispute prevention and resolution processes to improve tax certainty.
Amount A: this amount is set as 20%–30% of profit in excess of the 10% of revenue.
Amount A will be allocated to market jurisdictions where the MNE derives at least
€1m in revenue; or if the country has a GDP of less than €40bn, at least €250,000.

The amount allocated to such countries will be determined by a formula based on

revenues from the goods or services used or consumed, with double taxation
elimination measures.
Amount B: is a return based on the arm's length principle for an in-country baseline
marketing and distribution activity. The OECD has noted this will have a particular
focus on the needs of low-capacity countries.
The 2020 OECD pillar one blueprint proposals are complex and envisage multiple
elements with regard to the evaluation and calculation of the two amounts, and
the resolution of any disputes. The 2021 OECD statement has simplified matters
somewhat by no longer requiring application to specific digital service businesses.
Currently financial services businesses and the extractive industry are out of scope
of the proposals. More specific details of the updated proposals are planned to be
produced in the latter part of 2021 (you should review any further documents and
details published by the OECD in this regard).
Opinions on the Pillar One proposals
The authors of ‘The G7 Tax Deal’ article above, noted that this agreement might
not have even been possible at the start of 2020. They note the fact there has
been such widespread agreement in 2021 is likely result of two significant factors:
the political impetus driven by the financial need to rebuild public finances
following pandemic expenditure; and a new, more outward-facing US
administration.
The author of the International Review for June 2021 above also highlights the
probable impact from the Pillar One proposals. It is commented that the shift in
taxing rights is likely to be most pronounced for tech giants which have been
operating based on no PE or cost-plus transfer pricing models. However, it is
important to note that many of these tech giants are already subject to digital
services taxes in a number of territories (see discussions below). Thus, when
considering the overall impact on such companies’ worldwide tax liabilities, there
could be a trade-off with tax liabilities in the parent jurisdiction (for example, a
number of large US tech companies had already changed their models so as to
report foreign income as part of US taxable income). The biggest winners are likely
to be the US and European countries with large wealthy consumer markets, at the
expense of small wealthy trading hubs, such as Switzerland, Ireland and Singapore.
Other territories, such as China and Japan, may well see little difference in their tax
take as a result.
Other commentators (on the basis of the October 2020 proposals) have raised the
question – what is this whole exercise really for? It is not clear which countries will
be the losers in any reallocation. The OECD may be relying on the expectation
that countries will value the taxes reallocated to them more highly than the
unilateral taxes they are giving up (as those reallocated taxes are more likely to be
passed on to consumers). Countries may take positions on issues such as thresholds
and allocation percentages based upon what most benefits them. There is a
danger that under pillar one most countries will raise roughly the same amount of
tax as before but in a highly expensive and more complex way.
Pillar Two
While the Pillar One proposals look at the allocation of taxing rights between
jurisdictions, pillar two is concerned with the global anti-base erosion (GloBE)
proposal, a set of rules for ensuring minimum effective rates of tax.

The July 2021 OECD statement on the proposal under pillar two is built around
three key rules:
• an ‘income inclusion rule’ to impose tax on the parent in respect of profits of a

foreign-controlled entity or branch, if those profits are subject to a low
effective tax rate; and
• an ‘undertaxed payments rule’ which is a back-stop for when the income

inclusion rule does not apply – it can be used to deny deductions, or require
an equivalent adjustment, for payments made to related parties that are not
subject to tax at or above a minimum effective rate. These first two rules are
termed the ‘GloBE’ rules; and
• a ‘subject to tax rule’, which is a separate treaty-based rule, which can be

applied in priority, and subjects payments to withholding or other taxes at
source for income not subject to tax at or above a minimum effective rate.
A key element of the proposals is that they allow for a jurisdiction to ‘top up’ the
tax to the minimum level. For example, under the ‘income inclusion rule’, once a
jurisdiction has been identified as undertaxed, the difference between its effective
tax rate and the minimum rate, expressed as a percentage, is applied to the
appropriately identified parent entity's share of the undertaxed income.
The two GloBE rules will have the status of a common approach. This means that
countries within the Inclusive Framework are not required to adopt the GloBE rules,
but, if they choose to do so, they will implement and administer the rules in a way
that is consistent with the outcomes provided for under Pillar Two, including in light
of agreed model rules and guidance. They will also accept the application of the
GloBE rules applied by other Framework members including agreement as to rule
order and the application of any agreed safe harbours. It is intended the GloBE
rules will apply to MNEs that meet the €750 million threshold as determined under
BEPS Action Point 13 (country-by-country reporting). However countries are free to
apply the ‘income inclusion rule’ to MNEs headquartered in their country even if
they do not meet the threshold.
The OECD July statement notes that it is recognised that the ‘subject to tax rule’ is
an integral part of achieving a consensus on pillar two for developing countries.
Inclusive Framework members that apply nominal corporate income tax rates
below the minimum tax rate to interest, royalties and a defined set of other
payments, would implement this rule into their bilateral treaties with developing
Inclusive Framework members when requested to do so.
The ‘subject to tax rule’ would require treaties to be amended and the OECD
statement refers to a multilateral instrument to facilitate its implementation.
However, both the ‘income inclusion rule’ and the ‘undertaxed payment rule’
could be implemented through changes to domestic law. Model legislation may
need to be produced by the OECD to ensure consistent application.
One of the most important questions to be answered with respect to the

workability of the pillar two proposals is the level of the minimum rate of tax that
would be applied. The G7 and later agreements reached in 2021 on the proposals
refer to a minimum tax rate of at least 15% (see below) – the minimum rate for the
‘subject to tax rule’ is 7.5% to 9%.

Opinions on the Pillar Two proposals
It has been commented that the proposals are designed to impact the behaviour
of both taxpayers and tax administrations. For taxpayers, by introducing a global
minimum rate of tax, the incentives to shift profits to no or low tax jurisdictions are
to be taken away. Although it is noted that profit shifting was ‘particularly acute’ in
the context of intangibles which are prevalent in the digital economy, this initiative
is not limited to digital businesses and could potentially impact all types of
multinationals across the world.
For governments, the proposals are said to prevent a ‘race to the bottom’. The
concern appears to be that governments might seek to encourage investment
through tax incentives at the cost of public spending on infrastructure and public
services, or perhaps shifting taxes to fund such public spending onto workers and
consumers. It is claimed that this measure is needed to protect tax sovereignty,
although those jurisdictions that do not support a minimum tax rate might argue it
has the opposite effect.
The authors of ‘The G7 Tax Deal’ article consider that the pillar two minimum tax
rate proposals would mean that the principal tax take is at the headquarter
jurisdiction level. Therefore, the jurisdictions where many large MNEs are
headquartered will benefit most in terms of increased revenue. This includes (but is
not limited to) the US, where many of the giant ‘tech’ companies are
headquartered. Jurisdictions in which MNEs are not typically headquartered are
unlikely to see any increase in their tax revenue in the short term. That is unless they
choose to unilaterally increase their domestic tax rates, which is arguably the
secondary result that the proposals aim to achieve (i.e. the effective eradication
of ‘tax havens’, which in this context are those jurisdictions imposing very low or no
corporation tax rates).
The UK Chartered Institute of Taxation submitted its comments on the OECD’s

original proposals for pillar two, describing the GloBE proposal as ‘far too complex’.
They commented that ‘it is not clear whether the fundamental principle underlying
the proposal is to achieve a minimum effective tax rate for any entity, either in that
entity or at shareholder level; or whether it is to allow countries to protect their own
tax base from base eroding payments’. With the input of the US in 2021, it would
appear to be the achievement of a minimum tax rate.
Issues: Minimum Tax Rate and Unilateral DST Provisions
As we have noted, in June 2021 the G7 and then a wider group of over 130
countries announced they had come to agreement with regard to the
implementation of the OECD’s Two Pillar proposals (dubbed BEPS 2.0). This came
on the back of the Biden administration’s new tax proposals, and the renewed
focus on the need to stop the ‘race to the bottom’ in tax rates.
The proposals have been agreed in principle, with the OECD releasing a
statement with an overview of the updated approach in July, and with more work
on the details to be completed before the end of 2021. Notably a minimum tax
rate for Pillar Two of ‘at least 15%’ is proposed, even though the US had been
suggesting a rate of 21%. It is intended that the minimum rate will be applied on a
jurisdiction by jurisdiction basis. This is supported by the US, although no formal
consensus has yet been reached (mid 2021).
Applying the tax rate on a jurisdiction by jurisdiction basis will ensure that
multinationals pay tax of at least 15% in each country in which they operate, as
opposed to cross-jurisdictional blending of high tax and low tax to an average of
at least 15%. It is interesting that nine countries did not agree at this stage,

including Ireland (which has a 12.5% corporate tax rate) and Hungary (which has
a 9% corporate tax rate), amongst others such as Estonia, Barbados, Kenya and
Nigeria.
As highlighted by Sarson (June 2021 article above) it is notable that the G7 also
made a further commitment to coordinate the implementation of BEPS 2.0 with
the removal of all unilateral digital services taxes. This suggests that G7 countries,
such as the UK, France and Italy, which have already implemented digital services
taxes, might be prepared to eliminate these taxes, regardless of whether every
company currently within the scope of its digital services tax would fall within the
scope of the taxing right afforded under the final pillar one.
The UN Model Treaty – alternative approach – new Article 12B
Not all countries were happy with the ongoing work of the OECD in this area and it
is notable that the UN finally approved an alternative approach to the taxation of
digital businesses in April 2021 – which would negate the need for the OECD pillar
one proposals. Problems are likely to arise if the UN treaty article approach also
gets support (and it may do so, particularly with developing nations), although
currently the widespread ‘in principle’ agreement to the Two Pillar approach
would seem to imply the UN treaty approach is not likely to have many takers.
The UN approach involves a new double tax treaty article, Article 12B in the UN
Model Treaty. Article 12B permits a withholding tax to be imposed by source states
on gross payments for ‘automated digital services’ made by a resident of one
contracting state to a resident of the other contracting state. In essence it catches
income received with little human involvement from the service provider. The tax
would apply to income derived from online advertising services, the supply of user
data, online search engines, online intermediation platform services, social media
platforms, digital content services, online gaming, cloud computing services, or
standardised online teaching services. The payments for automated digital
services are sourced to the jurisdiction in which the services are used.
The withholding tax is subject to a rate limitation to be agreed between the

contracting states, although the UN has recommended a rate of between 3% and
4%. As an alternative to the withholding, the recipient (beneficial owner) of the
payments can elect to be taxed on the profits sourced from the provision of the
automated digital services. In effect, this is designed to permit the beneficial
owner of the income to declare a limited PE in the jurisdiction where the payer is
located in order to be subject to tax on a net income basis.
Unlike Pillar One, the taxing right is not based on first meeting revenue thresholds
and it only applies to companies engaged in the digital economy. The UN
approach is intended to simplify administration allowing greater adoption,
particularly by developing countries. One of the issues with the OECD’s proposals is
that they are complicated to apply and a global minimum tax rate is likely to
disadvantage developing jurisdictions. However, if the approaches taken by the
OECD and the UN cannot be reconciled, then it increases the risk for double
taxation. Notably, the UN approach is contained in the 2021 update to the UN
Model Treaty, so states are not obliged to include the article in their existing or
future treaties. With so many countries agreeing in principle to the OECD proposals
in 2021, it may be that Article 12B does not find its way into many (if any) actual
treaties.
You should monitor developments in this area of taxation.

25.6 Unilateral Action on the Digital Economy
We have been seeing different countries and groups of states unilaterally move in
a direction of their own, frustrated by the lack of actual coordinated action from
the OECD (the ongoing programme of work did not provide the immediate
solutions that some states would have liked). Notably in 2018 both the OECD’s TFDE
(above) and the EU (see below) issued significant reports on the digital economy,
however, the approaches taken in each are very different, and some would say
inconsistent.
Due to the lack of firm action on a global cross-border scale, some countries have
unilaterally introduced their own rules to target these types of businesses. There
were concerns this would lead only to multiple independent approaches to this
problem, rather than the originally hoped for collective global stance under the
BEPS Project. However, as noted above, the US and the OECD pushed forward
with their Two Pillar approach, which was agreed mid 2021.
The EU Approach
By contrast to the OECD, the EU originally took the approach that this is all about
‘fair’ taxation, and put forward concrete suggestions to capture income and tax
it. However the problem they have been faced with is a lack of unanimous
agreement, which is required for any tax provisions within the Union.
In March 2018 the European Commission published its digital tax package. This
included an explanation of why it was considered that the ‘digital economy’ was
undertaxed, in addition to two draft directives and one recommendation.
In the long term the package put forward draft proposals for a directive which laid
down rules relating to the corporate taxation of a significant digital presence. This
would mean changing existing PE thresholds where ‘digital services’ are provided.
It included thresholds such as annual revenues from digital services in a Member
State in excess of €7m; or more than 100,000 users of the digital service in a
Member State in a taxable year.
In the short term a draft directive was proposed which introduced a digital
services tax (DST), which would be a tax on revenues at 3% where users play a
major role in value creation. For example, taxable revenues created by selling
online advertising space; or where a digital intermediary allows users to interact
with other users; or where revenues are derived from the sale of data generated
from user-provided information.
However, the EU proposals faced various challenges and by the end of 2019 no
agreement had been reached as to an ‘EU’ way forward. It was agreed to wait
until global action was put forward by the OECD – which has taken some time.
However, the need to finance the Covid-19 recovery effort led to the European
Council considering the introduction of measures to generate additional ‘own
resources’ for the EU – with one key proposal involving a digital levy. In November
2020, the Commission reaffirmed its commitment to putting a formal digital levy
proposal forward by mid 2021, although this was later put back to October 2021
with a view to adopting it by 1 January 2023 at the latest.
An inception impact assessment was released in January 2021 which sets out the
purpose and potential implications of the levy. It is notable that it states the
initiative is ‘intended to not interfere with the ongoing work at the G20 and OECD
level on a reform of the international corporate tax framework’. Possible options

under consideration include: a corporate income tax top-up to be applied to all

companies conducting certain digital activities in the EU; a tax on revenues
created by certain digital activities conducted in the EU; and a tax on digital
transactions conducted business-to-business in the EU.
Even post the widespread agreement on the Two Pillar proposals, the EU continues
to examine the digital levy. The EU maintains such levy will be designed to co-exist
with pillar one and be compatible with WTO and international obligations.
It is notable that the consultation on the digital levy raised the possibility that such
a tax could distort competition in the market, if businesses in the same sector begin
to be taxed differently depending on how they interact with their customers. Some
responses suggested that unilateral action to tax the digital economy is
incompatible with the EU’s stated commitment to the BEPS project and should be
avoided. Interestingly other commentators took the opposite view and noted it
might in fact provide additional leverage for international reform. You should
consider the proposals in more detail once they are published.
In the meantime, various EU Member States, such as France, Spain, Italy and
Romania, amongst others, have moved ahead unilaterally with their own versions
of a DST. France’s DST which taxes direct sales and also imposes a levy on
advertising revenue, websites and the resale of private data, was passed in July
2019. The US soon after queried France on its implementation and France
eventually agreed to delay implementation until the end of 2020, and such
matters will be revisited in the period post the agreement on the Two Pillar
approach.
The UK Approach
In March 2018, the UK published its own proposals with regard to corporate
taxation and the digital economy.
This paper looked more closely, and provided more detailed explanation, on how
user participation creates value for digital businesses, in particular looking at the
generation of content and engagement with platforms which then contributes to
the brand. This is very different from the role that a customer serves in a traditional
business and the mere collection of customer data. The customer is effectively
performing supply-side functions which the company would have previously
undertaken itself. The paper proposed a solution which would involve reforming
the international framework and using profit reallocation rules to benefit the user
jurisdictions, to reflect the value of user participation.
In the absence of a unified international approach, the UK brought in DST at 2%,

applicable from April 2020. The tax applies to the revenues of search engines,
social media platforms and online marketplaces where this is connected to UK
users irrespective of where the owner of those revenues is based. It will specifically
apply to revenues earned from intermediating online sales rather than the online
sale itself. Where the business is providing a search engine, social media platform
or online marketplace, then it will also catch revenue from online advertising.
To be within the tax the business needs to generate in-scope revenues of at least
£500m globally (with at least £25m derived from the UK). There is an allowance for
the first £25m of relevant UK revenues – which means small businesses are out of
scope. It is intended that the DST would be reviewed in 2025 to see whether it was
still appropriate or required, depending on whether there is an agreed
international approach by then.

You should also review the Tax Journal article ‘Is the DST compatible with the UK’s
International Obligations’ Rupert Shiers, 1 November 2019, Issue 1463, 12.
The Indian Approach
By way of contrast it is interesting to note that in its Finance Act 2018 India
amended the definition of ‘business connection’ to include a new nexus which
taxes business profits based on a ‘significant economic presence’ (SEP). In
addition, India opted into the key BEPS Action Point 7 proposals on agency PEs in
the MLI, for its covered agreements, and introduced similar provisions in its
domestic law.
Income is deemed to accrue or arise in India if it accrues or arises, directly or

indirectly, through or from any ‘business connection’ in India. This includes the
current updated agency PE approach. From 1 April 2018 the new test lowers this
business connection threshold and allows the profits of a foreign company to be
taxed in India where the company has a SEP there (subject to the operation of an
applicable tax treaty, and the definitions contained therein).
The Indian authorities made it clear that such a change is in response to recent
advancements in information and communication technology which has enabled
new business models to carry on business activities in the source state remotely
and through digital means, without the need for establishing a physical or
representative presence. On this basis the traditional nexus to tax of a PE does not
‘hold good anymore’ and it was considered necessary to supplement such a
connection with a new nexus, based on a digital or virtual economic presence,
which would be in line with the approach already seen in the Dell Spanish case
from 2012.
A SEP is defined as: a transaction in respect of goods, services or property carried

out by a non-resident in India; a transaction involving provision of download of
data or software in India; and a systematic and continuous soliciting of business
activities or engaging in interaction with a number of users. The activity must
exceed the prescribed number of users. In addition, as noted above, this also
reflects one of the possible options discussed in the 2015 Final Report on BEPS
Action Point 1, although it was not a recommendation.
There have been some concerns that the provisions are a step too far, as the
international tax framework has not been amended to take into account such
changes to the nexus definition. However, as noted, the application of the
traditional PE definition in India’s existing double tax treaties with other countries
will continue to apply.
India also applies an equalisation levy at 6% which is imposed for online advertising
services, on the payments made to non-resident beneficial foreign owners for
providing digital advertising (rather than being charged on the income arising
from transactions) where they don’t have a PE in India. The objective of such a tax
is to equalise the tax burden, by imposing it on payments made to overseas
owners who have an unfair advantage over Indian competitors who provide a
similar service.
India’s Finance Act 2020 expanded the scope of the equalisation levy from 1 April
2020 to cover ‘e-commerce supplies or services’. So it covers a range of digital
transactions including business-to-business transactions, business-to-consumer
transactions, e-commerce marketplaces and digital services. This version of the
levy is chargeable at the rate of 2% on consideration received or receivable by
non residents who operate digital businesses targeting, among others, the India
market. The Finance Bill 2021 was supposed to provide clarification of terms and

application of the equalisation levy. However, commentators have noted that the
proposed changes could potentially make the equalisation levy applicable not
only to e-commerce retailers/marketplace operators, but also to ‘normal’
businesses that happen to have a fair degree of digitisation. The impact of the
2021 proposals remains to be seen.
Indirect Tax
One area where developments have been very evident, and governments are
themselves moving ahead, is in the introduction of legislation to apply VAT/GST on
the provision of cross-border business to consumer digital services. This has become
more commonly known as the ‘Netflix’ tax – and is an attempt to capture revenue
that is being lost on the local level to untaxed, non-local digital streaming services
(in particular). The EU introduced provisions which came into force at the start of
2015 in this area, but other countries around the world have also introduced
provisions with regard to the taxation of such cross-border services.
For example in May 2016 Australia passed a cross-border digital GST which requires
businesses outside the country to register and collect GST on sale of digital services
to consumers in Australia – effective from 1 July 2017. New Zealand had already
introduced similar rules which went into effect on 1 October 2016, and on 1
January 2017 Russia passed similar tax rules (under Russian VAT legislation) on cross-
border digital services sales to consumers, joining other countries such as Japan,
Korea and India which have also introduced such rules.
In the last couple of years this trend has continued and even picked up pace.
Singapore has introduced a tax, effective from 1 January 2020. Quebec province
in Canada has also introduced a similar GST on digital services (effective from 1
January 2019). Thailand has proposed an e-business tax which would be a levy on
online purchases, advertisements and website rent in Thailand earned by foreign-
based digital service providers (effective from 2021). Many other countries have
followed suit such as Bangladesh, Colombia, Taiwan, Switzerland and some of the
Gulf States, such as Saudi Arabia and the United Arab Emirates, to name but a
few.
This approach also fuels the debate with regard to e-commerce, as to whether
taxation should be on the basis of source, or on the basis of residence. Clearly the
source approach (based on where the services are being consumed) is being
seen as a practical and viable first approach. It is arguable that these approaches
are themselves outdated with regard to e-commerce – which seamlessness and
intangibility itself challenges the suitability of the traditional tax system. In any
event, whilst digital service providers themselves may not be too keen on such
VAT/GST provisions, it is notable that they are now becoming widespread - both
globally and also at the level of state and city sales taxes (for example within the
US).
Clearly work with regard to the digital economy on the part of the OECD, and
globally, continues to develop.

CHAPTER 26
OTHER ASPECTS OF TRANSFER PRICING
In this chapter we are going to look at the broader role of transfer pricing, beyond tax
planning and compliance, within a multinational organisation, including:
– transfer pricing and public affairs
– the impact of taxation on business decisions
– transfer pricing as a management tool
– customs duties and transfer pricing
26.1 Introduction
Most of the analysis of transfer pricing so far in this manual has focused on its role
for corporate tax purposes. Broadly speaking, two key dimensions have been
identified: the challenges for compliance created by complexities surrounding
transfer pricing, and the opportunity for tax planning due to the relationship
between transfer prices and taxable profit.
However, transfer pricing is not only important for corporate tax purposes. It also
has an important role to play across a range of important aspects of the internal
operation of multinational groups. In particular, in this chapter, we will consider the
relationship between transfer pricing and each of the following:
• Public affairs
• Business decisions
• Management incentives
• Customs Duties
26.2 Transfer Pricing and Public Affairs
For a long time, transfer pricing had very little profile amongst the broader public,
with few people being aware of the term, much less what it meant. That situation
has changed dramatically in recent years. One of the main reasons for this is the
increased focus on the level of corporation tax paid by subsidiaries of global
multinational groups.
Amazon, Google, Apple, IKEA, Starbucks and many other household names have
all been subject to high profile tax investigations.
As a result, in the short term, there are strong associations for the public between
transfer pricing and tax evasion. In many parts of the media, transfer pricing has
been portrayed as purely a planning tool, ignoring the complex compliance issues
faced by any company seeking to operate in multiple jurisdictions. Thus, the
consequences for any transfer pricing controversy issues that arise are likely to be
magnified.
In the longer term, transfer pricing seems to have shifted from a policy issue for tax
authorities to a political issue. As you have seen throughout this manual, transfer
pricing issues have played an important role in the BEPS Action Plan, in addition
many governments are implementing fundamental overhauls of tax legislation.

The consequence of all this is that for most multinationals, transfer pricing has been
elevated from purely an issue to be resolved within the tax function to a board-
level consideration. The potential consequences of a tax strategy that might be
perceived as overly aggressive is no longer limited to the risk of double taxation,
plus interest and penalties. It is instead a potential threat to brand value, revenue,
profitability and ultimately share price. There is much to be played out before the
tax environment can again be said to be in a stable state. In the meantime, there
are a much broader set of parameters and stakeholders for taxpayers to consider
in respect of their transfer pricing than simply seeking an appropriate balance
between compliance and tax optimisation.
26.3 The Impact of Taxation on Business Decisions
In order to survive, and ultimately thrive, companies must be profitable. At a basic

level, companies exist to generate an economic return for their shareholders.
There are a range of strategies for achieving this, from high growth strategies
designed to increase the underlying share price, to profit maximisation in more
stable businesses to fund dividend payments to shareholders. A key part of
delivering such strategies relates to managing both costs and uncertainties within
businesses.
There is a clear link between transfer pricing and the tax cost to a business. The
transfer prices that are acceptable to the tax authorities will affect the tax charge
in each section of the business. Non-compliance with transfer pricing legislation
can lead to penalties which in turn will impact on the cost to the business.
Thus, we can see a direct link between the transfer price for taxation purposes and
the company’s overall performance. Managing tax cost is as important to a
company as any other costs, since dividends are paid out of post-tax profits.
Transfer pricing has an important role to play in managing cash flow within the
business, since it determines where profits are earned. By managing transfer
pricing to ensure that profits are earned in the right location, companies can
make sure they have cash available to fund dividends, capital investment or
acquisitions without need for recourse to external borrowing. Furthermore, it is
important to consider that some countries, such as China and India, have tight
foreign exchange controls, making it difficult to take cash out of those markets. An
appropriate transfer pricing strategy is required to avoid having trapped cash in
those countries.
In addition to being a potentially high cost to the business, tax and transfer pricing
carry a significant degree of uncertainty. As new strategic initiatives are
developed and rolled out by companies, it is important for them to manage the
tax consequences. This means considering the most likely tax treatment, the
possible alternatives and related costs, and the mitigation of unnecessary risk
through appropriate business activity. Failure to do so can lead to significant
additional cost to the business through unmanaged transfer pricing exposures. It is
impossible to decouple significant business decisions from tax (and typically
transfer pricing) consequences. Consider a number of examples of decisions that
a company might face:
• Expansion into a new territory – As companies seek to grow, they will often
seek new markets for their products. In setting up a new business, transfer
pricing considerations will affect the legal structure (branch or subsidiary), how
the new operations will be financed (debt or equity), where the set up costs
will be borne, how the new entity will transact with the rest of the group and
forecast expected benefits (based on the tax impact of where profits will be
earned). In addition, if the business decides to enter a new territory through a

strategic relationship with a third party, this will be of relevance to transfer

pricing as the terms agreed may create a CUP (which may support or
undermine existing transfer pricing arrangements).
• Business acquisition – Where companies grow through merger or acquisition,

transfer pricing also has a key role to play. During the due diligence process, it
is important to understand the historic transfer pricing position within the
business to be acquired, as the acquirer will be taking on the risk associated
with this. Post-acquisition integration can be a challenging exercise involving
merging of two transfer pricing policies. Where policies are conflicting (for
example, one group operates a centralised business model with a Principal,
and the other uses a decentralised model with key decisions taken at a local
level), it can be a long and costly process to align the operating models.
• Closing of facilities – In the event that a company chooses to close a facility,

or even entire operations in a country, transfer pricing will help to determine
the tax cost of doing so, and specifically where those costs should be borne.
• Change of brand name – Companies that maintain a brand portfolio will need
to actively manage it, and in some cases that might involve brand refreshes.
Transfer pricing will inform the company where costs relating to brand
refreshes should be borne and will help to determine the value of the new
brand to the brand owner through the mechanism by which it is remunerated
by other group entities.
In relation to significant business decisions, there should be a two-way flow of

information. The business needs to inform the tax function, as changes will impact
on transfer pricing risk, and may need to be managed through amendments to
policy. However, transfer pricing will also need to provide input to key decisions, as
it can be an important part of delivering returns on investment, both in terms of
absolute level of post-tax profit earned and also in controlling variability around
those returns through tax risk management. Transfer pricing should not be driving
key business decisions. Nevertheless, it is a significant factor to be considered.
26.4 Transfer Pricing as a Management Tool
To be successful, companies rely upon individuals within the organisation making

the decisions that are in its best long-term interest. In a small company, this is
relatively easy for senior management to monitor, because they will have more
visibility over the decisions being made. However, this is much harder in a larger
company with a global reach. In those cases, senior management must delegate
a lot more responsibility, and put in place infrastructure to ensure that the decisions
made align with the broader business strategy.
A key part of the infrastructure is the series of incentives given to management of

different parts of the business. Transfer prices form part of this incentive process in
that they help to determine profitability at a local level. When a product is
transferred between different parts of the business, the transfer price will determine
the profit for that division/unit. Likewise, there will be an impact from service
charges or royalty flows. Within an organisation the use of the optimal transfer
price for remuneration purposes will enhance cooperation and transparency
within the business.

Incentive Conflicts Created by Transfer Pricing
Nevertheless, the challenges of a transfer pricing model and its impact on

behaviour are best demonstrated through a number of illustrations. The scenarios
set out below relate to manufacturing, but the same issues arise in other parts of
value chain.
 Illustration 1
Contract Manufacturer
Take the case of a simple contract manufacturing operation, producing products

solely on behalf of a Principal company.
Product
Contract →→→→→→→
Principal
manufacturer ←←←←←←←←
Payment
Typically under these circumstances, the manufacturing operation would be

remunerated on a cost plus basis. However, challenges arise if the management
of the manufacturer is assessed based on plant profitability. The revenue earned
by the manufacturer comes from sale of products to the Principal, and the transfer
price of products is set based on the cost of production. As such, the only way that
a manufacturer can increase its absolute level of profit is to increase costs. Whilst
this may create more profit for the manufacturer, this reduces overall profitability
for the group.
It may be that transfer prices are based on budgeted costs. In such

circumstances, the manufacturer could strive to operate more efficiently in order
to lower actual cost compared to budget. However, such efforts would only
temporarily increase profits as the new efficiencies would be factored into prices in
the following budget cycle.
The natural conclusion is therefore that plant management shouldn’t be

incentivised on profitability, but instead be measured on performance against
factors that they can control, such as quality and efficiency. These factors would
contribute to the group’s overall profitability. This is generally straightforward where
a plant is initially set up as a contract manufacturer. However, where a plant is
converted to contract manufacturer as part of a business restructuring, this can
cause conflict. Under such circumstances, the role/responsibility of the
manufacturer is reduced and as such the scope of responsibility for the
management of the manufacturer is reduced (with potential impact to salary and
business), creating potential conflict.

 Illustration 2
Fully-fledged Manufacturer
Consider now the case of a fully-fledged manufacturer selling direct to distributors.
Fully-fledged
manufacturer
Product ↓
↑Payment ↑Payment ↑Payment

(Price A) (Price B) (Price C)
Related Party Related Party Third Party Distributor
Distributor (Country A) Distributor (Country (Country C)
B)
In such a scenario, it is conceivable that a resale price method would be applied.

Under this method, Price A and Price B shown above would be set separately with
reference to operating costs and end market price for the product. In all
likelihood, this would produce two different transfer prices for the same product.
Let’s assume that this results in a higher transfer price for Country A.
If the management of the manufacturer were incentivised purely on plant

profitability, this would influence behaviour. In the event that there was limited
capacity, the manufacturer would be incentivised to prioritise production for
Country A. Whilst this may be consistent with short-term profit maximisation for the
group as a whole, it may not align to the business strategy. It may be the case that
Country B is the long-term strategic priority representing a higher growth market.
Some mechanism is therefore required to align manufacturer behaviour with
overall strategic priorities.
Now let’s assume that the manufacturer also sells to Country C through a third
party distributor. If Price C is higher than both Price A and B (for reasons of different
economic circumstances, rather than creating a CUP), this will further impact
behaviour. In the event of capacity constraints, the manufacturer may be
incentivised to prioritise sales to Country C rather than either of the two related
party markets. This would maximise profit for the manufacturer, but reduce overall
group profit as profits earned in Country A and Country B would be lost to the third
party in Country C.

 Illustration 3
Procurement Company
Now let’s consider the impact of a procurement company
Third Party Supplier

Material ↓
↑Payment ↑Payment
(Price A) (Price B)
Fully-fledged ← Material Related Party
Manufacturer → Payment (Price C) Procurement
Company
↓Material
↑Payment
(Price C)
Related Party
Manufacturers
Assume that the fully-fledged manufacturer requires Material as part of the

manufacturing process, which it sources from the Supplier. Given its size, it is able
to negotiate Price A as the price for Material.
Now assume that the group establishes Procurement Company to strategically

manage purchasing and leverage from the group’s global spend. Although the
fully-fledged manufacturer in this case is the largest manufacturer in the group, by
combining its requirements with those of other manufacturing sites, it is able to
negotiate a slightly lower price (Price B).
The Procurement Company then purchases Material from the Supplier and sells on
to all manufacturers at Price C. This price includes a margin for the Procurement
Company that covers its costs and provides it with an economic return for its
activities. For all other manufacturers, Price C represents a saving on what they
could have purchased Material for on their own. However, for the fully-fledged
manufacturer, the addition of a margin means that Price C exceeds Price A – the
price it was able to negotiate by itself.
Such a scenario creates a conflict. If the management of the fully-fledged

manufacturer in question are seeking to maximise their own profit, they will
negotiate their own supply at Price A. However, if their requirements are removed
from the spend managed by the Procurement Company, then the Procurement
Company will not be able to negotiate as good a price, and as a result overall
group profit will reduce.
Split of Management and Statutory Accounts
In each of the illustrations above, the conflict that arises could be alleviated by a
different transfer price that creates an alternative incentive for the manufacturer.
The question therefore arises: is it possible to operate with two sets of transfer prices
– one for statutory reporting purposes and the other for management accounts to
assess performance? Management accounts may not reflect certain
intercompany charges, such as management fees or royalties, and may set
product prices on a different basis than the statutory accounts (for example, with
reference to total system profit).

The use of two sets of accounts has some appealing features. It allows greater
control for management over their area of the business. It also reduces the time
spent by senior management on dealing with internal charges. Ultimately,
intercompany pricing determines how profit is shared within a group, but does
nothing to directly increase the amount of pre-tax profit that a group earns. As
such, management time and resource would seemingly be better spent
elsewhere. The tax function could put in place charges necessary to meet
statutory and fiscal requirements, and the business could be run based on
management charges.
However, this approach has two discernible drawbacks. Firstly, it is very time-
consuming and burdensome to run two sets of accounts. It adds complexity to the
management of the business that may be difficult to justify.
Secondly, and arguably more importantly, a transfer pricing framework should

reflect the economic reality of a business. If there is a need for a set of
management accounts that significantly diverges from the statutory accounts, it is
a strong indication that the pricing method used for the statutory accounts is not
arm’s length.
Consider the case of a limited risk distributor (LRD). If an LRD is remunerated based
on a low stable operating margin, it would be assumed that its activities could be
characterised as relatively routine, and that the key economic decisions
managing risk and driving profit would be made elsewhere. If however, there is a
second set of management accounts whereby the management of the distributor
is incentivised to maximise profit in the country (rather than for example maximising
sales within certain profit parameters), this implies that management is able to
exercise control over that profit. It suggests there is local decision making that is
capable of materially influencing local profit, and therefore undermines the
characterisation on which the transfer pricing is based.
Therefore, caution should be exercised before using separate management and

statutory accounts. In some cases, it is unavoidable. However the very existence of
separate management accounts and incentivisation structures can create
significant risk to the transfer pricing framework.
26.5 Customs Duties and Transfer Pricing
Much of the focus on transfer pricing is on the impact for corporate tax purposes.
However, as we have seen, many related party transactions will involve cross-
border transactions. Customs duties are normally applied when goods enter a
territory.
Customs duties are a further way of raising tax revenues; the higher the value of
the goods at importation, the greater the customs duty, which is normally applied
on an ad valorem basis. This means that the value of the product imported (ie. the
transfer price) is directly correlated with the customs duty payable. As such,
strategic consideration of the customs duty impact is required when determining
transfer prices.
Regulatory Overview
The OECD TPG recognise that the valuation methods for customs purposes may
not be aligned with the OECD recognised transfer pricing methods (see
paragraph 1.137 (1.157)). Having recognised this, the OECD TPG go on to state
that the customs valuation may still provide useful information as customs will have
contemporaneous information on the taxpayer and documentation.

There are of course some key differences between the valuation methods used for
customs duties and the transfer pricing methodologies. Customs pricing does not
take account of related party transactions; they deal with both related and
unrelated parties and include individuals. For customs purposes the focus is on the
value at the point of import; for transfer pricing there is more focus on profits that
are made and how they need to be split across an organisation.
The World Customs Organisation (WCO) and the OECD organised two
conferences (2006 and 2007) to discuss how the gap could be bridged between
direct and indirect taxation on the valuation of transactions between related
parties, and to explore possible areas for strengthening coordination between
customs and tax specialists. The conferences concluded that more analysis is
needed and that harmonisation will require changes and adjustments on all sides.
The OECD TPG at paragraph 1.138 (1.158) suggest greater cooperation between
income tax and customs administrations within a country.
Countries are recognising the need for co-ordination and cooperation. Canada
and the US have provided guidance on the acceptability by customs authorities
of transfer pricing valuations and adjustments. In the UK HMRC can conduct audits
into both taxes.
However the need for caution is demonstrated by the decision in Hamamatsu

Photonics Deutschland GmbH v Hauptzollamt München (Case C-529/16) (20
December 2017) where the CJEU found that the Customs Code does not allow
transfer pricing adjustments, made after the end of the relevant accounting
period, to change the customs value.
Hamamatsu purchased imported goods from its parent company, which charged
it intra-group prices in accordance with the advance pricing agreement
concluded between the group of companies and the German tax authorities. The
total of the amounts charged to Hamamatsu by its parent were regularly checked
and, if necessary, adjusted, in order to ensure the conformity of the sale price with
the ‘arm’s length’ principle laid down in the OECD transfer pricing guidelines.
Following such transfer pricing adjustments, Hamamatsu had applied for a

repayment of customs duty, which had been rejected by the German tax
authorities.
The CJEU observed that the cases in which it had allowed a subsequent
adjustment of the transaction value had been limited to specific situations relating
to quality defects or faulty workmanship in the goods, discovered after their
release for free circulation. The CJEU also noted that the Customs Code does not
impose any obligation on importer companies to apply for adjustment of the
transaction value where it is subsequently adjusted upwards; and that it does not
contain any provision enabling Customs to safeguard against the risk that those
undertakings only apply for downward adjustments. The CJEU concluded that the
Customs Code does not allow account to be taken of a subsequent transfer
pricing adjustment of the transaction value
Practical Considerations
From the perspective of strategically managing transfer prices, there are a range
of practical issues that need to be taken into consideration:
• Competing objectives of authorities – Customs and corporate tax authorities in

any given country have diametrically opposed objectives. Where a product is
imported into a country, customs authorities are looking to increase transfer

prices where products have been underpriced to increase duty yield.

Conversely, corporate tax authorities are seeking to decrease transfer prices
that are too high in order to increase taxable profit. This creates significant
challenge in creating analysis that justifies pricing from both perspectives.
• Timing – Customs duty is payable upon importation of products. It is unlikely

that the arm’s length nature of the prices applied will be challenged, adjusted
and resolved before the financial year is complete, much less before the
corporate tax return is submitted and assessed. This makes it very challenging
to manage corporate tax and customs duty risk in parallel. Although there is a
natural tension between the requirements of customs and corporate tax
authorities, in practice there will often be a gap of several years between
them considering the same issue.
• Unilateral nature of customs – The challenge in managing transfer pricing

compliance is balancing the needs and expectations of different tax
authorities in order to avoid double taxation. However, there is no such
concept of double taxation for customs purposes, since an increase to the
import price does not reduce the customs yield of another country. It may
have an impact on the corporate tax yield of that country if the adjustment
were reflected in the statutory transfer prices, but for the reasons described
above, that does not place much constraint on the customs authorities.
• Adjustments – Many companies operate transfer pricing models that rely on

periodic retrospective adjustments to ensure profitability stays between certain
parameters. Where these adjustments are made to product prices,
consideration needs to be given to whether the adjustments will have a
customs impact. Where the adjustment is an increase in product price, it is
likely for many countries that additional duty would be payable, which would
need to be disclosed to the customs authorities. This may in turn attract
additional costs such as interest, penalties and increased scrutiny from
customs officials going forward. Where the adjustment is a decrease in price,
then in theory a refund may be owed. However, in practice it can be very
difficult to obtain refunds from customs, and therefore the additional duty paid
will likely be lost.
• Dutiable items – For physical items imported, it is clear that duty is payable,
subject to rates applied under local regulations. However, there may also be
other transactions that are dutiable. For example, if separate payments are
made for IP or services that might otherwise be considered an inherent part of
the product, these may well also be subject to duty. For example, if a
distributor imports chocolate bars from a related party manufacturer, and
separately pays the distributor a royalty for the brand name used on the
wrapper, most customs authorities would consider the royalty payment
dutiable.
• Business restructuring – Where companies undertake business restructuring, this

often results in a change to the legal flow of products, even where the
physical flow is unaffected. Consideration needs to be given to the customs
impact of this. Not only might it create risk by increasing the import price
(creating risk for historic prices), but it may also jeopardise benefits arising from
Free Trade Agreements (FTAs) that the company may have been enjoying.
FTAs serve to reduce the duty for products moved between certain countries,
and analysis should be undertaken to consider whether interposing a principal
entity taking legal title will threaten this.
In reality, managing customs is a complex exercise and is often handled in a

separate part of the organisation to transfer pricing, despite the common starting
point of related party transactions. Nevertheless, in managing transfer pricing and

making decisions that will change prices and transaction flows, there needs to be
awareness of the customs impact to avoid creating unnecessary risk.
26.6 World Customs Organization (WCO) Guide to Customs Valuation and

Transfer Pricing
In 2015 the World Customs Organization (WCO) published the WCO Guide to
Customs Valuation and Transfer Pricing. The aim of the publication is to assist
customs officials responsible for customs valuation policy or who are conducting
audits and controls on MNEs in understanding the relationship between customs
valuation and transfer pricing. The Guide was updated in 2018. The 2018 version
consists of six chapters including an explanation of transfer pricing and technical
information on customs valuations.
In the background explaining transfer pricing it is referred to as a ‘neutral concept

that simply refers to the determination of transfer prices for transactions between
related parties.’ The guide does then go on to recognise that manipulation may
take place to decrease taxes including customs duties.
As with transfer pricing, customs valuation seeks to ensure the price between
related parties has not been influenced by the relationship. The guide emphasises
some of the similarities between the aim of customs valuations and transfer pricing
seeing the ability to use transfer pricing information and documentation as positive
where it is possible.
In Chapter 5 it is stated that it would be desirable that the customs community

strive to achieve a more consistent approach to the treatment of transfer pricing
adjustments. Chapter 6 also encourages production of more advance rulings for
customs valuation.
Chapter 6 includes guidance on good practices for customs valuation policy

makers, business and tax administrations. These include:
• Ensure the customs and tax advisors (either internal or external) of an MNE
communicate with each other regarding the mutual needs of the customs
and tax authorities in respect of transfer pricing and customs valuation;
• Consider the needs of customs when preparing transfer pricing

documentation;
• Consider customs needs in the development of advance pricing agreements;
• Depending on national procedures, ensure customs are given advance

notification where post-importation adjustment may occur at a later date;
• Consider requesting advance rulings from customs, where available; and
• Work with customs to provide and help interpret transfer pricing analyses and
data related to imported goods.
Annex X to the guide gives an overview of the key elements that may be found in
transfer pricing documentation which may be relevant for customs valuation. The
key elements are listed as:
1. Description of the MNE Group, its business activities and the industry in which it
operates;
2. Key financial information relevant to the controlled transactions;

3. Relevant aspects of the group’s transfer pricing policy, including details of how
prices are set and reviewed and whether the group has any relevant APAs;
4. Description of the related party transactions, including functional analyses;
5. Explanation as to why the transfer pricing method selected was selected, with
reference to local legislative requirements (where applicable);
6. Explanation of the process undertaken to try to identify comparable

uncontrolled transactions, including details of sources of information and
search criteria used. Comparability analysis of selected comparable
uncontrolled transactions, including analysis with respect to the 5
comparability factors and details of any further research conducted.
The guide can be found at:
http://www.wcoomd.org/-/media/wco/public/global/pdf/topics/key-
issues/revenue-package/wco-guide-to-customs-valuation-and-transfer-
pricing.pdf?la=en

Tolley® Exam Training ADIT PAPER 3.03 APPENDIX 1
APPENDIX 1
CASE LAW
For ease of reference we have listed some important transfer pricing cases below. You will
have come across many of these cases as you worked through the manual. All of the
following cases are long and complex. No attempt has been made to provide a full
analysis of each case; instead the following focuses on the aspects of the case that have
been most talked about in the Transfer Pricing press and that are likely to be most relevant
to the exam. There are of course other cases that you may have come across that you
can equally use to demonstrate a point in the exam room where necessary.
1.1 AB LLC and BD Holdings LLC v SARS ([2015] ZATC 2)
In this case, the South African Tax Court had to determine whether a US company,
providing strategic and financial advisory services for the South African airline
industry had created a PE in South Africa. The Court reviewed Article 5 of the US-
South Africa double tax treaty, (which differs from the OECD Model Treaty) in
particular the second paragraph of the article (para (5)(2)(k) in this case), which
‘includes especially’ as a PE ‘the furnishing of services including consultancy
services’
In its judgment of May 2015, the Court held that a PE had been created, on two
counts. Firstly, on the basis the US company employees had been in South Africa
for over 183 days. The Court considered that the wording of the second
paragraph of Article 5 in the treaty, which (as noted above) is preceded by the
words ‘includes especially’, creates a stand-alone definition of a PE by way of an
extension to the first paragraph of the Article. Secondly, it found there was in any
event a fixed place of business created by the use of the boardroom (the US
employees had exclusive use during relevant working hours). The US company was
liable for tax on the profits attributable to the services rendered. In addition, the
Court held that the penalties imposed by the tax authorities (100% of the tax) were
not disproportionately punitive and that the US company had been negligent in
not considering the tax consequences of the agreement it had entered into. The
company had argued that it did not deliberately ignore South African law, but just
misunderstood it. Interestingly the Court commented that the enterprise had a
global reach and that it was ‘not a novice in the area of tax liability’, and
therefore should accept responsibility for its own error.
1.2 Amazon State Aid: Luxembourg and Amazon State Aid [SA.38944V]
In October 2017 the EU competition commissioner Margrethe Vestager declared

that Amazon’s tax benefits in Luxembourg were illegal under EU state aid rules. The
European Commission published a non-confidential version of its decision in
February 2018.
The decision concludes that Luxembourg provided an economic advantage to

Amazon by allowing the application of a transfer pricing methodology not in line
with the ‘arm’s length principle’. The finding included that a royalty paid for the
use of the IP was a function of the residual profit of Amazon EU and not based on
sales. The Commission also questioned the functional analysis of the two entities
who were parties to the royalty in light of the pricing of the royalty and questioned

the application of the ruling given by Luxembourg for a period of more than 10
years without an apparent obligation to review the facts.
Luxembourg has put forward its defence against the decision, mainly based on
purported illegitimate ‘covert fiscal harmonisation’, wrong selectivity frame of
reference and violation of the legal certainty principle. It is likely that the case will
go to the Court of Justice of the European Union.
1.3 Apple State Aid: Ireland/Apple State Aid [SA.38373]
This case concerned two Irish subsidiaries of Apple Inc managed and controlled in
the US, so their tax residence at the material times was not in Ireland. As Irish
corporations, they were not US residents either. Each subsidiary had a branch in
Ireland. The essential issue concerns the profits properly attributable to those
branches and whether they were undertaxed selectively, giving rise to unlawful
state aid.
The Irish Government granted two APAs to Apple. The tax treatment in Ireland
enabled Apple to avoid taxation on almost all profits generated by sales of Apple
products in the entire EU Single Market. This was due to Apple’s decision to record
all sales in Ireland rather than in the countries where the products were sold.
Under the agreed method, set down in the APAs, most profits were then internally
allocated away from Ireland to a ‘head office’ within Apple Sales International.
This ‘head office’ was not based in any country and did not have any employees
or own premises. Its activities consisted solely of occasional board meetings. Only
a fraction of the profits of Apple Sales International were allocated to its Irish
branch and subject to tax in Ireland. The remaining vast majority of profits were
allocated to the ‘head office’, where they remained untaxed.
Apple Operations Europe benefitted from a similar tax arrangement over the same
period of time. The company was responsible for manufacturing certain lines of
computers for the Apple group. The majority of the profits of this company were
also allocated internally to its ‘head office’ and not taxed anywhere.
The European Commission (EC) argued that the tax rulings issued by Ireland
endorsed an artificial internal allocation of profits within Apple Sales International
and Apple Operations Europe, which had no factual or economic justification. As
head office had no physical presence or employees only the Irish branch of Apple
Operations Europe could be allocated the profits relating to intellectual property
as it alone had the capacity to generate any income from trading, ie from the
production of certain lines of computers for the Apple group. Therefore, sales
profits of Apple Operation Europe should have been recorded with the Irish
branch and taxed there.
The EC stated that Ireland must allocate to each branch all profits from sales
previously indirectly allocated to the ‘head office’ of Apple Sales International and
Apple Operations Europe, respectively, and apply the normal corporation tax in
Ireland on these re-allocated profits. The decision did not ask for the reallocation
of any interest income of the two companies that could be associated with the
activities of the ‘head office’.
Apple had to repay 13 billion Euro of tax plus interest.
In July 2020 the General Court of the European Union annulled the decision of the
European Commission that Ireland had granted Apple €13bn in unlawful tax
advantages.

The court held that the reasoning that the lack of physical presence or employees
in the head office meant that all the profits should go to the Irish branch was
based on erroneous assessments of normal taxation under Irish law. In particular,
the Commission had not attempted to show that income deriving from the
intellectual property was representative of the value of the activities actually
carried out by the branches themselves. It had simply identified the functions
performed by the company as a whole and presumed that they had been
performed by the branches when they could not be allocated to the head
offices.
The Commission had also concluded that, even if it was wrong on the main line of
reasoning, the rulings departed from a reliable approximation of a market-based
outcome in line with the arm’s length principle as a result of inappropriate
methodological choices. Those choices conferred a selective economic
advantage on the companies. The court held that, although the rulings were
incomplete and occasionally inconsistent, the Commission had not succeeded in
demonstrating that the methodological errors to which it had referred had
resulted in a reduction in the chargeable profits of the companies in Ireland. It was
held that The Commission's decision, that a selective advantage had been
granted, was not based on the actual activities of the branches.
In summary General Court:
• Concluded that, there were errors in applying the Irish law (which essentially
was in line with OECD authorised approach to profit attribution to a branch) as
there wasn’t sufficient focus, and consequently evidence collected, on the
activities carried on through the branches. As a result the Commission had
wrongly adopted the ‘exclusion approach’, attributing to the branches what it
considered was not attributable to the US head office without establishing,
with supporting evidence, that the branch managers actively managed those
functions or risks. Apple’s evidence indicated that personnel around the world
and in the US were involved in managing those functions and risks outside
Ireland and the branches essentially provided support for the R&D function.
The ‘centre of gravity’ of the Apple Group was in the US; and the court found
that the strategic decisions, in particular concerning the development of the
Apple IP, appeared to be in the US. Further, the subsidiaries only had one
director based in Ireland.
• Determined that the same deficiencies in the Commission’s approach were

found in relation to the allocation of procurement activities to an Irish branch.
• Rejected the EC view that choice of wrong tested party had created a tax
advantage.
• Accepted Ireland made some errors, but these were not sufficient to create a
tax advantage .
• Said the EC failed to perform evidence based comparative analysis on

functions performed by branches.
• Stated that operating costs was also an acceptable basis of PLI under OECD
approach, not just sales (as per EC).
• Stated that the EC had failed to show the approach taken did reduce tax.
This case shows that while it is possible for the Commission to target tax rulings
under state aid rules, the Commission needs to prove its case.

According to George Peretz QC and Tarlochan Lall of Monckton Chambers (Tax

Journal, 31 July), the decision ‘shows the difficulties the European Commission
faces in proving selective tax advantages that may constitute unlawful state aid’.
As was established in Portugal v Commission, ‘the very existence of an advantage
may be established only when compared with “normal” taxation’. The focus of this
case was state aid however, the discussion on the allocation of the profits to the
Irish branch does focus a lot on the attribution of profits to PEs.
This case also shows that the EU will look carefully at TP reports when deciding if
there has been illegal State Aid (basically State Aid without permission). APAs
cannot be used as a means of attracting companies into a country by offering a
tax advantage not otherwise available.
The European commission appealed the decision in September 2020.
The Commission contends that: ‘The General Court's failure to properly consider
the structure and content of the decision and the explanations in the
Commission’s written submissions on the functions performed by the head offices
and the Irish branches is a breach of procedure’. Further, the court showed
‘contradictory reasoning, which amounts to a failure to state reasons’.
In a press statement announcing the decision, Executive Vice-President Margrethe

Vestager said that the General Court judgment raises important legal issues that
are of relevance to the Commission in its application of state aid rules to tax
planning cases, and went on to say that the Commission respectfully considers
that in its judgment the General Court has made a number of errors of law. For this
reason, the Commission is bringing this matter before the European Court of
Justice.
1.4 Aska Gmbh (SKM2017.213.SR)
The facts of the case were that a Scandinavian sales manager was required under
the terms of his contract to work from home in Denmark. Although he spent the
majority of his employment duties travelling and visiting clients, the Danish tax
board ruled that the manager’s use of a home office for administrative work (for
which he was not reimbursed) constituted a PE of his German employer.
It was held to be irrelevant whether the home office was owned or rented or in
any other way made available to the foreign corporation. As long as the business
of the non-resident entity was carried out effectively and habitually (and these
activities were not preparatory), then a PE was in existence. It was also
emphasised that there must be recurring work from the home office not just
sporadic or occasional.
The Danish tax board determined that the administrative work was directly related
to the main business of the company and could not be said to be non-core or
preparatory.
1.5 Aztec Software & Technology Services Limited v ACIT 294 IT

(Bangalore)
Aztec Software & Technology Services Limited (Aztec-India), an Indian company

providing software development services, has a wholly-owned subsidiary in the
United States of America (Aztec-US). Aztec-US was appointed as Aztec-India's
marketing agent in the US. Aztec-US rendered marketing services in relation to sale
of products of Aztec-India, ensuring minimum orders and prompt payment by
customers. It also performed certain ‘onsite services’ such as identification of client

requirements, installation of software at client's location and acceptance testing.

Designing and development of software (offshore services) was undertaken (in
India) by Aztec-India. In consideration of such services, Aztec-US received
remuneration based on a cost plus mark-up basis (5 percent for onsite services
and 10 percent for marketing services) from Aztec-India.
The point was raised that in India the transfer pricing law was designed as an anti-
avoidance tax measure, thus the Revenue could invoke the provisions only under
specific circumstances where there is existence of material evidence to suggest
avoidance of tax. Since Aztec-India enjoyed a tax holiday, there was no plausible
reason or motive for avoidance of tax (in India). Thus, the application of the
transfer pricing regulations (per se) under such circumstances was entirely
misplaced.
The Tribunal held that there was nothing in the statutory provisions to require that
the Assessing officer must demonstrate avoidance of tax.
It was further found the TP law can apply even if the income is exempt as was the
case here because of the tax holiday.
The case looked in detail at the methodologies used. It observed that the while
the use of cost plus was not disputed, not enough attention had been paid to the
computation of the cost. In this particular case it was stated that single year data
was more appropriate. The Tribunal approved of the Revenue’s contention that for
the purposes of arm's length analysis, profits earned by Aztec–India (and not
Aztec-US) should have been benchmarked under TNMM analysis.
Reference was made to the OECD Guidelines where appropriate.
1.6 Centrica India Offshore Pvt Ltd v CIT [WP(C) No. 6807/2012] Heard at
the Delhi High Court
Employees seconded to India from group companies in the UK and Canada were
deemed to constitute permanent establishments (PE) of those companies in India.
A wholly owned subsidiary of the UK company, Centrica PLC, resident in India

provided back office services to overseas group companies, charging cost plus
15% for those services. To help support the establishment of the Indian back office
services in the first year of their operation, some overseas employees with
managerial and technical expertise were seconded to India. The court decided
that under the terms of the India-UK and India-Canada tax treaties, these
secondees constituted a PE in India that was providing technical services.
The decision arose because of the way the secondments themselves were
structured. The court distinguished between the concepts of economic employer
and legal employer. It also referred to the OECD Commentary which says that the
foreign company will not constitute a service PE if the seconded employees work
exclusively for the Indian enterprise and are released for the period in question by
the foreign enterprise – that was not the case here. The court further observed that
there was no purported employment relationship between the Indian company
and the secondees and, notably, that the company had no right to terminate the
employment contract and the employees had no right to sue the company for
non-payment of salary.

1.7 Cameco Corporation [2020 FCA 112]
This case looked at the transfer price between Canadian affiliates that produced
uranium and its distribution affiliates.
Cameco’s Canadian mines produced a significant amount of uranium at only $10

per pound, which was the lowest cost of production globally. In 1999, Cameco
established the Swiss affiliate Cameco Europe AG (CEL) to distribute the uranium
produced by these mines and established Cameco Europe S.A. (CESA) to
purchase the uranium that Russia had formerly used in its nuclear arsenal. CEL later
assumed the role of CESA thereby selling Russia's inventory of uranium and
Cameco production of uranium. This litigation involved the period from 1999 to
2006. During the early years of this intercompany arrangement, uranium prices fell
below $10 per pound as Russia inventory sales dominated the market. In 2003 and
2004, uranium prices rose with the average price over the 1999 to 2004 period
being $11 per barrel. Uranium prices continued to rise going from an average of
$27.93 per pound to $47.68 per pound in the period considered. As such, CEL
received substantial profits under the intercompany pricing policy while the
Canadian affiliates earned very little profit. The CRA challenged this intercompany
pricing arrangement using several approaches to argue that all of these Swiss
based profits should represent Canadian taxable income.
The CRA presented essentially three different positions. Its legal position was that
the restructuring was a sham and all of CEL’s profits should be taxable in Canada.
Its second position was based on the claim that CEL was a limited function
distributor so its gross margin should be very modest. Its third position was the
recharacterisation rule.
The first two approaches were later abandoned The CRA sought to rely on the
recharacterisation rule (now non-recognition) The CRA reassessed Cameco’s 2003,
2005 and 2006 taxation years, reallocating approximately $480 million of the Swiss
Subsidiary’s profits under this rule.
For recharacterisation to apply under the Canadian law the CRA needed to show
that the transaction or series of transactions (i)would not have been entered into
between persons dealing at arm’s length, and (ii) can reasonably be considered
not to have been entered into primarily for bona fide purposes other than to
obtain a tax benefit (Section 247(2)(bi) and (bii) of the Act). The CRA sought to
interpret the first condition very widely.
In June 2020, the Federal Court of Appeal (FCA) released its decision upholding
the Tax Court’s decision in favour of the taxpayer and confirmed a restrictive
interpretation of the recharacterisations provisions contained in sections 247(2)(bi
and bii) The FCA confirmed that the test in section 247(2)(b)(i) imposes an
objective test (ie whether any arm’s-length persons could enter the transactions
on any terms), and not a subjective test, taking into consideration the situation of
the particular taxpayer.
The FCA also stated that if the recharacterisation rule had applied, the
consequence should not have been drastic transfer-pricing adjustments, such as
the total reallocation of profits to Cameco Canada, because this would amount
to disregarding the existence of the other parties to the transactions (CEL and
CESA).
An appeal was lodged in October 2020 but was dismissed in February 2021.

The implications from this case include:
I. That in Canada the recharacterisation provisions will not apply where the
taxpayer’s arrangements are commercially rational and where nothing
prevents the determination of a transfer price, even if there is a tax-oriented
purpose to the particular structure utilised.
II. That where recharacterisation does apply, the CRA is required to substitute
arm’s-length terms and conditions for the transaction it has determined would
have been entered into between arm’s-length persons, rather than
substituting the terms and conditions with nothing on the presumption that the
no transaction would have occurred.
III. Given the Tax Court’s willingness to ascribe commercial rationality to any
transaction or series that could be transfer priced under sections 247(2)(a) and
(c), it remains difficult to predict the types of transaction that would be so
commercially irrational that the recharactersation provisions in Canada would
apply.
1.8 Chevron Australia Holdings Pty Ltd v Commissioner of Taxation [2017]

FCAFC 62
Chevron Australia Holdings Pty Ltd – ‘CAHPL’ (Borrower), is a wholly owned

subsidiary of the US-based Chevron Corporation (Parent) entered into a Credit
Facility Agreement in June 2003, under which it borrowed from a special purpose
subsidiary (Lender) which had been established to raise funds in US bond markets
(with the benefit of a guarantee from the Parent). The key features of this credit
facility were that:
• the Borrower borrowed the AUD equivalent of US $2.5 billion;
• interest would be paid at 1-month AUD LIBOR + 4.14% pa;
• the loan was repayable in full after 5 years, but could be repaid earlier at the
Borrower's option;
• the loan was not subject to financial covenants; and
• no guarantee or security was provided by the Borrower.
The Commissioner of Taxation disallowed deductions for interest paid under the
Credit Facility Agreement in the 2004 to 2008 period, based on his view that:
• the arm’s length rate of interest for the loan was lower than that charged by
the Lender; and
• an arm’s length loan agreement would have contained other terms which
reduced the amount payable by the Borrower (such as that the loan would
have been in USD and would have contained financial covenants).
The underlying issue was how one should construct the hypothetical transaction
between independent parties for the purposes of applying the transfer pricing
provisions. The potential tax adjustments that may arise under the transfer pricing
provisions will be affected by the features that are to be taken into account in
constructing the hypothetical.
One of the key issues was whether the features of the hypothetical transaction
should include the provision of security or other covenants in favour of the lender.
As is clear from the decision of the lower court, where the hypothetical loan

between an independent borrower and lender is to include security provided in

favour of the lender to secure repayment of the loan, then the inclusion of this
feature can affect the conclusion as to the arm's length interest rate for the
purposes of applying the transfer pricing rules.
The question at issue was whether the consideration actually provided by CAHPL
(ie the interest rate and nothing else) exceeded the arm’s length consideration for
that property. This task required one to assess what the consideration would be
under a hypothetical transaction between independent parties dealing at arm's
length.
On 21 April 2017, the Full Court of the Federal Court of Australia (full federal court)
handed down its decision, dismissing the taxpayer’s appeal from the lower court.
The full federal court unanimously upheld the ruling that the loan was not a
genuine ‘arm’s-length’ transaction and the taxpayer had breached transfer
pricing provisions of tax legislation. Chevron have said they will not appeal this
decision further.
The court held that the borrowing by CAHPL would not have been sustainable if
obtained from an independent party. As a standalone company, severed from
the financial strength of its ultimate parent and corporate group, CAHPL could not
secure a loan for an amount equivalent to $US2.5 billion at the rate obtained by its
subsidiary with the backing of the ultimate parent.
The evidence amply supported the reasonable expectation of a borrowing by

CAHPL being supported by security.
In the court’s decision it was stated that:
The conditions operating between CAHPL and CFC if they were independent of
each other would not include the direction by Chevron Treasury of the officers of
both for the benefit of the group as a whole. The conditions between mutually
independent CFC and CAHPL could, however, include CAHPL situated within the
Chevron group and CAHPL being subject to the direction of Chevron for the
benefit of the Chevron group.
In such circumstances, were CAHPL seeking to borrow for five years on an

unsecured basis with no financial or operational covenants from an independent
lender, in order to act rationally and commercially and conformably with the
interests of the Chevron group to obtain external funding at the lowest possible
cost consistently with any relevant operational considerations, it would do so with
Chevron providing a parent company guarantee, if such were available.
In the light of the evidence as to Chevron’s policy concerning external funding

and its willingness to provide a guarantee to achieve that end the above is the
natural and commercially rational comparative analysis when one removes the
controlled conditions operating between CAHPL and CFC and replaces them with
the condition of mutual independence. In the circumstances there would have
been a borrowing cost conformable with Chevron’s AA rating, which, on the
evidence, would have been significantly below 9%.
This decision means that the full federal court has now made it very clear that an
Australian subsidiary of a multinational group is not to be treated as if it were an
‘orphan’ when undertaking transfer pricing arm’s length calculations. In fact, some
level of parental support may need to be assumed to exist depending on the
facts/situation.

If a borrower is part of a group that has a policy to borrow externally at the lowest
cost and that has a policy that the parent will generally provide a third party
guarantee for a subsidiary that is borrowing externally, then in the case of related
party debt, the interest rate payable by an Australian subsidiary should probably
be set at a level that assumes a parent company guarantee has been given.
As a result, the appropriate interest rate on internal debt will likely be closer to the
parent’s global cost of funds for the relevant currency.
It remains to be seen whether pretending that the debt was raised with the benefit
of a guarantee given by the parent carries the further implication that the
borrower can also pretend it paid a fee for the benefit of that fictitious guarantee.
1.9 Dell Case (Ruling of Mar. 15. 2012, Central Tax Court, Spain)
Dell Spain is part of a multinational group (MNE) that manufactures products

outside Spain, with an entity of the group, IrelandCo, operating as distributor for
most of Europe. Dell Ireland has appointed related entities that operate as its
commissionaires in several countries; Dell Spain and Dell France are part of this
commissionaire network.
The MNE operates through a direct sales model so that purchase orders are
placed in a web page or call centre. Dell Spain had operated as a fully-fledged
distributor, undertaking strategic activities in Spain until 1995, when it entered into
the commissionaire agreement with Dell Ireland to serve medium and large sized
customers of the MNE. These latter customers, in many cases, require specialised
attention and Dell Spain’s relationship personnel are available to serve them. Sales
to Spanish small sized customers are made by Dell France, through a call centre
and a web page.
The Spanish court ruled that Dell Ireland had a Spanish PE, as the online store (the
website through which Spanish sales were made) could qualify as an ‘online PE’,
even though the server was situated outside Spain. Whilst the server was not
located in Spain, and there were no employees of the Irish company in Spain, Dell
Spain did employ people to translate the website, review the contents and
administer the site, which was considered a factor. In addition, the Court noted
the observations on the Commentary made by Spain in the 2003 and 2005
versions, which stated that Spain had a number of reservations on the OECD
approach. However, it is notable that on appeal, the Spanish National Appellate
Court (decision of June 2015) and the Supreme Court (in June 2016) decided the
matter with regard to the fixed place of business PE and the agency PE concept
(Dell Spain acted as commissionaire and Dell Ireland was bound by the contracts
under Spanish law), without referring to the creation of a ‘virtual PE’. It is interesting
that the Appellate Court did comment that according to the OECD Commentary
to the model treaty, an online website does not in itself have a location that can
constitute a PE, although the place where the server is located could constitute
one. The Supreme Court confirmed the point as to the commissionaire
arrangement, which is the issue of most importance in this case. In so concluding
the Spanish court differs from courts in France and Norway which have concluded
that commissionaire arrangements do not give rise to permanent establishments,
although notably such conclusions have been overtaken by the changes to the
Agency PE definition in the treaty put forward under the BEPS Action Point 7.

1.10 Dell (Norwegian Case) Supreme Court Ruling November 2011
The Norwegian subsidiary Dell AS (Dell Norway) performs two kinds of activities: on
the one hand, it sells accessories on its own account to end users (sales to small
companies and consumers in the Scandinavian market through a call centre in
Denmark) and, on the other hand, it sells Dell Ireland’s products under a
commissionaire agreement with it. This second activity is performed by Dell Norway
in its own name, but at the risk and on account of the principal (ie Dell Ireland).
This commissionaire agreement covers sales to large customers and customers in
the public sector. The commission for these services amounts to approximately 1%
of sales.
The Norwegian tax authorities assessed the taxpayer (ie Dell Ireland) on the basis
that it had a dependent agent PE in Norway, under Article 5, paragraph 5 of the
Ireland–Norway tax treaty, and that the profits to be attributed to the PE could be
determined by an indirect method of apportionment in accordance with Article 7,
paragraph 4 of the Ireland–Norway tax treaty. Therefore, they allocated 60% of
the profits to the Norwegian PE and 40% to the Irish Company.
Dell appealed.
On December 2nd, 2011, the Norwegian Supreme Court overturned the two
previous judgments and stated that Dell Ireland did not have a PE in Norway. The
arguments of the Supreme Court were based on the Vienna Convention, on
Article 5, paragraph 5 of the tax treaty, on the OECD Model and its commentaries,
and on case law. First of all, the court looked at the wording of the Ireland–Norway
tax treaty (in the English version) and stated that the expressions ‘acting on behalf
of an enterprise’ and ‘authority to conclude contracts in the name of the
enterprise’ clearly suggest that the contracts must be legally binding for Dell
Ireland, in order to have a PE in Norway.
Second, the court approached the OECD Commentary on Article 5, para. 5 of the
OECD Model and sustained that paragraph 32.1 was introduced under a
common law system point of view irrelevant to the Norwegian civil law approach.
Then, the court also referred to the above-mentioned French Zimmer case:
According to the French Supreme Court, a commissionaire acts in its own name
and cannot bind its principal. As a result, a commissionaire cannot constitute a
dependent agent PE of its non-resident principal, even if the commissionaire is
clearly not independent.
Finally, it was also stated that using a different approach from the legal and
formalistic one could involve substantial practical and technical difficulties related
to the uncertainty of applying a uniform practice to the other similar
commissionaire arrangements. Therefore, based on this reasoning, the Norwegian
Supreme Court affirmed that Dell Ireland did not have any PE in Norway and,
consequently, no income should have been assessed.
1.11 Director of Income Tax v M/S E-Funds IT Solutions
In February 2014 in a case before the Indian High Court it was held that an
(indirect) Indian subsidiary of a US company was not a PE, notwithstanding the
close association between the US company and the Indian company, the division
of functions, the assets used and the risks assumed. The High Court held that a
consideration of these factors was not the appropriate test to determine whether
a PE was present. The Court considered the PE definition in the US-Indian double
tax convention and focused on the conditions necessary to fulfil these PE tests. As
the Indian Revenue was not able to prove the ‘right to use’ test and having space

at ‘the disposal’ of the company test were satisfied, no fixed PE was found to exist.
Similarly, in the absence of any evidence that employees of the Indian subsidiary
were under the control or supervision of the US parent and providing services on
behalf of the US company, or that the Indian employees were participating in
negotiations with customers, no service PE or agency PE were found to exist either.
1.12 DSG Retail and Others v HMRC (DSG Retail 2009 UK FTT31 (TC))
DSG Retail and others v HMRC (TC00001) related to the sale of extended
warranties by one of the largest electrical goods retailers in the UK. As the first
instance of a tribunal considering application of ICTA88/SCH28AA, this case has
proved something of a watershed for HMRC, and while detailed analysis of the
supporting information presents a convoluted set of facts, a number of key
principles were identified in relation to profit allocations within a group.
Encompassing the brands Dixons, Currys and PC World, the Dixons Stores Group
(‘the Group’) included DSG International plc (‘the Parent’) and subsidiaries DSG
Retail Ltd (‘DSG’), Mastercare Coverplan Service Agreements Ltd (‘MCSAL’),
Mastercare Services and Distribution Ltd (‘MSDL’), and Dixons Insurance Services
Ltd (‘DISL’), the latter being the Group’s captive insurance company which was
resident in the Isle of Man. While not licensed by the Isle of Man regulators to write
insurance in the UK, DISL was authorised to write reinsurance business. Under the
Income Tax (Exempt Insurance Companies) Act 1981, DISL was exempt from Isle of
Man Corporate Tax.
Point of sale insurance-backed extended warranties were routinely offered on

electrical goods for a fixed premium, as an optional supplement to any
manufacturer's warranty. Two sets of arrangements were considered by the
Tribunal. The first, referred to as the ‘Cornhill period’, was in place between 1986
and 1997, and involved the sale of policies in DSG stores written by an
unconnected 3rd party insurer, Cornhill Insurance plc (‘Cornhill’), through its
fronting agent Coverplan Insurance Services Ltd (‘CIS’). CIS was remunerated by
an initial sales commission as well as a profit commission based on the eventual
level of underwriting profit. Any repairs under the extended warranty were carried
out by MSDL, which was paid an administration fee, deducted from the CIS
warranty contract price. Cornhill however only ultimately held 5% of the relevant
risk, reinsuring 95% with the DSG captive insurer DISL, which paid Cornhill a
commission of 1.5% of the premium being ceded. In 1993 the contract was
renegotiated, with Cornhill agreeing to increase the sales commission paid to CIS,
while the ceding commission received from DISL remained at 1.5%. The critical
assumption made by the taxpayer in relation to this arrangement was that DSG
had no contractual relationship with DISL, forming a fundamental part of DSG’s
defence against application of transfer pricing and CFC rules.
The second arrangement was established from April 1997, in response to the
increase in the rate of Insurance Premium Tax from 2.5% to 17.5% that was
announced in November 1996, and which would have reduced premium revenue
by 15%. In response to this the Group stopped offering insurance cover as part of
its extended warranties and started selling service contracts via a new 3rd party
company Appliance Serviceplan Ltd (‘ASL’), an Isle of Man company, with MCSAL
acting as its agent in the UK. As with the Cornhill period, the sales company
received a sales commission and passed on net premiums to an unconnected
third party company, now ASL. In this instance 100% of ASL’s liability was
subsequently insured by DISL, which changed its regulatory status to allow it to
undertake insurance business with ASL. This effectively circumvented the IPT rate
hike as the insurance element of the structure no longer fell into the scope of UK

tax, with the counterparties to the insurance policy both being Isle of Man
companies.
HMRC contended that the terms of the arrangement did not sufficiently recognise
the advantage gained by DSG at the point of sale, and that DISL subsequently
gained more of an advantage than would have been available in the case of an
arm's length agreement. While DSG had no direct legal contractual relationship
with DISL, the interposition of a third party (Cornhill/ASL) into the overall transaction
did not preclude the application of ICTA88, S773(4), whereby the opportunity for
DISL to able to enter into such an attractive insurance arrangement amounted to
DSG providing a ‘business facility’ to DISL. Additionally, on analysing the series of
transactions as a whole, the Commissioners found that the arrangements were
effectively integrated by means of an ‘understanding’ between the Group and
Cornhill/ASL – although the series of contracts were not in themselves technically a
‘provision’ for the purposes of Schedule 28AA, the means by which they were
applied equated to a provision ‘as between’ DSG and DISL, and the arm's length
principle therefore applied to the advantage gained by any party.
In summary from a transfer pricing perspective, the following issues were relevant:
1. Was there a provision between DSG and DISL for the purposes of Schedule
28AA ICTA?
2. Did the actual provision differ from the provision that would have been made
between independent enterprises?
3. Did the provision confer a potential advantage in relation to UK tax?
4. What adjustment is required?
Looking at the transfer pricing method used, DSG argued for the use of
comparable uncontrolled prices, and relying on transfer pricing reports produced
by a number of accounting firms, put forward a number of comparables in
support of its argument.
HMRC rejected the comparables for reasons such as:
• Time – the first comparable proposed concerned an agreement made in 1982

when the market for extended warranties was quite different to the market in
1997;
• Product – the second comparable proposed concerned an agreement

related principally to satellite equipment;
• Termination – the second comparable was terminable on one week’s notice

which was interpreted as demonstrating that the parties were not
contemplating a long term relationship, unlike the DSG arrangements;
• Terms – DSG sought to rely on a 1994 report from the Office of Fair Trading,
‘Extended Warranties on Electrical Goods’, which gave the commission rates
paid by three unidentified retailers. There was no evidence as to whether the
rates agreed were between arm’s length parties or what the other terms were;
and
• Bargaining position – DSG put forward a comparable involving a bargain

between equals. This comparable was rejected because this was not the
position with DISL.

On the basis that no comparables could be identified, HMRC’s expert witness set
out a profit split approach whereby DISL’s profits would be compared to a
notional ‘normal rate of return on investors’ capital’, based on the capital asset
pricing model. A ‘bottom up’ approach was preferred whereby it was assumed
any profits in excess of capital needed for solvency requirements was paid out as
a dividend each year. Solvency requirements were based on those contractually
agreed with Cornhill.
Following on from this case, the concept of ‘relative bargaining power’ now forms
an important element of HMRC’s operational guidelines for transfer pricing
enquiries. On this basis, particular care needs to be taken when considering
pricing of transactions, as well as identification of potential comparable
unconnected 3rd parties for comparability, to the extent that a tested party may
have a particularly strong or weak bargaining power when undertaking a
transaction at arm’s length rather than with a connected party.
What is apparent from the case is that transfer pricing issues are never clear-cut;
this particular hearing lasted 15 days and involved a lengthy and complex analysis
of the contracts and transactions involved. It has been suggested that it signals a
change in HMRC’s previous reluctance to litigate transfer pricing issues, which
serves to highlight the importance for groups to re-examine their transfer pricing
policies and consider whether the most appropriate pricing methodology has
been used. It also reinforces the importance of performing a sound economic
analysis of commercial agreements. HMRC will construe the arrangements as a
whole, and if it can show that the third party was engaged by a group entity only
on the understanding that the third party would, in turn, enter into a contract with
another entity from the group the transfer pricing legislation will apply.
1.13 Eli Lilly v Commissioner 856 F.2d 855
AUS corporation transferred patents and know-how to a Puerto Rica

manufacturing subsidiary. The IRS asserted that this transfer should be disregarded
on the basis that the US company could have retained the revenue streams from
the intellectual property transferred. That was rejected by the Tax Court and Court
of Appeals.
1.14 Ford Motor Company of Canada v Ontario Municipal Retirement

Board
In this case the minority shareholders brought the case questioning the level of the
transfer prices.
This case focused on the transfer pricing system between Ford U.S. and Ford
Canada. The system is a product of the Canada-United States Auto Pact and,
more recently, the free trade agreements that created a single, integrated market
for vehicles made and sold in Canada and the United States. To a large extent,
the structure of the system is tax driven. Canadian and U.S. tax regimes require
entities that do not deal with each other at arm’s length to attribute arm’s length
transfer prices to their goods and services to prevent entities from artificially
allocating losses in the high-tax regime and profits in the low-cost regime. For Ford
U.S. and Ford Canada, the transfer pricing system is the mechanism utilised to
comply with the tax laws in the two countries.
The transfer pricing system can impact on shareholders, such as the minority
shareholders of Ford Canada. If the system is unfairly skewed to assign losses to the
Canadian subsidiary, the subsidiary's minority shareholders will be deprived of their
fair share of Ford Canada’s profits. The parent, Ford U.S., will not be injured since it

will offset the loss from its Canadian holdings through increased profits in its U.S.
operations. With minor exceptions, the taxing authorities in the two countries have
not faulted the Ford transfer pricing system.
Following a lengthy trial, involving numerous experts and voluminous documentary

evidence, the court found that the transfer pricing system benefited Ford US, the
majority shareholder, by approximately C$3 billion between 1985 and 1995, while
depleting Ford Canada of those assets. The court held that the value of a Ford
Canada share in the absence of the flawed transfer pricing system would have
been between C$555 and C$610 per share, rather than the C$185 offered.
The court concluded that a poor transfer pricing mechanism that had been
adopted amounted to oppression of the minority shareholders since it frustrated
their reasonable expectation that the company would operate to maximize
profits. This case is indicative of how transfer pricing can be a factor in a matter
that is essentially one of corporate governance.
1.15 Formula One World Championship Ltd. v Commissioner of Income

Tax, International Taxation 3, Delhi & Anr. Appeal 3850 and 3851
(2017)
The case related to the UK company holding the commercial rights to stage, host
and promote the Indian Grand Prix in 2011 through to 2013 at a location owned by
an Indian company. The existence of a PE was upheld as the court found that the
racing circuit constituted a fixed place from which a business/economic activity
was conducted. In addition, the various agreements between the Indian
company and the UK company (together with its affiliates) could not be looked at
in isolation. In the Court’s view, the facts pointed to the conclusion that the UK
company had made its earnings in India through the racing circuit over which it
had complete control during the event. The duration of the event and the number
of days for which the UK company's personnel had access to the circuit would not
make any difference.
1.16 GE Energy Parts Inc v ADIT (ITA No. 671/Del/2011)
In July 2014 the ITAT (Indian tax tribunal) held that the LinkedIn profiles of the
employees of the GE group submitted by the tax authorities were admissible as
evidence in determining the existence of a PE in India. The tribunal’s view was that
the LinkedIn profiles were not hearsay but were akin to admissions made by a
person; and they had considerable bearing on the subject matter of this appeal.
However, the taxpayer is free to rebut the information contained in their LinkedIn
profile by bringing on record contrary facts to dislodge the claims made.
1.17 General Electric Capital Canada Inc. v. The Queen (2010) FCA 344
In this case, the Canada Revenue Agency denied the deduction of $136 million in
guarantee fees paid over several years by General Electric Capital Canada Inc.
(‘GEC’) to its US-based parent. The court found that the guarantee fee of 1% per
annum of the principal outstanding debt amount did not exceed an arm’s length
price. Consequently, the court ruled in favour of GEC and allowed the deduction.
The Parent company resident in the US guaranteed debt securities issued by a

Canadian subsidiary GEC in consideration for a guarantee fee equal to 1% per
annum of the principal amount of debt securities outstanding during a year. The
Minister disallowed the deduction of a total of $136 million in guarantee fees paid
to the Parent during its 1996 to 2000 taxation years.

The Minister argued that the guarantee provided no economic benefit to GEC
and, therefore, its arm’s length price would have been nil. GEC enjoyed a so-
called ‘implicit support’ from its Parent. The Parent could never allow its like-named
affiliate, such as GEC, to default on its debt because it would damage the
Parent’s own AAA credit rating and increase its borrowing costs significantly. Due
to this ‘implicit support’ GEC would have had the same credit rating, and could
have borrowed the same amount of money at the same interest rate without the
guarantee.
The Minister also added 5% withholding tax to the guarantee payments on the
basis that the payment of the guarantee fees was deemed to be a payment of
dividends which should be subject to withholding tax.
GEC argued that it paid an arm’s length price for the guarantee. The transfer
pricing rules invite the court to analyse a transaction between related entities as
though they were unrelated parties. Consequently, all distortions that arise from
the parties’ non arm’s length relationship, such as ‘implicit support,’ must be
ignored and the Minister’s position is unfounded.
GEC argued that the Minister’s approach is flawed because it did not adduce
evidence on what would be a price of a comparable arm’s length guarantee
arrangement, as required by the transfer pricing rules; the basis of the Minister’s
position was simply that the guarantee was not necessary in GEC’s business.
Finally, GEC presented expert evidence that without the guarantee its credit rating
would have been BB as opposed to AAA and its cost of borrowing would have
been higher.
After taking into account several expert opinions, the Tax Court of Canada
determined that the guarantee fee did not exceed an arm’s length price.
The court held that the factor of ‘implicit support’ is relevant in an arm’s length
analysis under the transfer pricing rules. It also noted that transfer pricing rules can
apply if the value of the benefit received from a transaction is nil and,
consequently, the transaction cannot be compared with an arm’s length
transaction. In its analysis, the court compared the borrowing cost of guaranteed
debt as opposed to the borrowing cost of unguaranteed debt. By relying heavily
on expert witnesses, the court determined that:
• GEC’s stand-alone credit rating without the guarantee and without the
‘implicit support’ from the Parent would have been B+ to BB-;
• When the factor of ‘implicit support’ from the Parent is taken into account,
GEC’s rating without the guarantee would only go up to BBB-/BB+; and
• The GEC’s interest cost savings afforded by the guarantee was approximately
1.83%. In this regard, the court also took into account that GEC’s investors
would not welcome the removal of the guarantee and would not have the
same confidence in unguaranteed debt. The court recognised that the
guarantee existed for many years and served a bona fide business purpose.
The ‘implicit support’ of the Parent could not replace the need for a
guarantee as few investors are foolish enough to believe that the it is
equivalent to a guarantee.
Based on this evidence, the court concluded that the 1% guarantee fee did not
exceed an arm’s length price because GEC ‘received a significant net economic
benefit from the transaction.’

Three transfer pricing-related issues in this case provided taxpayers with the
following lessons to be learned:
• whether implicit support from the parent-subsidiary affiliation would reduce

the net economic benefit to the subsidiary of an explicit financial guarantee,
• whether consideration of implicit support in determining the credit risk of the

subsidiary is consistent with the concept of the arm’s length principle, and
• provided the most appropriate transfer pricing methodology to determine an

arm’s-length guarantee fee for a financial (or loan) guarantee.
It is clear from this case that, in the assessment of the credit risk of the subsidiary,
the parent-subsidiary affiliation must be considered. This does not mean that this
assessment will conclude that the parent-subsidiary affiliation will always result in
the subsidiary’s credit risk being improved or that its credit risk could be equalised
with that of its parent’s external credit rating. The assessment of implicit support is
whether or not the debt market participants would, by reason of the parent-
subsidiary affiliation, charge the subsidiary a lower rate of interest (or some other
net economic benefit) without an explicit financial guarantee. It is not whether or
not the credit rating agencies would notch the stand-alone credit rating. This case
offers a glimpse into the subjective criteria that credit rating agencies use for
notching the stand-alone credit rating of the subsidiary for the parent-subsidiary
affiliation. However, this subjective process may not be entirely representative of
what lenders would consider in setting the interest rate.
1.18 GSK Canada 2010 [FCA 201] also known as The Queen v.
GlaxoSmithKline Inc
The Canadian Authorities argued that Glaxo Canada had overpaid its Swiss
affiliate Adechsa S.A. (‘Adechsa’) under a supply agreement for ranitidine.
Ranitidine is the active ingredient in the ulcer medication Zantac, which was
manufactured, marketed and distributed by Glaxo Canada under a licence
agreement with Glaxo Canada's parent, Glaxo Group Ltd. (‘Glaxo Group’). During
the years at issue, Glaxo Canada had paid Adechsa between $1,512 and $1,651
per kilogram of ranitidine. The Minister used subsection 69(2) (a precursor to
subsection 247(2)) of the Income Tax Act (‘ITA’)[2] to argue that this transfer pricing
was unreasonable based on the fact that other Canadian pharmaceutical
companies were paying, at a maximum, just over $300 per kilogram to their non-
affiliated suppliers for generic versions of the drug.
Under a separate licence agreement, Glaxo Canada was granted certain

intellectual property rights, including the right to sell the medication under the
Zantac trademark. In return, Glaxo Canada paid an annual royalty of 6% of its net
sales to Glaxo Group.
In the unanimous decision of the Supreme Court, Justice Rothstein’s reasons

support expansion of the factors that may be taken into account when
determining the reasonableness of transfer pricing. Since reasonableness is
relative: one must compare the amount paid by the entity to its affiliate to that
which would have been reasonable had the parties been dealing at arm's length
in similar economic circumstances. In order to assess this hypothetical amount,
however, a court should not be restricted to considering a particular transaction in
isolation if the facts do not warrant such a narrow inquiry. On these facts, for
example, any definition of reasonableness would need to account for agreements
between the parties (such as the licence agreement) outside the transaction for
the purchase of ranitidine itself.

Noting that subsection 69(2) of the ITA does not define ‘reasonable’, the Supreme
Court considered the reliance by the lower courts on the 1995 OECD Transfer
Pricing Guidelines for Multinational Enterprises and Tax Administrations (the ‘OECD
Guidelines’). The Supreme Court cautions that the OECD Guidelines are ‘not
controlling as if they were Canadian statute’; determination ultimately rests with
the provisions of the ITA. Justice Rothstein acknowledged that the OECD
Guidelines suggest four methods for determining the reasonableness of a transfer
price: (i) the comparable uncontrolled price (‘CUP’) method; (ii) the cost plus
method; (iii) the resale-price method; and (iv) the transactional net margin
method. The Supreme Court made no definitive determination as to the
appropriateness of any of these four methods in the circumstances, leaving any
determination for re-hearing at the Tax Court.
Justice Rothstein rejected the strict application of a ‘transaction-by-transaction’

approach stating: ‘while a transaction-by-transaction approach may be ideal,
[the OECD Guidelines] themselves recognize that it is not appropriate in all
cases.’[5]
Furthermore (and again quoting the OECD Guidelines), Justice Rothstein noted
that ‘an arm’s length comparator is only useful if economically-relevant
characteristics… [are] sufficiently comparable.’
The Supreme Court concluded that the licence agreement between Glaxo Group
and Glaxo Canada must be taken into account to determine whether the transfer
price for the ranitidine purchase is reasonable. It was because of the licence
agreement that Glaxo Canada was purchasing its ranitidine from its parent’s
preferred source: Adechsa. Without the licence agreement, Glaxo Canada would
not have been able to manufacture, market or distribute the ranitidine it received.
Also, the licence agreement allowed Glaxo Canada to avoid the risk involved with
product development. The Supreme Court concluded that these rights and
benefits ‘could not be irrelevant’ in an assessment of whether the higher price was
reasonable as compared to generic peers. However, the Supreme Court stopped
short of determining whether the transfer pricing was, in fact, reasonable.
The Supreme Court offered the following guidance for the Tax Court in
redetermining whether the transfer prices in this case are reasonable. First, as
transfer pricing ‘to use the words of the [OECD Guidelines] “is not an exact
science”,’ ‘reasonable amount’ should be interpreted as ‘reasonable range’. If
the transfer price used by Glaxo Canada is not within this reasonable range, then
‘the court might select a point within a range it considers reasonable in the
circumstances based on an average, median, mode or other appropriate
statistical measure, having regard to the evidence that the court found to be
relevant’. Second, the Tax Court must look at the respective roles of Glaxo
Canada and Glaxo Group to determine whether a higher price was justified on
the basis of compensation for intellectual property rights. Third, both the interests of
Glaxo Canada and Glaxo Group must be considered by the Tax Court having
regard to their independent interests. Lastly, the fact that arm’s length parties
have purchased ranitidine from Glaxo Group suppliers (rather than from a generic
source) indicates that some amount of price increase will be justified.
In dismissing the cross-appeal, the Supreme Court held that while Glaxo Canada
had successfully demolished the Minister's assumption that the transfer price paid
to Adechsa for the purchase of ranitidine should not be greater than the prices
paid for ranitidine by the Canadian generic pharmaceutical companies, it had
not demolished the Minister’s assumption that the amounts paid to Adechsa were
unreasonable. On this basis, the Supreme Court remitted the matter to the Tax
Court for redetermination.

Generally, the Supreme Court’s requirement for a fact-driven, holistic approach to

assessing the reasonableness of transfer pricing will give MNEs some leeway in
structuring arrangements within their corporate groups.
Based on Justice Rothstein’s reasons, it seems clear that the OECD Guidelines are
not determinative for purposes of interpreting the transfer pricing rules in the ITA. In
fact the only statement from the OECD Guidelines that the Supreme Court
appears to have confirmed was that determining arm’s length transfer prices is not
an exact science. As a result, even though the Tax Court favoured the CUP
method of transfer pricing analysis, the Supreme Court indicated that this may not
be the best method for analysis in this case, due to its narrow focus on particular
transactions. Since the Supreme Court declined to elaborate further, the door
appears to be open to any, or none, of the four OECD methods being acceptable
for determining a reasonable transfer price, depending on the circumstance of
the particular taxpayer as compared to that taxpayer’s economic comparator
group.
1.19 GAP International Sourcing (India) PvT. Limited v CIT (2012)
GAP International Sourcing provides procurement services for its group in India. The
Indian tax authorities sought to challenge the company’s transfer pricing policy of
a mark-up on value added expenses, preferring a commission of 5% of the Free on
Board price. The taxpayer’s position was upheld as the Tribunal found no evidence
of local intangibles that would move its transfer pricing method away from cost
plus and that any location savings would be passed on to customers by a third
party.
1.20 GlaxoSmithKline Holding (Americas) Inc v Commissioners
The Glaxo group recently settled a transfer pricing dispute in the US for $3.4 billion.
The magnitude of this settlement helps illustrate the scope of the problem in
valuing IP and exploiting it correctly without triggering potential tax avoidance.
Glaxo is headquartered in the United Kingdom and holds several subsidiaries in the
US. Glaxo's primary business is the development and manufacturing of
pharmaceutical drugs. Cross-border transactions of valuable pharmaceutical
drugs generating large profit margins have attracted the attention of revenue
authorities.
In 2000, when the predecessor of Glaxo (GlaxoWellcome) merged with SmithKline

Beecham to form Glaxo, the merger triggered a transfer pricing audit in the United
States. Glaxo also faced transfer pricing audit adjustments in Canada and Japan.
Glaxo’s sales of drugs in the United States generated almost $30 billion in revenues
from 1989 to 1999. During this period, Glaxo paid about $1.3 billion in U.S. taxes.
Glaxo claimed that the United Kingdom had already taxed the MNE Group’s
profits under dispute with the IRS, arguing that any reallocation by the United
States would result in double taxation of Glaxo.
Approximately seventy-five percent of Glaxo’s income in the United States was

attributable to Zantac.
The drug had been patented in the UK and hence, the US subsidiary was acting as
distributor for the US market. However, the IRS argued that the US subsidiary of
Glaxo overpaid its UK parent for the patent it held. The IRS also argued that
marketing efforts by the US subsidiary were the determining factor in the success of
Zantac. Also, as the US was the largest market for the drug, which was also

manufactured in the US, the economic ownership of the IP was challenged. The
IRS demanded about $8 billion in tax adjustments and penalties.
Glaxo tried to reach settlement with the IRS by referring the dispute to a
competent authority under the MAP procedure. The governmental discussions did
not reach common ground and the IRS took Glaxo to court to preserve evidence
in preparation for the anticipated trial.
In settling the Glaxo case, IRS Commissioner Mark Everson stated that transfer
pricing issues ‘are one of the most significant challenges’ tax agencies face.
1.21 Glencore [2020 FCAFC 187]
This is an Australian case that looks at an intra-group agreement for the purchase
of copper concentrate by Glencore International AG (GIAG), a Swiss company,
from its subsidiary, Cobar Management Pty Ltd (CMPL), an Australian company
which owned and operated a mine in Australia.
Pricing arrangements under the contract for the sale of the copper concentrate
were changed in 2007; the taxpayer stated that change was necessary as there
was uncertainty in the price of copper and an increase in costs associated with
the mine.
Before 2007, the intercompany contract between GIAG and CMPL only
compensated GIAG for its very modest marketing functions as CMPL was
responsible for paying for the actual cost of smelting. From 2007 to 2009, GIAG
paid third-party smelters and charged CMPL 23% of revenue in a price
participation arrangement. The taxpayer contended that the new arrangements
would give CMPL a better risk position. The Australian Tax authority (ATO) thought
the change was excessive and that it shifted too much profit to GIAG.
The ATO wanted to amend the pricing mechanism to what they thought
independent enterprises would have entered into. In addition to changing the
pricing mechanism the ATO sought to add additional clauses to the contract.
Amended assessments were issued to reflect the mechanism previously used and
increased the consideration paid under the contract by AUD241 million over the
years 2007 to 2009.
The taxpayer defended the change in arrangements by providing evidence of

similar contracts from 2003. The ATO argued that these contracts were not
comparable to a contract entered into in 2007. The court recognised that the
comparables were not perfect but stated that they still provided appropriate
reference points.
The ATO and the taxpayer presented evidence to the court from experts. The
experts did not agree on the pricing that independent parties would have agreed
on. The court took the view that a taxpayer did not have to demonstrate its pricing
is more likely than that put by the ATO or ‘to lead evidence to negate positive
claims put up by the Commissioner’, they just have to show through the evidence
that its position is within a range of ‘commercially acceptable arm’s length
outcomes’.
A key aspect of the case was how far the tax authority could go in making
changes to the intra-group agreement. In this case the transfer price was based
on a formula; the court stated that if required an alternative formula could be put
forward by the tax authorities to meet the arm’s length principle however

additional clauses that didn’t affect the calculation of the consideration may not
be added. The characteristics of the hypothetical independent parties were also
examined; it was determined that the characteristics assigned to the hypothetical
parties must only be those that would impact on the consideration payable under
the hypothetical contract only – objective attributes and features should be
assigned. Any features of the agreement that arose from the special relationship
that exists between group members shouldn’t be included. Risk was an important
part of the formulation of the consideration thus it was appropriate to take
account of how a hypothetical party would take account of the risk when
determining the mechanism for pricing. It was agreed that there could be a range
of arm’s length outcomes that could meet the requirements of the law in Australia
as what controls the range of acceptable arm’s length outcomes is the concept
of what might reasonably be expected; and that a degree of flexibility and
pragmatism is required in determining if the taxpayer has met the requirements of
the law.
The court stated that the OECD TPG are merely ‘guidelines’, stating that the
language in them is ‘highly generalised’ and ‘frustratingly opaque’. It was noted
that ‘reconstruction’ (now recharacterisation) was not limited to circumstances in
the OECD TPG.
The final decision on this case was made by the Federal court in November 2020.
The court agreed that the evidence of the taxpayer’s expert should be accepted
and that the rate of 23% used in the agreement was within an arm’s length range.
(An appeal was denied by the High court in May 2021).
Important points that arise from this case include:
• Substituting pricing methodology is permitted (as discussed above).
• Substituting terms which do not define price is not permitted (as discussed
above).
• Commercial prudence is sufficient. A taxpayer will succeed if it can

demonstrate that its position is a ‘sufficiently reliable prediction’ which is
reasonable, even where the ATO attempts to assert that it is instead
unreasonable.
• Taxpayers are not required to maximise profitability at the expense of risk

mitigation. The court recognised that business decisions occur with a range of
considerations in mind, and that there can be a trade-off between risk and
profits.
• It is not appropriate to rely on estimates or forecasts as certainties. This will be

the case even if in hindsight they turned out to be correct.
• Comparables that are not identical are still important as they can provide a
valid reference point.
• In Australia the OECD TPG do not replace the text of the statutes. The OECD
TPG are only a guide as to how a revenue authority or a taxpayer might apply
the ‘arm’s length principle’. They do not replace the text of the legislation in
Australia.
• Prior contracts are not determinative of whether the contract in dispute is

arm’s length. It was held that the ultimate issue for determination is whether
the contract in dispute was priced at arm’s length, not whether an arm’s
length party would have agreed to amendments to a prior agreement.

• The Australian courts will take a rigorous approach to evidence (this is in line
with the Chevron decision – see earlier chapter), thus taxpayers should ensure
they have good documentary evidence of the analysis that was undertaken
to arrive at pricing.
1.22 Re Japanese Taxation of Internet Sales, (19 ITLR 346)
The Japanese High Court considered the position of a US resident company which
sold automobile accessories over the internet. The US company had an apartment
in Japan, the address of which was listed on its internet site. The apartment was
used for packaging products, but also for inserting Japanese-language manuals in
the products and for receiving returns of defective goods. Subsequently, it hired a
warehouse in Japan. The Japanese tax authorities concluded that it had a
permanent establishment in Japan through the apartment complex and the
warehouse, and, in the absence of documentation of the profits, attributed profit
to the notional taxpayer formed by the permanent establishment on a
comparison with other similar enterprises. The taxpayer appealed arguing that it
had no permanent establishment in Japan by virtue of Article 5(4) of the US-Japan
tax treaty which was identical to the OECD Model.
A central issue in the appeal was whether the use of facilities for storage, display or
delivery in para (a) of the Article also needed to be ‘preparatory or auxiliary’. This
was a matter of some discussion, and the Japanese court concluded that the
activities must be preparatory or auxiliary for the exemption to apply. There is a
contrary view that paras (a) to (d) are ‘per se’ preparatory and auxiliary, and that
only paras (e) and (f) require activities to be preparatory or auxiliary. The Court
also commented on the role of the OECD Commentaries, and of the 2012 OECD
Discussion Draft on Permanent Establishments in this context. In addition, the Court
discussed whether it was significant that Japanese-language manuals were
inserted in the products at the locations in Japan, and that the address of the
premises was listed on the website of the US business.
1.23 Kammarrätten i Göteborg, ref KRG 2276-15
The Swedish Administrative Court of Appeal considered the issue of a physical PE.
The Court determined that a foreign company which regularly conducted
business from the same place in Sweden had a PE there. The case concerned a
German company that developed and sold software for tyre inflation pressure
systems. The company annually performed tests in winter conditions in Sweden,
and the test results were then used for software development in Germany. The
annual testing period ranged from three to four months, but the company was
only on the ground for a few weeks at a time.
The Court decided that the company regularly conducted business from the same
place in Sweden; and that the activities that it undertook, ie the testing, could not
be considered to be of a preparatory or auxiliary nature. Therefore, the company
was considered to have a fixed place of business in Sweden through which part of
the company's core business was conducted. Consequently, it was determined
that there was a PE in Sweden.
1.24 KHO:2010:73
A Finnish company replaced its external borrowings, on which it was paying

interest of a little over 3%, with an internal loan from a Swedish member of the
same group, on which the interest rate was 9.5%, reflecting the cost of external
funding of the group. The court confirmed that the price of external financing for
the group was not a relevant basis for determining the interest rate that should be

paid by the Finnish company, when, on a stand-alone basis, the borrower would
have received significantly better terms given its own credit rating and other
circumstances. The borrower's financial position had not deteriorated and the
Swedish lender was not providing any additional services that would have justified
a higher rate.
1.25 Kwiat v Commissioner 64 TCM (CCH) 327
In this 1992 case a purported lease with reciprocal put/call options was
recharacterised as a secured loan. This was not a transfer pricing case as such
because the parties to the transaction were not associated enterprises. However,
it serves as a contextual reminder of the need to consider all possible legal tools at
the disposal of the tax authority which might ultimately result in the disregard or
recharacterisation of a transaction. In the Kwiat case, the appellant taxpayers
leased shelving equipment to another party. There was a put option permitting the
taxpayers to sell the equipment at a projected profit to the taxpayers.
The Tax Court held that the rights and responsibilities of ownership of the shelving
had passed to the purported lessee: the lease was in substance a sale and the
taxpayer was denied tax depreciation in respect of the assets in question.
1.26 Lankhorst Hohorst C-324/00
This case looks at thin capitalisation rules and compatibility with EU law. The case
involved payment of interest from a German company to a non resident (Dutch)
grandparent company. Thin cap rules applied to limit the deduction in the
German company where the company was thinly capitalised and the loan was
not on arm’s length terms. However these rules only applied where the interest was
paid to a non resident (and certain non-CT paying domestic entities), interest
payments to other German companies were not caught by the rules. The ECJ held
that such rules were contrary to the freedom of establishment, as they
discriminated against shareholder companies based in other EU States. However,
the tax authorities put forward various justifications. The tax avoidance justification
did not work, as the rules were too broadly drafted and did not target wholly
artificial arrangements.
1.27 LG Electronics India Pvt. Limited v ACIT (2013)
LG Electronics India, a wholly owned subsidiary of LG Electronics, Inc. (a Korean

company, ‘LG Korea’), was manufacturing and selling LG products in India. LG
Korea charged LG India a 1% royalty for the use of industrial property rights,
designs, and technical know-how to manufacture the LG Products. LG Korea also
provided LG India with certain marketing subsidies, such as funds toward
sponsorship of a global cricket event, to assist it with this presumably global
branding event. There were no intercompany agreements for these subsidies or
any provisions for brand-building expenses.
LG India was to market and exploit LG products in the Indian market.
LG India was quite profitable, and considered more so than its perceived
comparable companies, notwithstanding that it incurred significant Advertising
Marketing and Promotion (‘AMP’) expenses. For the years in issue, its AMP
expenses were 3.85% of its gross sales.
During the audit, the TPO questioned this level of AMP expense, specifically
whether LG India was incurring significant expenses that directly added to the
brand value of the trademark owned by LG Korea.

The TPO asserted that two comparables, Videon Appliances Ltd. and Whirlpool of
India, had ratios of AMP expenses to gross sales of .12% and 2.66%, respectively, or
an arithmetic mean of 1.39%.
The TPO applied the ‘bright-line’ test of DHL 6 and asserted that 1.39% of these
AMP expenses were routine and, therefore, deductible expenses, and the
balance of 2.46% (3.85%-1.39%) of these expenses were brand-building, non-
deductible expenses that LG India incurred on behalf of LG Korea. The issues
before the court were:
• Whether, on the facts and circumstances of the case, the Assessing Officer
(AO), who referred the international transaction to the TPO, was justified in
making a transfer pricing adjustment in relation to AMP promotion expenses
that LG India incurred
• Whether the AO was justified in holding that LG India should have earned a
13% mark-up on the AMP expenses alleged to have been incurred for and on
behalf of LG Korea.
In its detailed analysis, the ITAT New Delhi Branch essentially ruled in favour of the
government, holding that the transfer pricing adjustment and mark-up were
appropriate, but remanded the case to the TPO for a more detailed comparability
analysis. Although remanded, the court’s analysis provided more guidance than in
the seminal case of Maruti-Suzuki, yet still seemed to circumvent certain critical
issues.
The court’s view was that the AMP expenses that LG India incurred on behalf of LG
Korea to build its brand were more than what independent third-party entities
would incur proportionately in a similar ‘transaction’. The court was not persuaded
by the lack of a formal agreement for incurring these expenses. Rather, it inferred
this intent and agreement from the facts and circumstances and, more
significantly, from the parties’ conduct. Here, the court seemed to believe that the
parties’ common objective and purpose were controlling as to the economic
substance of the transaction, but it never spelled out what were the compelling
facts for ‘acting in concert’.
The bright-line test came under scrutiny in the cost analysis. Here, the court
rejected the parties' various arguments that the bright-line test was not a
sanctioned transfer pricing method, but rather merely saw the test as a tool in the
cost allocation analysis to determine routine versus non-routine brand-building
expenses. Presumably, the court found this tool necessary because the parties
failed to specifically earmark the non-routine portion of its own AMP expenses.
However, while the court upheld the concept of the bright-line test, it was
uncomfortable with the TPO offering just two firms as the comparability measure
and remanded this issue to the TPO for consideration.
In determining the actual cost base, the court also determined the scope of the
AMP expenses. Noting that a portion of the AMP expenses were perceived as
brand-building expenses, the court went to great lengths to find that a ‘bald
comparison of the ratio of AMP expenses to sales’ can lead to an incorrect result.
Here, the court focused on selling expenses and correctly noted that they should
not be considered AMP expenses. Further, the court said that any and all subsidies
to LG India should be given true effect, ie reducing the cost base for non-routine
intangibles. Indeed, this was a message to the TPO to consider on remand. Next,
the court focused on the arm’s-length price for the assessed non-routine brand-
building expenses. The court considered this a marketing service by LG India on
behalf of LG Korea and agreed that the cost-plus method was appropriate. The
TPO calculated a mark-up of 13% over those costs.

The court rejected the taxpayer’s argument that LG India’s operating profit was
higher than that of the comparable firms and that any additional profit from a
marketing service fee was, therefore, unnecessary and duplicative. The court
stated that each transaction must be considered separately notwithstanding the
overall level of profitability. Hence, that LG India’s operating profit exceeded that
of the comparable firms was irrelevant.
The court also rejected the 13% mark-up simply for lack of a reasonable if not
robust comparability analysis and remanded that issue to the TPO.
Finally, the court considered Maruti Suzuki and noted that the principles observed
in that case are still held to be correct: (1) brand promotion expenses are an
‘international transaction’; (2) AMP expenses incurred on behalf of a foreign
enterprise should be compensated at arm’s length; and (3) identification of the
factors required to be considered for determining an arm’s length price.
1.28 Maruti Suzuki India v ACIT
Maruti Suzuki India Limited (‘Maruti’) manufactures passenger cars and spare parts
in India. Suzuki Motor Corporation, Japan (‘Suzuki’) holds majority shares in Maruti.
The trade mark/logo ‘M’ is the registered trade mark of Maruti. In 1992 Suzuki
entered into an agreement under which Suzuki agreed to grant a license to Maruti
for manufacturing and sale of specified models of cars. Under the said agreement,
Maruti was obligated to use the trade mark ‘Maruti Suzuki’ on all the products and
parts manufactured pursuant to this agreement. Further, since 1993, Maruti
replaced the logo ‘M’, logo of Maruti by ‘S’, logo of Suzuki on the front of the cars
manufactured and sold by it. At the same time it started using the ‘Maruti’ mark
along with the word ‘Suzuki’ on the rear side of the vehicles.
Maruti made payments to Suzuki for the use of their brand. The Indian tax
authorities said that Maruti should receive payments for using the Suzuki brand as
they were effectively giving the Maruti brand to Suzuki. The Maruti brand was
stronger than the Suzuki Brand in India at the time. Secondly the authorities said
that as Suzuki were penetrating the Indian market by ‘piggybacking’ on to the
Maruti brand, they thought that Maruti should receive something like a royalty fee
as a result. Thirdly it also held that the advertisement expenses incurred by Maruti
had gone to benefit Suzuki.
Maruti contended that at no time had there been a transfer of the Maruti brand to
Suzuki. Suzuki did not have a right to use the trademark, and the trademark could
be transferred only by a written instrument of assignment.
Further, Maruti asserted that it had received a large benefit from Suzuki while
Suzuki had received no benefit, and that because it had a right to use the Suzuki
trademark in the future, Maruti received the benefit of its advertising.
The case talks a lot about the approach taken by the government official, the TPO
(transfer pricing officer). The court found that the TPO can reject the price
computed by the assessment person only if he finds that the data used by the
assessment person is unreliable, incorrect or inappropriate or he finds evidence,
which discredits the data used and/or the methodology applied by the
assessment person; further the Transfer Pricing Officer (TPO)/Assessing officer (AO)
is obliged to give the assessment person an opportunity to produce evidence in
support of the arm’s length price and before making adjustments, he is obliged to
convey to the assessment person the grounds on which the adjustment is
proposed to be made and give the assessment person an opportunity to
controvert the grounds on which the adjustment is proposed.

In considering payments made for using the S logo on the products the court
stated that it is important to consider whether the use of the trademark was
discretionary or mandatory. All factors of the agreement needed to be taken in to
account including the value of marketing intangibles.
Turning to the expenditure incurred by a domestic Associate Enterprise on

advertising of its products using a foreign trademark, this does not require any
payment or compensation by the owner of the foreign trademark/logo to the
domestic entity on account of use of the foreign trademark/logo in the advertising
undertaken by it, so long as the expenses incurred by the domestic entity do not
exceed the expenses which a similarly situated and comparable independent
domestic entity would have incurred. Where they exceed this amount all factors
need to be taken into account to determine the ALP. In this regard the TPO needs
to identify comparables and make the required adjustments.
1.29 Medtronic Inc. & Consolidated Subsidiaries v. Commissioner (T.C.

Memo. 2016-112)
In June 2016 the US Tax Court ruled in favour of the taxpayer in Medtronic Inc. &
Consolidated Subsidiaries v. Commissioner (T.C. Memo. 2016-112)
The IRS contented that the royalties payable from a Puerto Rico affiliate to
Medtronic, Inc should be increased. by $US1.4 billion. The Tax Court rejected the
IRS’s proposed transfer pricing method as arbitrary, capricious, and unreasonable.
The court instead accepted Medtronic’s use of the Comparable Uncontrolled
Transaction (CUT method) but made adjustments to take account of differences
between the licenses and the comparable transaction relied on by Medtronic.
The IRS argued that the Puerto Rico company was merely an assembler which only
had a minor role in design manufacturing and development. The court disagreed;
they found that the company in Puerto Rico was involved in every step of the
manufacturing process and that the functions it performed were critical to overall
profitability of the group. It was found that product quality was a critical factor. An
interesting point is that the court found that a license could add to the assets of
the licensee. The rights under the license can become an intangible asset for the
licensee.
1.30 Mentor Graphics (India) Private Limited
This case looked at the selection of appropriate comparables and related

analysis.
The taxpayer, a company incorporated under the Indian Companies Act, is the
wholly owned subsidiary of IKOS Systems Inc., a company incorporated in USA and
engaged in the business of software development and also rendering marketing
systems services to the parent company. The taxpayer filed its return of income for
the year under consideration on October 31, 2002 declaring total income at INR
3,99,080/- for the FA 2001-02. The income disclosed included profit from export of
computer software to its parent company for which deduction was claimed.
Software is developed only as instructed by its parent (associated enterprise (AE)).

It does not create/develop/sell software products and packages. The software
developed by the appellant is used by the parent AE in-house for integrating the
same with other software components developed by the parent AE itself. The
whole software in turn supports the hardware manufactured by the parent, and is
sold as a package in the open market by the parent AE. Therefore the appellant’s
business is limited to providing services of software development support.

The taxpayer selected TNMM as the most appropriate TP method. The Indian
authorities stated that the amount for export of software development services
was not arm’s length; they were happy with the amounts for export of marketing
support services.
The taxpayer searched public databases as required by law and carried out
quantative analysis narrowing the selection to 16 companies. As the NCP (Profit
Margins) of above companies as per average arithmetic mean was 13.41%
against 11.07% earned by the taxpayer in the relevant assessment years, it was
claimed that taxpayer carried out international transactions at arm’s length.
The Indian authorities (TPO) raised objections to the comparables chosen and the
years looked at as comparables.
The Delhi branch of the Indian tribunal system rejected the adjustments that the
TPO tried to make to the figures submitted by the tax payer.
Emphasis was placed on selection of comparable companies based on

comparison of economically significant activities and responsibilities of the
independent enterprises vis-à-vis the taxpayer. The Tribunal emphasised the
importance of comparability adjustments with specific reference to adjustments
for differences in working capital, risk and R & D. Further, the Tribunal observed that
as long as the taxpayer is within the arm’s length range, the onus to prove an
otherwise scenario rests with the Revenue. In the Tribunal’s view, it is not necessary
for the taxpayer to satisfy all points in the range; even if one point is satisfied, the
taxpayer has established its case.
Commentators have noted that by ignoring high profit and high loss making
companies in the comparable set, the Tribunal's observation on the arm's length
range is commensurate with use of an inter-quartile range as prevalent in other
jurisdictions.
1.31 Microsoft Corp v Office of Tax and Revenue
In this case the IRS contract auditor applied a CPM analysis comparing profit-to-
cost ratio of Microsoft with the profit-to-cost ratio of businesses chosen as
comparables. However, the auditor aggregated both controlled and uncontrolled
transactions of Microsoft. Columbia Judge found that there was no justification for
such aggregation which rendered the analysis ‘arbitrary, capricious and
unreasonable’.
1.32 National Semiconductor (NSC) and Consolidated Subsidiaries v IRS
NSC was engaged in the manufacture of a variety of electronic products for use
by consumers, industry, and Government. Those products included IC’s, discrete
devices, hybrid circuits, electronic displays, module components, calculators,
digital watches, and other similar products.
During the late 1970s, the management of NSC decided that NSC needed to
enter the large-die market in order to continue growing. Large amounts were
spent on R&D.
NSC and other U.S. semiconductor manufacturers began moving their

semiconductor packaging operations to subsidiaries in Asia in the late 1960s. This
allowed them to take advantage of lower-cost labour and overheads and of the
tax and other investment incentives provided by local Asian governments. It was
essential for NSC to achieve labour cost savings by locating its packaging

operations in Southeast Asia, and the Asian subsidiaries were dependent on NSC
for a secure source of semiconductor dies to justify their substantial investment in
assembly equipment, packaging methods, and personnel.
The Asian subsidiaries performed semiconductor packaging and associated

activities at several plants in Malaysia, Singapore, Hong Kong, Thailand, Indonesia,
and the Philippines. Unaffiliated IDM’s performed a small amount of packaging for
NSC.
The Asian subsidiaries were responsible for packaging some of the products.
Several of the Asian subsidiaries held dies (used in the packaging) and finished
goods inventory. The Indian Subsidiaries financed inventories, held by them, of the
dies and sometimes of the finished goods. In addition, the Asian subsidiaries bore
the cost of shipping finished devices. On the whole, the Asian subsidiaries had
successful digital and linear lines and were efficient cost-competitive packagers.
The Asian subsidiaries increased the efficiency of the packaging through process
improvements, which, among other things, improved output per operator.
Like most organisations that produce a large number of individual products using
processes that are both complex and relatively standardised, NSC used a
standard cost system for product costing. It assigned a specified cost to each
material component and labour operation that was required to complete each
stage in the production process. A standard amount of manufacturing overhead
costs was also applied.
The total standard cost was computed as the sum of the material, labour, and
overhead costs when the product was completed. ‘Overhead’ costs were those
indirect costs that were most directly identifiable with the manufacturing activities
and were allocated to production on a unit-by-unit basis. Indirect costs that were
related to manufacturing activities, but not identifiable with specific units of
production, were classified as ‘manufacturing period’ expenses. The balance of
indirect costs were commonly classified as non-manufacturing period expenses,
engineering, R&D, selling, and general and administrative expenses. Any
difference that was identified as a result of comparing the standard cost of
producing a specific quantity of a product with the actual cost of producing that
quantity was called a variance. Variances had two potential causes: either the
standard cost was not accurate or production process irregularities resulted in
changed actual costs of production. To determine which was the cause required
a detailed investigation.
For financial and tax reporting purposes, NSC treated as sales (1) the transfer of
semiconductor dies in wafer form and associated materials to the Asian
subsidiaries (outbound sales) and (2) the transfer of assembled devices from the
Asian subsidiaries back to sales and marketing affiliates in the United States
(inbound sales) or to affiliates in third countries.
The issue presented to the Court was whether the transfer prices that were
charged between NSC and its Asian subsidiaries met the arm’s length standard of
section 482.NSC claimed to have proven that the determinations made by the IRS
were unacceptable and to have presented comparable transactions between
unrelated parties and industry data which proved that its transfer prices satisfied
the arm's length standard.
The IRS claimed that NSC had not presented comparable uncontrolled prices to
prove that its transfer pricing system should be upheld.

Before trial, NSC filed a memorandum requesting that the burden of proof be
shifted to the IRS with regard to certain allegations in the IRS’s amendments to
answer, pertaining to methods of allocation based on outbound sales prices,
because they were beyond the scope of the notices of deficiency.
Both sides called on expert witnesses who gave detailed calculations and
explanations of how the transfer price should be calculated.
The court ruled that because evidence presented by each side demonstrated
that the notices were unreasonable, the determinations in the notices were
arbitrary, capricious, or unreasonable.
With regards to the expert witnesses the court said:
‘Because proper income allocations cannot be determined from the

transaction evidence presented by the parties, we must look to opinions of
their experts. As we have frequently stated, “we are not bound by the opinion
of any expert witness. We may accept an expert's opinion or we may reject
testimony that is contrary to our own judgment’. See, eg, Estate of Hall v.
Commissioner, 92T C. 312, 338 (1989). Further, ‘We are not restricted to
choosing the opinion of one expert over another, but may extract relevant
findings from each in drawing our own conclusions’. Bausch & Lomb, Inc. v.
Commissioner, 92 T.C. 525, 597(1989), affd. 933 F.2d 1084 (2d Cir. 1991).”
The court allowed some amendments to the transfer price; however the majority
of the location savings were allowed to remain in Asia.
1.33 Philip Morris Case
One of the best known cases on PEs heard before a European Tax court is the
Phillip Morris case heard by the Italian Supreme Court (L Ministry of Finance (Tax
Office) v Phillip Morris GMBH Corte Suprema di Cassazione 7682/02 25th May 2002).
Phillip Morris GMBH, a company tax resident in Germany, received royalties from
the Italian Tobacco Administration for a license to produce and sell tobacco
products using the Phillip Morris trademark. The execution of the agreement was
supervised by Interba SPA, a group company resident in Italy. The company
performed agency and promotional activities for Phillip Morris in duty free zones. Its
other main activity was the manufacture and distribution of cigarette filters.
The Italian tax authorities argued that Interba Spa was a PE of the group as it
participated in the royalty agreement negotiations as well as other group business
activities with no remuneration. Accordingly the royalty income should be
allocated to a PE of Phillip Morris Gmbh. They also argued that the Italian
subsidiary had been formed to avoid a PE.
The Italian Supreme court found a PE existed. The activity could not be considered
auxiliary for the purposes of Article 5 of the German/Italian tax treaty (similar
provisions are contained in the model tax treaty). It was found that participating in
contract negotiations can be construed as an authority to conclude contracts. A
PE will also be established where a principal entrusts some of its business operations
to a subsidiary.

1.34 Ranbaxy Laboratories Ltd (Delhi Tribunal) 110 IRTD 428
This case looked at the transfer pricing analysis conducted by the taxpayer, such
as selection of tested party, aggregation of transactions, selection of overseas
comparables, etc. The Tribunal also made comments on the disclosure norms in
the accountant’s report and the inadequate disclosures made by the taxpayer.
Ranbaxy Laboratories Limited is a company registered in India engaged in the

business of manufacture and sale of pharmaceutical products. During the
financial year (FY) 2003-04, the taxpayer exported goods and services to its
associated enterprises (AEs). The prices charged by the taxpayer from its AEs were
determined to be at arm’s length by using the Transactional Net Margin Method
(TNMM) with the profit level indicator (PLI) of operating profit margin on sales
(OPM).
The case involved transactions with some 17 AEs in various countries including
Malaysia, Africa, Brazil, Germany and Ireland. The transactions included sale of
products and payments for technical knowhow. The company looked for
comparable transactions and eight comparables were selected including
comparables from the USA, Europe and Malaysia. The average ‘net profit margin
on sales’ of these comparable companies was higher than the average net profit
margin earned by the 17 AEs.
The Indian tax authorities questioned the company's return on two issues:
1. The determination of the arm’s length price (ALP) was not referred by the
Assessing Officer (AO) to the Transfer Pricing Officer (TPO) as required by the
law; and
2. TNMM was used as the most appropriate method and the PLI of the AEs were
tested instead of the PLI of the taxpayer.
On the second point Ranbaxy contended that the AEs were the less complex
party and as a result it was correct that they be the tested party. The tribunal
disagreed as reliable data on the AEs was not made available to allow
benchmark analysis to be undertaken.
The tribunal also stated that it was not correct to aggregate the 17 AEs and treat
them as one tested party. The tribunal agreed that the least complex part should
be the tested party but stated that if comparable data was available relating to
the other party, in this case the Indian tax payer Ranbaxy, then that should be
used. Note that in India there is publicly available data on pharmaceutical
companies and the tribunal thought this should be used in preference to foreign
data.
The tribunal also commented that the OECD Guidelines should not be referred to
on a selective basis as this would be against the spirit of the Guidelines.
1.35 Roche Products Property [2008 AATA]
The case is notable for being the first decision in a court room on a substantive
transfer pricing issue under Australian law.
This case involved the judges in looking at the testimony of expert witnesses who
do not agree with each other.
Roche Australia is a subsidiary of Roche Holdings Ltd of Basel, Switzerland. Globally,

the Roche Group carries on the business of producing, selling and supplying

pharmaceutical and diagnostic products. Roche Australia comprised three

operating divisions over the audit period which was 11 years up to 2002.
The Australian Tax Office (ATO) questioned the transfer pricing methodology used
by the company.
The court disagreed with the evidence put forward by the ATO criticising a number
of ‘presumptions’ made by a number of the expert economists. The court thought
the economists were too US-focused in their approach and as a result they did not
provide analysis that specifically addressed Australia’s transfer pricing provisions.
The key points made in the case included the following:
• that arm’s length prices be determined for each separate year under
consideration, rather than a multiple-year average.
• questions were raised about whether Australia’s double taxation treaties be

applied by the ATO to effect transfer pricing adjustments (independent of the
‘domestic’ transfer pricing provisions set out in Div 13 of ITAA 1936).
• the ruling acknowledges the difficulty in finding available comparable data,

and uses a uniform gross margin to price the transfers of all pharmaceutical
products; a preference was expressed for transactional methods over profit
methods, such as the profit-based transactional net margin method (TNMM).
Following the case the ATO released as a statement that in their view Roche is
confined to ‘to the facts of the case’ and that ‘all things considered [Roche] is
seen as having limited significance for the administration of transfer pricing laws
generally’.
1.36 Roche Vitamins 2012 (STS 201/2012)
The Supreme Court of Spain came to a controversial decision, in that the subsidiary
of a non-Spanish company was considered to be the overseas company’s PE. The
Supreme Court confirmed, on appeal, the existence of a Spanish PE of Roche
Vitamins Europe, as a result of the activities carried on by the Spanish subsidiary. In
particular, the Court considered that the Spanish subsidiary operated as a
dependent agent of the Swiss entity, as it carried on, under two contracts, the
activity which could have been done directly through a fixed place of business
(being the sale and distribution of the goods produced).
The fact that Roche Vitamins SA had no capacity to contract or negotiate on

behalf of Roche Vitamins Europe did not exclude the application of the
dependent agency clause of the double tax treaty, as interpreted under the
Commentaries to the OECD Model Tax Convention (1997 version). In particular,
under the ‘Agency Contract’ the Spanish subsidiary was obliged to promote the
goods that were sold by Roche Vitamins Europe, which function was considered
by the Court as a greater involvement in the Spanish market. It was a key element
leading to the conclusion that Roche Vitamins SA was not merely processing
purchase orders issued by the Swiss company.
1.37 SNF (Australia) Pty Ltd v FCT (2010) FCA 635
An Australian company purchased products from related companies outside

Australia.
The Australian company had used the comparable uncontrolled price using the
pricing of transactions between the suppliers and their arm’s length customers. The

company incurred losses, in part due to commercial issues (including a low level of
sales per salesperson, competition in the Australian market, excessive stock levels,
and poor management) and partly due to a strategy to penetrate the Australian
market.
The Australian Taxation Office argued that the transactional net margin method
should be used, where a ‘benchmark operating profit’ should be determined with
reference to the operating profits achieved by other ‘functionally comparable’
distributors.
The Federal Court accepted the company’s pricing, and held that it could not be
concluded that the prices were artificially inflated.
1.38 SNF (Australia) Pty Ltd v Commissioner of Taxation (Full Federal Court
Decision) (2011) FCAFC 74
The case concerned the deductibility of purchases by SNF Australia which was a
wholly owned subsidiary of SPMC SA, formerly SNF SA (SNF France) a company
resident in France. SNF Australia carried on the business of manufacturing and
selling products (commonly called ‘flocculants and coagulants’) to end users in
the mining, paper and sewage treatment industries in Australia. During the years
1998 to 2004, SNF Australia purchased its supplies, at a price of some $71 million,
from other subsidiaries that were resident in France, the USA and China. SNF
Australia sustained trading losses in most income years between 1991 and 2004,
except in 1995 and 1997, when small accounting profits were made.
Following an Australian Taxation Office (ATO) transfer pricing audit of SNF Australia,
the Commissioner in March 2007 made determinations pursuant to the Australian
transfer pricing provisions with respect to the prices paid for the acquisitions from
these subsidiaries and issued notices of assessment with respect to the 1998 to 2000
and 2002 to 2004 income years.
The Full Federal Court (Court) unanimously held that SNF Australia had proved that
the prices paid by it were less than the prices paid by independent comparable
purchasers. Thus the Court said those prices were arm’s length and SNF Australia’s
prices did not therefore exceed arm's length consideration. Further, the Court
made several comments in relation to the interpretation of Australia’s domestic
law and international agreements which are discussed below.
In applying domestic transfer pricing provisions, the Court firmly rejected the
Commissioner’s approach in that applying the arm’s length principle, SNF Australia
must look at purchases in identical circumstances.
In accepting that SNF Australia had raised appropriate comparable transactions

involving the acquisition of the same or sufficiently similar products in the same or
similar circumstances, the court rejected a narrower approach to the application
of domestic law which the Commissioner of Taxation (Commissioner) argued was
supported, amongst other things, by the OECD’s TPG. This narrower approach may
have allowed the Commissioner to apply a profit based method (in this case, the
transactional net margin method (TNMM)) or more particularly make a
determination of arm’s length consideration based on the Commissioner’s
arguably very broad powers under domestic law.
The court generally accepted that there may be more than one arm’s length
price. However, it was held that the taxpayer had demonstrated that the prices
paid were arm’s length or less than arm’s length prices and thus it had absolved

itself of the burden of proof and effectively rendered the Commissioner powerless
in pursuing adjustments.
After evaluating expert evidence given by both parties, the court concluded that
on the basis of the comparable sales methodology, the actual prices paid by SNF
Australia for the acquisition of the products were in fact lower than the large
majority of prices paid by purchasers in comparable transactions on a global basis
over a similar period.
In relation to the fact that SNF Australia had incurred sustained trading losses over
the years in question, the court said that this was irrelevant to applying the test of
whether arm’s length consideration had been paid in respect of the acquisitions.
The key implications from this case may be summarised as follows:
• The court’s decision provides a clear and cogent analysis of the relevant tax
issues, in particular, the interpretation of Australia’s domestic transfer pricing
provisions and the relevance / application of international tax agreements,
commentary and other guidelines on their interpretation.
• As a starting point, domestic provisions should not be interpreted as if it were

an international agreement. However, where the domestic provisions are
obscure or unclear then the ambiguity may be resolved in a manner which is
consistent with Australia’s international obligations.
• It is appropriate to interpret domestic provisions in accordance with the

ordinary principles governing the interpretation of DTAs where the Australian
Parliament incorporated the whole text of a DTA in domestic law.
• The OECD TPG are not a legitimate aid to the interpretation of DTAs or
Australia’s domestic transfer pricing provisions.
• So long as appropriate comparable transactions are available, an analysis of

relevant comparables (e.g. the ‘comparable uncontrolled pricing’ method) is
an appropriate method to determine an arm’s length price for transfer pricing
purposes.
• If there is a global market for the relevant product or service, taxpayers may
use comparables from countries and regions other than Australia in
determining an arm’s length price.
• When identifying appropriate comparables to determine an arm’s length

price for transfer pricing purposes, the selected entity comparables should
carry out similar activities to the taxpayer (ie have 'functional comparability'
with the taxpayer).
1.39 Starbucks State Aid: Netherlands and Starbucks [SA.38374]
The European Commission (EC) opened an investigation into the transfer pricing
arrangements that Starbucks had in the Netherlands.
The case looked at an APA between the Dutch tax authorities and Starbucks. The
APA covered royalties to a UK limited partnership for coffee-roasting knowhow,
and payments to a Swiss affiliate for green coffee beans, thereby reducing the
group’s taxable profits in the Netherlands.
The EC stated that the methodological choices in the transfer pricing report
provided by the tax adviser for Starbucks, which were accepted by the Tax

Authorities in the APA, did not lead to a reliable approach to a market result and
thereby do not fulfil the arm’s length principle.
More specifically the EC was concerned by the use of TNMM when the OECD TPG
show a preference for CUP then went on to say that if TNMM was the correct
choice that it had been applied incorrectly. As a result it was argued that the
Dutch Co didn’t get any economic value for the use of the IP related to coffee
roasting hence no fee should have been paid. The price paid for the green beans
was too high.
This case is a warning that the APA cannot be deemed to provide a reliable
approach to a market result in accordance with the arm’s length principle.
This case makes it clear that APAs can give rise to State Aid also that The EU will
look carefully at TP reports when deciding if there has been illegal State Aid
(basically State Aid without permission). Illegal State Aid has to be repaid, interest
will be added to the amount repayable. APAs cannot be used as a means of
attracting companies into a country by offering a tax advantage not otherwise
available.
1.40 Sunstrand Corporation v Commissioner (1991). Bausch & Lomb Inc. v.

Comm’r, 92 T.C. 525 (1989)
We will look at these two cases together as they are similar
Sundstrand Corporation, a component manufacturer, set up a subsidiary (Sunpac)

in Singapore to manufacture constant speed drives (CSDs). Sunpac was treated as
a full cost manufacture. The TP was calculated in a way that left all the location
savings in Singapore.
The IRS challenged this saying that Sunpac was a ‘machine shop’ or a ‘contract
manufacturer’.
The Court determined that Sunpac was not, in fact, a contract manufacturer
because it operated under a licence from the parent company.
Sundstrand argued that Sunpac should retain any location savings because the
licence agreement gave it a ‘monopolistic position’ with respect to CSD spare
parts. This monopolistic position would, Sundstrand argued, have led Sunpac to
price in a way that caused all the location savings to remain in Sunpac. The Court
agreed with this argument. The court accepted that Sunpac has market power as
a result of the IP it owned.
In the case of Bausch & Lomb (B&L), B&L, a manufacturer of contact lenses,
developed and patented the spin cast method for manufacturing soft contact
lenses, which enabled production costs of approximately $1.50 per lens, while
alternative methods used by competitors cost at least $3.00 per lens. B&L
subsequently licensed the technology to wholly-owned Irish subsidiary B&L Ireland.
B&L Ireland manufactured the lenses at a cost of approximately $1.50 per lens and
then sold them to B&L for $7.50 per lens. The TP price was challenged by the IRS.
The IRS contended that the Irish company was a contract manufacturer because
sale of its total production was assured. Because it did not bear the risks of an
independent manufacturer, B&L Ireland is only entitled to cost plus a comparable
contract manufacturer mark-up.

The court found that CUP was the correct method. This was partly because B&L
Ireland was not contractually bound to sell the lenses it produced to B&L.
Therefore, it bore the risks of an independent producer, and it was entitled to the
market prices commanded by analogous independent producers. If B&L
committed to purchase the entire production, it would need to be compensated
for taking on that additional risk in the form of a discounted unit price.
Some writers have criticised this decision as the cost savings had been developed
in the US via development of the technology.
1.41 Thin Cap GLO C-524/04
This ECJ case considered the UK thin cap rules and held that whilst they were a
restriction on the freedom of establishment, such rules would be acceptable so
long as: they were aimed at purely artificial arrangements (identified using
objective and verifiable elements); they allowed taxpayers to produce evidence
of commercial justification for the transaction; and any disallowance only applied
to the interest which exceeded the arm's length amount.
1.42 Unilever Kenya (Income Tax Appeal) 753 of 2003
Unilever UK manufactured various products for Unilever Kenya. Unilever Kenya

charged lower prices to Unilever Uganda than it did to domestic buyers and to 3rd
party importers. The transfer pricing documentation included the following based
on the capital used for production of the goods:
• A charge on working capital will be made on average working capital

employed.
• The price will not be less than full variable cost + 10% profit mark-up plus actual
transport costs.
• Insurance will be for the account of the buyer.
• Prices may be adjusted taking into account increases or decreases in prices of

raw materials, exchange rate, import duty on raw materials and changes in
prices of other inputs.
The Kenyan Tax Authorities tried to impose a CUP. Two potential CUPs were put
forward: the average price of sales in Kenya by Unilever Kenya and the average
price of export sales by Unilever Kenya. No attempt was made to adjust the
potential CUPs for functional or other differences.
Unilever said that there was no CUP pointing out the functional differences in the
third party sales compared to the sales to Unilever Uganda. An important
difference was that for third party sales Unilever Kenya undertook the marketing
function and hence incurred costs. This was not the case with the sales to Unilever
Uganda, allowing a sale at a lower price.
The court agreed that there wasn’t a CUP.
It was held that cost plus was an acceptable method.
In reaching its decision the Court endorsed the use of the OECD Guidelines in the
absence of detailed guidance from the Kenyan Revenue Authority. The lack of
guidance resulted in a shift of the burden of proof from the taxpayer to the KRA
The government's response to this judgment was the introduction of Transfer Pricing
rules, which are largely based on the OECD Guidelines.

The main implications from this case include that a Tax authority that lacks
experience can take the wrong approach when dealing with complex transfer
pricing cases and that if CUP is to be used then it cannot be applied where there
are wide range of product and contract differences
1.43 VERITAS Software Corp., 133 TC No. 14, Dec. 58,016 (Dec. 10, 2009)
This was a US case looking at ‘buy-in’ costs for a cost contribution arrangement.
The IRS argued that what had taken place was akin to a sale or spin-off of Veritas
operations hence the sum to be paid should be valued on this basis.
Veritas Software, which is in the business of developing, manufacturing, marketing,

and selling software products, went through several corporate changes; mostly
notably, it was purchased by Symantec Corp. on July 2, 2005. Prior to that, on
November 3, 1999, Veritas Software assigned all its existing sales agreements with
its European-based sales subsidiaries to a new corporation – Veritas Ireland. In
addition, on the same date, Veritas Software and Veritas Ireland entered into a
research and development agreement, as well as a technology license
agreement.
Based on the licensing agreement, Veritas Software granted Veritas Ireland the
right to use certain ‘covered intangibles’, as well as the right to use Veritas
Software’s trademarks, trade names, and service marks. In exchange for the rights
granted by licensing agreement, Veritas Ireland agreed to pay royalties, as well as
a ‘prepayment amount.”
In 2000 Veritas Ireland made a $166 million ‘lump sum buy-in payment’ to Veritas
Software. This amount was later adjusted downward to $118 million. At issue, from
a tax perspective, is whether the buy-in payment was “arm's length’.
The court rejected the IRS approach agreeing with Veritas that the amount to be
paid should be based on comparable uncontrolled royalties payable over the life
of the agreement. Further the court said that the IRS determination was arbitrary,
capricious, and altogether unreasonable.
Veritas used agreements between Veritas Software and certain original

equipment manufacturers (OEMs) as comparables. The IRS contended that the
OEM agreements involve substantially different intangibles. But the court
disagreed: it concluded that, collectively, the more than 90 ‘unbundled’ OEM
agreements the parties stipulated were sufficiently comparable to the controlled
transaction.
In noting the comparability, the court also pointed out the following:
(1) Veritas Ireland and the OEMs undertook similar activities and employed similar
resources in conjunction with such activities, (2) there were no significant
differences in contractual terms, (3) the parties to the controlled and uncontrolled
transactions bore similar market risks and other risks, and (4) there were no
significant differences in property or services provided,
therefore, the court was happy that the unbundled OEM agreements were
sufficiently comparable to the transaction they were looking at thus giving the
result that comparable uncontrolled transaction method (CUT) (as set down in the
US regulations) was the best method to determine the appropriate buy-in price.
The buy-in payment charged met the arm's length standard and the IRS’s
contention was rejected.

1.44 Waterloo Plc and Others v CIR
In 2001, the Special Commissioners heard a case relating to loans made by a

parent company to a related trust to enable the trustee to purchase shares and
grant share options to employees of subsidiary companies. While the findings in
Waterloo plc and others v CIR (SPC301) related to a period prior to 1998, the point
at which transfer pricing regulations were introduced in the UK, and have
therefore to a certain degree been trumped by contemporaneous legislation,
there were a couple of particular points raised during the case that continue to be
of application.
On a general note, this case included analysis of precisely what constituted

‘business facilities’, an area which has been raised in following cases, and which to
a degree informed the wording of subsequent UK TP legislation – while the phrase
‘business facility’ is no longer included in the regulations, the regulations now
require analysis of any series of linked transactions. It was specifically noted in
Waterloo that ‘the phrase ‘business facility’ is a commercial not a legal term,
and…that where a commercial term is used in legislation, the test of ordinary
business might require an aggregation of transactions which transcended their
juristic individuality’. HMRC's stated position is now that it does not matter that
business facilities had been given and received, by way of a complex provision,
rather than being sold and bought, by way of a straightforward transaction.
Therefore there is no need to specifically identify a transaction between the
parent and the subsidiary, the relevant share scheme merely needs to provide a
defined, valuable and quantifiable benefit to the subsidiary employing the
relevant employees.
Secondly, Waterloo gave some guidance on the complex and frequently

contentious area of tax treatment and allocation of employee share option costs.
HMRC now considers that a facility is being provided no matter how the
arrangements are set up for administering and delivering a Group employee share
plan, and transfer pricing rules will subsequently apply – under Waterloo, that
facility should be priced accordingly, with the provider receiving or imputing
receipts that reflect the full value of the facility it is providing. Detailed
commentary based in part on the Waterloo findings now form part of HMRC's
guidance on the pricing of share plans under IFRS accounting rules that apply to
accounting periods commencing on or after 1 January 2005.
1.45 XILINX Inc. & Consolidated Subsidiaries v The IRS 2009
This case looked at whether related companies engaged in a joint venture to

develop intangible property must include the value of certain stock option
compensation one participant gives to its employees in the pool of costs to be
shared under a cost sharing agreement, even when companies operating at
arm’s length would not do so. The tax court found related companies are not
required to share such costs and ruled that the Commissioner of Internal Revenue’s
attempt to allocate such costs was arbitrary and capricious.
Xilinx, Inc. (‘Xilinx’) researches, develops, manufactures, markets and sells

integrated circuit devices and related development software systems. Xilinx set up
Xilinx Ireland (XI).
Xilinx and XI entered into a Cost and Risk Sharing Agreement (‘the Agreement’),
which provided that all right, title and interest in new technology developed by
either Xilinx or XI would be jointly owned. Under the Agreement, each party was
required to pay a percentage of the total R&D costs in proportion to the

anticipated benefits to each from the new technology that was expected to be
created. Specifically, the Agreement required the parties to share:
1. direct costs, defined as costs directly related to the R&D of new technology,
including, but not limited to, salaries, bonuses and other payroll costs and
benefits;
2. indirect costs, defined as costs incurred by departments not involved in R&D

that generally benefit R&D, including, but not limited to, administrative, legal,
accounting and insurance costs; and
3. costs incurred to acquire products or intellectual property rights necessary to

conduct R&D. The Agreement did not specifically address whether employee
stock options (ESOs) were a cost to be shared.
The IRS contended that ESOs issued to its employees involved in or supporting R&D
activities were costs that should have been shared between Xilinx and XI under
the Agreement.
The relevant sections of the US tax code apply the arm’s length standard, the
latter stating: ‘… the standard to be applied in every case is that of a taxpayer
dealing at arm’s length with an uncontrolled taxpayer’. (italics added)
Contrast that regulation with the regulation dealing with cost sharing agreements
which provides that participants in a cost sharing arrangement are to allocate all
costs of developing the intangible.
The parties agreed (and the Tax Court found as a fact) that unrelated persons
entering into a cost sharing arrangement would not include the cost of employee
stock options as a ‘cost’. Since the arm's length principle would require related
parties to only share costs that unrelated parties would share, that principle
dictates that ESOs should not be included as a shared cost.
Therefore, as stated above, there is a risk that the existing US cost-sharing

regulations, which require stock-based compensation to be included in the pool of
costs to be shared between the parties, are not in line with the arm’s length
standard as set out by the OECD Guidelines.
1.46 Zimmer Case
Zimmer SAS, a former distributor in France for Zimmer Ltd products, was converted
in 1995 into a commissionaire. The French tax authorities then assessed Zimmer Ltd
to French corporate income tax for the years 1995 and 1996 on the grounds that it
had a PE, contending that the UK Company carried out a business through a
dependent agent (ie, the French company Zimmer SAS) under Art. 4(5) of the
France-UK tax treaty.
The Paris Administrative Court of Appeal decided in February 2007 that the French
commissionaire of the UK principal constituted a French PE of that company.
Zimmer Ltd appealed against this decision before the French Supreme
Administrative.
The Supreme Court made its decision on a pure legal analysis of the provisions of
the French Commercial Code, according to which a French commissionaire has
no legal authority to conclude a contract in the name of its principal.
The Supreme Court referred to Article 94 of the former Commercial Code (L 132-1
of the new Code) which states that a commissionaire acts in its own name on

behalf of its principal. Contracts concluded by a commissionaire, even on behalf

of its principal, cannot directly bind the principal to the co-contracting parties of
the commissionaire. The Court concluded that a commissionaire cannot create a
PE simply as a result of the commission agreement with the principal.
However, that there may be exceptions to this rule, such as where the terms of the
commission agreement or other aspects of the instructions demonstrate that,
despite the qualification of the contract given by parties, the principal is bound by
contracts entered into by the commissionaire with third parties.
Key points arising from the case:
• Where the wording of the commissionaire agreement follows the legal nature
of a commissionaire, in accordance with French civil and commercial
regulations, it cannot be recharacterised by the tax authorities as a
contractual arrangement of a different nature.
• A commissionaire agreement can grant sufficient flexibility to the

commissionaire for carrying out its daily activities without constituting a PE of its
principal.
• The decision is based on legal principles and does not look at what is actually
happening in the business and how it actually operates.

Paper 3 03 TP Manual 2022

Uploaded by

Copyright:

Available Formats

Paper 3 03 TP Manual 2022

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Paper 3 03 TP Manual 2022

Uploaded by

Copyright:

Available Formats

Tolley® Exam Training

TRANSFER PRICING OPTION

Tolley Exam Training

Tolley Online Academy: http://www.tolleytraining.co.uk/

Printed in the UK by KnowledgePoint Limited

All rights reserved. No part of this publication may be reproduced, stored in a

Additional material was provided by:

CHAPTER 1 FUNDAMENTAL SOURCES ......................................................................................1

CHAPTER 2 ASSOCIATED ENTERPRISES...................................................................................21

CHAPTER 3 THE ARM’S LENGTH PRINCIPLE ............................................................................27

CHAPTER 4 TRANSFER PRICING METHODS.............................................................................45

CHAPTER 5 FUNCTIONAL ANALYSIS.......................................................................................67

CHAPTER 6 PRACTICAL ASPECTS OF PREPARING A FUNCTIONAL ANALYSIS .....................77

CHAPTER 7 RELATING THE FUNCTIONAL ANALYSIS TO SELECTION OF TP METHOD ............95

CHAPTER 8 ENTITY CHARACTERISATION..............................................................................107

CHAPTER 9 COMPARABILITY ANALYSIS: OECD PROPOSED PROCESS...............................121

CHAPTER 10 COMPARABILITY ANALYSIS: AGGREGATION AND USE OF THIRD PARTY NON-

CHAPTER 11 COMPARABILITY ADJUSTMENTS INCLUDING PRACTICAL ISSUES ...................145

CHAPTER 12 SPECIFIC TRANSACTIONS: INTRA-GROUP SERVICES .......................................151

CHAPTER 13 SPECIFIC TRANSACTIONS: LOANS AND OTHER FINANCIAL TRANSACTIONS 175

CHAPTER 14 TRANSFER PRICING FOR FINANCIAL TRANSACTIONS .....................................189

CHAPTER 15 SPECIFIC TRANSACTIONS: INTANGIBLE PROPERTY ..........................................207

CHAPTER 16 COST CONTRIBUTION ARRANGEMENTS ...........................................................223

CHAPTER 17 SPECIFIC TRANSACTIONS: BUSINESS RESTRUCTURING ....................................229

CHAPTER 18 NON-RECOGNITION ISSUES..............................................................................243

CHAPTER 19 PERMANENT ESTABLISHMENTS ..........................................................................255

CHAPTER 20 ATTRIBUTION OF PROFITS TO PEs.......................................................................273

CHAPTER 21 PE IN THE UN MODEL DOUBLE TAX CONVENTION ...........................................285

CHAPTER 22 COMPLIANCE ISSUES.........................................................................................289

CHAPTER 23 AVOIDING DOUBLE TAXATION AND DISPUTE RESOLUTION I ..........................307

CHAPTER 24 AVOIDING DOUBLE TAXATION AND DISPUTE RESOLUTION II .........................333

CHAPTER 25 E-COMMERCE AND TRANSFER PRICING ..........................................................345

CHAPTER 26 OTHER ASPECTS OF TRANSFER PRICING...........................................................363

© RELX (UK) Limited 2021 I 2022 Sittings

APPENDIX 1 CASE LAW ..........................................................................................................375

© RELX (UK) Limited 2021 II 2022 Sittings

CHAPTER 1 FUNDAMENTAL SOURCES ......................................................................................1

CHAPTER 2 ASSOCIATED ENTERPRISES...................................................................................21

CHAPTER 3 THE ARM’S LENGTH PRINCIPLE ............................................................................27

CHAPTER 4 TRANSFER PRICING METHODS.............................................................................45

CHAPTER 5 FUNCTIONAL ANALYSIS.......................................................................................67

© RELX (UK) Limited 2021 III 2022 Sittings

CHAPTER 6 PRACTICAL ASPECTS OF PREPARING A FUNCTIONAL ANALYSIS .....................77

CHAPTER 7 RELATING THE FUNCTIONAL ANALYSIS TO SELECTION OF TP METHOD ............95

CHAPTER 8 ENTITY CHARACTERISATION..............................................................................107

CHAPTER 9 COMPARABILITY ANALYSIS: OECD PROPOSED PROCESS...............................121

© RELX (UK) Limited 2021 IV 2022 Sittings

CHAPTER 10 COMPARABILITY ANALYSIS: AGGREGATION AND USE OF THIRD PARTY NON-

CHAPTER 11 COMPARABILITY ADJUSTMENTS INCLUDING PRACTICAL ISSUES ...................145

CHAPTER 12 SPECIFIC TRANSACTIONS: INTRA-GROUP SERVICES .......................................151

CHAPTER 13 SPECIFIC TRANSACTIONS: LOANS AND OTHER FINANCIAL TRANSACTIONS 175

CHAPTER 14 TRANSFER PRICING FOR FINANCIAL TRANSACTIONS .....................................189

© RELX (UK) Limited 2021 V 2022 Sittings

CHAPTER 15 SPECIFIC TRANSACTIONS: INTANGIBLE PROPERTY ..........................................207

CHAPTER 16 COST CONTRIBUTION ARRANGEMENTS ...........................................................223

CHAPTER 17 SPECIFIC TRANSACTIONS: BUSINESS RESTRUCTURING ....................................229

CHAPTER 18 NON-RECOGNITION ISSUES..............................................................................243

CHAPTER 19 PERMANENT ESTABLISHMENTS ..........................................................................255

CHAPTER 20 ATTRIBUTION OF PROFITS TO PEs.......................................................................273

CHAPTER 21 PE IN THE UN MODEL DOUBLE TAX CONVENTION ...........................................285