POLICY FRAMEWORK FOR INVESTMENT USER’S TOOLKIT

Chapter 5. Tax Policy

Introductory note

The Policy Framework for Investment (PFI) User’s Toolkit project responds to a need for specific and
practical implementation guidance building on the experience of the countries that have already
piloted or are planning to pilot the PFI.

The Toolkit was developed with the involvement of government users, in co-operation with other
organisations, OECD committees with specialised expertise in the policy areas covered by the PFI
and interested stakeholders.

This document is a revised draft of the guidance relating to Chapter 5 of the PFI on Tax Policy.
Underlined text in this document represents links which will be activated in the final web version.

The PFI User’s Toolkit is purposely structured in a way that is amenable to producing a web-based
publication. A web-based format: is a flexible approach for providing updates and additions; allows
PFI users to download only the guidance relevant to the specific PFI application they are
implementing; includes a portal offering users more detailed resources and guidance on each PFI
question. The website is accessible at www.oecd.org/investment/pfitoolkit.

© OECD – 2013. A publication of the Investment Division of the OECD Directorate for Financial and Enterprise Affairs. OECD freely authorises the use of
this material for non-commercial purposes. All requests for commercial use or translation of this material should be submitted to investment@oecd.org.
This work is published on the responsibility of the Secretary-General of the OECD. The opinions expressed and arguments employed herein do not
necessarily reflect the official views of the Organisation or of the governments of its member countries.
This document and any map included herein are without prejudice to the status of or sovereignty over any territory, to the delimitation of international
frontiers and boundaries and to the name of any territory, city or area.
Tax Policy
A country’s tax regime is a key policy instrument that may negatively or positively
influence investment. Tax Policy in the PFI relates to the formulation of a tax
strategy which is supportive to investment. It covers the advantages and
disadvantages of alternative tax policy choices in meeting the twin goals of offering
a tax system attractive to investment, while at the same time raising revenues to
support the key pillars of a business-enabling environment, such as infrastructure. A
poorly designed tax system, where the rules and their application are non-
transparent, overly complex or unpredictable, may discourage investment adding to
project costs and uncertainty. Systems that leave excessive administrative
discretion in the hands of tax officials tend to invite corruption and undermine good
governance objectives fundamental to securing an attractive investment
environment. Policy makers are therefore encouraged to ensure that their tax
system imposes an acceptable tax burden that can be accurately determined, and
which keeps tax compliance and tax administration costs in check.
This chapter seeks to assist countries in understanding the bottlenecks within their
current tax system and to propose changes to improve the efficiency of the system
in terms of its ability to mobilise revenue on the one hand and attract the right kind
of investment on the other. It identifies the nine most important questions relevant
for judging the effectiveness of a country’s tax policies and practices and offers
specific guidance in formulating a tax policy strategy which is supportive to
investment. The nine PFI questions on tax policy relate to:
 the consistency of a country’s tax burden with its broader development
objectives;
 an evaluation of the actual tax burden on domestic profits;
 a comparison of the actual versus the target tax burden;
 understanding the potential tax effects on investment;
 an evaluation of tax distortions to investment;
 the determination of taxable income;
 accounting for unintended tax incentive effects;
 tax expenditure reporting;
 international tax co-operation.

2
Consistency of tax burden with broader development
objectives
5.1. Has the government evaluated the level of tax burden that would be consistent
with its broader development objectives and its investment attraction strategy? Is this
level consistent with the actual tax burden?

Rationale for the question

Behind this question lies a central trade-off facing policy makers – by reducing taxes
on host country investment and subsidising investors, revenues are forgone that
could be used to build infrastructure, improve labour skills, strengthen governance
and address other elements of the business environment that in many country
contexts are the real impediments to investment. Given the country’s overall fiscal
policy position and objectives, policy makers should consider whether the actual tax
burden imposed on the economy is appropriate, or whether it should be adjusted in
order to attract additional investment, discourage capital flight and swing location
decisions in the country’s favour.

Key considerations
Most would agree that a host country tax burden that is very high relative to other
countries generally discourages investment and could, in certain cases, be a
deciding factor for not investing or reinvesting in a particular host country. A more
difficult issue is when, under which circumstances, can a relatively low host country
tax burden (e.g. reduced statutory tax rates or tax incentives) be expected to
attract additional investment?
Investors are generally willing to accept a higher host country tax burden if the
country offers attractive business-enabling and market conditions, a stable
framework, and above all, host country location-specific profit opportunities.
Indeed, in principle, the tax burden on location-specific profit could be increased up
to the point where economic profit is exhausted without discouraging investment.
Thus, where an economy offers an abundance of location-specific profit
opportunities, policy makers may understandably resist pressures to adopt a
relatively low tax burden, to avoid tax revenue losses.
In the context of economic profit that is not location-specific, comparisons of tax
burdens in competing locations would be expected to be factored in. If a given
business activity can be carried out in a competing location with a lower rate than
that in the host country, then, in theory, investors would be unwilling to bear a tax
burden in the host country above that rate. On the other hand, if the competing
low-burden location does not offer an attractive business infrastructure and stable

3
macroeconomic environment, investors may be willing to pay a higher tax burden in
the host country without being tempted to invest elsewhere.
Where the investment conditions in a given location are on balance more attractive
than those elsewhere, the question arises as to how much higher the tax burden
may be set without significantly impacting investment. And if investment is
expected to decline, at what rate and in what sectors?

Policy practices to scrutinise

A central challenge for policy makers aiming to encourage domestic and foreign
direct investment is carefully weighing the relative advantages and disadvantages of
alternative tax policy choices that would be attractive to investment. Global
experience and analysis show that tax incentives are one of many, and sometimes
not the most important, factors of potential investment. Critically important to
potential investors are questions over costs and risks associated with
macroeconomic and enabling-environment conditions; access to capital; cost of
compliance with laws, regulations and administrative practices; and securing
location-specific profit opportunities. Therefore, the following criteria must be
carefully considered in setting or evaluating the level of a country’s tax burden.

Framework Political and monetary stability: Investors require stability for

conditions planning purposes. How stable is the political system? How
stable are the monetary system and fiscal framework and
what is the accumulated public debt? What are expectations
over future inflation, interest and exchange rates?
State of infrastructure: What is the state of the host
country’s infrastructure covering transport (airports,
seaports, rail systems, roads), telecommunications
(phone/Internet services) and other services important to
business? Are private costs of using/purchasing
infrastructure services high relative to competing
jurisdictions?
Simplicity and cost of compliance: The tax system should
display a high degree of transparency and clarity. How
significant are the costs and risks to business associated
with the complexity of all tax codes and their accompanying
rules, regulations and procedures? In what areas is public
governance weak and where is corruption a problem? Is the
cost of tax administration compliance relatively low?
Tax system stability. Although revisions in the code to
accommodate both changing economic circumstances and

4
 taxpayer practices are to be expected, the main body of the
tax law should remain relatively stable over time. Is the tax
system stable or prone to revisions? Does the tax system
avoid fluctuations due to an over-dependence on too few
types of taxes? At the same time, is the total number of
taxes limited to avoid complications and burdens on
taxpayers that might otherwise occur?

Market Considerations of market demand and supply are also

characteristics centrally important to investors, including:
Market size: What is the size of the domestic market? How
large is the domestic consumer market (number of
households, average level and distribution of per capita
income)? How large is the domestic producer market
(number of firms, asset size, input requirements)? How large
and accessible are markets in other (e.g. neighbouring)
countries?
Labour market conditions: What labour force skills are
available in the host country and what employee benefits
(e.g. social security) are provided by the state? Are labour
costs (wages plus mandatory employer social security
contributions) high relative to competing jurisdictions?

Prevalence of For many, if not most, investments, the levels of profit and
location-specific risk associated with undertaking a given business activity
profits may vary significantly across alternative locations and may,
in certain cases, be “location specific” – that is, may require
a physical presence in a particular location. Location-specific
activities include privatisations, the extraction of natural
resources and the provision of restaurant and hotel services.
In such cases, if profits can be expected at levels of risk that
investors are willing to assume, the profits are location-
specific – that is, they cannot be realised by locating in
another country or jurisdiction. A central question for
investors is how location-specific are the potential profits
and risk of a given host country? Is the tax burden of a host
country highly relevant to an investment decision? How
much higher may the tax burden be set without significantly
impacting investment? If investment is expected to decline,
which sectors will be the most affected?

5
Resources for further study
For further reading on optimal taxation and how to minimise the excess burden of
taxation, see:
 Matthews, S. (2011), “What is a Competitive Tax System?”, OECD Taxation
Working Papers, No. 2, OECD Publishing.
 Brys, B. (2011), “Making Fundamental Tax Reform Happen”, OECD Taxation
Working Papers, No. 3, OECD Publishing.
 OECD (2008), “Tax Effects on Foreign Direct Investment”, Policy Brief,
February, OECD Publishing.
 Auerbach, Alan J. (1985), “The Theory of Excess Burden and Optimal
Taxation”, in: Alan J. Auerbach and Martin Feldstein (eds.), Handbook of
Public Economics, Vol. 1, North-Holland, Amsterdam, pp. 61-127.

6
Evaluation of the actual tax burden on domestic profits
5.2. What is the average current tax burden on domestic profits, taking into account
statutory provisions, tax-planning opportunities and compliance costs?

Rationale for the question

Policy makers should regularly assess the tax burden on profits to determine if the
tax system is supportive of investment without forgoing tax revenue that could be
used to fund public expenditure on infrastructure or other areas of critical
importance to investors. The main statutory provisions as well as the effects of tax-
planning strategies increasingly used by businesses to lower the tax burden (e.g.
transfer pricing, thin capitalisation) should be taken into account. Compliance costs
from excessive complexity, non-transparency and unpredictability should also be
factored in. If the tax burden on business income is judged to be inappropriate,
either too high in order to attract investment or too low in relation to the country’s
revenue needs, consideration should be given to adjusting the parameters of the
statutory tax burden.

Key considerations
Policy analysts have various measures available to assess the tax burden on business
profit and the tax disincentive for investment.
 Statutory tax rates are the most visible and often-cited measure of the tax
burden. These rates are relevant because they have an important signal
function to investors and are commonly used in cross-country comparisons
as an important factor in the decision-making process for new investment.
 Backward-looking indicators, using historical data. Two backward-looking
indicators are commonly used, the corporate tax-to-GDP ratio and the
average tax rate.
a. Tax-to-GDP ratio measures the actual corporate tax revenues in
relation to gross domestic product and is the main aggregate
indicator used by policy analysts.
b. Average tax rate measures the ratio of business tax revenue to
corporate profit and is usually calculated based on micro-level firm-
specific data.1 The use of micro-data allows for a measure of the tax

1. The average tax rate could be calculated using aggregate data; however, when assessed
on the aggregate level, the indicator poses certain measurement and interpretation
issues, such as, for example, a mismatch of numerator and denominator in relation to the
treatment of business losses and foreign income.

7
 burden on the economy-wide level as well as disaggregated by firm
size, sector, location or ownership structure.
 Forward-looking effective tax rate indicators. The tax burden on business
income may be assessed using forward-looking, parameter-based indicators
(marginal and average effective tax rates) that capture the net effect of
basic statutory tax provisions.
a. Marginal effective tax rate (METR) summarises the effect of the
complicated tax code on an incremental investment and shows the
impact of the tax system on the investor’s decision to infuse capital
into the business.
b. Average effective tax rate (AETR) is a more general tax burden
indicator in that it shows the effect of the tax regime not only on the
incremental increase in investment like the METR but the effect of
the tax regime on a total investment project as well.
Basic measures of the tax burden on business income do not specifically address
tax-planning strategies of resident multinational firms to lower host country tax.
Nor do they account for the costs to business of complying with the tax system,
including costs associated with registration, completing and filing tax returns, and
tax audits. Therefore, it is necessary to adjust the tax burden estimates to take tax
planning and compliance costs into account in order to better gauge the “true” tax
burden on business. Understanding the effects of multinational tax planning and
compliance costs may lead to a reassessment of the appropriate tax policy and may
encourage a greater focus on lowering compliance costs as a way of lowering the
tax burden in the host country.

Policy practices to scrutinise

Each commonly used tax burden measure varies in terms of relevance, data
intensity and complexity of use. More importantly, as each measure has its
limitations for reflecting the true burden of the tax system on businesses, policy
analysts must exercise care when interpreting the results.

Statutory tax rates As the data is readily available, statutory tax rates are easy to
use as a tax burden measure and can be applied to compare
tax burdens across different time periods and in cross-country
analysis. However, as many would be quick to point out,
statutory rates tell an incomplete story. The effective tax on
investment is usually lower than that suggested by nominal
rates due to specific provisions in the tax legislation, such as
tax incentives to promote investment.

8
Backward-looking Tax-to-GDP ratio. The main aggregate indicator used by
measures, using policy makers worldwide to measure the significance of
historical data corporate taxes is the corporate tax-to-GDP ratio, which
shows the share of total corporate tax revenues in gross
domestic product. While relevant in policy analysis, this
indicator could lead to ambiguous conclusions. The
corporate tax-to-GDP ratio depends on the degree of
incorporation in the country, which may vary over time. This
variation could mislead policy analysts by implying that the
tax burden has changed even when the corporate tax policy
remained the same.
Average tax rate, expressed as the ratio of actual tax
revenues to total corporate profits, using historical, firm-
specific financial data, offers a more precise measure of the
tax burden. Since actual revenue figures are used, the
effects of statutory tax provisions – income tax rates, tax
deductions, tax credits – and the effects of tax planning, as
well as tax relief from lax or discretionary administrative
practice, are taken into account. Moreover, the use of firm-
specific, micro-level data permits the analysis of the tax
burdens for various taxpayer groups with different taxpayer
characteristics relevant to policy analysis, for example, for
various sizes of firms, ownership structures, sectors or
locations.

Forward-looking Backward-looking tax burden indicators are measured with

effective tax rates a time lag due to the reporting delay of National Accounts
and revenue statistics data. This consideration increases
interest in forward-looking tax burden measures, which only
require readily-available tax parameter information which
can be found in tax legislation and regulations. As such,
forward-looking measures are easily used by policy analysts
in assessing the tax burden implications of current or
planned tax reform.
The marginal effective tax rate (METR) allows policy
analysts to assess the impact of the various tax incentive
measures on the rate of return for representative
investment projects (at the margin). The METR assesses the
amount of tax that arises when a firm decides to undertake
an incremental unit of activity. A positive METR is an
indicator of an activity that is discouraged by the tax system;

9
 a negative METR is an indicator of an activity that is
encouraged by the tax system. A zero METR indicates that
the tax system is neutral to the activity. The METR measures
the net effect of the main statutory provisions in
determining effective tax rates by type of capital asset
(machinery and equipment, buildings, inventories,
intangibles) and by investor type (taxable resident, tax-
exempt resident, non-resident). Such measures may be
finessed by factoring in the effects of tax-planning
strategies employed in the host country to strip out taxable
profits (e.g. thin capitalisation, non-arm’s length transfer
prices) to tax havens. Annex 5.1 presents the standard
methodology for calculating METRs.
The average effective tax rate (AETR). Many investment
decisions are discrete rather than marginal in nature. For
example, multinational firms may face a choice between
alternative locations for investment. In making a mutually
exclusive location decision, the firm will choose the location
that offers the highest post-tax profit. The tax-burden effect
on this decision can be measured by the extent to which the
pre-tax profit is reduced by taxation as reflected by the
AETR. Conditional on this location choice, the scale of the
investment will be determined by the METR. Annex 5.2
presents the standard methodology for calculating AETRs.

Tax burden measures Two approaches may be used to gauge the effects of
adjusted for tax corporate tax planning on the tax burden of firms.
planning
 One approach adjusts the denominator (profit
measure) of a backward-looking average tax rate.
While the numerator of a basic average tax rate
measure includes actual tax paid inclusive of tax
planning, the denominator domestic profit measure
may understate the “true” profit amount. A
measure of true profit would include profit on
business activity that is “artificially” shifted offshore
by various means.2 It is useful for policy makers to
consider adjusted average tax rate estimates that

2. Possible means include the use of non-arm’s length transfer prices on cross-border goods
and service transactions with related affiliates, and (arm’s length or non-arm’s length)
interest payments on inter-affiliate debt provided directly or indirectly by tax haven
finance affiliates.

10
 factor in common forms of tax planning.
 Another approach adjusts forward-looking
parameter-based METR and AETR to capture the
effects of various forms of commonly employed
corporate tax planning strategies. Chapter 5 of the
OECD’s Tax Policy Study No. 17, Tax Effects on
Foreign Direct Investment: Recent Evidence and Policy
Analysis, provides an excellent guide for calculating
METRs and AETRs that includes a number of tax-
planning structures, such as thin capitalization of
high-taxed subsidiaries, deferral of home country
tax, use of “triangular” structures involving the use
of tax haven finance affiliates, and the use of hybrid
entities and hybrid instruments.

Tax burden measures Too often, policy makers ignore the qualitative side of tax
adjusted for burden analysis. However, depending on the degree and
compliance cost sources of complexity, transparency and predictability of the
given tax system, the tax compliance burden could be quite
significant. Most recently, the usage of published
international indicators, notably the World Bank’s Doing
Business Indicators (DBI), has grown in significance. While
the limitations of the DBI are widely recognised and
discussed, they can be utilised as a broader measure of
business compliance costs. If resources and time permit, a
dedicated, well-tailored tax compliance cost survey could
provide policy makers with a wealth of information by
identifying the most onerous legislative and regulatory
provisions, from the business compliance cost perspective.
Survey estimates of the total number of hours required to
comply with the business tax system, combined with
estimates of the value (cost) of each hour devoted, may be
included in the calculation of the total tax burden.

Resources for further study

 For a detailed discussion of the various measures of tax burdens that assess
the impact of taxes on economic activity, see OECD (2000), Tax Burdens:
Alternative Measures, OECD Tax Policy Studies, No. 2, OECD Publishing.

11
 For use of micro-, taxpayer-level data in measuring average tax rates, see
OECD (2003), Using Micro-data to Assess Average Tax Rates, OECD Tax Policy
Studies, No. 8, OECD Publishing.
 For the methodology for calculating the marginal effective tax rate (METR),
including data underlying the computations, see McKenzie, K.J., M. Mansour
and A. Brûlé (1998), “The Calculation of Marginal Effective Tax Rates”,
Working Paper, Vol. 97, No. 5, Technical Committee on Business Taxation,
Department of Finance, Canada.
 For a detailed discussion of the AETR, investigating the role of taxation in
cases where an investor faces a choice between two or more mutually-
exclusive projects, see Devereux, Michael P. and Rachel Griffith (1998), “The
Taxation of Discrete Investment Choices”, Institute for Fiscal Studies Working
Paper Series, No. W98/16.
 For a discussion of forward-looking tax burden measures adjusted for tax
planning, see OECD (2007), Tax Effects on Foreign Direct Investment: Recent
Evidence and Policy Analysis, Tax Policy Studies, No. 17, OECD Publishing.
 For evidence on the scale of tax-planning activity, see Altshuler, R. and H.
Grubert (2006), “The Role of Governments and MNCs in the Race to the
Bottom”, Tax Notes International, Vol. 41, No. 5, pp. 459-474.
 For a discussion on how to construct measures of effective tax rates that
reflect opportunities for various multilateral strategies, see Grubert, Harry
(2004), “The Tax Burden on Cross-Border Investment: Company Strategies
and Country Responses”, in: Measuring the Tax Burden on Labor and
Capital, edited by Peter Birch Sorense, MIT Press (CESifo Seminar Series),
Cambridge, Massachusetts, pp. 129-170.

12
Comparison of actual versus target tax burden
5.3. Is the tax burden on the business enterprises of investors appropriate with
reference to the policy goals and objectives of the tax system?

Rationale for the question

In choosing the tax burden to levy on domestic profit, the analysis should weigh the
various objectives guiding overall tax policy design, including efficiency concerns,
equity, simplicity and stability, and revenue requirements.
 Efficiency: The efficiency concern of a tax system calls for as little
interference as possible with the market incentives for investment. This
objective is often equated with low statutory and effective tax rates, broad
tax bases and little or no differentiation across economic activities or
industries.
 Equity: Equity concerns generally call for an equal sharing of the tax burden
across different taxpayers with roughly the same income or purchasing
power. Fairness implies equal taxation of equal incomes (horizontal equity)
and higher taxation of higher incomes (vertical equity).
 Simplicity and stability: The tax code should display a high degree of
transparency and clarity. Further, investors require stability for planning
purposes. Although revisions in the code to accommodate both changing
economic circumstances and taxpayer practices are to be expected, the
main body of the tax law should remain relatively stable over time.
 Revenue requirement: The tax system should generate adequate revenue
(recognising that adequacy is not easily defined) at marginal tax rates low
enough to avoid discouraging economic activity.
Different goals will most likely suggest different tax burden levels. Harmonising
various competing objectives is never an easy, but always a carefully balancing, act.

Key considerations
Tax officials should regularly assess the tax burden on investment, alongside
analyses of tax revenues, as part of an ongoing assessment of the ability of the tax
system to meet competing policy goals and objectives. The assessment must take
into account the broader administrative, institutional and political considerations,
such as:
 Administrative factors: Tax avoidance and evasion opportunities, and the
costs of compliance, administration and enforcement.

13
  Institutional factors: The transitional costs of changing the tax system and
complex implementation, legal and administrative issues.
 Political economy factors, such as election-related issues, including its timing.
For example, politicians may use the tax system to favour particular interest
groups and increase their probability of re-election.
If the existing tax burden on business income is judged to be inappropriate, policy
consideration may be given to adjusting the provisions of the statutory tax burden,
e.g. through some combination of adjustment to the statutory corporate tax rate,
rates of depreciation for tax purposes or other measures.

Policy practices to scrutinise

Choice over an appropriate tax burden on investment, shaped by balancing various
considerations, may begin with a fixed overall revenue requirement (to fund a given
set of public expenditures). Given revenue requirements, policy makers would
normally rely on a mix of taxes to meet those needs. The tax mix will include both
direct and indirect taxes, including taxes on income and profits, taxes on property
and wealth, consumption taxes, trade and other taxes.
The sensitivity of investment to taxation will be one of the determining factors in
choosing the appropriate tax burden. When sensitivity to taxation is high, generally
lower levels of taxation would be called for to discourage capital flight and non-
reporting. Similarly, if the level of underground activity is significant and contributes
negatively to the development potential of the country, drawing non-compliant
firms into the tax net may require reducing the tax burden on unincorporated
businesses, for example through more generous treatment of business losses or
simplified tax rules that both reduce compliance costs and lower the effective rate
of tax.
On the other hand, policy makers may be motivated to increase the tax burden on
business by unwinding incentives found to be inefficient or strengthening tax base
protection measures that guard against aggressive tax planning, to increase tax
revenues available to help finance infrastructure projects important to business.

Resources for further study

For further reading on the various economic and political considerations that drive
policy decisions on the level of tax burden, see:
 OECD (2010), Choosing a Broad Base – Low Rate Approach to Taxation, Tax
Policy Studies, No. 19, OECD Publishing.
 OECD, “Tax and Economic Growth”, Economics Department Working Paper,
No. 620, ECO/WKP(2008)28, OECD Publishing.

14
Understanding potential tax effects on investment
5.4. If framework conditions and market characteristics for investors are weak, is it
reasonable to assume that a low tax burden can compensate by impacting favourably
on investment decisions?

Rationale for the question

Any simple claim that tax incentives are either highly effective or highly ineffective
could be challenged. Costa Rica, Ireland, Malaysia and Mauritius are habitually cited
as success stories in attracting investment by lowering tax burdens. What is often
missing from the discussion of the success of these countries is the reflection on the
framework conditions and market characteristics that these countries offer in
addition to tax breaks and generous incentives regimes – namely, stable economic
and political conditions, a well-educated labour force, good infrastructure, open
trade for exporters, dependable rule of law and effective investment promotion
systems.
To support this point further, ample worldwide evidence suggests that where
framework conditions or market characteristics are relatively weak, a low tax
burden has a limited effect on swinging investment decisions in the country’s
favour. Moreover, where framework conditions are weak, a low host country tax
burden achieved through the use of special tax incentives may operate to
discourage rather than encourage investment, by contributing to project cost and
risk. This is especially the case where the introduction of, and subsequent changes
to, incentive schemes give the impression of an unstable tax system. Many
transition economies have had a disappointing experience from their attempts to
rely on a low tax burden (typically targeted at foreign investment) to boost
investment.
The real goal is then to analyse the economic, fiscal, institutional and political
conditions that a country has to offer in order to understand if the introduction of
tax incentives has the potential to impact favourably on the investment decisions.
Given the country’s setting, if and when a lower tax burden is deemed to be
appropriate and potentially beneficial, further analysis is necessary to design the tax
incentive programme so it maximises the impact and minimises the costs, i.e.
encourages investment without forgoing significant tax revenues.

Key considerations
Despite analysis indicating a limited investment response to a lower tax burden
relative to revenue losses and administrative costs, tax incentives are routinely
chosen by governments to attract investment in general, and foreign direct

15
investment (FDI) in particular. Three simple reasons provide the rationale behind
this widespread practice, particularly in the context of developing countries:
 it is much easier to provide tax incentives than to correct deficiencies in, for
example, infrastructure or skilled labour;
 tax incentives do not require an actual expenditure of funds or cash
subsidies to investors;
 tax incentives are politically easier to provide than public funds.
Further, politicians and policy makers often cite the following two arguments to
justify their decision to lower tax burdens in order to attract investment.
 First, domestic savings, especially in emerging and developing countries,
could be so low that they are insufficient to finance economic expansion.
Similarly, weak financial intermediation can have a similar effect on
investment, effectively limiting business resources for investment. In such
environments, a lower tax burden is thought to attract FDI as a source of
external finance.
 Second, evidence suggests that investment may generate positive
externalities – “spillovers” – toward the host economy. Investment can:
o act as a trigger for technology and know-how transfers;
o upgrade workers’ skills and generally support human capital
formation;
o assist enterprise development and restructuring, especially in
connection with privatisation;
o nurture business clusters and contribute to fuller international
(trade) integration.
As noted above, the effectiveness of tax incentive measures to impact favourably
on investment decisions will ultimately depend on the specific country’s situation,
most notably the macroeconomic framework conditions, its market characteristics
and the existence of location-specific profits.

Policy practices to scrutinise

When introducing changes in the tax burden in order to achieve given investment
goals, policy makers must ensure that the cost of achieving the given investment
goals – namely, revenue loss – is kept at the lowest feasible level and does not
exceed the benefits of the implemented policy. As such, thorough analysis of the
effectiveness of proposed tax provisions should be conducted both prior to the
introduction of investment-promotion measures as well as systematically ex-post, to

16
assess the extent to which the measures meet their intended objectives. The
following criteria could help decision makers in distinguishing between beneficial
and wasteful measures:3
 Ineffectiveness. The proposed tax burden-reduction measures fail to produce
benefits to the host economy that exceed the budgetary costs. This
situation may also arise where authorities applied faulty cost-benefit analysis
(or no cost-benefit analysis at all) to their incentive programmes or where
promised benefits do not materialise.
 Inefficiency. This is the case where incentives produce benefits that
outweigh the costs, but authorities fail to properly maximise the benefits
and minimise the costs. In other words, similar results might have been
obtained at a lower cost.
 Opportunity costs. When the resources available to attract investment are
scarce, the issue of alternative usage of funds arises. Incentive schemes that
are both effective and efficient may nevertheless be wasteful if the funds
that are sunk into financing them could have been used more profitably.
 Deadweight loss. This term refers to a situation when:
o Investment projects that would have taken place in the absence of
incentives are subsidised by a generous incentive scheme.
o The intended recipients of targeted incentives are not adequately
specified, resulting in spillover to non-target groups.
o By offering particularly generous incentives to some projects, policy
makers effectively “raise the bar”, creating a reference point that
future investors will demand for a similar degree of generosity.
 Triggering competition. The long-term costs of an incentive scheme include
the economic burden that arises if other jurisdictions put in place matching
measures. This is of particular concern when new measures are introduced
or the existing measures are significantly augmented. Doing so without
properly assessing the likely reactions of other jurisdictions can, in many
cases, amount to a wasteful practice.
While stressing the need for cost-benefit assessments of tax incentives, it must be
recognised that systematically assessing tax measures is a data-, time- and resource-
consuming process. Often times, policy analysts do not have sufficient data to
evaluate the overall effects of tax measures. It is therefore highly advisable to
collect data systematically to support the assessment of the costs of tax

3. This list of wasteful criteria draws on OECD (2003), Checklist for Foreign Direct Investment
Incentive Policies, OECD Publishing.

17
expenditures and to monitor the overall effects and effectiveness of individual tax
incentives.

Resources for further study

The OECD’s Task Force on Tax and Development has developed a framework for tax
incentives for investment, whose purpose is to promote transparency in decision-
making processes, increase the information available on costs and benefits, limit
discretion and increase accountability. The Principles of Transparency and
Governance of Tax Incentives for Investment is emerging as an international
consensus on the governance of investment-related incentives.
For a detailed discussion on wasteful investment measures and the potential pitfalls
and risks of excessive reliance on incentive-based strategies, see:
 OECD (2003), “Checklist for Foreign Direct Investment Incentive Policies”,
OECD, Paris.
 OECD (2007), Tax Effects on Foreign Direct Investment: Recent Evidence and
Policy Analysis, Tax Policy Studies, No. 17, OECD Publishing.
 OECD (2007), “Tax Incentives for Investment – A Global Perspective:
Experiences in MENA and Non-MENA Countries”, OECD, Paris.
 OECD (2003), Tax Policy Assessment and Design in Support of Direct
Investment: A Study of Countries in South East Europe, OECD Publishing.
 Oman, C. (2000), Policy Competition for Foreign Direct Investment: A Study of
Competition among Governments to Attract FDI, Development Centre
Studies, OECD Publishing.

18
Addressing tax distortions to investment
5.5. Where the tax burden on business income differs by firm size, age of the business
entity, ownership structure, industrial sector or location, can these differences be
justified? Is the tax system neutral in its treatment of foreign and domestic investors?

Rationale for the question

Tax systems may impose a non-uniform effective tax rate on different businesses,
depending on their size, ownership structure (e.g. domestic versus foreign-owned),
business activity or location. Certain firms may be specifically targeted to receive
preferential tax treatment. Tax relief may be provided on a targeted basis to firms
depending on firm size, sector, industry, location or ownership structure.
 Size: Tax relief may be targeted, for example, at “small” firms – those with
income or assets below a threshold value.
 Sector/industry: Tax burden can vary for firms in a specific sector (e.g.
manufacturing) or more narrowly a specific industry (e.g. automobile
manufacturing).
 Location: Tax relief may be targeted at firms in a given location, for example
in a specific province or state, or in a region of the country with high
unemployment.
 Ownership: The ownership structure of a firm may determine tax relief.
Concessions (e.g. tax holidays) may be targeted at firms wholly or partially
owned by foreign (non-resident) shareholders, and may depend on the type
of financing. For example, some countries target tax relief to “new” equity
capital.

Key considerations
In some cases, there might indeed be good reasons for implementing targeted tax
incentives. The commonly used arguments for unequal treatment of investors
involve economic and administrative efficiency, and equity.
Economic efficiency. The standard justification for differential tax treatment on
efficiency grounds is that tax incentives can correct for market imperfections. These
“market correction” arguments are based on the assumptions that private investors
do not take into account the benefits to the larger society of certain types of
investment, which leads to under-investment. Another line of market failure
arguments suggests that asymmetric information on markets or products or
monopoly power of large firms could make entry difficult for SMEs or make it
difficult for SMEs to raise finance.

19
Administrative efficiency: The administrative argument is that it is often easier for
government to administer a tax incentives programme than to deliver a similarly-
targeted expenditure programme. For example, if a government wants to
encourage renewable energy development, the tax administration will have an
advantage over the energy sector government agencies as it already has
knowledge, systems and experience in administering tax incentive programmes.
Equity: Some investment incentives have redistributive goals, for example, policies
aimed at increasing investment and bolstering employment and growth in poorer
parts of a country. In such cases, assessing the overall distributional impact is quite
difficult, but targeted location incentives may indeed encourage investment in
inefficient locations and have positive distributional consequences.
Where tax relief is targeted, policy makers should examine and weigh arguments in
favour of and against such treatment, and ensure that the different treatment can
be properly justified. Where justifications are weak (e.g. where corporate tax relief is
targeted at foreign investors to the exclusion of domestic investors) consideration
should be given to a non-targeted approach.

Policy practices to scrutinise4

Tax burden measures that vary considerably from one investment type to another
must be explained. Policy makers, therefore, need a clear understanding of the tax
system’s impact on various investor groups. Analysts also want to know whether
their targeted investment approach is effective in meeting its intended policy
objectives (e.g. encourage investment in disadvantaged regions), or otherwise
efficiency gains may be realised if the system were more neutral in its treatment of
different investment types. Beyond this, efficient targeting requires accurate
estimates of the amount of tax revenue foregone in order to compare the realised
benefit against the costs associated with the targeted incentives.
Further considerations in targeting tax incentives involve limiting tax relief to
targeted firms/activities only – for example, to small businesses. Policy makers must
recognise that all taxpayers will analyse the targeting criteria and attempt to benefit
from the tax incentive. For example, a non-qualifying (medium or large) firm may
reorganise itself into two or more new business entities to access tax relief
conditional on firm size. Similarly, companies will attempt to characterise or re-
characterise certain activities so that they fall within the boundaries of qualifying
business activities, for example, to qualify for R&D tax incentives. Non-qualified
companies can channel asset purchases through qualifying companies. Likewise,
qualifying firms in a loss position can sell their balances of unused business losses

4. Draws on OECD (2009), Taxation of SMEs, Key Issues and Policy Considerations, OECD Tax
Policy Studies, No. 18, OECD Publishing.

20
and tax credits to firms outside the target tax incentive group that are profitable
and able to use them to reduce their tax liability. Inevitably, the government can
come under pressure to extend tax incentive relief to taxpayers/activities that were
not initially targeted.
Two approaches may be used to inform an assessment of whether tax-driven
variations across businesses of different size and industrial sector can be justified.
Backward-looking average tax rates for different investor groups. The first approach
computes average tax rates for different groups of firms, stratified to correspond to
targeting investor groups, e.g. firm size, sector, industry, ownership structure,
location. The analysis could include an assessment on the profits of:
 SMEs, including enterprises structured in corporate and unincorporated
form;
 large enterprises majority-owned by residents;
 large multinational enterprises controlled by foreign parent companies.
Forward-looking effective tax rates for different investments. The second approach
is to measure effective rates (METRs/AETRs) for groups of firms disaggregated by
size, sector, industry, location, ownership structure, taxable status. Computation of
effective rates for different domestic investment types – for example, those
qualifying for targeted tax relief and those who do not – allows one to examine
distortions to investment decisions introduced by the tax system. In particular, the
distorting effects of a given tax incentive can be examined by allowing the incentive
parameter (e.g. corporate tax rate, tax depreciation rate, investment tax credit rate)
to vary, while holding all other factors fixed.
Please refer to the discussion under Question 5.2 of this Toolkit for a more detailed
discussion on tax burden measures.

Resources for further study

 OECD (2010), Tax Policy Reform and Economic Growth, OECD Tax Policy
Studies, No. 20, OECD Publishing.
 OECD (2009), Taxation of SMEs, Key Issues and Policy Considerations, OECD
Tax Policy Studies, No. 18, OECD Publishing.
 OECD (2001), Corporate Tax Incentives for Foreign Direct Investment, OECD
Tax Policy Studies, No. 4, OECD Publishing.
 OECD (2007), “Taxation and Foreign Direct Investment, Report of Working
Party No. 2”, OECD, Paris.

21
 OECD (2005), “Incentives and Free Zones in the MENA Region: A Preliminary
Stocktaking”, document prepared for Working Group 2 of the MENA-OECD
Investment Programme, OECD, Paris.
 Oman, C. (2000), Policy Competition for Foreign Direct Investment: A Study of
Competition among Government to Attract FDI, Development Centre Studies,
OECD Publishing.

22
Determination of taxable income
5.6. Are rules for the determination of corporate taxable income formulated with
reference to a benchmark income definition (e.g. comprehensive income), and are the
main tax provisions generally consistent with international norms?

Rationale for the question

In dealing with any given corporate tax system, investors expect basic tax
provisions that adequately reflect business costs, including loss carry-forward
provisions that are consistent with those commonly employed elsewhere. Investors
also view negatively the double taxation of income within the corporate sector and
generally expect zero taxation of, or tax relief on, inter-corporate dividends
particularly when paid along a corporate chain. In short, policy makers are
encouraged to ensure that reasonable expectations of the main design features of
the tax system are recognised.

Key considerations
As a guide for assessing policy for determining the tax base for business income,
policy makers are encouraged to consider alternative design options, with regard to
those applied in other countries. Such references, along with consideration of the
pros and cons of alternative approaches, are helpful in deciding and defending an
appropriate set of tax base provisions and avoiding adjustments that could prove
unsatisfactory when assessed against a balancing of policy goals (e.g. raising
revenue, providing competitive tax treatment, limiting inefficiency, supporting
equity and avoiding undue complexity). Policy makers wishing to retain and attract
investment should be encouraged to explore and address a number of key tax base
provisions, including depreciation, inventories, business losses, inter-corporate
dividends, corporate capital gains and losses, and allowances for the cost of
corporate equity.

Policy practices to scrutinise

Investors may raise various concerns with respect to tax base rules based on the
following set of questions:
 Do tax depreciation methods and rates adequately reflect true economic
rates of depreciation of broad classes of depreciable property (serving as
benchmark rates) and account for inflation?
 Are possible time limits on the carrying forward (and possibly back) of
business losses, to offset taxable income in future (prior) years, sufficiently
generous/consistent with international norms? The case for generous carry-

23
 forward is particularly strong where depreciation claims are mandatory
rather than discretionary.
 Are inter-corporate dividends (paid from one resident company to another)
excluded from corporate taxable income to avoid double/multiple taxation?
 Are domestic dividends paid to resident individuals subject to classical
treatment or is integration relief provided in respect of corporate tax on
distributed income (e.g. partial inclusion of dividend income, or imputation
or dividend tax credit)? Is there evidence that such relief lowers the cost of
funds for firms? Or is such relief intended to encourage domestic savings?
Where integration relief is given in respect of distributed profit (dividends),
is similar relief provided in respect of retained profit (e.g. partial inclusion of
dividends and capital gains)?
 Where capital gains are subject to tax on a realisation basis, are taxpayers
allowed a deduction for capital losses (e.g. against corresponding taxable
capital gains)? Do “recapture” rules apply to draw into taxable income
excess tax depreciation claims on depreciable property?
 Is the tax treatment of wage income, as well as interest income, dividends
and capital gains (realised at the personal or corporate level) designed to
minimise incentives to: i) characterise one form of income as another; and ii)
choose one organisational form over another (incorporated versus
unincorporated) for purely tax reasons? In other words, are efforts made to
minimise tax arbitrage possibilities?
While addressing investors’ concerns, policy makers should be encouraged to:
 limit windfall gains (i.e. the provision of tax relief that does not achieve the
desired goals) to investors and, in the case of inbound direct investment,
foreign treasuries;
 minimise the scope for exploitation by business of the tax system (e.g.
through tax arbitrage);
 ensure single taxation of income sourced in the host country (e.g. through
enforcement of domestic tax rules and negotiation of tax treaties);
 keep tax administration costs in check.

24
Resources for further study
For further discussion on the various policy considerations that drive the
determination of corporate taxable income and are the main tax provisions, see:
 OECD (2010), Choosing a Broad Base – Low Rate Approach to Taxation, Tax
Policy Studies, No. 19, OECD Publishing.
 OECD, “Tax and Economic Growth”, Economics Department Working Paper,
No. 620, ECO/WKP(2008)28, OECD Publishing.

25
Accounting for unintended tax incentive effects
5.7. Have targeted tax incentives for investors and others created unintended tax-
planning opportunities? Are these opportunities and other problems associated with
targeted tax incentives evaluated and taken into account in assessing their cost-
effectiveness?

Rationale for the question

Unfortunately, tax incentives are all too often viewed as a relatively easy “fix” to
promote investment. Indeed, a tax incentive holds out the apparent advantage of
not requiring a cash-equivalent outlay, in contrast with infrastructure development,
manpower training or other programmes introduced to foster investment. The
reasoning goes as follows: by targeting tax relief at new investment, a tax incentive
will only reduce the amount of tax revenue raised on additional investment –
revenue that would not have been raised anyway in the absence of the incentive.
This perception misses the fact that tax incentive relief will create unintended and
unforeseen tax-planning opportunities. To varying degrees, tax incentives will result
in windfall gains – that is, tax relief to investors that does not result in additional
investment but supports investment that would have gone ahead in the absence of
that relief. Even when targeted at new investment, tax incentives will always be
sought by businesses outside the target group. Existing firms will attempt to
characterise themselves as “new”, and other similar tax-planning strategies can be
expected that will deplete tax revenues from activities unrelated to any new
investment attributable to the tax relief, with lost revenues often significantly
exceeding the original projections.

Key considerations
Tax policy features that give rise to unwanted and unintended outcomes might be
adjusted to curtail negative effects. This question encourages policy makers to
carefully consider how tax incentives, depending on their type and design, give rise
to certain unintended and unwelcome results, and what policy changes might be
considered to counter those unwanted effects.

Unintended effects Tax holidays or partial profit exemptions, in particular,

due to tax policy offer significant scope for tax relief unintended by tax
design features authorities. These incentives are typically targeted at
“new” companies. However, the Aladdin’s lamp
problem (“new firms for old”) commonly occurs as old
firms reconstitute as new ones towards the end of their

26
 holiday periods, so that they can continue to be tax-
exempt.
Further, partial or full profit exemption also opens up
transfer pricing opportunities to artificially shift taxable
income from non-qualified business entities to entities
that do qualify. Similarly, channeling asset purchases
through qualifying companies on behalf of non-
qualifying ones is also common. Aggressive transfer
pricing techniques essentially involve the use of non-
arm’s length prices on intra-group transactions and non-
arm’s length interest rates on intra-group loans, to shift
taxable income to low or non-taxed entities.
Targeted tax incentives may create unintended
distortions. For example, investment tax credits could be
abused through “churning” of qualified assets. They
also distort investor choice towards short-lived assets.
Similarly, reinvestment allowances would tend to
discourage investment financed by new equity and may
raise the overall cost of funds, implying welfare losses.

Unintended Unintended distortions may also be created where

interactions of interactions between tax incentives and other
multiple tax provisions of the tax code (e.g. depreciation treatment,
provisions loss treatment) are not adequately addressed. Further,
if not properly co-ordinated, “stacking” of multiple tax
incentives offered by different ministries at the same
level or by different levels of government could occur,
creating unintended distortions, including possible over-
investment.

Unwanted Tax incentives may also encourage corruption and

encouragement of aggravate concerns raised by poor public governance.
corrupt practices When used, targeted tax incentives should be designed
to be as automatic as possible in their application, to
avoid the involvement of tax officials in the
determination of the application of provisions to
individual taxpayers. Also to be avoided are situations
where tax officials undertaking audits have the power to
withdraw tax incentive relief, without special safeguards
against corrupt practices. Frameworks should be in
place to discourage bribery of tax and customs officials
in such cases.

27
Policy practices to scrutinise
Various forms of targeted tax relief may create unintended scope for tax planning,
and result in revenue losses well in excess of the levels originally anticipated (e.g.
where the relief spills over to benefit non-targeted taxpayer groups). It should be
noted, however, that when a previously unforeseen tax-planning opportunity for an
existing tax incentive becomes apparent, it is not without cost for the government
to withdraw the incentive. While cancelling incentive relief for future investment
may be accepted by investors, cancelling relief tied to prior investment decisions –
that may have been based on the expectation of tax incentives previously on offer –
can carry a significant cost. In particular, policy credibility is seriously undermined,
weakening government’s ability to influence investment behaviour in the future
through policy adjustment. Given this, where tax incentive relief linked to
investment expenditure (e.g. enhanced or accelerated depreciation, investment tax
credit) is cancelled, tax relief tied to prior investment generally should be respected
– unless the costs are so exorbitant that respecting past commitments would be
devastating to public finances.
Within the context of a general policy goal to avoid windfall gains (and losses),
transitional considerations related to the introduction and removal of tax incentives
should be addressed. Where tax relief is provided, a general aim is to target tax
relief to incremental investment, that is, investment that would not have occurred
in the absence of the incentive. Conversely, where tax relief is withdrawn, it is
important to attempt to ensure that past investments are not penalised.

Resources for further study

For further reading on unintended tax-planning opportunities and policy
choices/considerations to address unintended tax revenue impacts of targeted tax
incentives, see:
 OECD (2001), Corporate Tax Incentives for Foreign Direct Investment, OECD
Tax Policy Studies, No. 4, OECD Publishing.
 OECD (2011), Corporate Loss Utilization through Aggressive Tax Planning,
OECD Publishing.
 See also OECD (2011), Tackling Aggressive Tax Planning through Improved
Transparency and Disclosure, OECD Publishing.

28
Tax expenditure reporting
5.8. Are tax expenditure accounts reported and sunset clauses used to inform and
manage the budget process?

Rationale for the question

Tax expenditure analysis and reporting are a cornerstone of fiscal policy in countries
where attracting capital and addressing public governance issues remain high on
the political agenda. The primary purpose of the tax expenditure analysis is to
identify the “cost” of preferential tax provisions – revenue losses associated with
tax incentives and other departures from a “benchmark” tax system. The amount of
revenue loss attributable to tax incentives should be reported regularly, ideally as
part of an annual budget process (covering all of the main tax incentives).
Further, like all forms of public spending, tax incentives targeted at investment
should be assessed for their effectiveness. The assessment should be conducted
both in advance and following the implementation, to gauge whether a preferential
tax measure passes or fails a cost-benefit test. To enable such evaluations, the
specific goals of a given tax incentive need to be made explicit at the outset to
enable a proper assessment of the degree to which stated goals are met. When
introducing tax incentive legislation, “sunset clauses” should also be included calling
for the expiry of the incentive (e.g. three years after implementation). An
assessment of the effectiveness of a given tax provision will establish whether the
incentive should be extended or not.

Key considerations
There are several powerful reasons for policy makers to analyse, document and
track their tax expenditures.
 Input to cost-benefit analysis. Most importantly, tax expenditure reporting
serves as input to cost-benefit assessment of tax incentives. As such, it
allows policy makers to initiate steps in containing tax expenditure costs by
supporting decision making on which tax incentives to keep and which ones
to let go.
 Accountability. Publicly available tax expenditure estimates increase public
knowledge of government activities and objectives and permit the
legislature and civil society organisations to scrutinise and hold government
accountable for all aspects of its budget. Furthermore, it allows the public to
more easily track and assess changes in government policy.
 Equity. Since the benefits of a tax expenditure are directly related to both
the tax status of the potential investors and to other provisions in the tax

29
 code, their effect is frequently uneven. Tax expenditure quantification helps
to focus attention on the structure of a tax system and asks the question
“what system is most equitable and efficient?” It thereby forces the
question as to whether each of the various deviations is justifiable.
 Efficiency. Tax expenditure estimates permit a comparison of the indirect
costs of programmes with alternative means of achieving similar objectives.
These alternatives may be either direct expenditures or other tax
expenditures.

Policy practices to scrutinise

Several issues need to be scrutinised to fully assess a host country’s tax expenditure
analysis and reporting practices. The OECD’s Principles of Transparency and
Governance of Tax Incentives for Investment serve as a good framework for the
management issues of investment-related incentives.

Defining a To estimate tax expenditures, the “benchmark” tax

“benchmark” tax structure which contains no preferential tax treatments
structure needs to be established. Tax expenditures are then
defined as deviations from this benchmark tax system. It
is important to recognise that reasonable differences of
opinion exist as to the definition of the benchmark tax
system, and hence what constitutes a tax expenditure.
Labelling a provision as a tax expenditure presupposes
that there is a “normal” tax structure. However, a
normal tax structure is not a fixed concept. For all of
their detail, there are no fully defined and accepted
concepts of taxable income or taxable transactions.
Policy analysts should be aware that cross-country
comparison of tax expenditures might not be valid due
to acceptance of arbitrary definitions and conventions.
Similarly, in-country revisions of methodology and the
“benchmark” can result in substantial changes to the
reported values of tax expenditure over time and make
year-to-year comparisons of tax expenditures invalid.

Identifying redundant Policy analysis should attempt to distinguish between

vs. additional tax redundant versus additional tax expenditures. If the
expenditures investment would not have been made in the absence
of tax incentives, the direct revenue loss is effectively
zero (tax expenditure is “additional”). On the other
hand, if incentives have no effect on investment, then
the entire forgone tax revenue constitutes a revenue

30
 loss (tax expenditure is “redundant”). The true amount
of direct revenue losses is likely to be between these
two extremes.

Estimating tax A range of methodologies exists internationally to

expenditures estimate tax expenditures:
The Revenue forgone method is an ex post quantification
of the extent to which a provision reduces revenues.
Taxpayer behaviour is accepted as given and impacts
resulting from other investment and consumption
patterns or decisions regarding other tax expenditures
are excluded from the estimates.
Revenue gain method is an ex-ante estimate of the total
expected budgetary effect of the expenditure, taking
into account the impact of changes in taxpayer
behaviour and inter-relationships of tax provisions.
Outlay equivalent method estimates the outlay that the
government would need to expend in order to provide
the beneficiaries with the same after-tax economic
benefits that they receive from tax expenditures as
measured by the revenue forgone method. This method
takes into account the fact that taxpayers must often
pay income taxes on the direct government benefits
they receive. As a result, the equivalent outlay estimates
are generally larger in order to account for the portion
of benefits that recipients must pay back in taxes.
In practice, the revenue forgone method is primarily
used. Outlay equivalent and revenue gain estimates can
be quite useful for specific policy analysis, but their
estimation is much more complex than for revenue
forgone. Policy analysts need to be aware of the
drawback of the revenue forgone method. As a
consequence of the all-else-equal nature of revenue
forgone estimates, they are not additive. Many tax
expenditures interact with each other; therefore, a
simple sum of individual tax expenditures is not a
precise reflection of the overall tax expenditure of the
system. Eliminating a particular tax measure will not
translate into an equivalent amount of revenue gain as
the tax expenditure estimate indicates.

31
Accounting for the Often ignored, the indirect costs of tax incentives,
indirect costs of tax including the administrative costs from running them,
incentives could be quite substantial. To present a full picture of
their costs, policy analysts should attempt to quantify
the indirect costs and include them in the total tax
expenditure reporting.

Conducting periodic Once they are granted, tax incentives usually remain in
review of tax laws unless they are revoked or introduced with a
incentives “sunset clause”. Hence, there is a need to assess
performance on a regular basis. Performance reviews
should include the costs as well as the benefits of the
tax incentive and if it has met its intended goals. The
results of such periodic reviews would inform decision
making around the continuation or removal of individual
tax incentives. The review criteria and results should be
reported publicly. To the extent possible, behavioural
responses, both positive (e.g. additional incremental
investment) and negative (e.g. aggressive tax planning)
should be tracked and communicated.

Reporting the The principal beneficiaries of tax incentives for

principal beneficiaries investment by specific tax provision should be included
of tax incentives in the regular statements of tax expenditures. It may be
possible that a few investors, or sectors, benefit from
most tax expenditures. The tax expenditure statement
should have sufficient detail to enable policy makers to
identify which sectors benefit from specific tax
provisions. Where compatible with the requirement of
laws and regulations governing taxpayer confidentiality,
authorities may wish to consider detailing the main
beneficiaries and the amount by which they benefit
from tax incentives. Making such information public can
enhance the legitimacy of governments and their
revenue authorities in the eyes of citizens which, in turn,
can enhance compliance more broadly.

32
Resources for further study
For further reading on a variety of tax expenditure issues, including tax expenditure
estimation techniques and methods to evaluate their effectiveness, see:
 OECD, The Principles of Transparency and Governance of Tax Incentives for
Investment, Task Force on Tax and Development, OECD, Paris.
 Caiumi, A. (2011), “The Evaluation of the Effectiveness of Tax Expenditures –
A Novel Approach: An Application to the Regional Tax Incentives for
Business Investments in Italy”, Taxation Working Papers, No. 5, OECD
Publishing.
 OECD (2010), Tax Expenditures in OECD Countries, OECD Publishing.
 Kraan, D.J. (2004), “Off-budget and Tax Expenditures”, OECD Journal on
Budgeting, Vol. 4/1, OECD Publishing.

33
International tax co-operation
5.9. Are tax policy and tax administration officials working with their counterparts in
other countries to expand their tax treaty network and to counter abusive cross-
border tax planning strategies?

Rationale for the question

Tax treaties aim to improve the certainty of cross-border projects and lower the
costs and risks of international investment. A wide tax treaty network is helpful to
countries seeking to attract investment in several ways:
 Most importantly, tax treaties operate to avoid double taxation of cross-
border investment returns.
 Treaties reduce investor uncertainty over tax treatment by establishing
procedures to help resolve disputes over the allocation of taxing rights
between host and home countries.
 Tax treaties usually seek to stipulate lower non-resident withholding tax
rates on dividends, interest and royalties.
 Often, treaties limit tax paid in the home (resident) country to that of the
host country.
Further, tax treaties bring clarity into cross-broader dealings by providing a
framework to exchange information among tax authorities to counter more
aggressive forms of tax planning.

Key considerations
Tax treaties generally reduce the uncertainty and cost of international investment;
they achieve this in several major ways.

Avoid double taxation First, and perhaps foremost, tax treaties operate to
avoid double taxation of cross-border investment
returns. In the absence of a tax treaty between a host
and home country, double taxation of returns will
normally arise where the two countries treat a given tax
return differently. For example, countries may take
different views on the source or origin of income, or the
type of income paid (e.g. interest versus dividends), with
different characterisations triggering different tax
treatment. Tax treaties operate to avoid these different
characterisations and thereby minimise the scope for
double taxation.

34
Reduce investor Tax treaties help reduce investor uncertainty over tax
uncertainty treatment. Indeed, certain articles of tax treaties are
specifically aimed at establishing procedures to help
resolve disputes over the allocation of taxing rights
between host and home countries. A wide tax treaty
network therefore tends to make countries more
attractive, in relation to tax considerations, both as
locations for business activity and as places from which
to conduct global business operations, by lowering
project costs as well as project risks.
Lower non-resident Tax treaties generally stipulate lower non-resident
tax withholding tax rates on dividends, interest and
royalties. Indeed, treaty-negotiated rates are often
significantly lower than statutory withholding tax rates
that would otherwise apply. This aspect of tax treaties
also serves to lower project costs.
Enable the exchange Tax treaties provide a framework to enable the
of information exchange of information amongst tax authorities to
amongst tax counter more aggressive forms of tax planning in
authorities relation to foreign source income as well as domestic
source income (that may be stripped out to tax havens
through the use of special corporate structures and
financing and repatriation strategies).

Policy practices to scrutinise

Number of tax treaty partners. An assessment of the scope of a country’s tax treaty
framework requires consideration of the specific countries that are its treaty
partners. A long list generally implies broad access to investors, particularly where a
number of the partners are large capital-exporting countries. However, the scope of
a host country’s tax treaty network cannot be assessed by a simple count of the
number of its tax treaty partners. While a large number of treaty partners may
signal a large pool of potential investors, it is not necessarily the case that a small
number of treaty partners signals the opposite. Treaties with certain countries
(e.g. Belgium, the Netherlands, the United Kingdom) that are attractive conduit
locations for routing investment may provide the host country with access to
investors in a large number of countries – in particular, to countries that have a tax
treaty with one or more of these countries as hosts to intermediate financing
services.

35
Tax treaty negotiation dates. Similarly, the scope of a host country’s tax treaty
network in reaching investors cannot be assessed by reference to the dates on
which its tax treaties were negotiated or renegotiated. When a treaty was brought
into force many years ago, the conclusion cannot be drawn that it does not provide
benefits that are less (or more) attractive than a relatively new treaty.
Level of non-resident withholding taxes. One of the central benefits tax treaties
provide to investors is the reduction to non-resident withholding tax rates on
payments used to repatriate earnings – dividends, interest and royalties. Therefore,
an important consideration is the level of non-resident withholding tax rates
negotiated with key tax treaty partners. To guide the assessment of tax treaties, the
table below suggests withholding tax rates that would generally be regarded as
attractive by investors.

Payment type Treaty withholding tax rate

Interest 10%
Direct dividends 5%
Portfolio dividends 15%
Royalties 10%

Resources for further study

 The OECD has proposed a model treaty to use as a starting point in
international treaty negotiations. The OECD Model Convention on Income
and on Capital provides a means for settling (on a uniform basis) the most
common problems that arise in the field of international exchange of goods
and services and cross-border movements of capital, technology and
persons.

36
Annex 5.1. Methodology for calculating the METR
The methodology for estimating marginal effective tax rates (METRs) is extensively
documented.5 We follow the discussion in Chen and Mintz (2008).6
The standard theory of investment defines the METR as:
rG – rN
METR =
rG
where rG is the pre-tax rate of return (at the margin) required by an investor and rN is
the after-tax rate of the return (at the margin).
The after-tax rate of return, rN, is defined by the formula:

rN = βi + (1 − β)ρ − π

where β is the debt-to-assets ratio, i is the nominal interest rate on debt finance, ρ is
the nominal required rate of return on equity and π is the inflation rate.
One of the main components of the pre-tax rate of return, rG, is the real cost of
funds, rf, defined as:

rf = βi (1 − Uj) + (1 − β)ρ − π

where Uj is the statutory corporate income tax rate of sector j.

Next, rG is defined for four classes of assets: building, machinery, land and inventory.

5. See, for example:

 OECD (1995), Taxation and Foreign Direct Investment: The Experience of the
Economies in Transition, OECD Publishing.
 McKenzie, K.J., M. Mansour and A. Brûlé (1998), “The Calculation of Marginal
Effective Tax Rates”, Working Paper, Vol. 97, No. 15, Technical Committee on
Business Taxation, Department of Finance, Canada.
 Mintz, Jack M. (1995), “Tax Holidays and Investment”, in Shah, A. (ed.), Fiscal
Incentives for Investment and Innovation, pp. 165-194, Oxford University Press, New
York.
6. Chen, D. and J. Mintz (2008), “Taxing Business Investment: A New Ranking of Effective
Tax Rates on Capital”, prepared for the Foreign Investment Advisory Service of the
World Bank.

37
Depreciable assets: Building and machinery. For depreciable assets, rG is estimated as:
ct (1 – U)
1–A+
(rf + π + α)
rG = (1 + µ) (rf + δ) (1 – k) ( )–δ
(1 – U) (1 – tp – tg)
where µ is the non-creditable transaction tax (such as import duty and sales tax) on
capital goods; δ is the economic depreciation rate; k is the investment tax credit
rate; A is the present value of future tax savings from depreciation allowances,
defined below for various depreciation schedules, ct is the capital tax rate, α is the
tax depreciation rate, tp is the property tax rate and tg is the gross receipts tax rate
or presumptive tax that is based on the gross revenue.
Land. For land, rG is defined by:
ct (1 – U)
1+
(rf + π)
rG = rf (1 + µ)
(1 – U) (1 – tp – tg)
where µ is the property transfer tax.
Inventory. For inventory, rG is defined as:
(rf + Uπζ)
rG = (1 + µ) + ct
(1 – U) (1 – tg)
where µ is the sales tax on raw materials (when applicable); and ζ = 1 for the FIFO
accounting method, 0 for the LIFO method and 0.5 for the average cost method.

The expression for A (the present value of future tax savings from depreciation
allowances) for a declining-balance depreciation schedule is defined as:
Uα
A=
α + rf + π
The expression for A in case of a straight-line depreciation schedule:
U (1 – e –( rf + π)L)
A=
(rf + π)L
where L is the lifetime for each depreciable asset, with the asset depreciated at the
rate 1/L in each year.
The expression for A when a portion γ of an asset can be immediately expensed,
with the remaining (1 – γ) depreciated under a declining-balance depreciation
schedule is defined as:

38
 (rf + π) γ + α
A =U
α + rf + π
The METR for a given industry is the proportional difference between the weighted
average of the before-tax rate of return by asset type and the after-tax rate of
return; the latter is the same across asset types within a given sector. That is, the
METR for industry i (ti), is calculated as follows:
∑j rgij wij – rni
ti = ∑j rgij wij
where j denotes the asset type (that is, investments in buildings, machinery,
inventories and land) and wij denotes the weight of asset j in industry i.

Annex 5.2. Methodology for calculating the AETR

The methodology for estimating average effective tax rates (AETRs) has been
developed by Devereux and Griffith.7 We present here only the final formulas for the
calculations; please see the original publication for the detailed methodology notes.
The AETR is calculated as:
p–r γ (1 + r)(p + δ)(1 + π)(1 – τ) γ (1 + A)(1 + r) (1 – δ)(1 + π) F (1 + r)
AETR = – + (1 – )–
P p (1 + ρ) p 1+ρ P

with the variables and their definitions as presented below:

Variables Definition

p pre-tax net profit

r real interest rate
factor measuring the difference in treatment of new equity and
γ = (1−md)/ (1− z)
distributions
md personal tax on dividends
z tax on capital gains
ρ = (1−mi)i investor’s discount rate
mi personal tax rate on interest

7. See Devereux, Michael P. and Rachel Griffith (2003), “Evaluating Tax Policy for Location
Decisions”, CEPR Discussion Paper Series, No. 3 247, Centre for Economic Policy Research.

39
i nominal interest rate
τ corporate tax rate
δ economic depreciation rate
π inflation
present discounted value of depreciation allowances, which can be
calculated for any depreciation scheme.

1+ρ for declining

A A = τϕ
ρ+ϕ balance
depreciation
τϕ (1 + ρ) 1 in case of straight-
A= (1 – ( )1/ρ) line depreciation
Ρ 1+ρ

ϕ depreciation allowance
Financial effects

F=0 in case of
financing by
retained earnings
– ρ (1 – γ) in case of
F FNE = (1 – ϕτ) financing by new
1+ρ
equity
γ (1 – ϕτ) in case of
FD = (ρ – i (1 – τ))
financing by debt
1+ρ

40