International Tax Primer

International Tax Primer

Third Edition

Brian J. Arnold
Published by:
Kluwer Law International B.V.
PO Box 316
2400 AH Alphen aan den Rijn
The Netherlands
Website: www.wklawbusiness.com

Sold and distributed in North, Central and South America by:

Wolters Kluwer Legal & Regulatory U.S.
7201 McKinney Circle
Frederick, MD 21704
United States of America
Email: [email protected]

Sold and distributed in all other countries by:

Turpin Distribution Services Ltd
Stratton Business Park
Pegasus Drive, Biggleswade
Bedfordshire SG18 8TQ
United Kingdom
Email: [email protected]

Printed on acid-free paper.

ISBN 978-90-411-5981-6

© 2016 Kluwer Law International BV, The Netherlands

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 written
permission from the publisher.
Permission to use this content must be obtained from the copyright owner.
Please apply to: Permissions Department, Wolters Kluwer Legal &
Regulatory U.S. 76 Ninth Avenue, 7th Floor, New York, NY 10011-5201,
USA. Website: www.wklawbusiness.com
 Table of Contents

Preface

CHAPTER 1
Introduction
1.1 Objectives of This Primer
1.2 What Is International Tax?
1.3 Goals of International Tax Rules
1.4 The Role of the Tax Adviser in Planning International Transactions

CHAPTER 2
Jurisdiction to Tax
2.1 Introduction
2.2 Defining Residence
2.2.1 Residence of Individuals
2.2.2 Residence of Legal Entities
2.2.3 Treaty Issues Relating to Residence
2.3 Source Jurisdiction
2.3.1 Introduction
2.3.2 Employment and Personal Services Income
2.3.3 Business Income
2.3.4 Investment Income

CHAPTER 3
Taxation of Residents
3.1 Introduction
3.2 Taxation of Residents on Their Worldwide Income
3.2.1 Tax Policy Considerations
3.2.2 The Tax Consequences of Residence
3.2.3 Double Taxation
3.2.4 Computation of the Foreign Source Income of Residents
3.2.4.1 In General
3.2.4.2 Foreign Exchange Gains and Losses
3.2.4.3 The Treatment of Losses
3.2.5 Tax Administration Issues
3.3 Exceptions to Worldwide Taxation
3.3.1 Nonresident Companies and Other Legal Entities
3.3.2 Temporary Residents
3.4 Special Issues
3.4.1 Exit or Departure Taxes
3.4.2 Trailing Taxes
3.4.3 Resident for Part of a Year

CHAPTER 4
Double Taxation Relief
4.1 Introduction
4.2 International Double Taxation Defined
4.3 Relief Mechanisms
4.3.1 Deduction Method
4.3.2 Exemption Method
4.3.2.1 Participation Exemption
4.3.3 Credit Method
4.3.3.1 General Rules
4.3.3.2 Types of Limitations
4.3.3.3 Indirect or Underlying Credit
4.3.4 Comparison of the Exemption and Credit Methods
4.3.5 Treaty Aspects
4.4 Allocation of Expenses
4.5 Tax Sparing

CHAPTER 5
Taxation of Nonresidents
5.1 Introduction
5.2 Tax Policy Considerations in Taxing Nonresidents
5.3 Threshold Requirements
5.4 Source Rules
5.5 Double Taxation
5.6 Excessive Taxation of Nonresidents
5.7 Computation of the Domestic Source Income of Nonresidents
5.8 Taxation of Various Types of Income of Nonresidents
5.8.1 Business Income
5.8.1.1 In General
 5.8.1.2 Branch Taxes
5.8.2 Income from Immovable Property
5.8.3 Income from Employment
5.8.4 Investment Income: Dividends, Interest, and Royalties
5.8.5 Capital Gains
5.9 Administrative Aspects of Taxing Nonresidents
5.9.1 Introduction
5.9.2 Obtaining Information about Domestic Source Income of
Nonresidents
5.9.3 Collection of Tax from Nonresidents

CHAPTER 6
Transfer Pricing
6.1 Introduction
6.2 The OECD Transfer Pricing Guidelines
6.3 United Nations, Practical Manual on Transfer Pricing for Developing
Countries
6.4 The Arm’s-Length Standard
6.4.1 Introduction
6.4.2 Comparability Analysis
6.4.3 Comparable Uncontrolled Price Method
6.4.4 Resale-Price Method
6.4.5 Cost Plus Method
6.4.6 Comparison of Traditional Methods
6.4.7 Profit-Split Method
6.4.8 Transactional Net Margin Method
6.5 Sharing of Corporate Resources
6.5.1 Introduction
6.5.2 Loans or Advances
6.5.3 Performance of Services
6.5.4 Use of Tangible Property
6.5.5 Use or Transfer of Intangible Property
6.6 Cost-Contribution Arrangements
6.7 Disregarded Transactions
6.8 Transfer Pricing Documentation Requirements
6.9 Treaty Aspects of Transfer Pricing
6.10 Tax Policy Considerations: Formulary Apportionment and the Future of
 the Arm’s-Length Method

CHAPTER 7
Anti-avoidance Measures
7.1 Introduction
7.2 Restrictions on the Deduction of Interest: Thin Capitalization and
Earnings-Stripping Rules
7.2.1 Introduction
7.2.2 The Structural Features of Thin Capitalization and Earnings-
Stripping Rules
7.3 CFC Rules
7.3.1 Introduction
7.3.2 Structural Features of CFC Rules
7.3.2.1 Definition of a CFC
7.3.2.2 Designated Jurisdiction or Global Approach
7.3.2.3 Definition and Computation of Attributable Income
7.3.2.4 Nature and Scope of Exemptions
7.3.2.5 Resident Taxpayers Subject to Tax
7.3.2.6 Relief Provisions
7.3.3 Tax Treaties and CFC Rules
7.4 Nonresident Trusts
7.5 Foreign Investment Funds

CHAPTER 8
An Introduction to Tax Treaties
8.1 Introduction
8.2 Legal Nature and Effect of Tax Treaties
8.2.1 Vienna Convention on the Law of Treaties
8.2.2 The Relationship between Tax Treaties and Domestic Law
8.3 The OECD and UN Model Tax Treaties
8.4 The Process of Negotiating and Revising Tax Treaties
8.5 Objectives of Tax Treaties
8.6 Interpretation of Tax Treaties
8.6.1 Introduction
8.6.2 The Interpretive Provisions of the Vienna Convention on the
Law of Treaties
8.6.3 The Interpretation of Undefined Terms in Accordance with
 Domestic Law — Article 3(2)
8.7 Summary of the Provisions of the OECD and UN Model Treaties
8.7.1 Introduction
8.7.2 Coverage, Scope, and Legal Effect
8.7.3 The Distributive Rules: Articles 6 through 21
8.7.3.1 Introduction
8.7.3.2 Business Income
8.7.3.3 Employment and Personal Services Income
8.7.3.4 Income and Gains from Immovable Property
8.7.3.5 Reduced Withholding Rates on Certain Investment
Income
8.7.3.6 Other Types of Income
8.7.4 Administrative Cooperation
8.8 Special Treaty Issues
8.8.1 Nondiscrimination
8.8.2 Treaty Abuse
8.8.2.1 Introduction
8.8.2.2 Treaty Shopping
8.8.2.3 OECD BEPS Action 6: Preventing Treaty Abuse
8.8.3 Resolution of Disputes
8.8.4 Administrative Cooperation
8.8.5 Attribution of Profits to Permanent Establishments

CHAPTER 9
Emerging Issues
9.1 Introduction
9.2 Base Erosion and Profit Shifting (BEPS)
9.2.1 Introduction
9.2.2 The 1998 OECD Harmful Tax Competition Report
9.2.3 The G20/ BEPS Project
9.3 Hybrid Arrangements
9.3.1 What Is a Hybrid Arrangement?
9.3.2 Hybrid Entities
9.3.3 Hybrid Financial Instruments
9.3.4 OECD BEPS Proposals for Hybrids
9.4 Fees for Management, Technical, and Consulting Services
9.4.1 Introduction
 9.4.2 The Taxation of Income from Management, Technical, and
Consulting Services under Tax Treaties
9.4.3 Proposed Article on Fees for Management, Technical, and
Consulting Services in the UN Model Treaty
9.5 Arbitration
9.6 The Digital Economy – Electronic Commerce
9.6.1 Introduction
9.6.2 Tax Challenges Posed by the Digital Economy

Glossary of International Tax Terms

Index
 Preface

The first edition of the International Tax Primer began in the 1990s as a
project to develop materials for the OECD to use in its outreach activities
with emerging economies in Europe after the breakup of the Soviet Union. I
recruited Mike McIntyre of Wayne State University to work with me in
preparing those materials. Unfortunately, the materials we prepared proved to
be unacceptable to the OECD largely because, in our view at least, it did not
adhere sufficiently to the OECD positions on several international tax issues.
However, the OECD generously permitted us to use those materials as the
foundation for what became the International Tax Primer.
Much to my surprise and delight, the first two editions of the Primer
were well received by students of international tax from all over the world. I
hope that a new generation of students will find this third edition to be
helpful in the increasingly challenging task of understanding international
tax.
The third edition does not have the benefit of Mike McIntyre’s insight
and wisdom. Mike passed away in 2013 after a long illness. We discussed the
need for a third edition on several occasions, so I know that Mike would have
approved of a third edition in principle; but I also know that his involvement
would have made it a better book.
The third edition of the International Tax Primer has been expanded
substantially. A lot has happened in the international tax world since the
second edition was published in 2004. The chapters on transfer pricing, anti-
avoidance rules, and tax treaties have been substantially revised to reflect the
changes of the last decade. Similarly, the chapter dealing with recent
developments has been changed significantly to deal with the OECD’s Base
Erosion and Profit Shifting project, arbitration of tax disputes, and the
proposed article in the United Nations Model Treaty dealing with fees for
technical services, none of which were dealt with in the second edition. In
addition, new chapters have been added dealing with basic aspects of the
taxation of residents on their foreign source income and nonresidents on their
domestic source income.
Large ongoing projects, such as BEPS, pose a special challenge for
authors. Obviously, BEPS is so important that it cannot simply be ignored.
However, when we do not know what the final recommendations will be and
whether countries will act on them, it is difficult to decide how much detail to
provide on the draft proposals. I have opted to provide a somewhat detailed
discussion of the BEPS project in general in Chapter 9 dealing with
Emerging Issues, and at least a brief mention of each BEPS action item at
places in the book where a particular topic is dealt with. Needless to say, I
anticipate that a fourth edition of the Primer will be necessary in a few years
once the full impact of the BEPS project is clearer.
Readers will quickly notice that the Primer does not contain any
footnotes or a selected bibliography. Given the basic nature and purpose of
the Primer, I concluded that footnotes were unnecessary and would make the
text less readable; however, any references that I considered to be necessary
are provided in the text. I also concluded that a selected bibliography was
unnecessary, largely because there is an extensive literature dealing with
international tax and, with the Internet, it is relatively easy to find relevant
sources.
I wish to thank my assistant, Carol Hargreaves, who edited and
proofread the manuscript with her customary skill more than once.
The manuscript is up to date as of September, 2015.

Brian J. Arnold
Ailsa Craig, Ontario, Canada
September 2015
CHAPTER 1
Introduction

1.1 OBJECTIVES OF THIS PRIMER

This Primer on international taxation provides the reader with an introductory
analysis of the major issues that a country must confront in designing its
international tax rules and in coordinating those rules with the tax systems of
its trading partners. At one time, international tax issues were important to a
small circle of tax specialists, primarily the tax advisers of large multinational
corporations and their counterparts in the tax departments of developed
countries. As the countries of the world have become increasingly integrated
economically, the importance of these issues has mushroomed. Many small-
and medium-sized firms, as well as individuals, now engage in cross-border
transactions that cause them and their tax advisers to confront international
tax issues on a regular basis; and most national governments must care about
international tax, both to present a hospitable environment for foreign
investment and to protect their revenue base.
Although this Primer is intended mainly for students, government
officials and tax practitioners who are confronting international tax for the
first time, I fondly hope that those with considerable experience in
international tax may also find it useful. Many times in my work, I have been
forced to return to fundamental principles in analyzing complicated tax
issues. In essence, the objective of this Primer is to articulate these
fundamental principles.
International tax planning is firmly grounded, if not mired, in the
technical minutiae of a particular country’s tax rules. Thus, in this Primer it
has been necessary to provide some level of detail on some issues in order for
the discussion of these issues to have any practical significance. However, the
objectives of a primer would be lost if it did not focus on general principles
and fundamental structure. I have tried to balance the need for both the
specific and the general by illustrating general principles with frequent
references to the actual practices of a variety of developed and developing
countries.
 The many examples provided throughout this Primer are given for
illustrative purposes only and are not meant to be definitive statements about
the laws of particular countries. No attempt is made to survey the practices of
all countries. I have avoided writing from the perspective of any particular
country, including the countries with which I am most familiar. Instead, I
have tried to identify and discuss issues of international tax that are relevant
and important to many countries.
Section 1.2 of this introductory chapter describes the meaning of the
term “international tax”. Section 1.3 identifies the most important goals that
should guide countries in designing their international income tax rules.
Section 1.4 describes the role of the tax adviser in planning international
transactions and offers a few examples of typical international tax planning
techniques.
Chapter 2 describes the rules that countries have adopted for defining
the scope of their jurisdiction to tax. Most countries tax residents and
nonresidents differently. Chapter 3 examines the issues involved in taxing
residents of a country on their worldwide income. If one country taxes its
residents on their worldwide income and another country taxes the same
income because it arises, is earned, or has its source in that country, the
income will inevitably be subject to double taxation. The mechanisms used to
mitigate the risks to taxpayers of this and other forms of international
double taxation are addressed in Chapter 4. As a counterpart to Chapter 3,
Chapter 5 examines the issues involved in a country taxing nonresidents on
their income earned in or sourced in the country. Chapter 6 examines the
controversial issue of transfer pricing rules for adjusting intercompany
transfer prices to prevent the avoidance of tax by multinational corporations.
Chapter 7 discusses a variety of anti-avoidance rules dealing with
international transactions, such as controlled foreign company (CFC) rules
and thin capitalization or earnings-stripping rules. Chapter 8 provides an
overview and analysis of the provisions of bilateral tax treaties and the
OECD (Organisation for Economic Co-operation and Development) and
United Nations (UN) Model Treaties on which they are generally based.
Several important emerging issues that cut across the issues addressed in the
prior chapters are addressed in Chapter 9. Those issues include the OECD’s
initiative against base erosion and profit shifting (BEPS), the tax aspects of
hybrid entities and financial instruments, the taxation of fees for technical
services under the UN Model Treaty, the use of arbitration to resolve
international tax disputes, and the challenges posed by taxation of income
derived from the digital economy.
There is an extensive glossary of international tax terms after Chapter 9.
The first time a term included in the glossary is used in the text, it is shown in
bold-face type. The meanings of the terms in the glossary reflect their
meanings in an international context; some of the terms may have a slightly
different meaning in a domestic context.

1.2 WHAT IS INTERNATIONAL TAX?

The term “international tax” is used in this Primer for convenience because
international tax law is more correctly referred to as the international aspects
of the income tax laws of particular countries. With minor exceptions, tax
laws are not “international” – they are creations of sovereign states. Arguably
at least, there is no overriding international law of taxation, arising either
from the customary practice of sovereign states or from the actions of some
international body such as the UN or the OECD.
Tax treaties are perhaps the most obvious “international” aspect of a
country’s income tax system. Most developed countries have entered into tax
treaties with their major trading partners, and often with their minor trading
partners as well. Many developing countries also have extensive treaty
networks. The growth in the number of tax treaties over the past decade has
been exponential – there are now over 3,000 bilateral tax treaties in existence.
These treaties impose significant limitations on the taxing powers of the
signatories to the treaty (often referred to as the contracting states). Tax
treaties, however, do not generally impose tax; in most countries, they are
exclusively relieving in nature. Although tax treaties are binding agreements
between sovereign states, in many countries they do not have any effect on
taxpayers unless they are specifically incorporated into a country’s tax law.
The scope of what is called international tax in this Primer is extremely
broad. It encompasses all tax issues arising under a country’s income tax laws
that include some foreign element: for example, cross-border trade in goods
and services, cross-border manufacturing, production, and resource
development by a multinational enterprise, cross-border investment by
individuals or investment funds, and individuals working outside the country
where they usually reside. These activities usually present international tax
issues under the tax laws of at least two countries.
 Some international tax issues arise out of extremely complex situations.
The reorganization of a multinational corporation with foreign subsidiaries
in several countries is an example. Other situations may be quite simple. For
example, an international tax issue may arise under some countries’ tax laws
if a resident individual attempts to claim a deduction for the support of a
dependent spouse or child residing in a foreign country.
The international tax law of a country has two broad dimensions:

(1) the taxation of resident individuals and legal entities on income

arising in foreign countries; and
(2) the taxation of nonresidents on domestic income (i.e., income
arising or sourced in the country).

The first dimension is referred to in this Primer as the “taxation of

residents on foreign income”, and the second dimension as the “taxation of
nonresidents on domestic income”. Obviously, what is the taxation of
residents on foreign income for one country (generally referred to as the
residence country) is the taxation of nonresidents on domestic source
income for another country (generally referred to as the source country).
A transaction that involves the export of capital or other resources from
a country is often referred to by tax analysts as an outward-bound or
“outbound” transaction. Conversely, the term inward-bound or “inbound”
transaction is commonly used to refer to a transaction involving the import of
capital or other resources from a foreign country. A transaction that a country
considers to be an outward-bound transaction typically involves its rules for
taxing the foreign income of resident taxpayers. In contrast, inward-bound
transactions typically involve a country’s rules for taxing nonresidents on
domestic income. In some circumstances, a single transaction may have
consequences under both sets of rules. An example is the liquidation of a
foreign affiliate into a domestic parent corporation.
International tax extends beyond income tax. It may include estate taxes,
gift taxes, inheritance taxes, general wealth taxes, value-added taxes, customs
duties, and a variety of special levies. The international aspects of estate and
gift taxes are particularly important. For example, such wealth-transfer taxes
have important international implications when a resident receives a bequest
or gift from a nonresident or non-domiciled individual or when a person dies
owning property in a foreign country. These important issues are beyond the
scope of this book, which is restricted to international aspects of income tax
law.

1.3 GOALS OF INTERNATIONAL TAX RULES

In designing its international tax rules, a country should generally seek to
advance the four major goals described below. Often these goals conflict, so
that a country must try to achieve a balance between them. Some of the
policy goals of international tax can be pursued effectively through unilateral
action; however, other goals can be achieved only through cooperation with
other countries.
Revenue considerations. Governments raise tax revenues to fund public
goods and services. From a purely national perspective, every country wants
to maximize its tax revenues. However, this goal conflicts with other goals,
such as the need to attract foreign investment and other countries’ revenue-
raising goals. From an international perspective, each country should obtain
its fair share of the tax revenues from income generated by transnational
activities. To achieve this goal of inter-nation equity, a country must protect
its domestic tax base – that is, it must develop a good domestic tax system
and an effective tax administration to enforce its tax rules, and it must avoid
entering into tax treaties that inappropriately limit its right to tax the domestic
source income of residents and nonresidents.
Fairness. The primary advantage of an income tax over other potential
taxes is fairness. In general, fairness is achieved by imposing equal tax
burdens on individuals with equal income, without reference to the source or
type of the income (so-called horizontal equity), and by making those
burdens commensurate with the ability to pay of individuals (so-called
vertical equity – the more you make, the more you pay). Fairness is not a
relevant consideration with respect to taxes imposed on corporations and
other legal entities because such entities are legal fictions created by the law
that, unlike natural persons, do not have any tangible existence in the real
world. Although corporations and other legal entities may pay tax, that tax
must ultimately be borne by natural persons – the shareholders, employees, or
customers of a corporation. It is unclear to what extent the corporate tax is
passed on to its shareholders, employees, or customers; this lack of solid
information as to the incidence of the corporate tax makes it difficult for
countries to implement good tax policies for taxing resident corporations on
their foreign source income and nonresident corporations on their domestic
source income.
Often when commentators talk about fairness with respect to
corporations, they are really referring to considerations of economic
efficiency and neutrality, which are discussed below.
For resident individuals, fairness requires the full taxation of both
domestic and foreign source income; moreover, foreign source income must
be taxed whether the income is earned directly or through some foreign
entity. However, no country has the power to impose a fairness standard on
nonresidents earning domestic source income because no country can tax all
the income of nonresidents arising outside its borders. For example, an
individual resident in Country A may earn income in Country A, Country B
and Country C. In general, Country B has jurisdiction to tax only the
individual’s income arising in Country B. It will not have any information
about, and cannot take into account, the individual’s income earned in
countries A and C. Countries can promote fairness, however, by contributing
to the development of fair and appropriate international tax standards, by
imposing tax burdens that are consistent with these standards, and by
otherwise cooperating with other countries in the assessment and collection
of tax on their residents.
Competitiveness considerations. Every country should care about the
welfare of nonresidents. Nevertheless, each country has a primary duty to
advance the economic interests of its own citizens and residents. To this end,
a country should avoid tax measures that undermine its competitive position
in the global economy.
In the international context, countries compete; tax competition is one
way in which countries compete for jobs and investment. Some countries try
to attract jobs and investment by reducing or eliminating taxes generally or
on income from certain activities. However, a particular country’s
competitiveness depends on a wide variety of other factors, including an
educated labor force, modern infrastructure, political stability and an
established legal system with protection for investors, and natural resources.
Countries can enhance their competitiveness by removing provisions of their
tax law that tend to encourage the movement of investment and jobs out of
the country or that discourage the importation of capital and jobs. In the
medium and long run, a country’s competitiveness is not enhanced by tax
incentives; these and other beggar-thy-neighbor policies invite a retaliatory
response by foreign governments and a “race to the bottom”, to the detriment
of all countries. Such policies erode the ability of all governments to impose
fair and effective taxes on income from movable capital.
Capital-export and capital-import neutrality. The principles of capital-
export neutrality or capital-import neutrality usually figure prominently in
discussions of international tax policy. Readers should be aware of these
concepts, although their importance is doubtful.
The principle of capital-export neutrality is that a country’s international
tax rules should neither encourage nor discourage outflows of capital.
Capital-export neutrality would be achieved if a country taxes its residents,
including its resident corporations, on their worldwide income, including
income earned by their foreign subsidiaries. In practice, policymakers
typically treat capital-export neutrality as at best a secondary goal with
respect to corporations. In virtually every country, capital inflows are
generally considered to be desirable and are encouraged through tax and
other economic policies. In contrast, capital outflows are generally thought to
diminish a country’s national wealth. Many countries adopt measures
designed to discourage capital outflows, although their tax laws may also
contain provisions that have the unintended effect of encouraging outflows.
Prudent policymakers exercise caution in discouraging outflows because
limitations on capital outflows may discourage capital inflows. For example,
a country that imposes excessively high withholding taxes on dividends,
interest and royalties paid to nonresidents is likely to discourage nonresidents
from investing in that country.
According to the principle of capital-import neutrality, taxpayers doing
business in a country should be subject to the same tax burden irrespective of
where they are resident. Capital-import neutrality is generally achieved to the
extent that a country exempts its residents, including its resident corporations,
from tax on their foreign source income, including income earned by their
foreign subsidiaries. Thus, if Country A does not tax corporations resident in
Country A on the income earned by their foreign subsidiaries, the
subsidiaries will be subject to tax only by the countries in which they are
resident, and in the same way as other corporations resident in those
countries.
Most countries have adopted international tax rules that contain some
features that are consistent with both capital-export neutrality and capital-
import neutrality. For example, most countries tax resident individuals on
their worldwide income, which reflects capital-export neutrality. In contrast,
most countries do not tax foreign source income earned by foreign
corporations that are controlled by residents (except in special
circumstances), which reflects capital-import neutrality. Capital-import
neutrality is widely accepted with respect to foreign business income earned
by corporations, so that such income is taxable only by the country in which
it is earned. Further, such income is either exempt from tax by the country in
which the corporation is resident or that country’s tax is deferred until it is
repatriated, usually in the form of dividends.
More recently, several tax analysts have referred to another concept of
neutrality – capital-ownership neutrality. Whereas capital-export neutrality
and capital-import neutrality focus on the location of capital, capital-
ownership neutrality also focuses on the ownership of capital. Under an ideal
tax system based on capital-ownership neutrality, tax would not distort the
ownership of assets by taxpayers. Capital-ownership neutrality is achieved if
all countries tax on either a worldwide or territorial basis.
The fairness and efficiency of income taxation ultimately depends not
on the income tax laws of any one country but on the cumulative effects of
the income tax laws of all countries. Countries have little to lose and much to
gain by coordinating their income tax systems with the tax systems of their
trading partners. Tax treaties are the primary means for achieving such
coordination.
Income tax treaties have two primary operational goals – to reduce the
risk of double taxation of taxpayers engaged in cross-border transactions and
to ensure that income from cross-border transactions does not escape tax
entirely (sometimes referred to as double non-taxation or stateless income).
Both of these goals are advanced by measures that promote the
harmonization of international tax rules through the adoption of income tax
treaties that follow the same general pattern. Other ancillary objectives of tax
treaties include the prevention of discrimination against nonresidents and
foreign nationals and administrative cooperation in exchanging information,
collecting tax, and resolving disputes. Virtually all modern income tax
treaties are based in substantial part on the OECD and UN Model Treaties.
The UN Model Treaty is based heavily on the OECD Model Treaty, although
it contains some alternative and additional provisions that allow developing
countries to tax more income than is permitted under the OECD Model
Treaty. Tax treaties are discussed in Chapter 8.
1.4 THE ROLE OF THE TAX ADVISER IN PLANNING
INTERNATIONAL TRANSACTIONS
The tax adviser’s role with respect to international transactions is similar to
his or her role with respect to domestic transactions. Probably the tax
adviser’s most important obligation is to ensure that clients do not fall into
any traps or anomalies that result in levels of taxation beyond what might
reasonably be expected. Such defensive tax planning should not ordinarily
put the tax adviser in an adversarial role with a country’s tax officials, who
should also be seeking to impose appropriate tax burdens on taxpayers.
Domestic and international tax advisers are also expected to be acquainted
with international tax strategies that might be used to minimize taxes. These
schemes often involve the use of countries with low or no taxes, either
generally or on certain types of income; these countries are commonly
referred to as tax havens.
International tax advisers are likely to spend more of their time engaging
in defensive tax planning than their domestic counterparts, since taxpayers
engaged in international transactions frequently confront serious risks of
paying excessive levels of tax. These risks typically arise when two or more
countries claim the right to impose tax on the same items of income. Many
important international tax rules are designed to mitigate or eliminate such
double taxation. The measures commonly used to relieve double taxation are
discussed in Chapter 4.
International tax advisers are likely to spend more of their time engaging
in defensive tax planning than their domestic counterparts, since taxpayers
engaged in international transactions frequently confront serious risks of
paying excessive levels of tax. These risks typically arise when two or more
countries claim the right to impose tax on the same items of income. Many
important international tax rules are designed to mitigate or eliminate such
double taxation. The measures commonly used to relieve double taxation are
discussed in Chapter 4.
Although visible and newsworthy, offensive tax planning activities
occupy a relatively modest part of the practice of most international tax
advisers. However, these activities may constitute a major part of the practice
of some large law and accounting firms and have caused governments to
respond with increasingly complex anti-avoidance legislation. The most
important of the rules designed to combat international tax avoidance are
discussed in Chapters 6, 7, and 8. These rules have not driven the tax havens
out of business – opportunities for international tax avoidance are still widely
available to individual investors and multinational enterprises.
The role of the tax adviser depends on whether the transaction involved
is an outward-bound or an inward-bound transaction or investment. In the
case of an outward-bound investment by a resident taxpayer, the tax adviser
often has an ongoing relationship with the client and is familiar with the
client’s total affairs. Consequently, the client usually looks primarily to the
domestic tax adviser for advice concerning both the domestic and foreign tax
consequences of a transaction. Although the domestic tax adviser is not
generally qualified to provide advice concerning foreign tax law, the client
often expects the tax adviser to act as a filter with respect to foreign tax
advice. It is not unusual for foreign tax advisers to deal with the domestic tax
adviser rather than with the client directly. In contrast, when the tax adviser is
providing advice concerning an inward-bound investment by a nonresident,
the role is often more restricted. Usually the advice is limited to the tax
consequences in the tax adviser’s particular country, and the adviser may not
be involved in the overall tax planning for the nonresident on an ongoing
basis. Also, as indicated earlier, in this situation the tax adviser may deal with
the foreign tax advisers rather than directly with the client.
Whether an inward-bound or an outward-bound investment is involved,
domestic tax advisers consulting on an international transaction invariably
deal with foreign lawyers, accountants, or business persons. The role of tax
advisers in this regard may often be difficult because of differences in basic
legal concepts, tax laws, and accounting practices. These differences may be
exacerbated by language and cultural differences.
Although a tax adviser may not be legally qualified to provide advice
concerning foreign tax law, knowledge of foreign tax systems is an important
asset. This knowledge enables an adviser to deal more effectively with
foreign tax advisers and to suggest alternative methods for structuring
transactions to provide desirable tax results under the laws of both countries.
From the taxpayer’s viewpoint, the foreign tax consequences of any
investment or transaction are often as important as, or even more important
than, the domestic tax consequences. Consider, for example, an individual, T,
who is resident in one country and who plans to make a portfolio investment
in another country. Obviously, T is concerned about how her country of
residence will tax the foreign source income and the provisions available for
relieving double taxation. T is also concerned, however, about the level of the
foreign tax. If her residence country relieves double taxation by exempting
foreign source income, the foreign tax is the only tax she needs to be
concerned about.
If T’s country of residence provides a foreign tax credit, the situation is
more complex, for reasons explained in detail in Chapter 4. In brief, countries
that grant a credit for foreign taxes typically limit the credit to the amount of
the domestic tax imposed on the foreign income – they do not allow a refund
for any foreign tax in excess of the domestic tax on the foreign income. If T
expects to obtain a credit for foreign taxes imposed on her foreign income,
she needs to be concerned about the foreign tax only if it exceeds the
domestic tax, in which case T will be subject to an effective rate of tax equal
to the foreign tax rate.
To take a more complicated example of the importance of foreign tax
law to the tax adviser, suppose that a multinational corporation wants to
reorganize its multinational group of corporations for business reasons. In the
absence of special relief provisions, such a reorganization might result in
significant adverse tax consequences under the tax laws of many countries.
Many countries, however, provide for certain corporate reorganizations to
occur on a tax-free (or, more accurately, tax-deferred) basis. In deciding
whether to undertake the reorganization, therefore, the multinational
corporation will look to its tax advisers for advice on the tax consequences of
the reorganization under the tax laws of the country in which the parent
corporation is resident, and also under the tax laws of the foreign countries in
which the foreign subsidiaries of the parent corporation are located or carry
on business. Providing this advice is no easy matter because the tax rules
governing corporate reorganizations vary widely and often interrelate in
complex ways.
This intersection of domestic tax law and foreign tax law is one of the
most challenging features of the study and practice of international tax.
Although tax advisers are usually qualified only to give advice on their
domestic tax law, they must be sufficiently familiar with foreign tax laws to
be able to recognize potential problems and to deal efficiently with foreign
tax advisers. Further, the intersection of foreign and domestic law extends
beyond tax. Tax consequences of transactions often depend on the underlying
legal results of the transactions. For example, the tax consequences may
differ if income is earned by an individual, a trust, a partnership, or a
corporation. Similarly, the tax consequences may differ if a taxpayer is
considered to have transferred property or know-how, or to have rendered
services to another person.
The problem of determining the tax consequences of a proposed
transaction on the basis of the underlying legal situation is exacerbated in the
foreign context because domestic tax consequences often must be determined
on the basis of foreign legal concepts. For example, if a resident of one
country holds an interest in a limitada or limited liability company (which is
in essence an entity that provides limited liability for its investors and flow-
through treatment for income tax purposes) organized in another country, are
the ownership rights characterized as an interest in a partnership, as shares in
a corporation, or as something else? And is the characterization the same in
both countries? If the entity is characterized differently by the two countries,
it is known as a hybrid entity. The issues raised by hybrid entities are
addressed in Chapter 9, section 9.2.
Tax is often not a major factor in the initial decision of an enterprise to
make a direct investment abroad. Other factors, such as the return on
investment, political stability, labor costs, and access to foreign markets, are
much more important as far as the original investment is concerned. The tax
‘tail’ should not wag the commercial ‘dog’. Once the decision to invest has
been made, however, tax is an important factor in determining the way in
which the investment is structured and financed. Further, tax remains an
important factor in determining whether to reinvest or repatriate the profits
from the investment. Tax advisers are expected to provide advice concerning
the tax consequences of the various ways in which the profits of a foreign
enterprise might be repatriated to the domestic corporation. Similarly, they
are expected to provide advice concerning the tax consequences of providing
the foreign enterprise with additional capital to finance its activities.
One important point about tax planning in general that must be kept in
mind is that the client’s organization must be able to live with the operational
implications of the tax plan. If the tax plan is too complex from an operating
viewpoint, any tax savings may be offset by additional administrative costs.
Moreover, if the business is unable to operate, in fact, in accordance with the
tax plan, the effectiveness of the plan for tax purposes may be jeopardized.
For example, a tax plan might involve the establishment of a foreign
subsidiary in a tax haven to purchase goods from the domestic parent
corporation and resell them to customers abroad. Such a tax plan might be
conditional on the delivery of the goods to the tax haven subsidiary.
Therefore, if the goods are shipped by the parent corporation directly to the
ultimate customers because that is the sensible thing to do from a commercial
perspective, the success of the tax plan may be jeopardized, and indeed,
significant penalties may be imposed on the taxpayer.
There are many different ways of structuring foreign investments. For
example, a manufacturing enterprise might sell its goods in a foreign country
in one or more of the following ways:

– selling its manufactured goods directly to customers in the foreign

country through, for example, mail-order sales, sales over the
Internet, or sales by itinerant sales agents;
– selling its goods to an unrelated foreign distributor for resale to
customers;
– establishing a branch in the foreign country consisting of a
warehouse and sales employees or agents to sell its goods there;
– establishing a foreign sales subsidiary in the foreign country to sell
the goods;
– establishing a foreign holding company, which establishes a foreign
sales subsidiary in the country to sell the goods; or
– licensing an unrelated foreign corporation to manufacture and sell its
goods in the foreign country.

The tax consequences of these various alternatives may vary

considerably under the tax laws of a particular country (and from country to
country).
One of the fundamental choices that a corporation resident in one
country faces in structuring a foreign investment in another country is the
choice between a foreign branch and a foreign subsidiary. The essential
difference between a branch and a subsidiary is that a subsidiary is a separate
legal entity, whereas a branch is a part or division of the resident corporation.
As a result, when a resident corporation sells its products through a foreign
branch, the resident corporation may be taxed on the profits of the branch
because the branch is not a legal entity separate from the corporation. Further,
for general law purposes, the resident corporation is responsible for any legal
obligations (e.g., with respect to product defects) arising out of its foreign
sales activities. In contrast, if the foreign sales are made by a foreign
subsidiary corporation, that corporation, as a separate legal entity, is taxable
on its profits and is responsible for its own legal obligations. There are, of
course, exceptions to this general rule.
In summary, a tax adviser is expected to perform two functions with
respect to tax planning for international transactions. First, tax advisers must,
within a reasonable range, quantify the tax costs and benefits of carrying out
transactions, and assess the tax risks of obtaining the tax benefits. Second, tax
advisers are expected to provide advice for minimizing the amount of tax
payable. Often, this tax-minimization aspect of international tax planning
involves identifying various methods of structuring a transaction and
recommending one method over others in light of the tax consequences and
the compatibility of the proposed structure with the overall operating plan of
the enterprise.
Although tax planning for international transactions must be tailored to
each client’s particular situation, certain common types of tax planning can
be identified. Three types of international tax planning are described below to
give some flavor of the nature of the exercise. The following examples have
been simplified drastically.
Double-dip leases. Some cross-border transactions are structured to
exploit differences in the tax treatment of transactions by two countries.
Cross-border leasing provides an example of this type of international tax
planning.
Assume that an airline company resident in Country A wishes to
acquire, on credit, some new aircraft for use in its business. It can take out a
commercial loan and purchase the aircraft directly, or it can acquire the
aircraft by utilizing a so-called financial lease. In general, a financial lease is
a financing arrangement under which the lessee acquires substantially all
ownership rights to the leased property. In effect, the lessor sells its
ownership rights in the property and finances the acquisition of the property
by the lessee. Instead of receiving interest and repayments of principal as a
conventional lender would, the lessor receives ‘rental’ payments that reflect
both the sale price of the property and the financing aspect of the transaction.
Assume that under the tax laws of Country A, a financial lease is treated
as a sale. Accordingly, if the airline company resident in Country A leases the
aircraft, it will be treated as the owner of the aircraft for purposes of Country
A’s tax and will be entitled to deduct depreciation in respect of the aircraft
and the interest element of the lease payments. The depreciation deductions
may be quite large, since many countries provide accelerated depreciation
deductions as a tax incentive for investment in substantial equipment. The
airline company will also be permitted to claim any investment tax credits
that Country A provides for purchases of aircraft.
If the lessor is also a resident of Country A, it will be treated as having
sold the aircraft, with the appropriate gain or loss recognized on the sale and
the interest element of the lease payment included in its income. Assume,
however, that the lessor is a resident of Country B and Country B treats
financial leases for tax purposes as genuine leases. Under these assumptions,
the lessor will be treated as the owner of the aircraft by Country B and will be
entitled to take depreciation deductions and claim any investment tax credits
offered by Country B to owners of aircraft. The lessor will be taxable in
Country B on the payments of rent received from the airline company.
However, Country A treats the financial lease as a sale so that the airline
company is considered to be the owner of the aircraft. Thus, the payments
considered to be rent by Country B will be treated as payments of the
purchase price and interest by Country A.
This type of structure is often referred to as a double-dip lease because
the tax benefits of ownership of the aircraft are claimed in both countries as a
result of the inconsistent characterization of the transaction by the two
countries.
Tax haven entities. International tax planning focuses heavily on the use
of countries that levy little or no tax. Such tax havens can be used in a wide
variety of ways to reduce taxes of residents of high-tax countries. One
common way is to establish a controlled foreign corporation in a tax haven.
For example, assume that ACo is resident in and manufactures goods in
Country A, which levies corporate tax at a rate of 40 percent. ACo sells its
manufactured goods not only in Country A but also in several other countries.
ACo is taxable in Country A on its worldwide profits. ACo incorporates a
wholly owned subsidiary, THCo, in Country TH, which does not impose any
income taxes. THCo purchases manufactured goods from ACo at their
arm’s-length price and resells them to clients outside Country A. As a
result, the sales profits attributable to sales outside Country A will be earned
by THCo, not by ACo. Because THCo is a separate legal entity and because
the tax advisers will ensure that it is not resident in Country A, the sales
profits derived by THCo are not usually taxable by either Country A or
Country TH. Thus, assuming that the sales profits are 2 million, this
transaction will reduce the taxes payable to Country A by 800,000 (40
percent of 2 million).
If THCo does not have any employees and never takes delivery of the
goods acquired from ACo, Country A may disregard THCo as a sham and
consider the sales profits to be derived by ACo. Even if THCo actually
performs the sales function, some countries have rules to attribute the income
derived by THCo to ACo. These “controlled foreign corporation” (CFC)
rules are discussed in section 7.3 of Chapter 7 dealing with international tax
avoidance.
Most countries that market themselves as tax havens have traditionally
provided broad protection against disclosure to foreign governments of the
particulars of the transactions involving resident corporations and financial
intermediaries located within their borders. These secrecy regimes made it
difficult for the tax authorities of high-tax countries to discover attempts at
tax avoidance and evasion by their residents. This situation changed
drastically in the early 2000s as a result of a series of high-profile cases of
widespread tax evasion facilitated by large banks in tax havens such as
Liechtenstein and Switzerland . These incidents led to the elimination of bank
secrecy and the adoption of a common international standard for the
exchange of information between tax authorities, on request and
automatically. The significant improvements in exchange of information to
prevent tax avoidance and evasion are discussed in Chapter 8, section 8.8.4.
Treaty shopping. Another type of international tax planning involves the
use of tax treaties to reduce tax. One common example involves the
establishment by a resident of one country of a “conduit” company in another
country in order to take advantage of that country’s tax treaty network.
Assume that ACo, resident in Country A, has developed valuable
intangible property and intends to license the property for use by
manufacturers in several other countries. Country A does not have treaties
with some of the countries where the potential licensees are resident, and the
treaties that Country A has with the other countries provide for withholding
taxes on royalties of 15 percent. Country A provides an exemption for
dividends received by a corporation resident in Country A from foreign
corporations in which it has a substantial participation. ACo transfers its
intangible property to a wholly owned subsidiary, BCo, established in
Country B. Country B has tax treaties with all the countries where potential
licensees are resident, and those treaties provide an exemption from any
withholding tax on royalties.
The result of the above arrangement is that no tax will be imposed on
the royalties by the countries where the royalties arise. Country B may not
tax the royalties derived by BCo (or may tax them at a low rate), either
because it is a traditional tax haven or because it provides a special low-tax
regime for royalties in respect of intangible property. When BCo distributes
dividends to ACo, Country A will not tax the dividends because of its
participation exemption for dividends from foreign corporations. Even if
Country A treats the transfer of the intangible property by ACo to BCo as a
taxable transaction, it may have significant difficulty in taxing the appropriate
amount of gain on the transfer because of the problem of accurately
establishing the fair market value of the intangible property at the time of the
transfer.
This example illustrates the problem of treaty shopping. In effect, ACo
has taken advantage of Country B’s treaty network by the simple expedient
of establishing a corporation as a resident of Country B. BCo functions as a
conduit to convert the royalties into tax-exempt dividends to ACo. BCo may
have no employees and no place of business in Country B. The overall effect
of the arrangement is that the withholding taxes of the countries in which the
royalties arise are avoided. Treaty shopping is dealt with in Chapter 8, section
8.8.2.2.
CHAPTER 2
Jurisdiction to Tax

2.1 INTRODUCTION
Income may be taxable under the tax laws of a country because of a nexus
between that country and the income or the activities that generated the
income. According to international usage, a jurisdictional claim based on
such a nexus is called “source jurisdiction”. All countries that impose an
income tax exercise source jurisdiction: that is, they tax income arising or
having its source in their country.
A country may also impose tax on income because of a nexus between
the country and the person earning the income. Such a jurisdictional claim
over a person’s income is called “residence jurisdiction”. Persons subject to
the residence jurisdiction of a country are generally taxable on their
worldwide income, without reference to the source of the income – that is,
the person is typically taxable on both domestic source income and foreign
source income.
It is frequently said that most countries tax residents on their worldwide
income and nonresidents on the domestic source income. Although this
statement contains a grain of truth, it is a gross oversimplification of the
scope of the tax systems of most countries. Countries that are said to tax their
residents on their worldwide income never tax all of their residents on all of
their worldwide income. Instead, these countries usually exempt their
residents, especially resident corporations, on their income from active
business carried on in foreign countries and income earned by foreign
corporations in which their residents own shares. Similarly, countries that are
said to tax on a territorial basis – that is, they tax only income derived from a
source in their countries – often impose tax on certain items of foreign source
income, such as fees paid by residents to nonresidents for technical services
performed outside the country.
With few exceptions, countries that exercise residence jurisdiction do so
only with respect to the income of individuals and legal entities that they
consider to be residents: thus the term “residence jurisdiction”. A few
countries – the United States (US) is the primary example – exercise
jurisdiction to tax their citizens as well as their residents. They assert the right
to impose income tax not only on the worldwide income of their residents,
but also on the worldwide income of their citizens wherever they might be
resident.
When a resident of a country earns income derived outside that country
(foreign source income), the claim of that country to tax the income based on
its residence jurisdiction may overlap with the claim of the country where the
income is earned to tax based on its source jurisdiction. The claims of
countries for tax revenue based on residence jurisdiction may also overlap
with the claims of other countries based on citizenship or, in the case of so-
called “dual-resident taxpayers”, on residence. In addition, countries with
conflicting source-of-income rules may both claim to tax the same income.
Unless resolved satisfactorily, the competing claims for tax revenue based on
residence and source would discourage international commerce and
investment. In addition, the tax burdens imposed on individuals earning
income from cross-border transactions would be unfair under traditional
concepts of tax equity. The measures that countries have adopted in their
domestic legislation and tax treaties to mitigate international double taxation
are addressed in Chapter 4.
Although persons engaging in transnational activities face risks of
double taxation, they also have possibilities for international tax avoidance
(sometimes referred to as double non-taxation). These opportunities result
from certain gaps in the residence and source jurisdictions of most countries.
The under-taxation of income from cross-border transactions is both
inefficient and unfair. Under-taxation is inefficient because it distorts
economic behavior; it induces taxpayers to engage in the under-taxed
activities instead of taxable activities that may produce higher before-tax
rates of return. It is unfair because individual taxpayers earning equal
amounts of income do not pay the same amounts of tax.
Tax haven countries increase the risks of under-taxation of transnational
income. Although tax havens may obtain some revenue from foreign
taxpayers, the amount is small in comparison with the amount of tax revenue
that other taxing jurisdictions lose on account of their conduct. The tax laws
of many countries are replete with complex provisions designed to protect the
legitimate tax claims of countries against the beggar-thy-neighbor policies
and preferential regimes of tax haven countries. The most important anti-
avoidance rules are addressed in Chapter 7.
Unilateral action by countries to block tax haven abuses has often been
ineffective, due in significant part to the inability of the source and residence
countries to obtain information about transactions routed through tax havens.
Until recently, tax havens had strict bank secrecy rules and similar non-
disclosure rules that facilitated tax avoidance and evasion by multinational
enterprises and wealthy individuals resident in other countries. In the last
decade or so, however, through a sustained project initiated and led by the
OECD, bank secrecy has been eliminated and the exchange of information
between countries to prevent tax avoidance and evasion has been improved
significantly. The recent efforts to facilitate exchange of information for tax
purposes are discussed in Chapter 8, section 8.8.4.

2.2 DEFINING RESIDENCE

For the purposes of taxing residents (and nonresidents), a country must
provide rules that classify individuals and legal entities either as residents or
nonresidents. The rules for determining the residence of individuals and legal
entities are discussed below in sections 2.2.1 and 2.2.2, respectively. Certain
tax treaty issues involving the determination of residence are addressed in
section 2.2.3, below. The rules for determining residence (and, as a
consequence, nonresidence) are clearly necessary for taxing residents on their
domestic and foreign source income, but they are also necessary for taxing
nonresidents on their domestic source income. The source of some types of
domestic source income, such as dividends and interest, are usually based on
the residence of the payer.

2.2.1 Residence of Individuals

An ideal test of residence is one that individuals and tax officials can apply to
obtain a clear, certain, and fair result. Certainty is highly desirable because
the tax consequences for residents and nonresidents are different, and
individuals need to know whether they are resident or nonresident in order to
comply with a country’s tax law. Nevertheless, a simple and certain test for
residence may be arbitrary and unfair, and may result in many individuals
who engage in cross-border activities ending up as residents of more than one
country. Therefore, the most that can be expected is a test that is simple,
certain, and fair for the overwhelming majority of taxpayers but that is
supplemented by more refined rules to deal with special circumstances.
In many countries, the residence of individuals is determined under a
broad facts-and-circumstances test. The most significant manifestation of an
individual’s allegiance to a country is probably the maintenance in the
country of a dwelling that is available for the use of the individual and his or
her family. The following factors are also usually relevant:

– the location of the individual’s income-producing activities;

– the location of the individual’s family;
– the social ties of the individual to the country (e.g., bank accounts,
club memberships, driver’s license, etc.);
– the individual’s visa and immigration status; and
– the individual’s actual physical presence in the country.

Under this test, the tax authorities of a country decide, based on largely
objective facts, whether an individual’s economic and social connections
with the country justify taxing the individual as a resident.
Unless buttressed by some simple presumptions, a facts-and-
circumstances test is unsatisfactory because it is often excessively difficult to
apply. A facts-and-circumstances test that uses certain objective tests to
establish presumptions may provide a good balance between certainty and
fairness. It may be appropriate for such a test to apply more rigorously in
situations in which a taxpayer is attempting to give up residence in a country
than in situations in which a taxpayer is acquiring residence in the country.
The following presumptions might be used, separately or in combination, to
establish a prima facie case for residence:

– Individuals present in a country for 183 days or more in a taxable

year are residents for that year, unless perhaps they establish that they
do not have a dwelling in the country and are not citizens of the
country.
– Individuals having a dwelling in the country are residents unless they
also have a dwelling in another country.
– Citizens of a country are residents unless they have established a
dwelling abroad and are regularly outside the country for more than
183 days per year.
 – Individuals who are domiciled in a country may be considered to be
residents of that country.
– Individuals who are temporarily absent from a country but intend to
return and resume residence in the country may be presumed to
remain residents of the country despite their temporary absence,
subject to the rebuttal of that presumption.
– Individuals who have established residence in a country cannot
relinquish residence status until they have clearly established
residence status in another country.
– Individuals who have either resident or nonresident status for visa or
immigration purposes might be presumed to have the same status for
income tax purposes, although that presumption might be rebuttable.

Special rules may be necessary for certain individuals. For example, it

may be appropriate to deem diplomats, military personnel, and other
government employees to be residents of the country that employs them
despite the fact that these individuals might not be residents on the basis of a
facts-and-circumstances test because they spend most of their time outside
the country.
Some countries use an arbitrary test, often tied to the number of days of
presence in the country, for determining residence. Such a test may be used
as a supplement to the facts-and-circumstances test discussed above. A
common, but defective, rule or presumption is that an individual who is
present in a country for at least 183 days during the year is a resident for that
year. The 183-day test is probably enforceable in countries that exercise tight
control over their borders; however, it is extremely difficult for the tax
authorities of a country to enforce when many individuals are frequently
entering and leaving the country without border checks, as occurs in the
countries of the European Union. In most countries, the test probably cannot
operate effectively unless the burden of proof is put on the individual to
prove that he or she is not present for the 183-day period. Many individuals
with substantial economic ties to a country can easily avoid becoming
resident under the 183-day test by leaving the country before the 183-day
threshold is passed. As a result, a country using that test is likely to catch
mainly unsophisticated or ill-advised individuals, some of whom may not in
fact have substantial ties to that country.
Domicile is a legal concept under the law of some countries by which an
individual’s permanent connection with a country is established. In general,
domicile involves a more permanent connection with a country than
residence. A person’s domicile may be the country in which the person is
born or in which the person’s mother or father is domiciled.

2.2.2 Residence of Legal Entities

The residence of a corporation is generally determined either by reference to
its place of incorporation or its place of management, or both. The place-of-
incorporation test provides simplicity and certainty to the tax authorities and
the corporation. It also allows a corporation to freely choose its initial place
of residence. Countries that market themselves as tax havens typically offer
convenient and inexpensive arrangements for incorporating under their laws.
In general, a corporation cannot freely change its place of incorporation
without triggering a tax on gains that may have accrued in respect of its
property, including intangible property that may have a high market value.
Consequently, the place-of-incorporation test places some limits on the
ability of corporations to shift their country of residence for tax avoidance
purposes. Many countries use the place-of-incorporation test, although it is
often combined with another test.
The place-of-management test is less certain in its application than a
place-of-incorporation test, at least in theory. For many corporations engaged
in international operations, management activities may be conducted in
several countries during any particular taxable year. In practice, most
countries using that test employ practical tests, such as the location of the
company’s head office or the place where the board of directors meets, to
determine the place of management. The place-of-management test is used by
the United Kingdom (UK) and many of its former colonies. Some countries,
such as Australia, Canada, and the UK, use both the place-of-incorporation
test and the place-of-management test.
A place-of-management test is easily exploited for tax avoidance
reasons where a change in the place of management can be accomplished
without triggering any tax. Assume, for example, that ACo is a corporation
resident in Country A, which uses a place-of-management test. ACo has
developed valuable intangible property that it intends to license to taxpayers
located in Country B. To avoid tax in Country A on the expected royalties,
ACo shifts its place of management to Country H, a low-tax country. The
large accrued gain on ACo’s intangible property is not taxable in Country A
because no transfer of that property occurred. (However, as discussed in
Chapter 3, section 3.4.1, several countries have adopted exit or departure
taxes to prevent taxpayers from avoiding tax by changing their residence.)
ACo then licenses the technology to users in Country B. The royalties
received by ACo escape taxation in Country A because ACo is no longer
resident in Country A.
If Country A in the above example used the place-of-incorporation test,
ACo could not transfer its residence to Country H without undergoing a
corporate reorganization that would probably result in the transfer of its
assets to a corporation organized in Country H. Such a transfer would trigger
a realization of the accrued gain on the intangible property, thereby limiting
or even eliminating ACo’s opportunity for tax avoidance.
Countries that use a place-of-incorporation test exclusively (such as the
US) have encountered avoidance schemes, called corporate inversions,
whereby resident multinationals reorganize to avoid aspects of the residence
country’s tax system. Although inversions take many forms and are
invariably quite complex, the following simplified example illustrates the
general idea.
Assume that USCo, a widely held US-based multinational, wants to
avoid the effects of the US CFC rules (discussed in Chapter 7, section 7.3).
To do so, it establishes a subsidiary (Forco) in a country without CFC rules
and then arranges for its shareholders to exchange their shares of USCo for
shares of Forco and for the shares of all USCo’s foreign subsidiaries to be
transferred to Forco. The end result is that Forco is not a CFC in respect of
any of its US shareholders and the CFC rules are no longer applicable to
USCo because its foreign subsidiaries have become subsidiaries of Forco.
Many commentators argue that the proper purpose of the corporate tax is
to impose tax burdens on a corporation’s individual shareholders. According
to this view, the corporate tax is paid in advance on behalf of the individual
shareholders of the corporation to prevent them from deferring tax by
investing in corporations. Therefore, the test of residence of a corporation
might be determined, at least in theory, by reference to the residence of its
shareholders. However, the application of a residence-of-the-shareholders test
would present serious problems when residents of more than one country
hold large blocks of stock in the company or when the stock of the company
is publicly traded and the identity of the shareholders is difficult to determine.
Taking this view to its logical conclusion, in effect, corporations would have
to be taxed like partnerships, with each country taxing the share of the
corporate income attributable to its resident shareholders.
For legal entities other than corporations, residence is generally
determined either under a place-of-organization test or a place-of-
management test. Determining the residence of a partnership is sometimes
difficult because of the informality with which a partnership can be
established: for example, in some countries a partnership may be created by
virtue of the course of conduct of the relevant parties without the necessity
for any formal legal documentation. In many countries, partnerships are
treated as transparent or flow-through entities for tax purposes: in other
words, the partners are taxed on their share of the income of a partnership,
but the partnership itself is not taxed. For these countries, the residence of a
partnership is usually irrelevant because the partnership is not a taxable
entity.
Difficult problems also can arise under the laws of some countries in
determining the residence of trusts. These problems are especially difficult
when the country of organization, the country where the trustee or trustees
are resident, the country where the grantor or settlor is located, and the
countries where the beneficiaries are located are all different.

2.2.3 Treaty Issues Relating to Residence

Under Article 4(1) of the OECD Model Treaty, a “resident” of a country for
purposes of the treaty is a person who is liable to tax in that country “by
reason of his domicile, residence, place of management or any other criterion
of a similar nature”. Article 4(1) of the United Nations (UN) Model Treaty
adds “place of incorporation” to the list of connecting factors. To avoid
situations in which an individual is considered to be resident in both
countries, Article 4(2) of both Models provides a series of tie-breaker rules
to make the individual resident in only one country for purposes of the treaty.
The first tie-breaker is the place where an individual has a permanent home;
the second tie-breaker is the country in which the center of the individual’s
vital interests is located; the third is the place of the individual’s habitual
dwelling; and the fourth is the country of citizenship. If these tie-breaker
rules are ineffective in making an individual a resident of only one country
for treaty purposes, the “competent authorities” of the two Contracting
States are mandated to determine residence by mutual agreement. Most
modern tax treaties follow the tie-breaker rules in the OECD and UN Model
Treaties closely.
For legal entities resident in both Contracting States, Article 4(3) of the
OECD and UN Model Treaties make the entity a resident of the country
where its place of effective management is located. According to the
Commentary on Article 4, the place of effective management of an entity is
the place where key management and commercial decisions are in substance
made – the place where decision-making at the highest level on the most
important issues of management takes place. Moreover, according to the
Commentary, an entity can have only one place of effective management,
although it can have multiple places of management. In my view, the
Commentary dealing with the place of effective management is not
convincing. Given the flexibility in the structure of the management of
entities, it seems likely that key decisions can be made in multiple locations.
The place of effective management tie-breaker rule is not likely to be
acceptable to countries that use a place-of-incorporation test as the sole test of
residence for corporations. Many treaties attempt to resolve conflicts over the
residence of entities by leaving the issue to the competent authorities to
resolve. Some treaties use place of incorporation as the tie-breaker if the
company involved is incorporated in one of the treaty countries. Other
treaties provide that a dual-resident company is not considered to be a
resident of either country for most treaty purposes; therefore, such a company
is not entitled to any of the benefits of the treaty. This provision is intended to
prevent the use of dual-resident companies for tax avoidance purposes; for
example, a dual-resident company with losses would be entitled to relief for
such losses in both countries in which it is resident.
The US insists on the inclusion in its tax treaties of what is commonly
referred to as a “saving clause”. The typical saving clause provides, with
some exceptions, that the US reserves the right to tax its residents and its
citizens as if the treaty had not come into effect. For example, a US citizen
resident in a treaty country is not entitled to the reduced rate of withholding
provided in the treaty on dividends received from the US. The OECD’s BEPS
Action 6: Preventing the Granting of Treaty Benefits in Inappropriate
Circumstances proposes to add a similar saving clause to the OECD Model
Treaty (see Chapter 8, section 8.8.2.3).
2.3 SOURCE JURISDICTION

2.3.1 Introduction
By international custom, a country has the primary right to tax income that
arises in, has its source in, or is derived from that country. As discussed in
Chapter 3, under international custom the country of residence is generally
expected to provide relief from double taxation if its residence jurisdiction
overlaps the source jurisdiction of another country. In other words, the
country in which a taxpayer is resident has only a secondary right to tax the
taxpayer’s income that is derived from or sourced in another country. Most
tax treaties provide that the country in which income is sourced has the first
right to tax that income and that the country of residence has an obligation to
eliminate double taxation of that income. Although tax treaties give a country
the first right to tax income sourced in the country, they usually provide that
the source country must limit its rate of tax on certain categories of
investment income and preclude the source country from taxing certain
categories of income, even if the income arises in the source country.
Despite the priority given to source jurisdiction, the concept of source is
rather poorly developed in domestic tax legislation and tax treaties. Unlike
the term “residence”, the term “source” is not used or defined in domestic law
or in tax treaties. Thus, source rules are usually implicit in other rules. For
example, withholding tax imposed on amounts paid by residents to
nonresidents has an implicit source rule that such amounts are sourced in a
country if they are paid by a resident of the country.
Most countries have only sketchy rules for determining the source of
income, especially income derived from business activities. For example, in
the UK and countries that were former UK colonies, income from business
activities is considered to have its source where the real business is carried
on. Such a rule is too vague to provide any guidance to taxpayers or tax
officials.
The OECD and UN Model Treaties provide a mixture of implicit source
rules and rules that function effectively like source rules. For example, under
Article 11(4), interest income is taxable by the country in which the payer is
resident and under Article 6, income derived from immovable property,
including income arising from the operation of a mine or well, is taxable by
the country where the immovable property is located. However, the OECD
and UN Model Treaties do not contain any explicit source rules for business
profits. Under Article 7, business profits derived by a resident of one
contracting state are taxable by the other contracting state only if the
resident carries on business through a permanent establishment (“PE”)
located in that other state (Article 7) and the profits are attributable to the PE.
This rule is the functional equivalent of a source rule.
In general, a PE is a fixed place of business, such as an office, branch,
factory, or mine. The OECD and UN Model Treaties (Article 5) treat a
dependent agent as constituting a PE of its principal in some circumstances.
Activities relating to the purchase of goods for export generally do not cause
a taxpayer to have a PE. For additional discussion of PE rules, see Chapter 8,
section 8.7.3.2.
Some provisions of the OECD and UN Model Treaties, such as Article
21 (Other Income), contain general wording that refers to income arising in a
contracting state, without any further elaboration. In these circumstances, it
seems inevitable that the source of income for that particular purpose must be
determined under the domestic law of the country applying the treaty.
Good source rules should have the following characteristics:

– They should be broadly acceptable by many countries in order to

ensure that double taxation is eliminated and double non-taxation is
not facilitated.
– They should allocate income and tax on a reasonable basis that is
broadly acceptable by most countries.
– They should be relatively clear and simple for taxpayers and tax
officials to apply.
– They should be applicable on a reciprocal basis (i.e., a country should
not unilaterally adopt a source rule that it would object to another
country adopting).
– They should allocate income to a country where the income has a
substantial economic connection (in the language of the OECD’s
BEPS project, income should be taxable in the country where value is
added or created).
– They should not be subject to manipulation by taxpayers.
– They should not allocate income to countries that do not impose tax.

The source rules generally applicable to employment and personal

services income, business income, and investment income are discussed
below. The modifications of those source rules contained in the OECD and
UN Model Treaties are also discussed.

2.3.2 Employment and Personal Services Income

The general rule in many countries is that income derived from personal
services performed by employees, independent contractors, or professionals
has its source in the country where the services are performed. Difficult
allocation issues may arise when a taxpayer is paid for services performed in
more than one country. Allocation among the countries where an individual
performs services is typically based, at least in part, on the amount of time
spent by the individual performing the services in each country. Some
countries, including several South American and Latin American countries,
consider income from services to be derived in their countries if the services
are consumed or used (by customers or clients) in their countries even if the
services are performed outside their countries.
Under Article 14 of the UN Model Treaty, income from services derived
by professionals and other independent contractors is taxable by the country
in which the services are performed only if the service provider has a fixed
base (which is equivalent to a PE) regularly available in the source country or
spends at least 183 days in the source country. The OECD Model Treaty
provided a similar exemption until the elimination of Article 14 in 2000. In
tax treaties following the OECD Model Treaty, income from professional and
other independent services is taxable as business profits under Article 7 and
is exempt from tax by a country unless the service provider has a PE (rather
than a fixed base) in that country. Under the UN Model Treaty, a taxpayer
resident in one contracting state is deemed to have a PE in the other state if
the taxpayer furnishes services in the other state through employees or other
personnel with respect to the same or a connected project for a period of more
than 183 days in any twelve-month period beginning or ending in the year.
The Commentary on Article 5 of the OECD Model Treaty provides a similar
deemed services PE rule as an alternative provision that countries may adopt.
One effect of eliminating Article 14 from the OECD Model Treaty has
been to clarify that the various exceptions to PE status for preparatory and
auxiliary activities in Article 5(4) and the agency PE rules in Articles 5(5)
and (6) are equally applicable to income from professional and independent
services. Both of these issues remain unresolved under the UN Model Treaty.
Article 15 of both the OECD Model Treaty and the UN Model Treaty
provides that income from employment is taxable exclusively by the country
in which an employee is resident unless the employment is exercised in the
source country and the following conditions are met:

(1) the employee must be present in the source country for no more
than 183 days;
(2) the employee is not paid by or on behalf of an employer resident in
the source country; and
(3) the employee’s remuneration is not deductible in computing the
profits attributable to a PE in the source country of a nonresident
employer.

In other words, an employee resident in one contracting state will be

taxable by the other state on any income from employment duties performed
in the other state for an employer resident in that state or for a nonresident
employer with a PE in that state if the employee’s remuneration is borne by
the PE. Otherwise, income of an employee resident in one contracting state
from employment exercised in the other contracting state is taxable by that
other state only if the employee is present in the other state for 183 days or
more in any twelve-month period beginning or ending in the relevant year.
(See Chapter 8, section 8.7.3.3.)

2.3.3 Business Income

The taxation of business income by source countries varies considerably.
However, two general patterns can be noted. The most common pattern,
consistent with Article 7 of the OECD and UN Model Treaties, is that
business income is generally taxable by a country only if the taxpayer carries
on business through a PE in the country and the income is attributable to that
PE. In these systems, the PE rules serve not only as a threshold for taxation
but also as the means for identifying the income subject to tax, namely,
income “attributable” to the PE. Most European countries follow this general
pattern; however, the definition of a PE under domestic law is often broader
than the definition in tax treaties.
In many Latin American and South American countries, the PE concept
is used to differentiate between the taxation of income from services derived
by nonresidents on a net or a gross basis. Nonresidents who earn income
from services through a PE situated in a country are taxable by that country
on a net basis (i.e., the nonresidents are allowed to deduct expenses incurred
in earning the income). Nonresidents who earn income from services
performed in a country or consumed in the country but who do not have a PE
in the country are taxable on a gross basis (without the allowance of any
deductions) through a withholding tax.
According to Article 7 of the OECD and UN Model Treaties, the
amount of income attributable to a PE is determined by assuming that the PE
is a separate legal entity and that it deals at arm’s length with other parts of
the enterprise, including the head office, of which it is a part. Transactions
between a PE and the head office are subject to the arm’s-length transfer
pricing rules. Transfer pricing rules are discussed in Chapter 6. In practice,
most countries determine the income of a PE by relying heavily on the books
of account of the PE, with adjustments made to those books only in cases of
perceived abuse. The effect is to give substantial discretion to taxpayers to
determine the business profits attributable to a PE. The rules for the
attribution of profits to PEs are discussed in Chapter 8, section 8.8.5.
The other general pattern for taxing business profits is that the concept
of a PE (or some functional equivalent) is used as a threshold requirement for
the taxation of nonresidents but explicit source rules are used to determine
the extent of the income subject to tax. The US is the most prominent
example of this approach. Under US law, most categories of gross income are
assigned a source. Deductions are then associated with items of gross income,
generally in accordance with accounting conventions. Some items of business
income are assigned exclusively either to the US or to foreign countries. For
example, income derived from the purchase and sale of personal property is
considered to have its source in the country of sale. Other categories of
income are apportioned between foreign countries and the US, often by
formula. For example, income from the manufacture and sale of personal
property is apportioned between the country of manufacture and the country
of sale, typically by a two-factor formula (sales and property).
Telecommunications income generally is apportioned equally between the
country of origin of the telecommunication signals and the country of
reception.
A key feature of US source rules is the treatment of deductions. Many
deductions are linked with gross income under accounting rules comparable
to inventory accounting rules, under which deductions such as depreciation
and other fixed costs are allocated for purposes of determining the cost of
goods sold. However, special detailed rules apply to interest payments,
research and development expenses, and certain other expenses that are
difficult to link with specific items of gross income.
Most countries lack sophisticated source rules with respect to income
and deductions. Accordingly, income and deductions are allocated between
domestic and foreign income in accordance with general rules that give
considerable discretion to taxpayers.

2.3.4 Investment Income

With some exceptions, most countries tax investment income derived by
nonresidents, such as dividends, interest, and royalties, through withholding
taxes imposed on the gross amount of the payment at a flat rate. Capital gains
are not usually subject to withholding tax, although special enforcement
measures are used by some countries, as discussed in Chapter 5, section
5.8.5. The source of investment income is usually determined by implicit
source rules or their functional equivalents. With some technical exceptions,
the following rules have been adopted by most countries and are endorsed in
the OECD and UN Model Treaties.

– Interest, dividends, and royalties are considered to arise or be sourced

in the country of residence of the payer. (See Articles 11(4) and 10(4)
of the OECD and UN Model Treaties and Article 12(4) of the UN
Model Treaty.) It is notable that the source rule for investment
income relies on the residence of the payer of the interest, dividends,
or royalties. Therefore, even countries that tax exclusively on the
basis of the source of income (territorial taxation) require rules to
determine the residence of persons. Special rules may apply to
interest that is deductible in computing the profits attributable to a
PE.
– Royalties paid with respect to intangible property are considered to
arise in and are taxable by the country where the property is used and
that provides legal protection for the intangible property. Some types
of royalty income, such as royalties paid for the showing of motion
 pictures and royalties on computer software, may be classified as
business income under the laws of some countries. No source rule for
royalties is necessary in the OECD Model Treaty because exclusive
jurisdiction to tax royalties is given to the residence country.
However, under Article 12 of the UN Model Treaty the source
country is entitled to tax royalties, with the setting of a maximum
withholding rate left to negotiations between the treaty partners.
Under Article 12(4), royalties are considered to arise in the country
where the payer is resident or, in the case of royalties deducted in
computing the income of a PE, in the country where the PE is
located.
– Rental income derived from the operation of a business is typically
taxable under the rules applicable to business income discussed
above. Rental income from immovable property is taxable by the
country in which the property is located; therefore, implicitly the
source of the royalties is that country. Rental income from the use of
movable property is generally taxable by the country where the
property is used; therefore, implicitly the source of the income is the
country in which the payer is resident. Rental income derived from
movable property is taxable as business profits under Article 7 of the
OECD Model Treaty and as royalties under Article 12 of the UN
Model Treaty.
– The source of gains from the disposal of property varies considerably
and depends on the nature of the property. Gains from the disposal of
immovable property are almost invariably taxable by the country in
which the property is located. This is the rule in Article 13(1) of the
OECD and UN Model Treaties. Gains from the disposal of assets
used in carrying on a business in a country are often taxable by the
country in which the business is carried on. Under Article 13(2) of
the OECD and UN Model Treaties, gains from the disposal of
property used in carrying on a business through a PE in a country are
taxable by that country. The structure of the OECD and UN Model
Treaties allows the same country to tax both income and capital gains
from the disposal of immovable property and assets forming part of a
PE. Thus, the characterization of a gain as income or capital gain is
determined under domestic law and is not relevant for purposes of the
treaty.
 Gains from the disposal of shares of companies or interests in a
partnership or trust are generally subject to tax exclusively by the country in
which the taxpayer is resident. However, if a nonresident taxpayer owns a
substantial interest in a resident entity, some countries impose tax on the
gains. Article 13(4) of the UN Model Treaty allows a Contracting State to tax
such gains, but the OECD Model Treaty does not. Similarly, some countries
tax gains from the disposal of shares in a corporation or interests in a
partnership or trust if the value of the shares or interests is derived principally
from immovable property located in those countries and owned by the entity.
Both the OECD and UN Model Treaties allow the Contracting States to tax
such gains. In effect, this provision is an anti-avoidance rule designed to
prevent taxpayers from avoiding a country’s tax on gains in respect of
immovable property located in the country by transferring the property to a
corporation or other legal entity established in the country, and then
disposing of the shares of the corporation or interests in the entity.
As noted above, most countries entering into tax treaties agree to some
limitations on the withholding tax rates applicable to interest, dividends, and
royalties under their domestic laws. The intent of these limitations is to
ensure that the source country does not impose excessive taxes on investment
income and tax revenue is shared between the source country and the
residence country. For a discussion of withholding taxes, see Chapter 5,
section 5.9.2 and Chapter 8, section 8.7.3.5.
Some tax treaties eliminate source country taxation entirely for some
types of investment income by mandating a zero rate of withholding; almost
all of these treaties are between developed countries. The zero rate is
premised on two assumptions: that the flow of investment funds between the
treaty countries in the zero-rate categories is approximately equal, and that
the tax jurisdiction relinquished by the source country will be exercised by
the residence country. As discussed in Chapter 8, section 8.2.1, these
assumptions do not reflect reality in many circumstances.
Some commentators favor residence taxation of investment income over
source taxation on the ground that a withholding tax at source may operate in
some circumstances as an excise tax on the payer, whereas a residence tax
generally operates as an income tax on the recipient of the income. As an
example of this excise-tax effect of withholding taxes, consider a foreign
bank that requires a borrower to make interest payments net of any
withholding tax imposed by the source country. In such circumstances, the
borrower is likely to view the withholding tax as an additional cost of
borrowing.
In theory, zero rates of withholding simplify administration and promote
business efficiency by allowing intercompany transfers to be made without
tax consequences. In practice, zero rates may promote tax avoidance schemes
and, in the absence of complex anti-avoidance rules, may provide unintended
benefits through treaty shopping. Treaty shopping occurs, for example, when
treaty benefits are obtained by corporations that are nominally resident in a
treaty country but are owned beneficially by nonresidents. For a discussion of
treaty shopping, see Chapter 8, section 8.8.2.2.
CHAPTER 3
Taxation of Residents

3.1 INTRODUCTION
As discussed in Chapter 2, many countries tax persons – individuals and legal
entities – who are residents on their worldwide income and nonresident
persons only on their domestic source income. Thus, the essential difference
between the taxation of residents and nonresidents is that nonresidents are
taxed only on income derived from a country (domestic source income) while
residents of a country are taxed on both their domestic source income and
their income derived from outside the country (foreign source income).
Although a few countries tax only domestic source income (so-called
territorial taxation), most countries tax resident persons on at least some of
their foreign source income. Therefore, the typical pattern of taxation
internationally can be described as the taxation of residents on both their
domestic source income and at least some items of their foreign source
income. The taxation of nonresidents is dealt with in Chapter 5.
The distinction between resident persons and nonresident persons is a
fundamental one and has important consequences. As discussed in Chapter 2,
the determination of the residence of a person is usually based on the
person’s connections to a country, although the specific rules vary
considerably from country to country. The closer and more extensive a
person’s connections are with a country, the more likely it is that the person
will be considered to be a resident of that country for purposes of its tax
system.
This chapter examines the major issues involved in taxing residents on
their worldwide income other than the determination of residence itself,
which is dealt with in Chapter 2. Initially, it considers the tax policy reasons
for taxing residents on their worldwide income and then several more
practical issues, such as the computation of foreign source income, departure
or exit taxes, trailing taxes, the treatment of temporary residents, and foreign
exchange gains and losses.
3.2 TAXATION OF RESIDENTS ON THEIR
WORLDWIDE INCOME

3.2.1 Tax Policy Considerations

It may seem initially that countries that impose tax on the worldwide income
of their residents are exceeding their sovereign authority to tax because they
are taxing income that arises outside their territories (extraterritoriality).
However, there are no international law constraints on a country’s legal
authority to tax persons who have close connections to the country. There are
practical constraints on a country’s ability to tax, since it makes little sense
for a country to impose taxes that cannot be effectively collected. However,
tax on the worldwide income of residents can be effectively collected
because, by definition, the residents of a country are persons with significant
ties to the country.
The tax policy justifications for taxing resident individuals on their
worldwide income are equity and neutrality. If two residents have equal
amounts of income, they should be subject to the same tax burden, even if
one resident’s income is derived totally from domestic sources while the
other’s income is derived exclusively from outside the country. This equity
justification is based on the assumption that all residents of a country derive
significant personal benefits from the country in the form of public goods and
services that justify taxing them irrespective of the source of their income.
The neutrality justification is that a country should not create a tax incentive
for its residents to work or invest outside the country. If the foreign source
income of residents is not subject to residence country tax, residents have an
incentive to earn low-taxed foreign source income in preference to domestic
source income, and this incentive is detrimental to the domestic economy.
With respect to corporations and other legal entities, considerations of
equity are of little significance because such entities are not the ultimate
beneficial owners of their income. Accordingly, the justification for taxing
legal entities on their worldwide income must rest primarily on the neutrality
argument. This form of neutrality is referred to as capital-export neutrality – a
taxpayer should invest where the pre-tax return is maximized. To achieve this
type of neutrality, a country must tax its resident entities on income from
both foreign and domestic investments. If income from foreign countries is
not taxed, legal entities resident in a country will prefer to invest in foreign
countries with lower tax rates, and especially in tax havens with no or low
taxes.
There is, however, another side to the neutrality argument with respect
to business income – capital-import neutrality or international
competitiveness – which suggests that entities resident in one country must
compete in various countries with entities resident in those countries and with
third countries. If the entities resident in Country A are subject to tax by
Country A on their worldwide income, they will not be able to compete as
effectively in Country B as entities resident in Country B or resident in third
countries that are subject to tax only in Country B. The arguments concerning
international competitiveness, although complex and controversial to
academics, have proved irresistible to governments. As a result, it is fair to
say that the international norm is that active business income derived by legal
entities is generally taxed on a territorial basis; in other words, the foreign
source business income derived by a resident corporation is not usually
subject to residence country tax. This point is explained in more detail below.
It might be thought that one of the reasons for a country to tax its
residents on their foreign source income is the additional tax revenue that will
be generated. However, as explained in Chapter 2, to the extent that such
income is subject to tax in the country in which it is derived (the source
country), that country has the first right to tax the income. The residence
country is generally considered to be under an obligation (although not a
legal obligation unless there is a treaty in effect between the two countries) to
eliminate double taxation, either by exempting the foreign source income
from residence country tax or by providing a credit against residence country
tax for the source country tax on the foreign source income. Thus, additional
tax revenue will be generated only if the residence country eliminates double
tax through a foreign tax credit and only to the extent that the source
country tax is less than the residence country tax. The elimination of double
tax is explored in Chapter 4.

3.2.2 The Tax Consequences of Residence

In a worldwide tax system, an individual taxpayer’s income includes both
income from inside the country in which the taxpayer is resident and income
from outside that country. The individual is usually taxable on that
worldwide income at progressive rates, although the extent of the progression
in the rates may be limited; for example, an amount may be taxable only if
the taxpayer’s income exceeds a threshold amount. In some countries,
although foreign source income is not taxable, it is taken into account in
determining the rate of tax on the taxpayer’s other income (exemption with
progression). In some countries, certain types of foreign source income
(typically business income) may be exempt while other types are subject to
residence country tax. As a result, as noted above, just because a country is
described as taxing on a worldwide basis does not mean that it taxes all
foreign source income derived by residents.

Example

Ms X is a resident of Country X. She has income from employment in

Country X of 100,000. She also receives dividends of 10,000 from
corporations resident in Country Y and interest of 3,000 from a bank account
in Country Z. The dividends are subject to withholding tax in Country Y of
15 percent, or 1,500, and the interest is subject to withholding tax in Country
Z of 10 percent, or 300. Ms X’s income and tax payable to Country X might
look as follows:
Income from Country X 100,000
Foreign source income
10,000
Dividends from Country Y
Interest from Country Z 3,000
Worldwide income 113,000
Less: personal allowance 10,000
Taxable income 103,000
Tax payable (40%) 41,200
Less: single parent credit 1,200
foreign tax credit 1,800
Net tax payable 38,200

3.2.3 Double Taxation

If one country taxes its residents on their worldwide income and another
country taxes part of that income because it is derived from sources in that
country, the income is subject to double tax. Worldwide taxation of residents
inevitably results in double tax because most countries insist on taxing
income that is derived or has its source in their countries. The well-
established international norm is that the source country – the country in
which the income has its source – has the first right to tax the income and the
residence country has a secondary right to tax the income; however, if the
residence country does so, it must provide relief for the source country’s tax
in order to eliminate double taxation. The methods that residence countries
use to eliminate the double taxation of foreign source income earned by their
residents are discussed in Chapter 4.

3.2.4 Computation of the Foreign Source Income of Residents

3.2.4.1 In General

Since residents are taxable on their worldwide income, rules are necessary to
compute both their domestic source income and their foreign source income.
Typically, the same rules apply for the purpose of computing both types of
income. The same amounts are included in income; the same deductions are
allowed; and the same timing rules apply. However, tax incentives may be
restricted to domestic source income. For example, a country may provide
accelerated depreciation for investment in machinery and equipment used in
certain domestic industries or areas of the country or it may provide enhanced
write-offs for domestic research and development.
Source rules are irrelevant for purposes of determining the worldwide
income of residents since all income, domestic and foreign, is taxable.
However, source rules are required if any items of foreign source income are
exempt from tax. In addition, source rules are necessary for purposes of
determining the limitation on the foreign tax credit. As discussed in detail in
Chapter 4, a country that taxes the foreign source income of its residents is
obligated by its treaties (and also by international practice and fairness) to
allow a credit against its domestic tax for the foreign tax paid on the foreign
source income. However, this credit for foreign taxes never exceeds the
amount of domestic tax on the foreign source income.
Expenses incurred to earn foreign source income are clearly deductible
if the foreign source income is subject to tax. Sometimes there may be a
serious mismatch between the timing of the recognition of the income and the
timing of the deduction of expenses. For example, the taxpayer may borrow
funds to finance the earning of foreign source income. The interest will be
deductible currently but, with respect to some items of income, such as
dividends, the inclusion of the income may be postponed to subsequent years.
The same type of timing mismatch often occurs with respect to research and
development expenses.
If foreign source income is exempt from residence country tax, in
principle, any expenses incurred to earn such income should not be
deductible. Many countries, however, allow the deduction of interest expense
on borrowed funds used to acquire shares of foreign corporations, even
though dividends received from such corporations are exempt from residence
country tax. This issue has become increasingly important as more countries
have adopted participation exemptions. The OECD’s BEPS Action 4: Interest
Deductions and Other Financial Payments (December 18, 2014) deals with
this issue and is discussed in more detail in Chapter 7, section 7.2. Although
expenses incurred to earn foreign source income that is taxable are deductible
in computing a taxpayer’s worldwide income, these expenses should also be
deducted in computing foreign source income for purposes of the limitation
on the foreign tax credit. This issue is also discussed further in Chapter 4,
section 4.4.

3.2.4.2 Foreign Exchange Gains and Losses

Foreign source income is often earned in the currency of the country in which
it is earned. Similarly, expenses incurred to earn foreign source revenue are
often incurred in foreign currency. In a worldwide tax system, a resident’s
income must generally be reported in the currency of the country of
residence. As a result, amounts of revenue and expense expressed in foreign
currency must be translated into the domestic currency. In theory, each
amount should be translated at the exchange rate applicable at the time that
the amount is earned or incurred. For practical reasons, some countries allow
amounts denominated in foreign currency to be translated into domestic
currency using average exchange rates (monthly, quarterly, or annually).
For purposes of computing capital gains from the disposal of foreign
property, it is appropriate to translate the cost of the property in foreign
currency into domestic currency at the exchange rate applicable at the time
that the property was acquired, and to translate the proceeds from the sale of
the property at the exchange rate applicable at the time the property is sold.
This method of foreign currency translation results in the recognition of the
foreign currency gains and losses as part of the capital gain or loss from the
disposal of the property, as illustrated in the following simple example.

Example

X, a resident of Country R, acquires property in Country S on October 20,

2000 at a cost of € 100,000, the currency of Country S. On October 20, 2000
the exchange rate of the Euro relative to Country R’s currency was € 1 = 1.5
Country R dollars. X sells the property on December 7, 2015 for € 100,000.
The exchange rate relative to Country R dollars on that date is € 1 = 1
Country R dollar. The loss for purposes of Country R tax might be calculated
as follows:

100,000
Proceeds 100,000 =
dollars
150,000
Cost 100,000 =
dollars
50,000
Loss
dollars

Because the dollar has appreciated against the Euro, the disposal of the
property produces a loss, although in Euros there is no gain or loss. If the
dollar had depreciated relative to the Euro (i.e., the exchange rate was € 1 = 1
dollar on October 20, 2000 and € 1 = 1.5 dollars on December 7, 2015) the
taxpayer would have realized a gain of 50,000 dollars, although there would
have been no gain or loss in terms of Euros.
An alternative method of calculating the gain or loss, which is similar
but not as accurate, is to determine the gain or loss in the foreign currency
(i.e., the cost and proceeds) and then translate that amount into domestic
currency at the rate applicable on the date of sale, or perhaps the average rate
applicable for the year in which the sale occurs. Under this approach, foreign
exchange gains and losses are not realized at all. As shown in the previous
example, there would be no gain or loss despite significant movement in the
exchange rate.
 Foreign exchange gains and losses may also arise with respect to debt
obligations. For example, assume that a taxpayer resident in Country R,
whose currency is R dollars, borrows € 1,000. At the time of the borrowing
the exchange rate is € 1 = 5 R dollars. Interest on the loan is 10 percent
annually. Each time that interest of € 100 becomes payable (or accrues or is
paid), the taxpayer must convert that amount to R dollars at the exchange rate
applicable at the time the interest becomes payable or is actually paid. When
the loan is repaid, the taxpayer will realize a gain or loss depending on
whether the R dollar has appreciated or depreciated in relation to the Euro.
For example, if on repayment the exchange rate is € 1 = 4 R dollars, the
taxpayer will realize a gain of 1,000 R dollars. In effect, the taxpayer has
borrowed 5,000 R dollars, but needs only 4,000 R dollars to repay the loan.
Under domestic law, foreign exchange gain or loss on the repayment or
settlement of a debt obligation may be treated as a capital gain or loss or as
ordinary income or loss.
Foreign currency risks arise with respect to actual foreign currency
transactions and also with respect to other aspects of a taxpayer’s business.
For example, a taxpayer that sells its products to residents of another country
is exposed to the risk that the country’s currency may weaken against the
taxpayer’s currency, with the result that the taxpayer’s products become more
expensive for residents of the other country.
Businesses often try to manage or hedge their foreign currency risks
through natural hedges, such as borrowing in the currency of the country in
which they carry on business or own assets, or through derivatives, which are
financial products designed to produce a gain or loss that offsets a gain or
loss in respect of an underlying asset or liability. The treatment of these
hedges for tax purposes is dependent on a mixture of accounting rules and
domestic tax law. Tax treaties do not generally deal with hedging.

3.2.4.3 The Treatment of Losses

Under a worldwide tax system, residents are taxable on their foreign source
income. It follows that, in principle, they should be allowed to deduct any
losses from foreign sources. Although this treatment of losses is consistent
with the treatment of foreign source income from a theoretical perspective, it
gives rise to problems, since taxpayers can manipulate the deductibility of
business losses inappropriately. For example, a taxpayer will often
commence business operations in a foreign country through a branch rather
than a foreign subsidiary so that the start-up losses are deductible against the
taxpayer’s worldwide income in its country of residence. Once the business
becomes profitable, the taxpayer can transfer the assets of the business to a
foreign subsidiary – often on a tax-free basis – so that the future profits
derived by the foreign subsidiary are not subject to tax by the taxpayer’s
country of residence, as explained in section 3.3.1 below.
If the residence country exempts some foreign source business income
(e.g., business profits attributable to a foreign PE), then any foreign source
losses attributable to a foreign PE should not be deductible in computing the
taxpayer’s worldwide income.
Some countries try to protect their tax base against the inappropriate
deduction of foreign losses. Various rules are used for this purpose. For
example:

– deductions for foreign source losses may be limited to a taxpayer’s

foreign source profits; and
– deductions for foreign source losses may be recouped if the foreign
business is sold or transferred to a foreign subsidiary.

3.2.5 Tax Administration Issues

Taxing the foreign source income of residents presents special problems of
administration and enforcement for the tax authorities of the residence
country. The tax authorities require information concerning the taxpayer’s
foreign source income both to ensure that all of the income is reported and to
verify that the income is properly computed. Typically, the tax authorities
will attempt to obtain this information from the taxpayer in the first instance.
Taxpayers may also be required to provide information about their foreign
income-earning activities on a regular basis in their tax returns through
information reporting returns, or pursuant to a specific request from the tax
authorities. The tax authorities of the residence country may also obtain
information from the tax authorities of the country where the income is
earned by way of the exchange-of-information article of the tax treaty
between the two countries. Exchange of information under tax treaties is
discussed in Chapter 8, section 8.8.4.
Not surprisingly, taxpayers are sometimes tempted not to provide full
disclosure to the tax authorities concerning their foreign source income
because they know that it is much more difficult for the tax authorities to
obtain information that is located outside the country than to obtain domestic
information. Taxpayers may also be tempted to provide only favorable
information. As a result, some countries have adopted rules to discourage this
practice. Under these rules, if a taxpayer does not make full disclosure, the
taxpayer is precluded from introducing any further information in any
subsequent legal proceedings involving the foreign source income.
The tax authorities of one country are not generally allowed to visit
another country for the purpose of auditing a taxpayer’s reported foreign
source income unless invited to do so by both the taxpayer and the foreign
government. Thus, it is usually more difficult for the tax authorities to verify
information provided by a taxpayer concerning its foreign activities. Some
countries have entered into arrangements providing for joint audits in certain
circumstances.

3.3 EXCEPTIONS TO WORLDWIDE TAXATION

3.3.1 Nonresident Companies and Other Legal Entities

Although countries are said to tax on a worldwide basis, the reality is quite
different. In theory, if a resident of a country that taxes on a worldwide basis
earns income from another country, that foreign source income will be
subject to residence country tax. If, however, the resident establishes a
foreign corporation or other foreign entity to earn the foreign source income,
that income will not be subject to tax by the residence country in the absence
of special rules such as controlled foreign corporation (CFC) rules or
foreign investment fund (FIF) rules. The foreign entity is generally
considered to be a separate taxable entity from the resident who owns it and,
in most cases, the foreign entity will be considered to be a nonresident of the
country in which the resident shareholder or owner resides. As a result, it is
relatively easy, especially for corporations resident in a particular country, to
avoid paying tax to that country on foreign source income through the use of
foreign corporations or other legal entities such as trusts.

Example
Corporation A, resident in Country A, derives income of 1 million from
business activities in Country B. The tax rate in Country B (20 percent) is
lower than the rate in Country A (30 percent). As a result, Corporation A will
pay tax to Country B of 200,000 on the income of 1 million derived from
Country B. Assuming that Country A taxes on a worldwide basis,
Corporation A will also pay Country A tax of 100,000 on the income of 1
million earned in Country B (300,000 less a credit for 200,000 of Country B
tax). If Corporation A establishes a wholly owned subsidiary corporation in
Country B and that subsidiary earns the income of 1 million from Country B
that would have been earned directly by Corporation A, the subsidiary will
pay 200,000 of Country B tax. Corporation A will pay no tax to Country B or
Country A until it receives dividends from the subsidiary or sells its shares of
the subsidiary.
The immediate tax saving by Corporation A of 100,000 is easy and
inexpensive to achieve; it simply requires the incorporation of a foreign
corporation and the transfer of the income-earning assets to it. As a result, it
is not surprising that the use of controlled foreign corporations and other
foreign entities, such as trusts, as a tax-planning device is widespread. Nor
should it be surprising that several countries have responded to such planning
with rules to prevent the avoidance or deferral of domestic tax by the use of
such foreign entities. These rules are discussed in Chapter 7, sections 7.3 and
7.4.

3.3.2 Temporary Residents

Some countries have special rules for temporary residents to relieve them of
some of the more onerous aspects of residence taxation. Temporary residents
are persons, such as corporate executives, who may be clearly resident in a
country because their homes, families, and employment are located in the
country but who intend to be, and are, resident only for a limited period,
usually less than five years. As residents of the country, they become subject
to all of that country’s rules concerning the taxation of foreign source
income; but because they are only temporarily absent from their home
countries, these temporary residents often retain substantial economic
interests there. The application of the full range of residence country rules to
temporary residents can cause serious problems that may discourage talented
persons from taking short-term postings in other countries.
 Two specific examples may serve to illustrate the difficulties. Assume
that an executive resident in Country A is a member of his company’s
pension plan. Under the tax law of Country A, the executive is entitled to
deduct his contributions to the plan and is not taxable on the employer’s
contributions to the plan. Moreover, the income earned and accumulated in
the plan is not taxable. However, distributions from the plan are taxable in
full. Assume further that the executive takes a temporary posting in Country
B, which does not allow any deduction for contributions to the pension plan
and taxes the executive on the employer’s contributions to the plan as a fringe
benefit. Country B taxes distributions from the pension plan only to the
extent that they exceed the employee’s and employer’s contributions to the
plan. The executive plans to retire in Country A. To the extent of
contributions to the plan while he was resident in Country B, the executive
will be subject to double taxation: once in Country B when the income was
earned, because he was taxed on the employer’s contributions to the plan and
not allowed any deduction for his own contributions, and again in Country A
when he receives distributions from the plan. This double taxation is clearly
unfair and should be eliminated.
As a second example, assume that A, a resident of Country A,
establishes a trust under the laws of Country A for the benefit of his children.
Country A taxes the income accumulating in the trust, but at the tax rates
applicable to the children. A moves to Country B to take a temporary position
for a few years and becomes subject to special rules in Country B for
residents who have established foreign trusts. Most likely, A did not establish
the trust to avoid Country B tax because at the time the trust was established
he may not have known that he would be moving to Country B. Under
Country B’s rules, A is taxable on the income of the foreign trust even though
the trust is taxable on its income in Country A. To eliminate this double
taxation, some countries exempt temporary residents from their foreign trust
and foreign investment fund rules for a limited period of time.

3.4 SPECIAL ISSUES

3.4.1 Exit or Departure Taxes

When a person ceases to be resident in a country that taxes residents on their
worldwide income, the person will no longer be subject to tax on worldwide
income in that country. The consequences of ceasing to be resident are not,
however, quite so simple. What if, for example, the person owns shares of a
resident corporation that are worth significantly more than when they were
acquired? Under most tax treaties, capital gains derived by a taxpayer
resident in one country from the disposal of shares of a company resident in
the other country are taxable only by the country in which the taxpayer is
resident (unless the value of the shares is attributable primarily to immovable
property located in the country). Therefore, if the person moves to another
country that has a treaty with his former country of residence, the shares can
be sold without any tax imposed by the former residence country. If the new
country of residence has been chosen carefully, it may not impose any tax, or
little tax, on the gain realized on the sale of the shares. Not surprisingly, the
person’s former country of residence is unlikely to be pleased about the
result.
Several countries have adopted special rules, often called exit or
departure taxes, to prevent the avoidance of domestic tax by departing
residents. Countries that have adopted exit taxes applicable to all property
include Australia, Canada, and Norway. Other countries, such as France,
Germany and the Netherlands, have more limited exit taxes that apply only to
certain shares in resident companies. The United States (US) has an even
more limited exit tax that applies only to transfers of tangible property out of
the US by US citizens who give up US citizenship and US permanent
residents (green card holders) who give up that status.
Typically, these taxes operate by requiring the departing resident to pay
tax not only on the income and gains realized up to the date that the taxpayer
ceases to be resident, but also on any accrued but unrealized income or gains.
For example, assume that X, a resident of Country X, ceases to be resident on
September 30, 2016. At that time X’s worldwide income from January 1,
2016 to September 30, 2016 is 70,000. X also owns shares of a company that
he started several years ago and which is now very successful. The shares
currently have a value of 30 million. The cost of the shares to X is nominal.
X also owns an interest-bearing bond issued by a company resident in
Country X. The interest on the bond is payable annually on December 31. If
Country X does not impose any tax on departing residents, X will not pay tax
on the accrued gain on the shares or the interest accrued to September 30 on
the bond. When X sells the shares or receives the bond interest after
September 30, 2016, he will no longer be resident in Country X. Even if
Country X imposes tax on nonresidents deriving interest from Country X and
realizing capital gains on shares in companies resident in Country X, there
may be a tax treaty between Country X and X’s new country of residence.
Typically, tax treaties based on the OECD or UN Model Treaties
preclude a country from taxing residents of the other country on capital gains
from disposals of shares of resident companies unless the assets of the
companies consist primarily of immovable property located in the country.
Treaties also typically limit the tax imposed by a country on interest paid to a
resident of the other country to 10 or 15 percent, which may be significantly
less than the rate of tax applicable to interest derived by residents.
To avoid these results, Country X may decide to impose an exit or
departure tax on persons ceasing to be resident. Under these rules, X will be
deemed to have disposed of his shares for their market value and to have
received the interest accrued on the bond immediately before ceasing to be
resident. These amounts would be included in X’s worldwide income for the
period ending on September 30, 2016 and would be subject to tax in Country
X. The treaty with X’s new country of residence would not apply to preclude
Country X from taxing X on these amounts as long as Country X’s tax is
imposed on income derived or deemed to be derived at a time when X is still
a resident of Country X.
Special problems are encountered with departure taxes. Most important,
the taxpayer will often not have the funds to pay the tax because the income
has not actually been received or the property has not actually been sold.
Accordingly, some countries allow the departing resident to defer the
payment of the tax if appropriate security for the ultimate payment of the tax
is provided. A departure tax may also cause serious problems of double
taxation. Most countries without departure taxes measure the amount of a
gain on the disposal of property as the difference between the sale proceeds
and the historical cost of the property. If a taxpayer ceases to be resident in a
country that levies a departure tax, the taxpayer will be subject to tax on the
accrued gain in respect of capital property owned at the time of departure.
The accrued gain is the amount of the value of the property immediately
before departure (e.g., 1,000) in excess of the historical cost of the property
(e.g., 200), so that the accrued gain is 800. When the taxpayer sells the
property at a future date, the taxpayer’s new country of residence will usually
tax the entire gain (e.g., 1,800 computed as proceeds (2,000) less historical
cost (200)) and not just the portion of the gain (1,000) that was not taxed by
the former country of residence. Thus, a portion of the gain may be subject to
double taxation. In this situation, under the provisions of a typical tax treaty,
neither country is obligated to provide relief for the other country’s tax
because both countries impose tax on the basis of the taxpayer’s residence
(but for different years so that the tiebreaker rules for dual residents do not
apply).
Countries that impose departure taxes often deem taxpayers becoming
resident to have acquired property owned at that time for its market value. In
effect, the taxpayer is given a step-up in the cost of the property for tax
purposes from its historical cost to its value at the time the taxpayer becomes
resident. The overall result of this step-up in cost and the departure tax is that
the country taxes only gains and losses accrued while a taxpayer is resident;
gains and losses accruing before a taxpayer becomes resident or after a
taxpayer ceases to be resident are not taxable.

3.4.2 Trailing Taxes

An alternative or supplement to an exit tax that some countries have adopted
is a so-called trailing tax. Under a trailing tax, a country imposes tax on all
or certain items of income of a resident even if the resident ceases to be a
resident under the country’s ordinary rule for determining residence. These
trailing taxes take a wide variety of forms and may be broad or narrow in
scope. For example, some countries, such as Germany, have special rules
under which former residents who move to designated tax havens continue to
be subject to tax on their entire income as deemed residents. The US has a
similar rule for US citizens who give up their US citizenship for the purpose
of avoiding US tax. Such former US citizens continue to be subject to US tax
for ten years after they renounce their US citizenship.
One common type of trailing tax applies to capital gains. As noted above
in connection with exit taxes, residents of high-tax countries may avoid tax
on accrued capital gains by shifting their residence to a country that does not
tax capital gains, or taxes them at a low rate, and then disposing of the
property. An exit tax on departing residents captures only the gain accrued to
the date that the taxpayer ceases to be resident. Under a trailing tax, the
taxpayer would be subject to tax on the entire gain if property owned at the
time the taxpayer ceased to be resident is disposed of within a certain period
after the taxpayer ceases to be a resident (typically five to ten years). The
basic operation of a trailing tax on capital gains of former residents of a
country and the relationship between a trailing tax and an exit tax is shown in
the following example.

Example

Country A imposes tax on capital gains at a rate of 15 percent; however,

Country A does not impose tax on capital gains (other than gains from the
disposition of immovable property located in Country A) realized by
nonresidents. A, a resident of Country A, owns shares of a company resident
in Country A that originally cost 1 million and now have a value of 10
million. If A moves from Country A to Country B, which does not tax capital
gains, A can avoid Country A’s capital gains tax on the 9 million accrued
gain. In order to prevent this type of tax avoidance, Country A may adopt a
trailing tax. This trailing tax would tax any capital gains in respect of
property owned by former residents at the time they cease to be resident if the
property is disposed of within five years after they cease to be resident.
Therefore, under the trailing tax, if A disposes of the shares for 13 million
within five years of ceasing to be resident in Country A, the entire 12 million
gain would be subject to tax by Country A.
If Country A has an exit tax, A’s accrued capital gain of 9 million at the
time of A’s departure from Country A would be subject to Country A tax at
the time A ceases to be resident in Country A. When A sells the shares for 13
million, say, four years later, a capital gain of 3 million (proceeds of 13
million less deemed cost of 10 million) would be subject to tax under
Country A’s trailing tax. However, if A waits for more than five years after
ceasing to be resident in Country A to sell the shares, the trailing tax would
not apply.
The application of a country’s trailing taxes will be prevented by any tax
treaties that the country enters into unless those treaties contain special
provisions allowing the imposition of such a trailing tax. Article 13 of the
OECD and UN Model Treaties does not allow the imposition of trailing taxes
on capital gains realized by former residents. Therefore, on the facts of the
preceding example, if Country A and Country B have a tax treaty with an
article dealing with capital gains similar to Article 13 of the OECD or UN
Model Treaties at the time of the sale of the shares, A would be entitled to the
benefits of the treaty as a resident of Country B. Country A would not be
entitled to tax A’s capital gain from the sale of the shares (unless the value of
the shares was derived primarily from immovable property situated in
Country A or the shares represent a substantial interest in the company (UN
Model Treaty only)). Some countries put special provisions in their tax
treaties to allow them to impose trailing taxes on certain capital gains.
Some countries, such as the United Kingdom (UK), impose tax on
former residents who resume their residence within a relatively short period
(five years in the case of the UK). The tax applies to capital gains realized
during the taxpayer’s nonresidence but applies only on the resumption of UK
residence.
3.4.3 Resident for Part of a Year
Special rules may be necessary if a taxpayer is a resident of a country for
only part of a year. A taxpayer could be subjected to tax on worldwide
income for the entire year if the taxpayer is resident at any time during the
year. This result seems harsh, especially if the taxpayer’s new country of
residence follows the same practice, although the dual-residence tiebreaker
rules in tax treaties may provide relief in some circumstances. As a result,
some countries have rules to tax part-time residents on their worldwide
income for only the portion of the year during which they are actually
resident. These rules may be difficult to apply with respect to certain types of
income, such as business income, that are usually calculated on an annual
basis. A taxpayer’s personal deductions, allowances, and credits are usually
prorated to reflect the portion of time during the year that the taxpayer is
actually resident. In theory, the taxpayer’s new country of residence should
provide the taxpayer with personal allowances for the balance of the year.
The problems of part-time residents are different from those of deemed
residents. For example, as discussed in Chapter 2, section 2.2.1, many
countries have 183-day rules under which persons who are present in the
country for more than 183 days in any year are deemed to be residents. These
residents are typically deemed to be resident for the entire year, not just for
the portion of the year during which they are present in the country.
However, part-time residents are resident for part of a year and nonresident
for the remainder of the year.
CHAPTER 4
Double Taxation Relief

4.1 INTRODUCTION
As discussed in Chapter 2, most countries tax their residents on their
worldwide income and nonresidents on their domestic source income.
Consequently, foreign source income earned by a resident of a country may
be taxed by both the country in which the income is earned (the source
country) and the country in which the taxpayer is resident (the residence
country). If income tax rates are low, as they were in the early years of the
last century when income taxes were in their infancy, the inefficiencies and
unfairness caused by this double taxation may be tolerable. But when tax
rates reach the levels that now prevail, double-tax burdens can become
onerous and interfere substantially with international commerce. The
necessity for the relief of international double taxation is clear on grounds of
equity and economic policy. However, the type of relief that is appropriate is
a controversial question.
International double taxation can arise in a variety of ways. The
following three types of double taxation arise from overlapping claims by
two or more countries to tax the same income:

– Source-source claims. Two countries assert the right to tax the same
income of a taxpayer because they both claim that the income is
sourced in their country.
– Residence-residence claims. Two countries assert the right to tax the
same income of a taxpayer because they both claim that the taxpayer
is a resident of their country. A taxpayer that is a resident of two
countries is commonly referred to as a “dual-resident taxpayer”.
– Residence-source claims. One country asserts the right to tax foreign
source income of a taxpayer because the taxpayer is a resident of that
country, and another country asserts the right to tax the same income
because the income arises or has its source in that country.
 Of these three types of international double taxation, overlapping
residence-source claims are the most likely to occur. To some degree,
taxpayers can minimize their exposure to the other types of double taxation
through careful tax planning, but residence-source double taxation is difficult
for taxpayers to avoid through tax planning. Therefore, the attempts of the
international tax community to deal with international double taxation have
focused primarily on the elimination of residence-source conflicts.
International double taxation can also occur due to differences in the
way countries define income and in the timing and tax accounting rules they
adopt for computing income. As explained in Chapter 6, international double
taxation may also occur due to disputes between countries about the proper
arm’s-length prices for cross-border transfers of goods and services between
related parties. Other rules adopted to curtail tax avoidance can also produce
double taxation. For example, if one country denies the deduction of interest
paid by a resident corporation to a shareholder in another country pursuant to
thin capitalization rules and treats the interest paid as a dividend, the
amount may be taxable in both countries, as a dividend subject to
withholding tax in one country and as interest included in a resident’s income
by the other country.
Typically, tax treaties provide relief from the three major types of
international double taxation, and from some of the other types as well,
although the relief is sometimes limited. Some cases of double taxation
resulting from overlapping claims based on the source of income are dealt
with by explicit rules for the source of income. For example, Article 11(5) of
the OECD and UN Model Treaties provides a rule that interest is deemed to
arise (i.e., have its source) in the country in which the payer is resident. As
noted in Chapter 2, section 2.3.1, however, most tax treaties do not contain
extensive source rules. Cases involving source-source double taxation that
are not resolved by the specific provisions of a treaty may be resolved
through consultation between the competent authorities of the two treaty
countries under the treaty’s mutual agreement procedure. See Chapter 8,
section 8.8.3 for a discussion of the mutual agreement procedure. Resolution
of such issues is not easy because the competent authorities of most countries
are naturally reluctant to give up their country’s right to tax domestic source
income.
Individual taxpayers almost always obtain relief from international
double taxation resulting from dual residence through the tiebreaker rules in
tax treaties. Cases involving the dual residence of legal entities are also
resolved by treaty. As discussed in section 2.2.3, Article 4(2) of the OECD
and UN Model Treaties provides a series of “tie-breaker” rules to resolve
cases in which an individual is resident in both countries. The dual residence
of a legal entity is resolved under Article 4(3) the OECD and UN Model
Treaties by deeming the entity to be resident in the country where its place of
effective management is located. The mutual agreement procedure is
sometimes used to deal with dual-residence cases that are not resolved
explicitly in the treaty. Since dual-resident entities are often used to avoid tax,
some bilateral tax treaties deny treaty benefits to such entities.
Ordinarily, the residence country grants relief from double taxation
resulting from the imposition of tax on the same item of income by both the
residence country and the source country. In other words, the source
country’s right to tax on the basis of the source of the income has priority
over the residence country’s right.
Three methods – the deduction method, the exemption method, and
the credit method – are commonly used for providing relief from double
taxation. These methods are discussed in section 4.3 below after a brief
explanation of what is meant by the term “international double taxation”.

4.2 INTERNATIONAL DOUBLE TAXATION DEFINED

The term “double taxation” is used in so many different contexts that any
precise definition of the term is not appropriate in all contexts. The term is
not defined in the OECD or UN Model Treaties or in the Commentary on
those Models, although they identify one of their main objectives as “the
avoidance of double taxation with respect to taxes on income and on capital”.
“International double taxation” can be defined as the imposition of
income taxes by two or more sovereign countries on the same item of income
(including capital gains) of the same taxable person for the same period. This
juridical or legal definition of international double taxation is narrow and
does not cover many situations that commentators frequently refer to as
double taxation, although it does identify the essential ingredients of
international double taxation. Even so, under this definition, it is not always
easy to determine whether double taxation exists in a particular case. For
example, questions may arise as to whether the taxes levied by the two
countries are both income taxes or whether the items of income subject to tax
are the same.
The legal definition of international double taxation should be
distinguished from the broader economic concept of double taxation.
Economic double taxation occurs whenever there is multiple taxation of the
same item of economic income. Under the legal definition, taxation of a
subsidiary company by one country and taxation of the parent company on a
dividend from that subsidiary by another country is not international double
taxation because the two companies are separate legal entities. In the
economic sense, however, the parent and the subsidiary constitute a single
enterprise. Economic, but not legal, double taxation also may arise when
income is taxed to a partnership and to the partners or when it is taxed to a
trust and to the beneficiaries of the trust.
Methods for relieving international double taxation are primarily
focused on legal double taxation rather than economic double taxation. The
reason double taxation relief is limited to legal double taxation is that the
definition of economic double taxation is exceedingly broad and difficult to
specify with the precision needed for tax laws. For example, some economic
double taxation occurs when income is taxed when earned and again when
consumed, yet no country is prepared to extend double taxation relief to sales
taxes or other consumption taxes. Similarly, countries are not prepared to
grant relief from the economic double taxation resulting from the imposition
of both an income tax and an estate or wealth tax. However, double taxation
relief is sometimes extended to economic double taxation where taxes are
paid by foreign subsidiaries and other foreign affiliates of a resident parent
corporation despite the fact that the taxes are not paid by the parent.
International double taxation should be distinguished from the double
taxation of an item of income by a single country, which might be termed
“domestic double taxation”. Domestic double taxation may arise, for
example, with respect to income earned by a corporation and distributed to its
domestic shareholders under the so-called classical method of corporate
taxation. It may also arise when tax is imposed on the income of a person by
both the central government of a country and one or more of its political
subdivisions. Double taxation by national and sub-national governments is
not necessarily objectionable – indeed, when the levels of taxation are
properly regulated to avoid excessive tax burdens, such double taxation may
be an inevitable feature of fiscal federalism.
4.3 RELIEF MECHANISMS
No international consensus has been reached on the appropriate method for
granting relief from international double taxation. The following three
methods are in common use. Most countries use all three methods for
different types of international double taxation; a country may use only one
of these methods, or it may use some combination of methods:

– Deduction method. The residence country allows its taxpayers to

claim a deduction in computing income for taxes, including income
taxes, paid to a foreign government in respect of foreign source
income.
– Exemption method. The residence country exempts foreign source
income derived by its residents from residence country tax.
– Credit method. The residence country provides its resident taxpayers
with a credit for income taxes paid to a foreign country against
residence country taxes otherwise payable. Under the credit method,
foreign taxes are deductible in computing the tax payable to the
residence country but not in computing the taxpayer’s income.

Foreign source income earned by residents of a country that uses the

deduction method is taxable at a higher effective rate than it would be under
either the credit method or the exemption method. The exemption method
and the credit method typically give equivalent results whenever the effective
foreign tax rate is equal to or greater than the domestic effective tax rate. The
exemption method is generally the most favorable to the taxpayer when the
foreign effective tax rate is less than the domestic effective tax rate. The basic
results under the three methods are illustrated by the following example.

Example

R, a resident of Country A, earns 100 of income from Country B on which

she pays 40 of tax to Country B. Under the deduction method, R will pay tax
to Country A on net income of 60 (100 – 40). The foreign tax paid to Country
B of 40 is deductible in computing R’s income subject to tax in Country A.
Assuming that R is taxable in Country A at a rate of 50 percent, she will pay
tax of 30 to Country A and a total tax of 70 on her income of 100, for a
combined foreign and domestic rate of 70 percent. If Country A uses the
credit method, R’s tax liability to country A (before any foreign tax credit)
will be 50 percent on her total worldwide net income (100) with no deduction
for the taxes paid to Country B. However, she will receive a credit against the
tax otherwise payable to Country A for the taxes paid to Country B of 40.
The foreign tax paid of 4 is deductible against the tax payable to Country A.
The result is that R will pay tax to Country A of only 10 (50 – 40) and total
tax of 50, for a combined foreign and domestic effective tax rate of 50
percent. Finally, if Country A uses the exemption method, R will pay no tax
to Country A in respect of the foreign source income earned in Country B,
and the total tax payable on the income will be 40, for a combined foreign
and domestic rate of 40 percent. These results are summarized in Table 4.1.

Table 4.1 Comparison of Methods for Relieving Double Taxation

Deduction Credit Exemption
Method Method Method
Foreign source income 100 100 100
Foreign tax (40%) 40 40 40
Deduction for foreign tax 40 nil nil
Net domestic income 60 100 nil
Domestic tax before credit
30 50 nil
(50%)
Less: foreign tax credit nil 40 nil
Final domestic tax 30 10 nil
Total domestic and
70 50 40
foreign tax

Additional information on the operation of the deduction, exemption,

and credit methods is provided below in sections 4.3.1, 4.3.2, and 4.3.3,
respectively, and the exemption and credit methods are compared in section
4.3.4. Section 4.3.5 examines some of the effects of tax treaties on double-
taxation relief.

4.3.1 Deduction Method

Countries using the deduction method tax their residents on their worldwide
income and allow those taxpayers to take a deduction for foreign taxes paid
in the computation of their taxable income. In effect, foreign taxes – income
taxes and other types of taxes – are treated as costs or current expenses of
doing business or earning income in the foreign jurisdiction. As noted above,
the deduction method is the least generous method of granting relief from
international double taxation.
The deduction method was used by a number of countries in the
formative years of their tax systems when worldwide tax rates were low, and
at that time it was an acceptable approach. As tax rates increased in the post-
World War II period, however, most countries adopted either the exemption
method or the credit method as the basic method for relieving international
double taxation. The OECD and UN Model Treaties authorize only the
exemption method and credit method as methods for granting double-tax
relief.
The deduction method has not disappeared. Several countries that have
adopted the credit method have retained the deduction method as an optional
form of relief and as a way of dealing with foreign taxes that, for some
reason, do not qualify for the foreign tax credit. In addition, some countries
use the deduction method for taxes paid with respect to income derived from
foreign portfolio investments.
In effect, countries use the deduction method whenever they tax
residents on the net amount of the dividends they receive from a foreign
corporation, assuming that the foreign corporation has paid some foreign
income tax and a foreign tax credit is not allowed with respect to that tax. For
example, assume that FCo, a foreign corporation, earns 100 of foreign
income and pays foreign income tax of 20. FCo pays its remaining after-tax
income of 80 as dividends to its shareholders, including a dividend of 20 to
R, a resident of Country A who owns 25 percent of the shares of FCo. On
these facts, R has earned 25 of foreign source income through FCo on which
foreign income tax of 5 (25 percent × 20) is paid. If Country A taxes R on
income of 20, it is in effect allowing R a deduction for the 5 of income tax
that was paid by FCo. A country that requires the associated tax to be added
to net dividends is said to “gross up” the dividends to approximate the
before-tax income out of which the dividends were paid. The purpose of a
gross-up rule is to provide equivalent treatment to taxpayers earning foreign
income directly and taxpayers earning such income indirectly through a
foreign corporation. See the discussion of the indirect foreign tax credit in
section 4.3.3.3 below.
The effect of the deduction method is that residents earning foreign
source income and paying foreign income taxes on that income are taxable at
a higher combined tax rate than the rate applied to domestic source income.
As a result, the deduction method creates a bias in favor of domestic
investment over foreign investment whenever the foreign investment is likely
to be subject to foreign income tax. Thus, the deduction method is not neutral
with respect to the allocation of resources between countries. This treatment
may be justified from the viewpoint of national self-interest: not only is
domestic investment encouraged, but also residents with equal net worldwide
income are treated similarly in that they will pay the same amount of
domestic tax. Of course, from the perspective of the total (combined domestic
and foreign) tax burden on a taxpayer’s worldwide income, the deduction
method does not achieve equal treatment of residents. Although residents
with equal net worldwide income will pay the same domestic tax, they may
pay widely differing amounts of foreign tax.

4.3.2 Exemption Method

Under the exemption method, the country of residence taxes its residents on
their domestic source income and exempts them from domestic tax on some
or all of their foreign source income. In effect, the country of residence gives
up its right to tax foreign source income, which consequently is taxable
exclusively by the source country. The exemption method completely
eliminates residence-source international double taxation because only one
jurisdiction, the source country, imposes tax on the income.
Some countries – Hong Kong is a prominent example – have adopted
the exemption method with respect to most or all foreign source income
earned by their residents. In effect, these countries tax only income from
domestic sources. For this reason, they are often said to tax on a territorial
basis rather than a worldwide basis. For most countries using the exemption
method, however, the exemption of foreign source income is limited to
certain types of income, most commonly business income earned in foreign
countries and dividends from foreign affiliates. Further, the exemption
method is sometimes restricted to income that has been subject to tax, or
subject to a minimum rate of tax, by the foreign country.
 Although foreign source income may be exempt from residence country
tax by countries using the exemption method, the income may be taken into
account in determining the rate of tax applicable to the taxpayer’s other
taxable income. This practice is referred to as “exemption with progression.”
In such systems, the foreign source income is included in income for the
limited purpose of determining a taxpayer’s average tax rate as if the foreign
income were taxable; this average rate is then used to compute the actual tax
due on the taxpayer’s other (non-exempt) income. Several countries,
including Belgium, Finland, Germany, and the Netherlands, use the
exemption with progression method.

Example

Assume that Country A levies tax at a rate of 20 percent on the first 10,000 of
income and 40 percent on income in excess of 10,000. T, a taxpayer resident
in Country A, has 10,000 of domestic source income from Country A and
10,000 of exempt foreign source income. T would pay tax of 2,000 (20
percent of 10,000) under a regular exemption system. Under an exemption
with progression system, T must determine the average tax rate that would
apply if his entire income of 20,000 were domestic source income. In this
example, the average rate would be 30 percent ((10,000 × 0.20 + 10,000 ×
0.40) divided by 20,000). The tax payable to Country A would then be
determined by applying the 30 percent average rate to the domestic source
income of 10,000, resulting in tax payable of 3,000.
The exemption method is relatively simple for the tax authorities to
administer and is effective in eliminating international double taxation. The
exemption with progression system is more complex because it requires the
tax authorities to obtain information about the amount of foreign source
income earned by resident taxpayers.
Although the exemption method is widely used and is sanctioned by
both the OECD and UN Model Treaties (see Article 23A of both treaties), it
is inconsistent with the tax policy objectives of fairness and economic
efficiency. To the extent that foreign taxes are lower than domestic taxes,
resident taxpayers with exempt foreign source income are treated more
favorably than other residents. Moreover, an exemption system encourages
resident taxpayers to invest abroad in countries with lower tax rates,
especially in tax havens, and encourages them to divert domestic source
income to such countries. For example, a taxpayer residing in an exemption
country who earns interest on funds invested in that country has a strong
incentive to move the funds to a foreign country that imposes low or no taxes
on interest income.
Because of these deficiencies, as noted above, the application of the
exemption method for relieving double taxation to all foreign source income,
which is equivalent to taxing on a territorial basis, is difficult to justify and is
used by only a few countries. The exemption method can be justified if it is
used as a convenient and simple proxy for the credit method or is limited to
certain types of income. For example, a country might exempt resident
taxpayers on income derived from foreign countries that impose tax at rates
and under conditions that are roughly comparable to its own rates and
conditions. If such an exemption system is properly enforced, the results are
similar to those obtained under a credit system because, in such
circumstances, a country using the credit method would collect little or no tax
with respect to any foreign source income that is subject to foreign tax
comparable to the residence country’s tax. This point is illustrated in the
following example.

Example

ACo is resident in Country A, which levies income tax at a rate of 40 percent.

ACo earns income of 1,000 in each of Country B and Country C, which levy
tax at rates of 40 and 50 percent respectively. Country A has a foreign tax
credit system to relieve international double taxation. Consequently, the
credits for taxes paid to Countries B and C, 400 and 500 respectively, will
completely offset Country A’s tax of 800 on ACo’s total foreign source
income of 2,000. Country A will collect tax from ACo after allowing the
credit for foreign taxes only if ACo’s effective foreign tax rate is less than the
effective tax rate of Country A.
Of course, in the example above, the effective foreign tax rate may be
lower than the Country A rate even if Country B and Country C generally
impose substantial taxes on foreign corporations. For example, one or both
countries may offer some special tax incentives or their tax laws may contain
some loopholes that foreign corporations are able to exploit. In such
circumstances, Country A might collect some tax revenue from ACo in
respect of its foreign source income.
 Several countries use the exemption method for active business income
earned by resident corporations through a foreign branch or permanent
establishment. Several countries also exempt certain dividends received from
foreign corporations in which resident corporations have a minimum
ownership interest, usually 5 or 10 percent. This exemption for dividends is
often referred to as a participation exemption and is discussed in more
detail in section 4.3.3.1 below.
The alleged virtue of the exemption method for relieving international
double taxation is its simplicity: it minimizes compliance costs for taxpayers
and administrative costs for tax authorities. However, for an exemption
system to operate effectively, a country must be able to ensure that the
exemption is limited to foreign source income that is subject to foreign tax
comparable to domestic tax. Thus, an effective exemption system requires
vigorous source-of-income and expense rules. It also requires anti-avoidance
rules to prevent low-taxed foreign source income from qualifying for
exemption. Finally, it requires expense allocation rules or anti-avoidance
rules to prevent taxpayers from deducting expenses incurred to earn exempt
foreign source income against their domestic income.
One often-overlooked weakness of an exemption system is its likely
impact on the shifting of tax burdens from an income earner to the payer in
some circumstances. Assume, for example, that Country A, which has a
corporate tax rate of 50 percent, provides an exemption for foreign source
income. Country B imposes a withholding tax of 25 percent on interest
payments made to nonresidents. ACo, a resident of Country A, makes a loan
of 100,000 to BCo, a resident of Country B. If ACo can earn 10,000 of
interest free of tax by loaning money to a resident of Country C instead of
BCo, ACo is likely to demand that it receive annual payments of 10,000, net
of Country B’s withholding tax from BCo. Therefore, BCo must gross up its
payments to ACo so that ACo ends up with 10,000 after Country B’s 25
percent withholding tax. The effect of this arrangement is that the burden of
the withholding tax of 2,500 imposed by Country B on the payment to ACo is
borne by BCo. This economic effect would be avoided if Country A used the
credit method. In that case, ACo would pay taxes of 5,000 wherever it earned
the 10,000 of interest income. ACo would have no leverage to shift Country
B’s withholding tax to BCo because it would have no opportunity for earning
10,000 free of tax and Country B’s withholding tax would be creditable
against ACo’s tax payable to Country A.
4.3.2.1 Participation Exemption

Most foreign direct investment takes the form of equity or share investments
in foreign or nonresident corporations. Special considerations apply to the
relief of international double taxation with respect to dividends from foreign
corporations and capital gains from the disposition of shares of foreign
corporations. This section discusses the exemption of dividends and capital
gains with respect to substantial participations in foreign corporations. The
indirect or underlying foreign tax credit for dividends from foreign
corporations is discussed in section 4.3.3.3 below. The participation
exemption and the indirect credit are compared in section 4.3.4.
Several countries use the exemption method to eliminate the double
taxation of dividends from foreign corporations. The exemption method has
been the traditional method used by European countries; however, in recent
years Australia, Japan, and the United Kingdom have also adopted
participation exemptions. The United States (US) has been discussing the
possible adoption of an exemption for dividends for many years, but has not
yet done so.
There are 3 key elements in the design of a participation exemption:

– the level of share ownership necessary to qualify for the exemption;

– the nature of the income earned by the foreign corporation out of
which the dividends are paid; and
– the amount of foreign tax on the income of the foreign corporation.

These same three elements are also important in the design of an indirect
foreign tax credit, as discussed in section 4.3.3.3.
The participation exemption is limited to dividends received by a
resident corporation from a foreign corporation in which the resident
corporation has a substantial ownership interest or participation. The level of
share ownership required varies from 5 percent (e.g., in the Netherlands) to
25 percent (e.g., in Japan) and in the Parent-Subsidiary Directive in the EU.
Many countries use a 10 percent ownership threshold. The ownership
threshold can be based on voting shares, the value of shares (or both votes
and value) or all the shares of the foreign corporation.
In theory, an exemption for dividends should be limited to dividends out
of the active business income earned by a foreign corporation. Dividends out
of passive investment income should not qualify for exemption; otherwise,
resident corporations would have an incentive to divert passive income to
their foreign subsidiaries in order to reduce residence country tax. For
example, assume that ACo, a company resident in Country A, has funds
available for investment that could earn passive income of 1 million. If ACo
earns the income by investing in Country A, it will pay tax to Country A of
40 percent. However, if ACo uses the funds to acquire shares in its wholly
owned subsidiary, BCo, resident in Country B, which taxes at a rate of only
10 percent, and BCo earns passive income of 1 million, BCo will pay tax to
Country B of 100,000. BCo can then distribute its after-tax profits of 900,000
to ACo. Assuming that Country A exempts the dividend, this simple tax
planning would result in substantial tax savings for ACo.
Therefore, some countries limit the exemption to dividends out of active
business income of foreign affiliates. Such an approach imposes significant
compliance obligations on taxpayers to keep track of the type of income
earned by their foreign affiliates and requires rules to determine the type of
income from which dividends are considered to be paid. As a consequence of
these problems, some countries have abandoned any attempt to limit their
participation exemptions to dividends paid out of active business income of
foreign affiliates of resident corporations, and instead rely on CFC rules or
other anti-avoidance rule to prevent the abuse of the participation exemption.
For example, under CFC rules, any passive income earned by a controlled
foreign affiliate of a resident corporation is taxable to the resident corporation
when earned by the controlled foreign affiliate without waiting for the
income to be distributed in the form of a dividend. If the passive income is
taxable to the resident parent corporation when earned, any subsequent
dividend out of that income can be exempt from tax. CFC rules are discussed
in Chapter 7, section 7.3.
As noted above, if the income of a foreign affiliate in which a resident
corporation has a substantial participation is subject to foreign tax at a rate
that, when combined with any withholding tax on dividends, approximates
the tax rate imposed by the residence country, the residence country will not
collect any tax on dividends from foreign affiliates in that country even if it
uses the credit method. Therefore, from a theoretical tax policy perspective, a
participation exemption can be justified as a proxy for a foreign tax credit if
the exemption is limited to dividends out of income that is subject to foreign
tax (corporation tax and dividend withholding tax) at a rate that is comparable
to the residence country’s corporate tax rate.
Some countries have limited their participation exemptions to dividends
from foreign affiliates established in listed comparable-tax countries or to
countries with which they have concluded bilateral tax treaties that provide
an exemption for dividends. In the interests of simplicity, other countries
have abandoned any attempt to limit their participation exemptions to
dividends that are paid out of income that has been subject to foreign tax
comparable to residence country tax. In these countries, the participation
exemption is available even for dividends from foreign affiliates in low-tax
countries. Most of these countries rely on other rules, such as CFC rules, to
prevent abuses of the participation exemption. As noted above, if the income
of a CFC is taxable to its resident parent corporation when earned, any
subsequent dividends out of that income can be exempt from residence
country tax.
Some countries with a participation exemption for dividends from
foreign affiliates also extend the exemption to capital gains on the disposition
of the shares of those foreign affiliates. The rationale for extending the
participation exemption to capital gains is that, from an economic and
commercial perspective, dividends are often a substitute for capital gains with
respect to substantial participations. Thus, if dividends from a foreign affiliate
are exempt from tax by the country in which the shareholder corporation is
resident but capital gains on the sale of the shares of a foreign affiliate are not
exempt, the shareholder corporation can reduce the capital gain from the sale
of the shares of a foreign affiliate by requiring it to pay exempt dividends
before the sale.
For example, assume that ACo, resident in Country A, owns all of the
shares of BCo, resident in Country B. Country A has a participation
exemption for dividends from foreign corporations in which resident
corporations own at least 10 percent of the shares (by votes and value).
However, Country A imposes a tax of 20 percent on capital gains, including
capital gains from the disposal of shares of foreign corporations. ACo is
contemplating a sale of the shares of BCo to an arm’s length purchaser and
expects to make a capital gain of 10 million (proceeds of sale of 14 million
less the cost of the shares (4 million)). The gain would be subject to tax by
Country A of 20 percent of 10 million, or 2 million. If BCo pays a dividend
of 10 million to ACo before the sale, the dividend will reduce the value of the
shares, the proceeds of sale and the capital gain. However, the dividend may
be subject to withholding tax by Country B. If so, the payment of dividends
to reduce the capital gain would be beneficial only to the extent that the
source country’s withholding tax is less that the residence country’s tax on
the capital gain.

4.3.3 Credit Method

Under the credit method, foreign taxes paid by a resident taxpayer on foreign
source income generally reduce domestic taxes payable on that income by the
amount of the foreign tax. For example, if P pays a foreign tax of 10 on some
foreign source income and would otherwise be subject to domestic tax of 40
on that income, the foreign tax credit reduces the domestic tax payable from
40 to 30. Consequently, the credit method completely eliminates international
double taxation of the residence-source type. Under the credit method,
foreign source income is subject to domestic tax whenever the foreign tax
paid is less than the domestic tax payable. In such circumstances, the net
domestic tax is an amount equal to the foreign source income multiplied by
the difference between the two tax rates. In effect, assuming that the domestic
tax rate is lower than the foreign tax rate, the foreign taxes are “topped up” by
domestic taxes so that the combined domestic and foreign tax rate on the
foreign source income is equal to the domestic tax rate.
Invariably, credit countries do not refund foreign taxes paid by their
residents on foreign source income in excess of the domestic tax on that
income; see, for example, Article 23B of the OECD and UN Model Treaties.
Similarly, countries with foreign tax credit systems do not generally allow
excess foreign taxes to offset taxes imposed on domestic income. In other
words, the credit for foreign taxes paid is usually limited to the amount of the
domestic tax payable on the foreign source income. Various limitation rules,
sometimes quite complex in application, as discussed below, are used to
prevent what are perceived to be inappropriate uses of foreign tax credits. As
a result of these limitations on the credit, foreign income is typically taxed at
the foreign tax rate whenever the foreign rate is higher than the domestic rate.
In summary, under the credit method, foreign source income earned by
residents is generally taxed at the higher of the domestic and foreign tax rates.

4.3.3.1 General Rules

The credit method avoids the shortcomings of the deduction method
described in section 4.3.1: resident taxpayers are treated equally from the
perspective of the total domestic and foreign tax burden on their foreign
source income, except when foreign taxes exceed domestic taxes. Moreover,
subject to the same exception, the credit method is neutral with respect to a
resident taxpayer’s decision to invest domestically or abroad. These points
are illustrated by the following example.
X and Y, who are both residents of Country A, each earn 100 of foreign
source income. The foreign tax on such income is nil for X and 40 for Y. If
both X and Y are subject to tax by Country A at a rate of 50 percent, X will
pay 50 and Y will pay 10 of tax to Country A. In both cases, the combined
domestic and foreign tax paid will be 50. If the foreign tax paid by Y is 60,
however, the combined domestic and foreign tax rate on Y would be 60
percent because Country A would not provide relief for 10 of foreign taxes
paid (60) in excess of domestic taxes (50) on the foreign source income. As a
result, Y would pay tax of 60 and X would pay tax of 50.
Many countries allow foreign income taxes that cannot be credited in the
current year (excess foreign tax credits) to be carried forward and credited
against domestic taxes in future years. The carry forward period varies from
country to country. The limitations on the credit apply to the deduction of
these excess foreign tax credits in future years. Assume, for example, that R
is resident in Country A, which imposes tax at a rate of 30 percent. In year 1,
R earns foreign income of 100 and pays foreign tax of 50. The foreign tax is
allowed as a credit against the Country A tax to the extent of 30, thereby
eliminating completely the tax payable to Country A. To the extent that the
foreign tax exceeds the Country A tax (20), the foreign tax is not creditable,
and R has an excess credit of 20. In year 2, assume that R earns foreign
income of 100 and pays foreign tax of 25. R might be allowed a credit of 30 –
the current foreign tax of 25 plus 5 of the excess credit carried forward from
year 1 for use in future years. The amount of the excess credit from year 1
that is available for carry forward to year 3 and subsequent years would be
reduced from 20 to 15.
On tax policy grounds, the credit method is recognized by many tax
commentators to be theoretically the best method for eliminating
international double taxation. The credit method, however, is not free from
difficulties. Most importantly, the operation of a foreign tax credit system can
be complex from the perspectives of both the government and taxpayers.
Among the difficult questions that must be resolved are the following:

– What foreign taxes are creditable?

– How should the limitations on the credit be calculated? On a source-
by-source, an item-by-item, a country-by-country, or an overall basis,
with various special rules applicable to certain types of income? Or
some combination of these methods?
– What rules should be adopted for determining the source of income
and deductions?

Detailed, technical, and highly complicated legislative provisions are

needed to resolve these and other matters if the credit method is to operate
effectively. The compliance and administrative burdens imposed on
taxpayers and tax authorities as a result of these complex rules are probably
both necessary and justifiable in respect of income earned in no-tax or low-
tax countries – otherwise, domestic tax could be avoided by diverting
domestic source income to these countries.
When resident taxpayers are subject to foreign tax on their foreign
source income at a rate that is comparable to the domestic tax rate, it is
questionable whether the complexity of a credit system is worthwhile. In
such circumstances, a country is unlikely to collect a significant amount of
domestic tax from those taxpayers with respect to their foreign source income
after allowing them a credit for foreign taxes. A foreign tax credit system
used by one country may encourage other countries to increase their taxes on
income earned by residents of that country to the level of tax in that country
(so-called “soak-up” taxes). Such a tax increase would not affect the after-
tax return to nonresident investors and therefore would not discourage
investment from abroad. It would, however, result in a shift of tax revenues
from the country with the credit system to the country in which the income is
earned. For example, assume that Country A imposes tax at a rate of 40
percent and uses a foreign tax credit system, and that residents of Country A
have substantial investments in Country B. If Country B imposes tax on the
income earned by residents of Country A at 25 percent, the residents of
Country A will be subject to tax by Country A on the income earned in
Country B of 15 (40 – foreign tax credit of 25) and total taxes on the income
of 40 (25 to Country B and 15 to Country A). However, if Country B imposes
tax at 40 percent on the income earned by the residents of Country A, those
residents will still be subject to total tax on the income of 40, but the entire
tax will be paid to Country B.
A country is most likely to impose a discriminatory tax on residents of
credit countries when the overwhelming amount of foreign investment in the
country is owned by residents of a few foreign countries, and those foreign
countries have approximately equivalent tax rates. Some countries include
provisions in their foreign tax credit rules to prevent the soak-up taxes from
qualifying as creditable foreign taxes.
Many countries use the credit method to eliminate international double
taxation with respect to at least certain taxpayers and types of foreign source
income. Some countries grant a credit for foreign taxes unilaterally; others
grant a credit only pursuant to their bilateral tax treaties. Most credit
countries grant the credit both unilaterally and by treaty. Still others have
extended their foreign tax credit mechanisms to encompass “tax sparing”.
Tax sparing is discussed in section 4.5 below.

4.3.3.2 Types of Limitations

As noted above, countries that use the credit method limit the credit for
foreign taxes to the amount of domestic tax on the foreign source income. For
this purpose, countries use a variety of limitations.
Under an overall or worldwide limitation, foreign taxes paid to all
foreign countries are aggregated; in effect, the credit is limited to the lesser of
the aggregate of foreign taxes paid and the domestic tax payable on the total
amount of the taxpayer’s foreign source income. This method permits the
averaging of high foreign taxes paid to some countries with low foreign taxes
paid to other countries.
Under a country-by-country or per-country limitation, the credit is
limited to the lesser of the taxes paid to a particular foreign country and the
domestic tax payable on the taxpayer’s income from that particular country.
This method prevents the averaging of high and low foreign taxes paid to
different countries, but it permits the averaging of high and low rates of
foreign tax paid to a particular country on different types of income.
Under an item-by-item limitation, the credit is limited to the lesser of the
foreign tax paid on each particular item of income and the domestic tax
payable on that item of income. This method prevents averaging and is
probably the best method from a theoretical perspective, although few
countries use it in practice. In this context, an “item” of income is some
defined category of income, such as interest income or shipping income. In
principle, a country might define an item of income as any category of
income subject to a special tax regime in a foreign country. For example, a
country might treat business income and interest income arising in a foreign
country as separate items of income for purposes of imposing a limitation on
its foreign tax credit, especially if foreign countries tax interest income
derived by nonresidents at preferential (low) rates.
The results of the overall, per-country, and item-by-item limitations on
the foreign tax credit are compared in the following example. ACo, a resident
of Country A, earns foreign source income and pays foreign taxes on such
income, as shown in the following table.

Table 4.2 Example: Facts

Foreign Income Foreign Tax
Business income from Country X 100,000 45,000
Dividends from Country X 20,000 1,000
Business income from Country Y 50,000 10,000
Interest from Country Z 10,000 1,500

The corporate tax rate in Country A is 30 percent. ACo earns 200,000

domestic source income from its business carried on in Country A. If there is
no limitation on the foreign tax credit, the amount of tax payable to Country
A would be:

Example: No Limitation
Total income 380,000
Tax before credit (30%) 114,000
Foreign tax credit 57,500
Total tax 56,500

Therefore, the total tax payable would be 114,000 (foreign tax of 57,500
and Country A tax of 56,500). If Country A uses an overall, per-country, or
item-by-item limitation, the tax payable would be as follows.
 Table 4.3 Example: Overall Limitation
Overall Limitation
Country A tax before credit 114,000
Credit:
Lesser of:
(1) Foreign tax of 57,500
(2) Country A tax on foreign income (180,000 x 30% 54,000
= 54,000)
Country A tax after credit 60,000
Total tax (57,500 + 60,000) 117,500

Example: Per-Country Limitation

Per-Country Limitation
Country A tax before credit 114,000
Credit:
(a) Country X
Lesser of:
(1) Foreign tax of 46,000
(2) Country A tax on Country X income (120,000 x 30% 36,000
= 36,000)
(b) Country Y
Lesser of:
(1) Foreign tax of 10,000
(2) Country A tax on Country X income (50,000 x 30% 10,000
= 15,000)
(c) Country Z
Lesser of:
(1) Foreign tax of 1,500
(2) Country A tax on Country X income (10,000 x 30% 1,500
= 3,000)
Total creditable taxes 47,500
Country A tax after credit 66,500
Total tax (66,500 + 57,500) 124,000

Example: Item-by-Item Limitation

Item-by-Item Limitation
Country A tax before credit 114,000
Credit:
(a) Country X
(i) business income
lesser of:
(1) Foreign tax of 45,000
(2) Country A tax on business income (100,000 x 30% 30,000
= 30,000)
(ii) dividends
lesser of:
(1) Foreign tax of 1,000
(2) Country A tax on dividends (20,000 x 30% = 6,000) 1,000
(b) Country Y
Lesser of:
(1) Foreign tax of 10,000
(2) Country A tax on business income (50,000 x 30% 10,000
= 15,000)
(c) Country Z
Lesser of:
(1) Foreign tax of 1,500
(2) Country A tax on interest (10,000 x 30% = 3,000) 1,500
Total creditable taxes 42,500
Country A tax after credit 71,500
Total tax (71,500 + 57,500) 129,000

The three methods for limiting the foreign tax credit are not mutually
exclusive. For example, a country could use an overall limitation as the basic
method and also use the item-by-item method for certain types of income
such as active business income and passive investment income. Several
countries use this type of hybrid method, which is sometimes referred to as
the separate-baskets approach.

4.3.3.3 Indirect or Underlying Credit

Some countries, such as the US, provide what is often referred to as an

“indirect” or “underlying” foreign tax credit. The indirect credit is a credit
granted to a resident corporation for the foreign income taxes paid by a
foreign affiliated company when the resident corporation receives a dividend
distribution from its foreign affiliate. The amount allowable as a credit is the
amount of the underlying foreign tax paid by the foreign affiliate on the
income out of which the dividend was paid. Ordinarily, a foreign tax credit is
allowable only for foreign income taxes that a resident taxpayer pays directly.
In effect, the indirect credit rules ignore the separate corporate existence of
the resident and foreign corporations for the limited purpose of allowing the
credit. To claim a credit for taxes paid by a foreign affiliate, the domestic
corporation must usually own a substantial interest, varying from 5 percent to
25 percent, in the share capital of the foreign corporation.
The basic operation of an indirect foreign tax credit is illustrated in the
following example. Assume that ACo, resident in Country A, has a wholly
owned subsidiary BCo, resident in Country B. BCo’s income for the year is
800, and it pays tax to Country B at a rate of 30 percent, or 240, on its
income. BCo distributes all its after-tax profits of 560 (800 – 240) to ACo as
a dividend. ACo is taxable in Country A on 800 – the dividend of 560 and the
underlying tax of 240 (often referred to as the “gross-up amount”). Assuming
that Country A levies tax at a rate of 40 percent and there is no limitation on
the foreign tax credit, the tax payable to Country A would be 80 (320 minus a
foreign tax credit of 240 for the foreign taxes paid by BCo on the income out
of which the dividend was paid).

Table 4.4 Example: Indirect or Underlying Foreign Tax Credit

BCo’s income 800
Country B tax 240
After-tax profit 560
Dividend paid 560
ACo’s income:
 Dividend received from BCo 560
Gross-up amount 240
Total 800
Country A tax before credit (40%) 320
Credit for Country B tax paid by BCo 240
Net Country A tax 80

If the dividend received by ACo in the above example is subject to

withholding tax by Country B, the withholding tax would also usually be
creditable against ACo’s tax payable to Country A, subject to any applicable
limitation rule. The credit for withholding tax is a direct foreign tax credit,
not an indirect credit, because ACo is treated as paying the withholding tax.
The credit method may have the effect of discouraging domestic
corporations that have earned profits abroad through foreign affiliates from
repatriating these profits as dividend distributions. Assume that ACo, resident
in Country A, has a wholly owned affiliate, FCo, resident in Country F. The
tax rate in Country A is 35 percent and the rate in Country F is 10 percent.
FCo earns profits in Country F of 100 and pays tax to Country F of 10. If
FCo’s after-tax profits are repatriated to ACo as a dividend of 90, ACo will
get a foreign tax credit of 10 for the underlying foreign tax paid by FCo, but
it will be required to pay a net tax to Country A of 25, as shown below.

Table 4.5 Example: Effects of the Credit Method

Dividend received from ACo 90
Gross-up amount 10
Income of ACo 100
Country A tax before credit (35%) 35
Indirect foreign tax credit for taxes paid by FCo 10
Country A tax 25

By retaining the profits in FCo, ACo can defer indefinitely the potential
Country A tax of 25. This type of tax planning strategy has been adopted by
several US multinationals and has been sharply criticized by some US
politicians.
 To avoid creating a bias against the repatriation of profits, a credit
country could tax the income of foreign affiliates of resident corporations on
an accrual basis (i.e., as the income is earned by the foreign affiliates).
Accrual taxation would eliminate the deferral of residence country tax on the
foreign source income earned by residents through foreign affiliates.
Proposals for a comprehensive accrual system have surfaced from time to
time, but have not yet been adopted in any country, although accrual taxation
is used in some circumstances. Under the controlled foreign corporation
rules and the foreign investment fund rules described in sections 7.3 and
7.4, some countries impose domestic taxes currently on certain income
earned by foreign affiliates and foreign funds in what are perceived to be
abusive situations.
The rules designed to govern the indirect foreign tax credit are often the
most complex part of a foreign tax credit system. The indirect credit is
available only when a resident corporation receives a dividend from a foreign
affiliate. The amount allowable as a credit is the amount of foreign income
tax properly attributable to the dividend. Difficult timing and income
measurement issues must be resolved for this purpose. For example, the
resident corporation must determine the profits of the foreign affiliate out of
which the dividend was paid and the foreign tax attributable to those profits.
Those profits may have been earned over many years in the past and would
usually have been computed in a foreign currency under tax accounting rules
that may differ significantly from the tax accounting rules applicable to the
resident corporation. When these rules are combined with rules for limiting
the foreign tax credit discussed in section 4.3.3.2 above, the level of
complexity causes serious compliance and administrative problems. This
complexity has led several countries to adopt exemption systems for
dividends from foreign affiliates.

4.3.4 Comparison of the Exemption and Credit Methods

The debate about whether the exemption method or the credit method is
better for relieving international double taxation is often vigorous and
emotional. Few countries have either a pure exemption system or a pure
credit system. Costa Rica, Hong Kong and Panama are examples of
jurisdictions that tax on a territorial basis; they tax only income earned or
having its source in their territory and generally exempt all foreign source
income from tax. For most countries using the exemption method, however,
the exemption of foreign source income is often restricted to certain active
business income earned by resident corporations and dividends from foreign
affiliates. Thus, a corporation is often exempt only on its active business
income derived from foreign sources and dividends received out of the active
business income of its foreign affiliates. An exemption is not generally
available for investment income because such an exemption would make it
easy for resident taxpayers to avoid paying taxes on their investment income
by shifting the source of domestic investment income to a foreign country.
With respect to business income, an analysis of the exemption and credit
methods indicates, first, that the two methods raise essentially the same
structural issues, and second, that the two methods are reasonably
comparable if designed properly. The following material compares an
exemption for dividends received out of active business income of foreign
affiliates and an indirect credit for the underlying foreign taxes paid by
foreign affiliates on active business income.
The first point is that the results of these two methods for relieving
international double taxation are the same if the underlying foreign taxes paid
by the foreign affiliate, plus any withholding taxes on the dividends, are at
least equal to the domestic taxes on the dividends. Under the exemption
method, the dividends are exempt from domestic tax, so the total tax is the
sum of the underlying foreign taxes paid by the foreign affiliate on the
income out of which the dividend is paid and any foreign withholding taxes
on the dividend. Under the indirect credit method, the underlying foreign
taxes and the foreign withholding taxes are creditable against the domestic
tax on the dividend. Therefore, if the sum of those foreign taxes equals or
exceeds the domestic tax on the dividend, no domestic tax is payable. This
result is illustrated in the following example.
Assume that Parentco, a company resident in Country A, has a wholly
owned subsidiary, Forco, resident and carrying on business in Country B.
Country A imposes tax on corporate profits at a rate of 35 percent. Country B
imposes tax on corporate profits at a rate of 30 percent. Forco earns profits of
100, pays tax to Country B of 30, and distributes its entire after-tax income of
70 to Parentco as a dividend. The tax results, if Country A uses a
participation exemption or an indirect credit system for relieving international
double taxation of dividends, are shown in the table below.

Table 4.6 Comparison of Credit and Exemption Methods

 Table 4.6 Comparison of Credit and Exemption Methods
Credit Exemption
Forco
Income of foreign subsidiary 100 100
Foreign tax (30%) 30 30
Dividend to parent 70 70
Withholding tax (10%) 7 7
Parentco
Dividend received 70 70
Gross-up amount 30
Taxable income 100 0
Domestic tax before credit 35 –
Foreign tax credit 37 –
Net domestic tax 0 0
Total tax 37 37

Even if the sum of the foreign corporate tax and the dividend
withholding tax is less than the domestic tax, remember that the domestic tax
payable by Parentco is deferred until dividends are received. The longer the
payment of dividends is deferred, the lower the present value of the domestic
taxes on the dividends, assuming that the foreign affiliate can earn a higher
after-tax rate of return on the funds than its parent corporation.
The usual justification for a participation exemption is simplicity: the
reduction of the costs of administration and compliance for tax officials and
taxpayers. However, the benefits of simplification are often overstated or are
achieved only by sacrificing the integrity of the exemption.
If the participation exemption is intended to be a proxy for the indirect
credit, it should be designed to ensure that the exemption is restricted to
foreign source income that is subject to foreign tax rates that are comparable
to domestic tax rates. A properly designed exemption system for dividends
requires complicated rules to protect its integrity. Many of these rules are
strikingly similar to the rules with respect to an indirect foreign tax credit. For
example, both systems require rules dealing with:
 – the resident taxpayers qualifying for the exemption or credit;
(usually, the entitlement to the exemption or credit is limited to
foreign affiliates in which resident corporations have a substantial
interest; (often defined as 10 percent or more of the share capital of
the foreign affiliate))
– the type of income that qualifies for the exemption or indirect credit;
(usually, these rules distinguish between active business income and
other types of income)
– the source of income;
– allocation of expense rules; (see section 4.3.5 below)
– the current or accrual taxation of passive income of foreign
corporations
– controlled by residents; (CFC rules, which are discussed in Chapter 7,
section 7.3 below)
– the treatment of foreign losses; and
– the computation of the income of the foreign affiliate in accordance
with domestic tax rules.

The most important difference between the exemption and indirect

credit methods is that the indirect credit method requires a definition of
creditable foreign taxes, whereas the exemption method requires rules to
determine when foreign source income is subject to a level of foreign tax that
is comparable to domestic tax (assuming that the exemption system is a
proxy for an indirect credit system).
As noted above, many countries achieve the benefits of simplification
with respect to their participation exemptions for dividends from foreign
affiliates only by sacrificing the integrity of the exemption. In many
participation exemptions, the exemption is available for dividends received
from foreign affiliates that have not been subject to foreign tax rates
comparable to domestic tax rates. Sometimes this appears to happen by
inadvertence rather than intentionally. For example, a country may provide an
exemption for dividends from foreign affiliates resident in countries with
which it has entered into tax treaties. If the country enters into tax treaties
with low-tax countries or countries that provide preferential low-tax regimes,
the dividend exemption will be available for dividends from foreign affiliates
in these countries, even though the income out of which the dividends are
paid is not subject to foreign tax rates that are comparable to domestic tax
rates.
Several countries have intentionally adopted participation exemption
systems that do not even attempt to ensure that the income of the foreign
affiliates is subject to foreign tax rates comparable to domestic tax rates. For
these countries, the exemption method is not a proxy for the credit method.
The underlying policy of such an exemption system is not just to eliminate
international double taxation (although it accomplishes that result) but also to
promote the international competitiveness of a country’s resident
multinational corporations. Thus, several countries have adopted exemption
systems under which all dividends received by resident corporations from
foreign affiliates in which they have a substantial interest are exempt from
residence country tax. As a result, multinationals resident in such countries
are able to compete in other countries with corporations resident in those
countries and in third countries because they are subject to tax only by the
country in which the business is carried on.
For example, assume that a multinational corporation, MCo, is resident
in Country A, which imposes corporate tax at a rate of 35 percent. MCo has a
wholly owned subsidiary that is resident and carries on business in Country
B, which imposes corporate tax at a rate of 12.5 percent. If Country A taxes
dividends received by MCo from its subsidiary in Country B, that tax
represents a cost to MCo of carrying on business in Country B (although the
cost is deferred until the dividends are paid) that corporations resident in
Country B and resident in other countries that carry on business in Country B
do not bear (assuming, of course, that those other countries exempt dividends
from foreign affiliates from tax).

4.3.5 Treaty Aspects

As mentioned above, both the credit and exemption methods are authorized
by Article 23 of the OECD and UN Model Treaties. The deduction method is
not authorized by these model treaties. Article 23 of the OECD and UN
Model Treaties establishes the general principles of exemption and credit,
with each country left to establish detailed rules in its domestic law for the
implementation of the general principle.
Some countries provide an exemption for foreign source income or a
credit for foreign taxes paid (and paid by a foreign affiliate) under their
domestic law in addition to providing relief in any treaties that they enter
into. Treaty relief is still important, however, because it may be more
generous than the unilateral relief provided in domestic law and because it
constrains a country’s ability to amend its domestic law to withdraw the
double taxation relief afforded to its residents.
For example, assume that Country A provides an exemption in its
domestic tax law for certain foreign source income earned by residents of
Country A. Country A enters into a treaty with Country B that incorporates
the same exemption. If Country A subsequently repeals the exemption in its
domestic law, it must nevertheless continue to provide the exemption to its
residents that earn income in Country B unless the treaty with Country B is
modified or terminated.

4.4 ALLOCATION OF EXPENSES

Whether a country uses an exemption method or a credit method to provide
relief from international double taxation, it should have rules for allocating a
proper portion of the expenses incurred by its resident taxpayers between
their foreign source gross income and their domestic source gross income.
Most countries recognize the need for such rules for the purposes of taxing
nonresidents on their domestic source income. Thus, expenses incurred by
nonresidents will be denied unless those expenses are properly related to the
earning of the domestic income subject to tax. Similar rules are necessary to
properly apportion the expenses of resident taxpayers between domestic
source and foreign source income for purposes of the foreign tax credit.
For countries that exempt foreign source income, expenses incurred by a
resident taxpayer to earn that income should not be deductible. This result
follows from the fundamental principle of tax law that expenses incurred to
earn exempt income should not be deductible. For example, a taxpayer
should not be allowed to deduct interest expense on borrowed funds used to
earn exempt foreign source income. A country that allows such interest
expenses to be deductible provides its resident taxpayers with an incentive to
earn exempt foreign source income rather than taxable domestic source
income. In effect, the country is providing an exemption not only for foreign
source income but also for a portion of the domestic source income of its
resident taxpayers.
Most countries lack specific rules for attributing expenses to foreign
source income. Two approaches that might be used for that purpose are
tracing and allocation or apportionment. A tracing approach involves a
factual inquiry into the connection between the expenses and the foreign
source income. In contrast, allocation or apportionment involves the
attribution of expenses to foreign source income by formula, either on the
basis of the proportion of the taxpayer’s foreign assets to its total assets or the
proportion of its gross foreign income to its total gross income. Unlike
tracing, allocation or apportionment is based on an assumption that the
relevant expenses were incurred to support all of the taxpayer’s assets or
income-earning activities equally.
Countries that have a foreign tax credit system should allow resident
taxpayers to deduct expenses incurred to earn foreign source income because
those taxpayers are taxable on their worldwide income. As explained above,
however, the foreign tax credit is invariably limited to the amount of a
country’s domestic tax otherwise imposed on foreign source taxable income.
For this purpose, the amount of a taxpayer’s foreign source taxable income
must be computed properly or the limitation on the credit will be improperly
inflated. In order to compute foreign source income properly, the taxpayer
should be required to deduct from its gross foreign source income the
expenses incurred to earn that income.
The need for expense allocation rules can be illustrated with a simple
example. Assume that a resident corporation borrows 1,000 with interest at 8
percent annually and uses the loan proceeds to finance the business activities
of a foreign branch. The foreign branch produces gross income of 280. After
deducting the interest payment of 80, the branch’s net income is 200. The net
income of 200 is subject to foreign tax of 50 percent, resulting in a tax of
100. If the corporation’s domestic source net income is 2,000, the
corporation’s total net income is 2,200. Assuming that the domestic tax rate
is 40 percent, the tax payable, prior to subtracting the allowable foreign tax
credit, is 880 (40 percent of 2,200). Subject to the limitation rules, the
corporation is entitled to a credit for the foreign taxes paid. The credit is
limited, however, to the lesser of 100 (the foreign tax paid) and 80 (880 ×
200/2,200) (the domestic tax on the foreign source income).

Table 4.7 Allocation of Expenses

Gross foreign income 280
Interest expense 80
Net foreign income 200
Foreign tax (50%) 100
Net domestic income 2,000
Total income (200 + 2,000) 2,200
Domestic tax of 40% 880
Credit for foreign tax 80
Total tax 800

In the above example, the interest expense of 80 was applied or

allocated totally against the foreign source income. If it had been applied
against domestic source income, the entire amount of foreign taxes (100)
would have been credIn the above example, the interest expense of 80 was
applied or allocated totallyitable against the domestic tax because the credit
would be limited to the lesser of 100 and 112 (880 × 280/2,200). Assuming
that the interest is properly attributable to the foreign source income, it should
be allocated to that income in computing the limitation on the credit because
otherwise the domestic tax system would be giving a credit for foreign taxes
in excess of the domestic taxes on the foreign source net income. Therefore,
in order to protect the domestic tax base, it is crucial for interest and other
expenses to be allocated properly between domestic source and foreign
source income.
An appropriate amount of expenses should also be attributed to foreign
source income for purposes of computing the limitation on the indirect
foreign tax credit, discussed in section 4.3.3.3. In addition, the indirect credit
raises the issue of the timing of the deduction of expenses incurred by a
resident parent corporation to earn foreign source income through a foreign
affiliate. Residence country tax on foreign source income earned through a
foreign affiliate is generally postponed or deferred until the resident parent
receives a dividend (or other taxable distribution). Interest and other expenses
incurred by the resident parent to earn that deferred income should not be
deductible, at least theoretically, until the income to which the expenses
relate is subject to residence country taxation. These payments should be
deductible when the resident taxpayer receives a taxable distribution out of
the related income from its foreign affiliate. Few countries currently attempt
to deal with this timing issue because of its complexity.
4.5 TAX SPARING
Some tax treaties provide for “tax sparing”,typically through a tax sparing
credit. A tax sparing credit is a credit granted by the residence country for
foreign taxes that, for some reason, were not actually paid to the source
country but that would have been paid under the source country’s normal tax
rules. The usual reason for the tax not being paid is that the source country
has provided a tax holiday or other tax incentive for foreign investors to
invest or conduct business in the country. In the absence of tax sparing, the
actual beneficiary of a tax incentive provided by a source country to attract
foreign investment might be the country in which the investor is resident
rather than the foreign investor. This result occurs whenever the reduction in
source country tax is replaced by an increase in residence country tax.
The shifting of the benefit of an incentive from the foreign investor to its
home country’s treasury in the absence of tax sparing is illustrated by the
following example. Country A, a developing country whose normal corporate
tax rate is 30 percent, offers foreign corporations a ten-year tax holiday if
they establish a manufacturing enterprise in Country A. BCo, a resident of
Country B, establishes a manufacturing plant in Country A. Country B
imposes corporate tax at a rate of 40 percent and uses a foreign tax credit
system to provide relief from international double taxation. BCo earns
income in Country A of 1,000 in the first year. In the absence of the tax
holiday, Country A would impose a tax of 300 on BCo and Country B would
impose a tax of 100 on BCo, determined by subtracting from the tax of 400
otherwise payable a foreign tax credit of 300. The tax holiday eliminates
Country A’s tax of 300. Therefore, BCo’s tax liability to Country B becomes
400 minus the allowable credit, which is zero because BCo did not actually
pay any tax to Country A due to the tax holiday. Thus, the tax revenue of 300
forgone by Country A in granting the tax holiday to BCo goes to the benefit
of Country B and not to BCo.
If Country B were willing to give a tax sparing credit to BCo for the
taxes forgone by Country A, then BCo would get the benefit of the tax
holiday. Its income in Country A would be 1,000, and it would pay no tax to
Country A. It would have an initial tax obligation of 400 in Country B but
would be allowed to reduce that amount by the 300 of tax forgone by
Country A, for a total tax liability of 100. The results are shown in the
following table.
 Table 4.8 Example: Tax Sparing
Country A
Income from Country A 1,000
Country A tax before holiday 300
Tax holiday credit 300
Country A tax 0
Country B
Income from Country B 1,000
Country B tax 400
Tax sparing credit 300
Total Country B tax 100

As the example shows, in the absence of the tax sparing credit, Country
A’s tax holiday will not have any impact on potential investors resident in
Country B because their residence country tax will increase to offset the
benefit of the tax incentive provided by Country A.
Tax sparing is primarily a feature of tax treaties between developed and
developing countries. In the past, many developed countries granted some
form of tax sparing to developing countries by way of treaty as a matter of
course, with some voluntarily granting tax sparing in their treaties with
developing countries as a way of encouraging investment in those countries,
but others granting tax sparing credits only reluctantly. Some developing
countries traditionally have refused to enter into a tax treaty with a developed
country unless they obtain a tax sparing credit.
The US is adamantly opposed to tax sparing and has not granted it in
any of its tax treaties. Consequently, for many years it concluded very few
tax treaties with developing countries. The US position is that the grant of a
credit for phantom taxes – taxes not actually paid – is inconsistent with the
efficiency and fairness goals of its foreign tax credit and encourages
developing countries to engage in beggar-thy-neighbor bidding wars through
their tax incentive programs. This position has been characterized as
“arrogant”, “imperialistic”, and “patronizing”, but it is a defensible
assessment of the effects of tax sparing. In recent years, the hard-line view of
many developing countries has softened, and the number of US tax treaties
with developing countries is growing rapidly. Several other developed
countries have recently agreed to tax sparing provisions in their treaties with
developing countries only under stringent conditions.
The merits of tax sparing credits cannot be divorced from the merits of
the tax incentives that they encourage. Although tax incentives have some
enthusiastic supporters in the political arena, they are difficult to justify on
the basis of tax policy principles. Certain targeted incentives aimed at
achieving some identified goal may be justified, but those incentives are so
narrowly drawn to prevent abuse that they tend to generate little political
support. The general conclusion to be drawn from the voluminous tax
literature dealing with tax incentives is that the costs of tax incentives are
typically large, the benefits are always uncertain, and only rarely do the
potential benefits justify the likely costs.
A developing country wishing to use tax incentives to attract foreign
investment is not stymied by a failure to obtain a tax-sparing article in its tax
treaties. Tax sparing obviously is not needed if the country of residence of the
potential investors uses the exemption method to avoid double taxation – for
investors resident in an exemption country, the source country tax is the only
tax. Thus, any reduction in source taxation automatically accrues to their
benefit. Even investors from credit countries may benefit from a source
country incentive with a little tax planning, as the example below illustrates.
BCo, the investor resident in Country B in the preceding example, wants
to obtain for itself the benefits of the tax holiday offered by Country A. To
that end, it organizes ACo, a wholly owned subsidiary, in Country A. ACo
engages in manufacturing activities that qualify for the tax holiday. ACo
earns 1,000 from its manufacturing activities in Country A, which is not
taxable by Country B because ACo is not resident in Country A and the
income is not earned in Country A. BCo would be taxable by Country B on
any dividends received from ACo, but ACo has no obligation to pay such
dividends. Indeed, Country A may benefit more from its tax holiday under
this arrangement than it would from tax sparing because the potential tax on
dividends paid to BCo will provide a strong incentive for ACo to reinvest its
profits in Country A. However, BCo may be reluctant to invest in Country A
if it cannot repatriate any profits generated by its investment without paying
tax.
The above example illustrates only one of several ways that tax
incentives may benefit residents of countries using the credit method in the
absence of tax sparing. Many US multinationals benefit from host country
investment incentives because of the way the US overall limitation on the
foreign tax credit operates. Under this overall limitation, US corporations that
pay high foreign taxes to one country can use what would otherwise be
excess foreign tax credits to offset the US tax otherwise imposed on foreign
business profits that are subject to low foreign taxes in another country.
Assume, for example, that PCo, a US corporation, has an excess foreign tax
credit of 35 from operations in Country A. PCo earns profits of 100 in
Country B that ordinarily would be subject to tax in Country B of 35, but that
tax is eliminated because of a tax holiday provided by Country B. PCo
benefits from that tax holiday because it can eliminate the US tax of 35 that
would otherwise be imposed on its profits in Country B with the excess credit
of 35 from Country A.
Another problem with tax sparing is the potential for abusive tax
avoidance. For example, generous tax sparing credits in a particular treaty
often encourage residents of third countries to establish conduit entities in the
country granting tax sparing. Tax sparing also puts pressure on the
enforcement of a country’s transfer pricing rules because taxpayers are
encouraged to shift profits to the country providing the tax incentives.
In 1998 the OECD published a report, Tax Sparing: A Reconsideration,
which suggests that the case for tax sparing is not persuasive. It recommends
that tax sparing be restricted to countries whose economic development is at
a considerably lower level than that of OECD member countries. It also sets
out some best practices for the design of tax sparing provisions to ensure that
the provisions are limited to genuine business investments and are not
susceptible to abuse. See paragraphs 72-78.1 of the Commentary on Article
23 of the OECD Model Treaty.
CHAPTER 5
Taxation of Nonresidents

5.1 INTRODUCTION
As noted in Chapter 2, most countries tax their residents on their worldwide
income and nonresidents on their domestic source income (i.e., income
earned or derived in a country’s territory). A few countries impose tax
exclusively on domestic source income (territorial taxation) irrespective of
whether the income is derived by a resident or a nonresident. Thus, it is fair
to say that all countries, other than pure tax havens, tax the income earned or
derived in their territory by nonresidents. For countries that tax on a
worldwide basis, it is necessary to have rules that distinguish between
residents and nonresidents because nonresidents are taxable only on their
domestic source income, not on their worldwide income. The rules for
determining whether a person is a resident of a country for income tax
purposes are discussed in Chapter 2, section 2.2.
As discussed in Chapters 2 and 3, the international consensus is that
countries are entitled to tax any income that arises or has its source in their
territory. The rules for determining the source of income are dealt with in
Chapter 2, section 2.3. A country’s right to tax domestic source income takes
priority over the right of another country to tax that income based on the
residence of the person deriving the income. For this reason, the residence
country has an obligation to relieve international double taxation in
recognition of the source country’s prior right to tax.
This chapter examines the major issues involved in taxing nonresidents
on their domestic source income. The chapter begins with a brief discussion
of the policy justification for taxing nonresidents and then deals with
practical issues such as threshold requirements, the taxation of business
profits and investment income of nonresidents, and the collection of tax from
nonresidents.
It is convenient for conceptual purposes to divide the taxation of
nonresidents into the following stages:
 – A country must determine what type of connection (nexus) a
nonresident must have to the country (activities in the country, the
ownership of property in the country, physical presence in the
country, etc.) in order for the country to be able to exercise its
jurisdiction to tax.
– Once a country has decided that it has jurisdiction to tax, it must
decide whether it should exercise that jurisdiction to tax only if the
nonresident meets some minimum threshold such as a permanent
establishment or fixed base.
– If the threshold is met or the country decides that a threshold is
unnecessary, a country must have rules to determine what amounts
derived by nonresidents are subject to tax; these rules are usually
referred to as source rules.
– Rules are necessary to compute the nonresident’s income and tax
payable.
– Finally, rules are necessary with respect to the collection of tax from
nonresidents.

These stages are intimately connected and often overlap. For example, if
a country decides to tax any interest or dividends paid by a resident to a
nonresident, the source of the income as represented by the residence of the
payer is the connection that gives the country the jurisdiction to tax;
accordingly, that country has rejected the necessity for any threshold
requirement. Similarly, transfer pricing rules can be viewed as source rules or
as computational rules. These stages are set out here to assist in the analysis
of the taxation of nonresidents; they do not attempt to describe the ways in
which countries actually tax nonresidents.
The distinction between business profits and investment income is
particularly important with respect to the taxation of nonresidents. Business
income is typically taxed on a net basis at the same rates applicable to
resident taxpayers, so that individuals earning business income in another
country are often subject to tax at progressive rates. In contrast, investment
income is typically taxed at a flat rate on the gross amount; moreover, the tax
is usually imposed by way of a withholding tax (i.e., there is an obligation on
the resident person paying the amount to the nonresident to withhold the
amount of the tax from the payment to the nonresident and to remit the tax to
the tax authorities).
5.2 TAX POLICY CONSIDERATIONS IN TAXING
NONRESIDENTS
It may be recalled from Chapter 3, section 3.2 that the tax policy
justifications for taxing residents on their worldwide income are equity and
neutrality. It is difficult to justify taxing nonresidents on the basis of equity
because the source country does not have complete information about the
nonresident’s tax situation; for example, income earned in the source country
may be offset by losses incurred in other countries. It is generally impossible
for a country to determine whether residents and nonresidents are similarly
situated for tax purposes except in situations where all or almost all of a
nonresident’s income is derived from one country.
In general, however, it is reasonable to say that, to the extent possible,
nonresidents should not be treated better or worse than residents in similar
situations. The taxation of nonresidents on their domestic source income can
be justified on the basis that nonresidents derive benefits from the source
country; for example, nonresidents doing business in a country take
advantage of the country’s infrastructure and its legal system in the same way
as residents. It can also be argued that, even if a nonresident simply sells
goods in a country, the nonresident is benefiting from the market provided by
that country and that benefit is sufficient to justify taxation.
The principle that nonresidents deriving income from a country should
not be treated less favorably than residents of that country – the
nondiscrimination principle – is an important principle that most countries
follow, at least in part. Although it may be tempting for a country to tax
nonresidents more harshly than residents – after all, nonresidents do not vote
– the likely response of other countries would be to do the same, thus putting
the first country’s residents at a disadvantage. In practice, there is
surprisingly little discrimination against nonresidents in the tax systems of
most countries. Many countries do, however, discriminate in favor of
nonresidents in certain circumstances by providing them with tax holidays
and other tax incentives in order to attract foreign investment. Discrimination
in favor of nonresidents is not considered to be offensive, although it is
widely criticized by tax policy commentators. The nondis-crimination
principle is recognized in Article 24 of both the OECD and the UN Model
Treaties. The nondiscrimination article in tax treaties is dealt with in
Chapter 8, section 8.8.1.
 From a revenue perspective, it makes obvious sense for countries to tax
nonresidents. However, the need for tax revenue must be balanced against the
need for foreign investment. If a country taxes nonresidents too harshly, the
effect may be to discourage nonresidents from investing in the country;
moreover, other countries can be expected to respond by taxing that country’s
residents equally harshly. Thus, countries that import and export capital and
tax on a worldwide basis have interests as both residence countries and
source countries that must be balanced. As residence countries, they want to
minimize tax imposed by source countries on the foreign source income of
their residents and to ensure that their residents are not discriminated against
relative to the residents of source countries. As source countries, they want to
attract foreign investment but also want to tax nonresidents as heavily as
possible. These competing interests cannot all be achieved fully because of
the inevitable retaliation by other countries that would result.
Another important consideration in the taxation of nonresidents is
enforcement. On the one hand, it obviously makes no sense for a country to
impose tax on nonresidents that cannot be enforced effectively. On the other
hand, it may not make sense for a country to tax all the income derived by
nonresidents that can be enforced effectively. Most countries do not follow
the practice of taxing nonresidents on everything that they can tax, probably
because, as noted above, they do not want other countries to do the same and
they want to attract foreign investment. Nevertheless, it is probably fair to say
that countries seriously consider taxing nonresidents to the maximum extent
possible unless there is some good reason not to.

5.3 THRESHOLD REQUIREMENTS

Although there is no restriction on the authority of a country to tax any and
all domestic source income derived by a nonresident, few countries do so –
most countries tax nonresidents on certain types of income only if a
minimum threshold is met. For example, many countries tax nonresidents on
their business income only if the income is attributable to a PE in the country.
This threshold for the taxation of business profits is also used in Article 7 of
the OECD and UN Model Treaties. Even countries that do not use the PE
concept in their domestic law usually tax nonresidents on their business
income only if their business activities exceed some threshold; for example,
in the United States nonresidents are taxable on their business income only if
they are engaged in a trade or business in the United States.
There are several reasons for the establishment of a threshold
requirement for the taxation of nonresidents. First, serious compliance and
enforcement problems arise when nonresidents are taxable on all domestic
source income. It is difficult for tax authorities to identify all nonresidents
earning income from the country and to get information about that income.
(Consider, for example, the difficulties in taxing a consultant who performs
services in a country for a few days.) Moreover, unless a nonresident has
some type of substantial and continuing presence in a country, it may be
difficult or impossible for the country to collect its tax. Second, as noted in
Chapter 2, section 2.3, few countries have detailed source rules; as a result, a
threshold requirement can provide more certainty for nonresidents as to when
they become subject to tax by a country. Third, requiring nonresidents to file
tax returns and pay tax on relatively small amounts of income is likely to
discourage cross-border trade and investment or result in nonresidents
ignoring their tax obligations.
Threshold requirements for taxing nonresidents are provided by
domestic law and tax treaties and differ depending on the type of income.
Some common thresholds are described below:

– Business profits: The threshold provided by tax treaties is the

existence of a PE in a country. In general, a PE is a fixed place of
business or a dependent agent with authority to contract on behalf of
the nonresident. Certain types of business profits, such as income
derived by entertainers and athletes, are usually subject to a lower
threshold. The UN Model Treaty uses a 183-day threshold for the
taxation of income from services derived by nonresidents and has a
special provision for insurance businesses.
– Income from immovable property: The immovable property must be
located in the country.
– Employment income: As a general rule, the threshold is the physical
presence of the employee in the country and the performance of the
duties of employment in the country, although under tax treaties the
source country is precluded from taxing a nonresident employee of a
nonresident employer without a PE in the source country, unless the
employee is physically present for more than 183 days.
– Investment income: Typically, there is no threshold for source
 country taxation of dividends, interest, and royalties under domestic
law or treaties. As discussed below, the source country tax is
imposed as a final withholding tax at a flat rate on the gross amount
of the payment.

In general, thresholds for the taxation of nonresidents take the form of a

fixed place (either a fixed place of business or immovable property) or the
physical presence of the nonresident in the country (sometimes for a specified
period). Thresholds based on the amount of revenue or income derived by a
nonresident are rare in both domestic law and treaties.

5.4 SOURCE RULES

Once it has been determined that a country has jurisdiction to tax a
nonresident and that any threshold for taxation has been met, it is necessary
to have rules to determine what amounts are subject to tax and how those
amounts are taxed. In general, countries tax nonresidents only on their
domestic source income. As a result, source-of-income rules are necessary to
determine whether a nonresident’s income is derived from sources inside the
territory of the country. Sometimes these source rules are explicit: for
example, a country’s tax law might provide that a nonresident is taxable on
domestic source income and then list items or amounts that are considered to
be from domestic sources. More often, however, countries simply prescribe
the amounts derived by nonresidents that are subject to tax, without explicit
reference to the source of those amounts. For example, a country might
impose tax on dividends paid by a resident corporation to a nonresident. The
source rule in this case is implicit – in effect, dividends are considered to
have their source in the country in which the company paying the dividends
is resident. Except in cases where the income is taxed on a gross basis, it is
also necessary to determine what expenses are deductible in determining the
domestic income subject to tax. Source rules are discussed in more detail in
Chapter 2, section 2.3.

5.5 DOUBLE TAXATION

In situations where a resident of one country earns income sourced in another
country, by international consensus the country in which the income is earned
has the first right to tax the income, and the residence country has a
corresponding obligation to relieve international double taxation by
exempting the income from tax or providing a credit for the source country
tax. Therefore, in taxing nonresidents, countries do not need to be concerned
about eliminating double taxation of this type. The only type of double
taxation that source countries should be concerned about is where two
countries both claim that the relevant item of income has its source in their
country. Accordingly, the more expansive a country’s source rules are, the
more likely it is that its source claims will overlap with other countries’
source claims.

5.6 EXCESSIVE TAXATION OF NONRESIDENTS

Although source countries do not need to be concerned about the elimination
of double taxation except in the case of overlapping source rules, they should
be concerned about the excessive taxation of nonresidents. As discussed
below, certain types of income are typically taxed by withholding at a flat
rate on the gross amount of the payment. In these situations, there is a risk
that the source country tax may be excessive relative to the net income
derived by the nonresident. For example, consider a situation in which a
nonresident incurs substantial expenses to earn royalties in a country. If the
source country taxes the royalties at a flat rate of 30 percent without any
recognition for the expenses, the nonresident may realize little, if any, after-
tax profit from the transaction. If the residence country exempts foreign
source royalties, it will not provide any relief for the source country tax; and
even if the residence country provides a foreign tax credit, the limitation on
the credit (see Chapter 4, section 4.3.3 for a discussion of the limitations on a
foreign tax credit) will likely result in the taxpayer getting only partial relief
for the source country tax. The overall result is that the royalties may be
taxable at an effective rate that is considerably higher than the residence
country tax rate. Sometimes nonresidents may be able to avoid excessive
source country taxation by requiring the resident payers to effectively absorb
the tax by grossing up the payments. Such excessive source country tax may
be borne by residents or may discourage foreign investment.

5.7 COMPUTATION OF THE DOMESTIC SOURCE

INCOME OF NONRESIDENTS
In general, the rules for computing the income of nonresidents are the same
as the rules applicable to residents. Thus, the rules that determine what
amounts are included in income, what deductions are allowable, and the
timing of income and deductions are applicable equally to residents and
nonresidents. For example, if a country allows a deduction for only a portion
of a taxpayer’s entertainment expenses, that rule will apply equally to
nonresidents. Although in general the rules for computing the income of
residents and nonresidents are the same, there are some exceptions. For
example, transfer pricing rules apply to transactions between a resident and a
related nonresident and not to transactions between related residents. Transfer
pricing rules are discussed in Chapter 6. Similarly, thin capitalization rules
are typically applicable only to interest paid by a resident corporation to
nonresidents, although in a few countries the rules also apply to interest paid
to tax-exempt residents. Conversely, controlled foreign corporation (CFC)
rules apply only to nonresident companies that are controlled by residents of
a country. Thin capitalization rules and CFC rules are dealt with in detail in
Chapter 7, sections 7.2 and 7.3 respectively. It should be noted in this regard
that the case law of the European Court of Justice has severely restricted the
ability of an EU member country to have rules, such as thin capitalization
rules or CFC rules, that apply differently to residents of that country and
residents of another EU member country.
As discussed below in section 5.8, the nondiscrimination article of an
applicable tax treaty requires the source country to allow nonresidents to
deduct expenses in computing the profits attributable to a PE on the same
basis as residents engaged in similar activities. However, where nonresidents
are subject to tax on a gross withholding tax basis, no deductions are allowed.
Therefore, the distinction between amounts such as business profits, which
are subject to net-based taxation, and amounts such as investment income,
which are subject to withholding tax, is very important. This distinction is
discussed in section 5.8.1 below.
With respect to nonresident individuals, personal deductions, reliefs,
allowances and credits are not customarily provided by source countries. For
example, many countries provide a basic personal or family exemption from
tax so that if an individual’s or family’s income does not exceed the
minimum amount, no tax is payable. Similarly, many countries provide
deductions or allowances for family members who are dependent on the
taxpayer for support. These and other similar personal allowances are not
generally provided by countries to nonresidents, and the typical
nondiscrimination article in tax treaties does not require such allowances to
be extended to nonresidents.
As mentioned above, there are no legal constraints to prevent a source
country from treating nonresidents more favorably than residents. In
particular, many developing countries provide nonresident investors with
special tax incentives that are not available to residents.
If a country imposes a final withholding tax on the gross amount of
certain payments, such as dividends, interest, and royalties, no computational
rules are necessary since the gross amount is taxable. If, however, the
withholding tax is imposed on an interim basis on account of a nonresident’s
final tax liability, rules for the computation of net income are necessary.
Some South American countries impose final withholding taxes on a wide
range of payments made to nonresidents. In many cases, the tax is imposed at
a fixed rate on a fixed percentage of the payment rather than on the gross
amount. Taxing a presumptive amount in this way represents an attempt to
give relief for the expenses incurred to earn certain types of income without
the necessity for either the taxpayer or the tax authorities to calculate a
particular nonresident’s actual income.

5.8 TAXATION OF VARIOUS TYPES OF INCOME OF

NONRESIDENTS

5.8.1 Business Income

5.8.1.1 In General
Because business income earned by nonresidents is usually taxed on a net
basis and investment income is taxed on a gross basis, it is important to
distinguish between the two types of income. In some civil law countries, all
the income earned by a legal entity is characterized as business income, and
therefore it is necessary to distinguish between business and other income
only with respect to individuals. In other countries, however, both legal
entities and individuals can earn various types of income. Some countries tax
on a schedular basis, which means that they tax different types of income in
accordance with different rules, and sometimes even at different rates. Even
countries that tax on a global basis often have different rules for business
income and other income. Typically, the characterization of an amount as
income from business or other income arises with respect to capital gains
from the disposal of property, interest, rent, and royalties.
How is the distinction between business and other income made? In
many Commonwealth countries, there is a substantial body of case law
concerning the distinction between capital gains and ordinary business
income. This case law is usually equally applicable to nonresidents.
Countries may also have statutory rules that distinguish between the two
types of income. For example, some countries have rules that limit capital
gains treatment to property that is held or owned for a minimum period; other
gains are treated as ordinary business income. More generally, some
countries may have rules that define business income.
The distinction between business and other types of income is also
important for purposes of tax treaties because tax treaties deal with various
types of income on a schedular basis. As a result, for example, business
income, interest and capital gains are subject to different rules. The OECD
and UN Model Treaties, on which most bilateral tax treaties are based, do not
provide a comprehensive definition of “business”. Since 2000, the OECD
Model Treaty has defined business to include the performance of independent
and professional services (Article 3(1)(f)) because Article 14 dealing with
such activities was deleted at that time. Because of the absence of a complete
definition in the treaty, it is necessary to refer to the meaning of the term
“business” under the domestic law of the country applying the treaty.
Under tax treaties based on the OECD and UN Model Treaties, it is also
necessary to distinguish between various types of business income. In
general, under Article 7, business income derived by a resident of one
country in the other country is taxable by the other country only if the
business is carried on through a PE and the income is attributable to the PE.
Under Article 8, however, income from international shipping and air
transportation income is taxable only by the country in which the enterprise
has its place of effective management. In contrast, under Article 17, business
income from personal services performed by a resident of one country in the
other country as an entertainer or athlete is taxable by that other country
without the need for a PE in the country. In effect, under tax treaties, a PE is a
threshold requirement for most business income; the rule that only profits
attributable to the PE are taxable is the functional equivalent of a source rule.
In the interests of accuracy, it must be noted that under Article 7, the profits
attributable to a PE can include profits from outside the country in which the
PE is located. These treaty rules are discussed further in Chapter 8, section
8.8.5.
Business income derived by a nonresident is usually taxed by the
country in which the income is earned on a net basis, and the rules for the
computation of business income are generally the same as the rules for
residents. The nondiscrimina-tion article of tax treaties (Article 24(3) of the
OECD and UN Model Treaties) requires the source country to tax a PE of a
resident of the other state no less favorably than a resident of the source
country carrying on the same activities. The nondiscrimination article is
discussed in more detail in Chapter 8, section 8.8.1. Once a nonresident has
met the minimum threshold requirement for taxation in the source country
(usually the existence of a PE in the source country), domestic tax rules will
apply in order to determine the amount of income from the business that is
subject to tax. In some countries, once a nonresident has a PE in the source
country all the nonresident’s income from the source country becomes
taxable. However, most countries do not follow this force-of-attraction
principle. Instead, only the income from the business carried on in the source
country is taxable (although other amounts, such as investment income, may
be taxable on a different basis). Under the OECD Model Treaty, there is no
force of attraction; only income that is attributable to the PE is taxable by the
source country. Under the UN Model Treaty, there is a limited force-of-
attraction rule: the source country is authorized to tax any business profits
from the source country of the same or similar kind as those derived through
the PE. The treaty rules for the attribution of profits to a PE are discussed in
Chapter 8, section 8.5.

5.8.1.2 Branch Taxes

As noted several times in this Primer, taxpayers generally have the choice of
doing business in a country in the form of a branch or a separate legal entity,
usually a company. For corporate taxpayers, this choice is usually described
as a choice between a branch and a subsidiary. If a corporation resident in
one country forms a subsidiary corporation in another country, the subsidiary
will likely be treated as a separate legal and taxable entity, and therefore as a
resident of the other country (assuming that the subsidiary’s place of
management is located in that country). If a corporation establishes a branch
in another country, the branch is simply a part of the corporation. As a
resident of the source country, a subsidiary of a nonresident corporation is
ordinarily taxable on its worldwide income, whereas in respect of a branch
only the domestic source income attributable to the branch is usually taxable
by the source country. If, however, the subsidiary earns exclusively domestic
source income (i.e., all of its income is earned in the country in which it is
resident), there is no difference between the subsidiary and a branch in this
regard.
There is a significant difference with respect to the tax consequences of
the repatriation of funds from a branch or subsidiary. If a subsidiary pays a
dividend to its nonresident parent, the country in which the subsidiary is
resident may impose a withholding tax on the gross amount of the dividend.
In contrast, if funds are withdrawn from a branch and repatriated to the head
office of the nonresident corporation, there is no dividend or other payment
on which the source country can levy tax. Therefore, nonresidents may prefer
to do business in a country through a branch rather than a subsidiary in order
to avoid withholding taxes on dividends and other intercorporate payments.
Some countries (e.g., Canada and the United States), have adopted special
branch taxes in order to equalize the treatment of branches and subsidiaries.
These branch taxes can be quite complicated because of the need to impose a
tax that is equivalent to a withholding tax on dividends on the basis of some
type of proxy for dividends. Other countries impose a slightly higher rate of
tax on nonresidents carrying on business in the form of a branch in order to
make up for the lack of withholding tax on the repatriation of funds from
branches. Both of these measures appear to violate the nondiscrimination
article of a typical tax treaty, although they are arguably justifiable on tax
policy grounds.
If borrowed funds are used to finance the activities of a branch, the
interest expense incurred with respect to the funds is ordinarily deductible in
computing the profits of the branch under both domestic law and tax treaties.
Such interest expense erodes the tax base of the country in which the branch
is located; however, the interest is not usually subject to withholding tax
because the interest is paid by a nonresident. A few countries attempt to
subject such interest to withholding tax.

5.8.2 Income from Immovable Property

The ownership or use by a nonresident of immovable property situated in a
country is clearly sufficient to justify jurisdiction to tax by that country. In
addition, the existence of the immovable property operates as a threshold
requirement and as a source rule. The country in which immovable property
is situated is entitled to tax the nonresident owner on any income derived
from the property or any capital gains derived from the disposal of the
property. Articles 6 and 13(1) of the OECD and Model Treaties confirm the
source country’s right to tax income and gains from immovable property on
this basis.
Income from immovable property is treated differently from income
from business under the OECD and UN Model Treaties. Although the
location of immovable property in a country may be seen as the equivalent of
a PE in terms of nexus, under Article 6 there is no requirement for income
from immovable property to be taxed on a net basis, as there is for business
profits under Article 7. In any cases of conflict between Articles 6 and 7,
Article 6 clearly prevails (see Article 6(4) and Article 7(4) of the OECD
Model Treaty and Article 6(4) and Article 7(6) of the UN Model Treaty). As
a result, there is nothing in a typical tax treaty to prevent a country from
taxing income from immovable property on a gross basis, and, in fact, some
countries tax some types of income from immovable property on a
presumptive basis. It is important, therefore, to understand how a country
defines and taxes income from immovable property under its domestic law.

5.8.3 Income from Employment

A country has jurisdiction to tax nonresident employees if the employment
activities are performed in the country or the income is derived from the
country, or even if the benefits from the employment activities are used or
consumed in the country. In most countries, there is no minimum threshold
for the taxation of nonresident employees, although the United States
provides an exemption for nonresident employees who are paid by a
nonresident employer, earn not more than USD 3,000, and are present in the
United States for not more than ninety days. If a nonresident is employed by
a resident employer, any tax on the employee is relatively easy to enforce by
requiring the employer to withhold the tax. Even if the employer is a
nonresident, the tax can be effectively enforced if the employer has a PE in
the source country. In other situations, any tax imposed on a nonresident
employee who spends a few days in a country performing services may be
difficult to enforce.
Under the OECD and UN Model Treaties, income from employment
derived by a nonresident employee is taxable by the source country only if
the employee is present in the country to perform the employment services.
An exemption is provided for nonresident employees if they are present in
the source country for less than 183 days, they are not paid by a resident
employer, and their remuneration is not deductible for purposes of computing
the income of a PE that the nonresident employer has in the source country.

5.8.4 Investment Income: Dividends, Interest, and Royalties

Most countries tax certain investment income derived by nonresidents. For
this purpose, it is necessary to define the types of investment income that are
taxable and, in particular, to distinguish between business income and
investment income, as discussed in section 5.8.1.1. This distinction cannot be
made solely on the basis of the nature of the income because, for example,
interest earned by a financial institution from lending money is clearly
income from business, whereas interest earned by an individual investor is
investment income.
The distinction is important because, typically, investment income is
taxable by source countries through a withholding tax at a flat rate on the
gross amount paid to nonresidents, while business income is usually taxable
on a net basis by way of an assessment. It may be questioned whether a gross
basis withholding tax is appropriate as part of an income tax; however, in
practice, the difficulty of enforcing tax imposed on the investment income of
nonresidents makes withholding taxes generally acceptable if the rate is
limited so that the withholding tax approximates the tax that would be
imposed at ordinary rates on net income. This explains why withholding tax
is generally limited to amounts in respect of which the nonresident is unlikely
to have incurred substantial expenses to earn those amounts.
The imposition of tax on the gross amount of interest, rent, or royalties
can be excessive in certain circumstances, as explained in section 5.6 above.
Investment income derived by nonresidents is typically taxed without
any threshold requirement under either domestic law or tax treaties; in
general, it is considered to have its source in the country in which the payer is
resident. The same source rule is used in Article 10 through 12 of the OECD
and UN Model Treaties. Several countries, however, have special source
rules for certain types of investment income. For example, interest and
royalties may be considered to be earned where the funds or property are
used.

5.8.5 Capital Gains

The taxation of capital gains realized by nonresidents presents special
problems for countries that tax such gains differently from business income.
Many countries tax capital gains derived by nonresidents from the disposal of
immovable property situated in their countries, property of a business carried
on (often through a PE) in their countries, and substantial participations in
resident companies, partnerships, and other legal entities. Most countries do
not tax nonresidents on capital gains from the disposal of shares of resident
companies other than land-rich companies (companies whose assets consist
primarily of immovable property located in the country) and substantial
participations. This pattern is reflected in Article 13 of the OECD and UN
Model Treaties, except that, under the OECD Model Treaty, source countries
are not allowed to tax capital gains from substantial participations.
The enforcement of tax on the capital gains of nonresidents raises
special difficulties. If a nonresident sells immovable property situated in a
country, that country’s tax on the capital gain can be enforced by requiring
the purchaser to withhold an amount on account of the seller’s tax from the
purchase price, unless the nonresident prepays the tax or provides security for
the payment of the tax, such as a bank guarantee. If the prepayment is
excessive, the nonresident is usually entitled to file a return to claim a refund
of the excess. Even if the nonresident sells the property to another
nonresident, the tax can usually be enforced in this way by refusing to allow
the purchaser to register the property unless the tax has been paid.
In the case of the sale of the property of a business carried on in a
country by a nonresident, the tax on any capital gains from the disposal of the
business assets can be enforced in the same way as tax on the income from
the business, although the tax may be difficult to collect where the
nonresident sells all the assets of the business.
The taxation of capital gains in respect of shares of resident companies
holding immovable property is necessary to prevent the easy avoidance of the
tax on gains from the disposal of immovable property by holding the property
in a resident company and then selling the shares of the company rather than
the immovable property itself. The rule in Article 13(4) of the OECD and UN
Model Treaties provides that a country is entitled to tax gains from the sale of
shares of a resident company if more than 50 percent of the value of the
shares of the company is attributable, directly or indirectly, to immovable
property situated in the country. For capital gains from the sale of shares of
land-rich resident companies, or substantial participations in resident
companies, the only effective method of enforcing the tax is to place an
obligation on the purchaser to withhold the tax from the purchase price. This
method of enforcement is not as effective for shares as for the disposal of
immovable property because in the case of immovable property, the tax can
be registered as a lien against the property, whereas in the case of shares, the
immovable property is owned by the company.

5.9 ADMINISTRATIVE ASPECTS OF TAXING

NONRESIDENTS

5.9.1 Introduction
The difficulties of tax administration are exacerbated with respect to the
taxation of cross-border transactions and investment involving both residents
and nonresidents. Collecting tax from nonresidents earning domestic source
income is different from collecting tax from residents because, unlike
nonresidents, residents are usually present in a country (or have substantial
connections with the country) and are subject to its legal system. It makes
little sense for a country to impose a tax on nonresidents that it cannot collect.
In this section, two major problems are examined: obtaining the necessary
information and collecting the tax.

5.9.2 Obtaining Information about Domestic Source Income of

Nonresidents
The basic question here is: what information do source countries need to
collect tax effectively from nonresidents? First, they need basic information,
such as name, address, and taxpayer identification number if available, to
identify nonresidents earning domestic source income. Second, they need
information to determine whether nonresidents are carrying on business in
their countries, or have PEs there, or are earning investment income there.
Third, information is necessary to determine or verify the computation of a
nonresident’s domestic source income (revenue and expenses). Fourth,
information is necessary concerning transactions with related persons,
especially with related persons who are residents of the source country. Fifth,
if the nonresident is claiming a reduction of or exemption from source
country tax under an applicable tax treaty, the source country should have
sufficient information to verify whether the benefits of the treaty should be
granted.
In some cases, the necessary information can be obtained by imposing
reporting requirements on the nonresidents themselves. Ideally, however, the
tax authorities should have independent information to verify information
provided by nonresidents.
In other cases, information can be obtained from residents who have
relationships or transactions with nonresidents. For example, residents paying
dividends, interest, royalties or other amounts to nonresidents can be required
to report basic information about the nonresident recipients and the amount
and nature of the payments. Imposing these types of reporting requirements
on third persons may result in their incurring significant compliance costs. As
a result, in adopting such reporting requirements, a source country must
carefully balance the need for information against the compliance costs
imposed on third parties. A country should not ask for information that it
cannot use effectively.
Information should be provided in electronic format if the tax authorities
have the necessary technology to use information in this format. If the tax
authorities have information in electronic format and have taxpayer
identification numbers, they will be able to match information from various
sources. The information should be provided in a consistent format from year
to year. It can be filed with the nonresident’s tax return (assuming that a
return is required) or filed separately, or retained by the taxpayer for possible
inspection by the tax authorities.
In many situations, the necessary information is located outside the
source country. If the information is in the possession of the taxpayer, a
penalty can be imposed on the taxpayer for failing to produce the information
on a timely basis. Some countries have adopted special rules to preclude a
taxpayer from introducing in any subsequent legal proceedings foreign-based
information that is not disclosed to the tax authorities when requested; such a
rule is ineffective if the taxpayer discloses all of the information favorable to
the taxpayer’s case. If the information is in the possession of an unrelated
third party, the taxpayer should not be penalized for not producing the
information. If the information is held by a related party, however, it may be
appropriate to impose penalties in certain circumstances.
If there is a treaty in place between the source country and the country in
which the nonresident is resident, foreign-based information may be obtained
through the exchange-of-information provision in the treaty. Bilateral tax
treaties based on the OECD or UN Model Treaties contain an exchange-of-
information article (Article 26), which authorizes the tax authorities to
exchange many types of information in response to a specific request from
the other country, and to do so automatically. In addition, several countries
have recently entered into Tax Information Exchange Agreements (TIEAs)
with countries (such as tax havens) with which they do not have
comprehensive tax treaties. Comprehensive tax treaties with these countries
are not necessary, but exchange of information can be useful for both
residence and source countries. Exchange of information is also covered by
multilateral agreements among the Nordic countries, the European Union,
and a joint Convention on Mutual Administrative Assistance in Tax Matters
of the Council of Europe and the OECD, which entered into force in 1995;
after a slow start, over ninety countries have now been signed this
Convention. Exchange of information under tax treaties is discussed in more
detail in detail in Chapter 8, section 8.8.4.

5.9.3 Collection of Tax from Nonresidents

There are two basic ways of determining the tax payable by nonresidents:
assessment and withholding. Assessment is typically used in situations in
which nonresidents are taxed on a net basis, whereas withholding is typically
used for passive investment income. Assessment involves the determination
of a nonresident’s income subject to tax and tax payable, usually by way of
the filing of a tax return. If the nonresident does not pay any tax owing, the
source country can take action to enforce the tax in accordance with its
domestic law.
Withholding operates by the imposition of an obligation on persons
(usually residents) paying certain amounts to nonresidents to withhold tax at
a specified percentage from those payments and remit the tax to the tax
authorities on behalf of the nonresident. There are two types of withholding.
Provisional or interim withholding is purely a collection device. The amounts
withheld are remitted to the tax authorities on account of the nonresident
taxpayer; they are treated, in effect, like installments of tax paid by the
nonresident. The nonresident is under an obligation to file a return and either
pay any tax owing in excess of the amount withheld or receive a refund of the
amount withheld in excess of the tax payable. If the nonresident does not file
a return, the tax authorities have access to the amounts withheld to satisfy the
nonresident’s tax liability.
Provisional withholding by employers is often used to collect tax from
salary and wages paid to employees. Final withholding is a tax imposed on
the gross amount (or a percentage of that amount) of the payment to a
nonresident. It is final because the nonresident is not entitled to file a return
on the basis of its actual net income. Although a final withholding tax is not,
in form, an income tax, if the rate of withholding tax is set appropriately, it is
recognized as an internationally accepted proxy for an income tax because of
the difficulties in collecting tax from nonresidents.
Obviously, if a nonresident has assets in the source country or is
physically present in the source country, collection action can be taken by the
source country directly against the nonresident. In these circumstances, which
may include nonresidents carrying on business in a country, source countries
typically impose tax by means of an assessment levied on the nonresident. In
countries with self-assessment systems, nonresidents are expected to file tax
returns in which they report their revenue and expenses and determine their
tax payable. Source countries can audit nonresidents’ tax returns and enforce
tax payable by nonresidents in largely the same way as with residents. Often
such nonresidents may be required to pay periodic installments of tax
throughout a taxation year; they may also be subject to interim or provisional
withholding on certain amounts paid to them and by them.
If, however, the nonresident is not present and does not have significant
assets in the source country, the source country must take special measures to
collect its tax. First, the source country might consider obtaining a court
judgment for the unpaid tax against the nonresident from the country’s courts
and then seeking enforcement of that judgment by the courts in the
nonresident’s country of residence. The problem with this course of action is
that many countries will not enforce other countries’ criminal and tax
judgments (this is widely known as the revenue rule). Second, the source
country may consider requesting assistance in the collection of the unpaid tax
from the country of residence pursuant to the tax treaty between the two
countries, if that treaty has an article dealing with Assistance in Collection
based on Article 27 of the OECD and UN Model Treaties. However, Article
27 was added to the OECD Model only in 2002 and to the UN Model in
2011, and has been included in relatively few treaties to date. Third, if the
source and residence countries are parties to a multilateral convention dealing
with administrative assistance in tax matters, such as the Convention on
Mutual Administrative Assistance in Tax Matters, referred to above in
section 5.9.2 in connection with exchange of information, the source country
can request the residence country to collect the source country’s tax as if it
were tax owing to the residence country.
Many countries have concluded that withholding is the most effective
method of collecting tax from nonresidents that do not have a significant
presence in the country. The withholding is usually provisional with respect
to amounts such as employment income, other income from services, and
sometimes rents and royalties, because the net income may be significantly
less than the gross amount. Nonresidents have the right to file a return and
obtain a refund of any excess tax withheld. Often, however, as a practical
matter, provisional withholding operates as a final withholding tax because
the nonresident may choose not to file a return unless the amount withheld is
substantially in excess of the amount of tax payable. Also, unfortunately,
some countries may make it difficult, either deliberately or inadvertently
through inefficient tax administration, for nonresidents to obtain refunds.
A final withholding tax on the gross amount of certain payments is a
convenient and effective method of collecting tax from nonresidents,
especially for developing countries that lack sufficient administrative
resources. However, unless the rate of tax is quite low, a final withholding
tax is inappropriate for amounts in respect of which a nonresident is likely to
have incurred significant expenses. This may be the case where amounts such
as dividends, interest, or royalties constitute profits of a business carried on
by a nonresident.
Another method of providing relief from excessive withholding taxes is
allowing nonresidents to elect to pay tax on a net basis; if the election is
made, the nonresident must file a return and pay tax on the net income.
Several countries provide this type of election with respect to income from
immovable property. A nonresident deriving rent from immovable property
located in a country will often be subject to withholding tax on the gross
amount of the rent. Although such a withholding tax on rental income from
immovable property is in accordance with Article 6 of the OECD and UN
Model Treaties, the nonresident may have incurred significant expenses with
respect to the immovable property, such as mortgage interest, property taxes,
and maintenance. As a result, a gross basis withholding tax, even at a
relatively low rate, may well be excessive and the ability for the nonresident
to make an election to pay tax on a net basis can provide relief.
An example of the tension between the effectiveness and the
appropriateness of final withholding is the taxation of consulting, technical,
and management fees by developing countries. Under the OECD and UN
Model Treaties, such fees are business profits that are taxable by the source
country only if the nonresident spends more than 183 days in the source
country or has a PE or fixed base in the source country and the fees are
attributable to the PE or fixed base. In most cases, taxpayers can arrange their
affairs so that they can earn substantial fees without having a PE or fixed
base in the source country (or without spending more than 183 days in the
source country under Article 14(1)(b) of the UN Model Treaty).
This result is often unacceptable to developing countries, and some of
them have taken the position that technical and management fees are royalties
subject to withholding tax. Several countries, such as India, Jamaica, Kenya,
Mongolia, Tanzania, and Vietnam, have included special articles in their tax
treaties allowing them to tax consulting, technical, and management fees on a
gross basis, but at a limited rate. The UN Committee of Experts is currently
working on a new article dealing with consulting, technical, and management
fees to be included in the UN Model Treaty. This new article would permit
source countries to tax such fees through a withholding tax on the gross
amount, but at a limited rate to be agreed on by the parties to the treaty. Such
fees would be taxable by a country if the payer of the fees is a resident of the
country or a nonresident with a PE or fixed base in the country. The
nonresident would not be required to meet any threshold (such as a PE or
fixed base) and the source country would be entitled to tax even where the
services are rendered outside that country. The proposed article on
consulting, technical, and management services is dealt with more
extensively in Chapter 9, section 9.3.
CHAPTER 6
Transfer Pricing

6.1 INTRODUCTION
A transfer price is the price established in a transaction between related
persons. For example, if ACo manufactures goods in Country A and sells
them to its foreign affiliate, BCo, resident in Country B, the price at which
that sale takes place is called a transfer price. A transfer price may be
different from the market price, which is the price set in the marketplace for
transfers of goods and services between unrelated persons. In a sale of goods
or services between unrelated parties, the conflicting interests of the parties
usually ensure that the price charged for the goods or services is neither
artificially high nor low. However, related parties do not have conflicting
interests, and therefore the prices charged in transactions between related
parties may be significantly different from market prices.
Multinational companies use transfer prices for sales and other transfers
of goods and services within their corporate group. These intercompany
prices are the most important category of transfer prices. Transfer prices may
also used by individuals dealing with corporations or other entities under
their control and by individuals dealing with close family members.
Unless prevented from doing so, related persons engaged in cross-border
transactions can avoid tax through their manipulation of transfer prices. For
example, in the example above, ACo might avoid paying income taxes in
Country A by setting a price on the sale of its manufactured goods to BCo
that results in its earning little or no profit. If the effective tax rate in Country
B is lower than the effective tax rate in Country A, then the total tax burden
on ACo and BCo would be reduced through the use of inappropriate transfer
prices. If Country B is a low-tax country, then ACo and BCo would pay little
or no tax on their combined profits.
In a well-designed income tax system, the tax authorities should have
the power to adjust the transfer prices set by related persons if those prices
differ from the market prices. This authority should include the power to
allocate revenue, deductions, credits, and other allowances among related
persons so that the country is able to prevent the erosion of its domestic tax
base and collect its fair share of tax revenue.
In general, related persons include persons that are owned or controlled,
directly or indirectly, by the same interests. A good indicator of such a
relationship is the ability to set transfer prices that differ from market prices.
As suggested above, the tax authorities should be given the power to
adjust transfer prices to prevent taxpayers from shifting income to related
persons resident in tax havens or in countries where they enjoy some
preferential tax treatment. Examples of preferential tax treatment include a
relatively low tax rate, a tax holiday or other tax incentive, and a tax-
deductible loss. Although taxpayers generally do not seek to deflect income
to a country that has high statutory tax rates, they may do so when a member
of their affiliated group has losses in that country or if they are able to exploit
some loophole or preferential tax regime in the high-tax country’s tax system.
The tax authorities of a country also need the power to adjust transfer
prices in order to prevent other countries from obtaining an unfair share of
the tax revenue on income derived from cross-border transactions through
overly aggressive enforcement of their transfer pricing rules. When one
country is aggressive in making transfer price adjustments and another
country is not, taxpayers engaged in transactions in both countries may divert
income to the aggressive country in order to mitigate their risks of double
taxation.
Double taxation is a serious possibility when multiple countries apply
their transfer pricing rules. For example, assume that ACo manufactures
goods in Country A at a cost of 60 and sells them to a related company, BCo,
that is resident in Country B. BCo then sells the goods to retail customers in
Country B for 150. ACo is taxable in Country A on its manufacturing profits
and BCo is taxable in Country B on its sales profits. The corporate group
(consisting of ACo and BCo) has a net profit of 90 (150 – 60). Assume that
Country A concludes that the proper transfer price on the sales from ACo to
BCo is 90, whereas Country B takes the position that the proper price for the
sales is 50. In that event, double taxation will result because the combined
group will have income of 90 but will be taxable on income of 130, as
follows:

Income of ACo taxable in Country A:

Sales 90
Cost of goods 60
Income 30
Income of BCo taxable in Country B:
Sales 150
Cost of goods 50
Income 100
Total income of ACo and BCo 130

The double taxation illustrated in the above example would be

eliminated if both countries had uniform rules for adjusting inappropriate
transfer prices and applied those rules in the same way in all cases. Thus, on
the facts of the example, double taxation would be eliminated either if
Country B accepted Country A’s transfer price of 90 (because Country B
would then tax BCo on income of only 60) or if Country A accepted Country
B’s transfer price of 50 (because then Country A would treat ACo as having a
loss of 10).
In an attempt to achieve some degree of uniformity, Article 9 of both the
OECD and UN Model Treaties provides that transfer prices should be
adjusted to reflect the prices that would have been used in the same
transaction between unrelated enterprises acting independently. This so-
called “arm’s-length method (standard, principle, or approach)” has been
adopted by most countries of the world. The wide acceptance of the arm’s-
length method, however, masks substantial disagreements over the way the
method is applied in practice. The main transfer pricing methods employed
for implementing the arm’s-length standard are described in section 6.4
below.
Some countries try to reach agreement with taxpayers on the
methodologies to be used by them in setting their transfer prices before a
transfer pricing dispute actually arises. A major objective of this type of
approach is to reduce the high costs that taxpayers and the tax authorities
typically incur in litigating disputes over transfer prices. A taxpayer that
wants advance approval of its pricing methodology with respect to one or
more transactions typically submits a request to the tax authorities for what is
generally known as an “advance pricing agreement” or APA. The taxpayer
must give details about the transfer pricing methodology that it intends to
apply to the transactions covered by the APA and must explain why that
methodology would produce an appropriate result. In some instances, two or
more governments may use the dispute-resolution mechanism in their tax
treaties to agree jointly on the pricing methodology to be used by a taxpayer.
The OECD has issued guidelines for countries in developing joint APAs.

6.2 THE OECD TRANSFER PRICING GUIDELINES

Beginning in the 1960s, the United States (US) took the lead in developing
techniques for limiting transfer pricing abuses. The definition of the arm’s-
length standard under section 482 of the US Internal Revenue Code was
initially controversial, but is now widely accepted. The regulations under
section 482 promoted three methods for determining the arm’s-length price:
the comparative uncontrolled price (CUP) method, the resale price method,
and the cost plus method. In 1995, the US adopted new transfer pricing
regulations that endorsed some additional methods, to be applied primarily
when products embodying intangible property are sold or licensed. Although
initially quite controversial, those methods have now been accepted by many
other governments and, with some qualifications, have been endorsed by the
OECD.
The OECD has been working steadily for many years to achieve an
international consensus on transfer pricing rules. In 1979, the OECD
published a report, Transfer Pricing and Multinational Enterprises, which
advocated the adoption of the arm’s-length standard to determine the prices
of transactions between associated enterprises. The 1979 report was
supplemented by a 1984 report, Transfer Pricing and Multinational
Enterprises: Three Taxation Issues, which dealt with the mutual agreement
procedure, banking, and the allocation of central management and service
costs. In 1992, the OECD established a task force to review transfer pricing
developments in the US; another task force was established in 1993 to revise
the 1979 and 1984 reports on transfer pricing. These efforts culminated in a
comprehensive and fundamental review of transfer pricing issues and the
publication, in 1995, of the OECD’s Transfer Pricing Guidelines for
Multinational Enterprises and Tax Administrations (OECD Guidelines).
The OECD Guidelines were published in loose-leaf format to
accommodate subsequent revisions. Revisions to the Guidelines have been
made on several occasions since 1995 (most recently in 2010), and more
revisions are proposed as part of the OECD’s BEPS project. For example, the
chapters dealing with intangibles, cost-contribution arrangements, and
documentation requirements are likely to be revised.

6.3 UNITED NATIONS, PRACTICAL MANUAL ON

TRANSFER PRICING FOR DEVELOPING COUNTRIES
In 2013, the United Nations Committee of Experts published a Practical
Manual on Transfer Pricing for Developing Countries (New York: United
Nations, 2013). This Manual is intended to provide practical assistance to the
tax authorities of developing countries in applying the arm’s-length standard
while recognizing the particular needs and administrative capacities of those
countries. It is generally consistent with the OECD Transfer Pricing
Guidelines. In addition to material on transfer pricing methods,
documentation requirements, comparability analysis, and similar material that
is also dealt with in the OECD Transfer Pricing Guidelines, the Manual deals
with building capacity to handle transfer pricing issues in developing
countries, audits and risk assessment techniques, and dispute resolution
procedures, as well as a description of transfer pricing practices in Brazil,
China, India, and South Africa.

6.4 THE ARM’S-LENGTH STANDARD

6.4.1 Introduction
According to international custom, an appropriate transfer price is one that
meets the so-called arm’s-length standard. This standard is met if a taxpayer
sets its transfer prices in its dealings with related persons so that those prices
are the same as the prices used in comparable dealings with unrelated
persons.
The above definition of the arm’s-length standard provides little
guidance as to how transfer prices should be established in concrete
situations. Some of the basic rules that countries have adopted to give content
to the arm’s-length standard are summarized below. Section 6.4.2 describes
the identification of comparable arm’s-length transactions or enterprises that
are used to determine the appropriate arm’s-length price for transactions
between associated enterprises. Section 6.5 describes rules applicable to
setting transfer prices when a group of related corporations shares common
resources. Section 6.6 describes the rules that apply to cost-contribution
arrangements, under which related persons share the profits from the
exploitation of intangible property that they have developed jointly.
The OECD Guidelines on transfer pricing strongly endorse the arm’s-
length standard. At the same time, they acknowledge frankly that the
application of that standard sometimes presents serious difficulties for
taxpayers and tax administrations. The Guidelines provide a valuable
discussion of the factors to be considered in determining whether transactions
between unrelated persons are comparable to the transactions actually entered
into by members of a corporate group. Like most of the literature on the
arm’s-length approach, however, the OECD Guidelines are better at
highlighting the problems of establishing the comparability of controlled and
uncontrolled transactions than they are at giving practical advice to tax
administrators on how to cope with these problems.
Many methods are used throughout the tax world for determining the
arm’s-length price for sales of tangible personal property. Five methods are
discussed below. The first three methods – the comparable uncontrolled price
(CUP) method, the resale price method, and the cost plus method – are
sometimes referred to as the traditional methods and are widely accepted by
the international tax community. Unfortunately, these methods are extremely
difficult, if not impossible, to apply in many important cases, especially cases
in which the products being sold incorporate valuable intangible property.
The traditional methods are described in section 6.4.3 through 6.4.5 and
compared in section 6.4.6.
The two other arm’s-length methods can be applied in more situations.
The profit-split method is frequently used on an informal basis by tax
authorities in settling disputes with taxpayers through internal appeal
procedures. The transactional net margin method (TNMM), also known as
the comparable profit method (CPM), was formally approved by the US in
revisions to the section 482 regulations, finalized in 1994. The OECD, in its
1995 Report on transfer pricing, suggests that the profit-split and TTNM
methods should be used only as a last resort. They methods are described in
sections 6.4.7 and 6.4.8.

6.4.2 Comparability Analysis

The essence of transfer pricing is determining the appropriate price for a non-
arm’s-length transaction based on a comparable arm’s-length transaction. The
arm’s-length transaction used for comparison may be a transaction between
the taxpayer (the person whose non-arm’s-length transaction is being priced)
and an arm’s-length person (internal comparable) or between unrelated
arm’s-length persons (external comparables). For example, an enterprise may
sell the same goods to both related and unrelated persons; therefore, the
prices charged for the sales to unrelated persons may provide good
comparables for determining the arm’s-length price for the related-party
sales. However, if the enterprise does not sell the same goods to unrelated
parties, it would be necessary to search for an independent enterprise
operating in the same market or industry that sells the same goods to arm’s-
length parties.
In most cases, it is impossible to find perfect comparables because
arm’s-length prices reflect many considerations (e.g., the quantity sold, the
quality of the goods, terms of sale, the conditions of the market, the location
of the market, etc). However, it is not necessary to identify precise
comparables as long as it is possible to make adjustments to a transaction so
that, with the adjustments, it becomes comparable.
All the transfer pricing methods discussed below require some type of
comparability analysis. In the case of the traditional methods – CUP method,
resale price method, and cost plus method – the key is to find a comparable
transaction between arm’s-length parties; however, in the case of a profit split
or TNMM, the search is for a comparable independent enterprise rather than
a comparable transaction. In both cases, comparability analysis is used to
select the most appropriate transfer pricing method and apply that method to
determine the arm’s-length price.
The comparability of a related-party or controlled transaction and an
arm’s-length or uncontrolled transaction is usually based on five factors:

(1) the characteristics of the transferred property or services;

(2) a functional analysis of the parties to the transaction;
(3) the terms of the contract;
(4) the economic circumstances; and
(5) the business strategies pursued by the parties.

It is apparent from these factors that comparability analysis requires

good information about the taxpayer and its business as well as the
transactions involved.
Functional analysis is a key element in identifying useful comparable
transactions. (It is also a key element in the attribution of profits to PEs under
new Article 7 of the OECD Model Treaty (see Chapter 8, section 8.8.5)).
Functional analysis involves an examination of the functions performed,
assets used, and risks assumed by the parties to the relevant transaction, since
these factors should determine the returns that the parties should expect from
the transaction. Functions include research and development, manufacturing,
purchasing, transportation, storage, marketing, and management services;
assets include both tangible and intangible assets; risks include financial,
product, collection, market, country/locational, and general business risks.
In many situations, reliable comparable transactions or enterprises may
not be available for one reason or another, such as a lack of information or
transactions involving new technology, for which there are no comparables.

6.4.3 Comparable Uncontrolled Price Method

The CUP method establishes an arm’s-length price by reference to sales of
similar products made between unrelated persons in similar circumstances.
The CUP method is the preferred method if comparable sales exist. It is
widely used for pricing oil, iron ore, wheat, and other goods sold on public
commodity markets. It is also useful for pricing manufactured goods that do
not depend substantially for their value on special know-how or brand names.
However, it is not suitable for pricing many intermediate goods, such as
custom-made automobile parts, that are not generally sold to unrelated
parties. Nor is it suitable for setting the price on sales of goods that are highly
dependent for their value on the trade name of the producer. The operation of
the CUP method is illustrated by the following example.
Assume that XCo is a corporation organized and resident in Country X.
It manufactures wooden chairs in Country X at a cost of 40 and sells them to
unrelated foreign distributors for 47 each. It also sells nearly identical chairs
to YCo, a wholly owned foreign subsidiary resident in Country Y. YCo
resells the chairs purchased from XCo to unrelated consumers at 70. If the
conditions of the sales to YCo and the unrelated distributors are essentially
equivalent, the arm’s-length price for the sale of the chairs to YCo is 47.
Thus, XCo would have a profit of 7 (47 – 40) from the intercompany sales,
and YCo would have a profit of 23 (70 – 47).
The CUP method may be used even if the terms and conditions of the
related-party sales are not identical to the sales to unrelated parties, as long as
adjustments can be made to account for those differences. For example, if in
the previous example the sales by XCo to unrelated distributors do not
include delivery costs, whereas the sales to YCo do include delivery costs,
the sales may still be considered to be comparable, although some adjustment
must be made for freight and handling costs.

6.4.4 Resale-Price Method

The resale-price method sets the arm’s-length price for the sale of goods
between related parties by subtracting an appropriate markup from the price
at which the goods are ultimately sold to unrelated parties. The paradigm case
for its application is the sale by a taxpayer of its manufactured goods to a
related party acting as a distributor, followed by a resale to unrelated
customers without any further processing of the goods. The appropriate
markup is the gross profit, expressed as a percentage of the resale price that
distributors would typically earn from similar transactions with unrelated
parties.
Assume that in the previous example XCo does not make any sales of
furniture to unrelated parties and that no CUP is available. Assume also that
the only activity performed by YCo is to resell the chairs in foreign markets.
Under these assumptions, the resale price method might be appropriate for
determining an arm’s-length price for the chairs. Under the resale-price
method, the first step is to determine the normal markup for a distributor
engaging in activities similar to those performed by YCo. If independent
foreign distributors earn commissions of 20 percent on the purchase and sale
of products comparable to the wooden chairs, a 20 percent markup might be
used to determine the arm’s-length price on sales of chairs by XCo to YCo.
Thus, if YCo sells the chairs to customers for 70 (the resale price), then the
arm’s-length price of the controlled sale between XCo and YCo under the
resale-price method would be 56 (70 – 14 (20 percent of 70)). Thus, under
the resale-price method, XCo would have an arm’s-length profit of 16 (56 –
40) and YCo would have a profit of 14 (70 – 56).

6.4.5 Cost Plus Method

The cost plus method uses the manufacturing and other costs of the related
seller as the starting point in establishing the arm’s-length price. The seller’s
costs are then multiplied by an appropriate profit percentage and the result is
added to the seller’s costs to determine the arm’s-length price. The profit
percentage is determined by reference to the gross profit percentage earned
by the seller in transactions with unrelated parties, or by comparable
unrelated parties in similar transactions with unrelated parties. A paradigm
case for the application of the cost plus method is a sale by a manufacturer of
goods to a related party, with the related party affixing its brand name to the
goods and reselling them to unrelated customers.
Assume that in the previous example XCo sells furniture to YCo without
any brand name affixed to the furniture. YCo affixes its valuable brand name
to the furniture and resells it to customers in foreign markets. In such
circumstances, the cost plus method may provide the appropriate arm’s-
length price. Assume, for example, that the standard gross profit margin
practice in the furniture manufacturing industry is 25 percent of the costs of
production. XCo’s average cost of producing furniture, determined under
generally accepted accounting principles (GAAP), is 40. Under these
assumptions, the arm’s-length price under the cost plus method on sales of
furniture by XCo to YCo is 50 (125 percent of 40).
It is not necessary for the gross profit margin to be based on the gross
profit margins of other taxpayers engaged in the same activities, as long as
adjustments are made to take account of the differences.

6.4.6 Comparison of Traditional Methods

In the examples above, XCo and YCo were engaged in entrepreneurial
activities that could result in an overall gain or an overall loss. Under the
CUP method, the entrepreneurial gain or loss is allocated between XCo and
YCo by reference to comparable transactions between arm’s-length parties.
Under the resale-price method, the distributor, YCo, is guaranteed a profit
and all the entrepreneurial gain or loss is allocated to XCo, the manufacturer.
In the cost plus method, XCo is guaranteed a profit and the entrepreneurial
gain or loss is allocated to YCo. Table 6.1 summarizes the income
attributable to PCo and SCo under the three traditional methods.

Table 6.1 Income Attributable to PCo and SCo under the Three Traditional
 Methods
CUP Resale Price Cost Plus
Method Method Method
(1) XCo’s cost of goods sold 40 40 40
(2) YCo’s sales price to related
70 70 70
customers
(3) Arm’s-length transfer price 47 56 50
(4) Profit for XCo (Line (3) –
7 16 10
Line (1))
(5) Profit for YCo (Line (2) –
23 14 20
Line (3)
(6) Total profits to PCo and SCo 30 30 30
(7) Recipient of entrepreneurial
shared PCo SCo
profit

When a multinational group of corporations is engaged in the

manufacture and sale of products that embody valuable intangible property, it
usually earns substantial entrepreneurial profits. The CUP method generally
cannot be applied when the goods sold embody valuable intangible property
because the goods sold are usually unique – there are no comparable
transactions. In some cases, however, the conditions required for applying the
resale-price method or the cost plus method may be met. If the manufacturer
(XCo in the above example) is producing the goods in a low-tax country and
the distributor (YCo in the above example) is selling those goods in a high-
tax country, the corporate group is likely to favor the application of the
resale-price method because that method allocates the entrepreneurial profits
to the manufacturer in the low-tax country. In contrast, if the country of
production is a high-tax country and the country of sale is a low-tax country,
the corporate group would prefer to use the cost plus method, which allocates
the entrepreneurial profit to the country of sale.

6.4.7 Profit-Split Method

Under the profit-split method, the worldwide taxable income of related
parties engaging in a common line of business is allocated among the related
parties in proportion to their contributions to earning the income. This
method typically is employed when none of the three traditional methods can
be applied. If a group of affiliated companies has more than one product line,
the profit-split method might be applied separately to each product line.
Indeed, there is a wide variety of ways that a profit-split method might be
applied. A distinctive feature of the method is that it applies to aggregate
profits from a group of transactions and not to individual transactions. The
traditional methods, in contrast, are all based on individual transactions. The
following example illustrates the application of a profit-split method.
XCo and YCo are related companies engaged in the production and sale
of pharmaceuticals. XCo engages in extensive research activities and uses
patented processes to manufacture the pharmaceutical products, which it sells
to YCo. YCo repackages the products for retail sale, attaches its valuable
trade name, and resells them through an extensive marketing operation. XCo
does not make any sales to unrelated parties, and there are no comparable
sales of equivalent products by other pharmaceutical companies.
Under these conditions, some countries might use a profit-split method
to establish an appropriate transfer price for the pharmaceuticals. Assume that
XCo incurs costs of 300 in manufacturing the drugs and YCo incurs costs of
100 in packaging, marketing and selling them. Assume also that the sales
proceeds from aggregate sales by YCo to unrelated customers is 600. On
these facts, the corporate group has net profits of 200 (600 – (300 + 100)). If
XCo’s contribution to the enterprise accounts for approximately 75 percent of
the total net profits, then a 75/25 split of the profits might be appropriate.
Thus, XCo would have profits of 150 and YCo would have profits of 50.
There are many possible variations of the profit-split method, including
combining it with one or more of the traditional methods. For example,
traditional methods might be used to allocate average profits from routine
activities, and the profit-split method might be reserved for dividing
entrepreneurial profits from the exploitation of valuable intangible property.
Assume in the example above that XCo engages in routine production
activities and YCo engages in routine sales activities. XCo has gross costs of
production of 300. Unrelated companies engaged in comparable
manufacturing activities earn gross profit margins of 20 percent of costs. On
these facts, XCo would have profits of 60 (20 percent of 300) allocated to it
under the cost plus method. YCo has gross sales revenue of 600. Unrelated
companies engaged in similar activities earn gross profit margins of 10
percent. Under the resale price method, therefore, YCo would have profits
allocated to it of 60 (10 percent of 600). The remaining profits of 80 (200 –
(60 + 60)) would be allocated under the profit-split method. Assuming that a
75/25 split is applied, then XCo would be considered to have profits of 60 (75
percent of 80) under the profit-split method, and total profits of 120 (60 +
60). YCo would be considered to have profits of 20 (25 percent of 80) under
the profit-split method, and total profits of 80 (20 + 60).
For the profit-split method to operate fairly and effectively, some fair
and effective method must be applied to determine the appropriate profit
split. One approach used by the OECD is to examine profit splits between
uncontrolled persons that are engaged in comparable activities.
Unfortunately, such information is typically not available. Because the profit-
split method is most likely to be applied when valuable intangible property is
involved, a profit split based on the relative contributions of the related
parties to the development of that intangible property might be appropriate.

6.4.8 Transactional Net Margin Method

Under the TNMM, (which can be viewed as the OECD equivalent of the US
(CPM), the taxpayer must establish, either for itself or a related party (the
tested party), an arm’s-length range of profits for a set of transactions. If the
tested party’s reported profits from those transactions fall within that range,
its transfer prices will be accepted by the tax authorities. If its profits fall
outside that range, the tax authorities may adjust transfer prices so that the
profits fall within the range, typically at the midpoint. Despite the use of the
term “transactional” in the name TNMM, TNMM is not a transaction method
for determining arm’s-length transfer prices. Instead, TNMM is based on an
entity’s profits from a group of transactions and not on the prices for
particular transactions, which is the focus of the traditional transfer pricing
methods.
In general terms, the profits of a tested party are determined by
determining the ratio of profits to some economic indicator for an unrelated
person and then applying that ratio to calculate the profits of the tested party.
Assume, for example, that an unrelated person has taxable income of 80 and
invested capital of 800, and that invested capital is the economic indicator
used in applying TNMM. The ratio of taxable income to invested capital for
the unrelated person is 80:800, or 10 percent. If the tested party has invested
capital of 500, then under a simplified version of TNMM, its arm’s-length
profits will be 50 (500 × 80/800).
To refine the application of TNMM, the taxpayer or the government
might make TNMM calculations for more than one unrelated person. The
more such calculations are made, the more reliable the results are likely to be.
The arm’s-length profits of the tested party are an amount falling within the
range of profits determined under the several calculations. Statistical
techniques might be applied to select the point within that range that would
be deemed to be the tested party’s arm’s-length profits. If the tested party is a
related corporation rather than the taxpayer, then the profits of the taxpayer
are determined by subtracting the profits of the tested person, as determined
under TNMM, from the combined profits of the two corporations.
Whether the taxpayer or a related person is used as the tested party
depends on the facts and circumstances of each particular case. The objective
is to have, as the tested party, the related corporation that is most similar with
respect to its business functions to the unrelated corporations that are used as
comparables. For example, assume that ACo manufactures goods in Country
A, sells the goods to BCo, its wholly owned subsidiary, and BCo sells the
goods in Country B after affixing its valuable trade name to those goods. If
information necessary for applying the traditional pricing methods or for
applying TNMM to ACo is not available, such information for applying
TNMM to BCo might be available. If so, BCo would be the tested party,
whether or not it is the taxpayer.
To apply TNMM, a taxpayer must determine a range of profits that
unrelated persons would be expected to earn from engaging in comparable
transactions. The taxpayer can establish this range in a variety of ways. One
way, illustrated above, is to determine the rate of return on capital employed
by two or more unrelated parties engaging in activities that are broadly
similar to the activities of the taxpayer. This rate of return on capital for each
unrelated person is then multiplied by the amount of capital of the taxpayer
(or tested party, as the case may be). An alternative approach is for the
taxpayer to determine the ratio of operating profits to gross sales receipts for
two or more comparable related persons and then apply these ratios to its own
(or the tested party’s) sales. A third way is to determine the ratio of gross
profit to operating expenses for two or more related persons and apply these
ratios to the taxpayer’s or the tested party’s operating expenses. Other
economic indicators might also be used.
 Assume, for example, that ACo, the tested party, is engaged in business
activities similar in complexity and character to the activities of XCo and
YCo, corporations unrelated to ACo and to one another. XCo and YCo have
ratios of operating profits to gross receipts of 0.2 and 0.3, respectively. ACo
has gross receipts of 200,000. Under TNMM, ACo’s arm’s-length range of
profits would be from 40,000 (200,000 × 0.2) to 60,000 (200,000 × 0.3).
Assuming that the various conditions for application of TNMM are met,
ACo’s arm’s-length profits would be considered to be in the range of 40,000
to 60,000.
Once the TNMM range has been established, it is necessary to select
some amount within that range as the arm’s-length profits of the tested party.
In general, the tax authorities would probably accept the transfer prices
shown in the taxpayer’s books and records if the profits determined by using
those prices fall within the TNMM range. If the taxpayer’s reported profits
fall outside the range, the tax authorities are likely to treat the midpoint of the
range as the arm’s-length profits. If data for more than two unrelated persons
is used to establish the TNMM range, then a weighted average of the
resulting profit numbers would be used to establish the midpoint of the range.
TNMM can be manipulated, by the taxpayer or the tax authorities,
through the choice of comparable companies. To prevent systemic biases in
favor of either the taxpayer or the tax authorities, criteria need to be
developed for selecting appropriate comparable companies. In addition,
neutral rules should be applied to eliminate comparable companies that yield
unreasonable results and to select the arm’s-length profits from within the
TNMM range.

6.5 SHARING OF CORPORATE RESOURCES

6.5.1 Introduction
Related corporations frequently share funds, credit lines, corporate
headquarters, know-how, trade names, employees, and other corporate
resources. The arm’s-length standard requires that the owner of the shared
resources charge related parties an arm’s-length fee for their use. In theory,
the fee should equal the amount that the owner of an equivalent resource
would charge an unrelated party for its use. In practice, the appropriate arm’s-
length price is difficult to determine, in part because unrelated corporations
do not often share comparable resources.

6.5.2 Loans or Advances

A group company engaged in the business of making commercial loans
should be required to use a rate of interest for loans or advances to related
parties that reflects the current cost of borrowing. For a group company that
is not in the business of making loans, some countries provide a safe-harbor
interest rate so that the interest rate charged on the loan will not be adjusted if
it is within the safe harbor. For example, a country may allow taxpayers to
use an interest rate pegged to the average cost of government borrowing.

6.5.3 Performance of Services

If marketing, managerial, administrative, technical, or other services are
performed by one person for the benefit of a related person, the person
providing the services should charge an amount that an independent service
provider would charge for the same services. If no comparable arm’s-length
services are available, the arm’s-length fee might be based on the cost of
providing the services plus an appropriate profit. However, the problem of
setting an appropriate arm’s-length price in these circumstances is
formidable.

6.5.4 Use of Tangible Property

If tangible property, such as an office or equipment, is made available by a
person to a related person, the owner of the property should charge the user
an arm’s-length rental fee. The same rule should apply to subleases of
tangible property.

6.5.5 Use or Transfer of Intangible Property

If intangible property, such as a patent or trademark, is made available to a
related party, the owner of the property should charge whatever amounts
would be charged to an unrelated person for the use of the property in similar
circumstances. This charge might be set by reference to royalty rates charged
by the owner on the same or similar property made available to unrelated
parties. However, obtaining the data necessary to determine the proper arm’s-
length royalty is often difficult, both for the tax authorities and the taxpayer,
and multinational companies are commonly accused of avoiding tax through
the use of inappropriate royalty rates.
If intangible property is sold to a related person, the arm’s-length sales
price can be established by reference to the discounted value of the arm’s-
length royalties anticipated over the life of the property.
In 1986, the US adopted legislation requiring that royalty rates charged
between related parties be commensurate with the income from the intangible
property. Under the arm’s-length standard, royalty rates generally are based
on facts known or knowable at the time the license for the use of the property
is concluded. The commensurate-with-income standard requires periodic
adjustments in royalty rates to reflect the actual experience of the parties in
utilizing the intangible property. For example, if XCo, a US corporation,
transfers patents and know-how to its Irish subsidiary that allows it to
manufacture plastic contact lenses, under the commensurate-with-income
standard, the parties may be required to make periodic adjustments in the
royalty charged for use of that property to reflect the level of profits earned
by the Irish subsidiary from the manufacture and sale of the contact lenses.
Despite their potential, the US commensurate-with-income rules have
not been effective in curbing transfer pricing abuses. The OECD has
endorsed the limited application of a commensurate-with-income standard
where unrelated persons operating at arm’s length would not have made an
outright sale or long-term license of intangible property. For example, a
multinational enterprise with unique and valuable intangible property would
never sell or lease that property to an unrelated party. In these circumstances,
it is more appropriate to ignore the related-party sale or license and substitute
a different arrangement, under which the transferor is entitled to a substantial
share of the actual profits earned through use of the transferred intangible
property. See section 6.7 below, describing the circumstances in which the
tax authorities are permitted to recharacterize the actual transactions entered
into by associated enterprises.
The OECD’s BEPS Action 8: Guidance on Transfer Pricing Aspects of
Intangibles proposes extensive revisions to the OECD Transfer Pricing
Guidelines, especially Chapter 6 dealing with the basic treatment of
intangibles for transfer pricing purposes. These revisions clarify the
definition of intangibles and provide guidance for identifying transactions
involving intangibles and determining the arm’s-length conditions for cases
involving intangibles. The revisions also include several examples that
illustrate the application of the guidelines to intangibles. This work will not
be finalized until the related BEPS work on intangibles is completed.

6.6 COST-CONTRIBUTION ARRANGEMENTS

If a group of corporations intends to develop valuable intangible property and
share the benefits among two or more of its members, it can avoid transfer
pricing issues by having all the prospective users of the intangible property
jointly develop that property through a cost-contribution or sharing
arrangement. In that situation, all the contributors to the development of the
property have rights to the profits generated by the property in proportion to
their contributions, and no transfer of the property or rights to the use of the
property between members of the corporate group is required. Cost-
contribution arrangements are dealt with in Chapter 8 of the OECD Transfer
Pricing Guidelines. In general, the OECD rules are designed to recognize
bona fide cost-contribution arrangements, but limit their use for tax
avoidance purposes.
For a cost-contribution arrangement to be consistent with the arm’s-
length standard, it should have the following characteristics:

– The arrangement should be embodied in a legally enforceable written

contract entered into when the arrangement was initially established
that clearly establishes the nature of the arrangement, its duration,
and the terms for its enforcement and amendment.
– Only persons with a legitimate expectation of benefiting under the
arrangement should be permitted to be participants.
– The contract should require the participants to the arrangement to
contribute to the costs of development of the intangible property in
proportion to the benefits that they might reasonably be anticipated to
derive from the use of that property.
– The participants to the arrangement should be required to keep
adequate records documenting their costs and explaining how their
anticipated benefits were calculated.

Assume, for example, that ACo and BCo are related corporations
engaged in manufacturing small electrical appliances. ACo is engaged in
business in Country A and BCo is engaged in business in Country B. ACo
and BCo intend to develop new technology that would allow them to
manufacture their appliances at a lower cost. They enter into a written
contract that assigns to ACo the rights to exploit any intangible property
developed under the agreement in Country A. BCo receives similar rights
with respect to Country B. ACo has current sales of appliances in Country A
of 400, and BCo has sales in Country B of 600; that general pattern is
expected to continue in the future. Under the cost-contribution arrangement,
ACo agrees to pay 40 percent of the costs and BCo agrees to pay the
remaining 60 percent. They both agree to keep detailed accounting records
with respect to their costs and their sales.
Assuming that County A has adopted rules for cost-contribution
arrangements similar to those in the OECD Guidelines, the arrangement
between ACo and BCo would qualify as a bona fide cost-contribution
arrangement. Country A should permit ACo to take a deduction for payments
made under that arrangement under the rules generally applicable to amounts
paid to develop intangible property. In addition, ACo should be treated as a
coowner of the resulting intangible property; therefore, ACo should not be
treated as paying a deemed royalty to BCo under the transfer pricing rules of
Country A. If ACo makes an actual royalty payment to BCo, that payment
should not be deductible.
As part of BEPS Action 8 dealing with transfer pricing, the OECD has
proposed revisions to Chapter 8 of the Transfer Pricing Guidelines dealing
with cost-contribution arrangements. In general terms, the arm’s-length
standard requires each participant’s contribution to a cost-contribution
arrangement to be consistent with its share of the expected benefits from the
arrangement. For this purpose, the revised Guidelines will require each
participant’s contribution to be based on its value rather than on its cost
unless, in certain limited circumstances, cost is a reliable indicator of the
value of contributions.
The rules governing cost-contribution arrangements among related
persons should permit the participants to modify an arrangement to reflect
changed economic circumstances. To conform to the arm’s-length standard,
however, those rules should require that the participants receive a payment
equal to the fair market value of any rights that they may have relinquished
and that they pay a fair market fee for any new rights obtained. If a new
participant is brought into a cost-contribution arrangement, that participant
should be required to compensate the other participants for the fair market
value of the dilution of their interests in the arrangement. Any revision of a
qualifying cost-contribution arrangement should be in writing and otherwise
conform to the requirements for a new cost-contribution arrangement.
To prevent tax avoidance, governments should have the authority to
treat related persons as if they had entered into a cost-contribution
arrangement when their economic behavior is consistent with such an
arrangement. Assume, for example, that ACo and BCo are related persons
and have jointly developed some intangible property. However, they have not
entered into a cost-contribution arrangement. ACo makes use of the
intangible property in its business in Country A and pays a royalty for that
use to BCo, a resident of Country B. Under the tax treaty between Country A
and Country B, the royalty is not subject to withholding tax by Country A
because the treaty contains an article similar to Article 12 of the OECD
Model Treaty, giving the exclusive right to tax royalties to the residence
country. ACo claims a deduction for the royalty payment in computing its
income taxable in Country A. Country A should have the authority to treat
ACo and BCo as having entered into a constructive cost-contribution
arrangement, with the result that ACo would not be permitted to take a
deduction for the royalty payment to BCo.
Cost-contribution arrangements have been widely used to shift profits to
low-tax countries. This result is possible largely because multinational
enterprises have access to information about their intellectual property and
research and development activities that is not available to the tax authorities.
Therefore, for example, a group company in a low-tax country might enter
into a cost-contribution arrangement with another group company in a high-
tax country; the high-tax country would allow the first company to acquire
rights to research that is disproportionately profitable compared to all the
research and development activities of the group.

6.7 DISREGARDED TRANSACTIONS

The OECD Transfer Pricing Guidelines emphasize that the tax authorities
must accept the actual transactions entered into by associated enterprises and
apply the transfer pricing rules to those transactions. However, in two narrow
circumstances, the OECD Guidelines permit the tax authorities to ignore the
legal transactions entered into by related enterprises and allocate the profits
between the enterprises on a different basis. First, if the economic substance
of the actual transaction differs from the legal form of the transaction, the
transaction may be recharacterized by the tax authorities in accordance with
its substance and taxed on that basis. For example, depending on the
circumstances, including its terms, a loan to an associated enterprise might be
recharacterized as equity. Second, if the arrangements entered into by the
associated enterprises would not have been entered into by arm’s-length
parties, the tax authorities may recharacterize the arrangements in accordance
with what arm’s-length parties would have done. For example, a
multinational enterprise with valuable intangible property would not sell or
lease that property on a long-term basis to an unrelated enterprise without
some price-adjustment mechanism allowing the transferor to share in future
profits.
These exceptional circumstances, in which the tax authorities can
disregard actual transactions, are strikingly similar to anti-avoidance rules. It
is questionable whether the tax authorities have the power to recharacterize
transactions, as provided in the OECD Guidelines, without explicit statutory
authority in domestic law. Part of the OECD’s BEPS Action 10 involves
clarifying the circumstances in which transactions between associated
enterprises can be recharacterized in order to prevent base erosion.

6.8 TRANSFER PRICING DOCUMENTATION

REQUIREMENTS
Many countries have attempted to deal with perceived transfer pricing abuses
by requiring multinational companies to provide the tax authorities with
extensive, contemporaneous documentation to support the methods used to
establish their transfer prices. The idea is that by forcing multinational
companies to establish their transfer prices in advance, a country can prevent
after-the-fact shifting of income for tax avoidance purposes. Taxpayers
failing to provide the requisite documentation may be subject to substantial
penalties.
For example, assume that ACo, operating in Country A, is selling goods
to BCo, a related corporation operating in Country B. The corporate tax rate
is 40 percent in Country A and 20 percent in Country B. ACo and BCo set
the transfer prices for their intercompany sales so that there is a small profit
in Country A and a large profit in Country B. After setting those prices and
providing the documentation to Country A, ACo discovers that it will suffer a
large tax loss in Country A from some unrelated operations. If its pricing
methodology had not been fixed, ACo might have been strongly tempted to
revise the methodology to deflect some income from Country B to Country A
so that it could fully utilize the Country A loss. The contemporaneous
documentation rules, however, may prevent such a revision.
The OECD Guidelines support the dual strategy of requiring
contemporaneous documentation and penalizing taxpayers for failure to
comply with the documentation requirements. However, tax administrators
are advised to pursue that strategy with extreme caution so as to avoid
imposing unfair or excessively burdensome obligations on taxpayers acting in
good faith.
Action 13 of the OECD’s BEPS project proposes important
enhancements to the existing transfer pricing documentation requirements.
Multinational enterprises will be expected to provide the tax authorities of all
countries in which they do business with a master file containing general
information about their business operations and their transfer pricing
practices and policies. In addition, each country must be provided with a
local file containing information with respect to any related-party transactions
occurring in the country. Most important, multinationals will be required to
provide an annual report to each country in which they operate setting out the
amount of revenue, profit, and taxes accrued and paid with respect to that
country. Although these country-by-country reports will not require
information to be provided on an entity-by-entity basis, they should
nevertheless provide the tax authorities with an important tool with which to
apply their transfer pricing rules more effectively.
Countries can limit the most egregious forms of transfer pricing abuses
by giving appropriate discretion and resources to their tax departments and by
imposing stiff penalties on taxpayers that have not set their transfer prices in
good faith. They can take away some of the incentive for transfer pricing
abuses by setting their marginal tax rates at levels that are moderate by
international standards. Adoption of specific legislation targeted at tax havens
can also take away some of the incentive for transfer pricing abuses because
taxpayers commonly set improper transfer prices in order to deflect their
income to affiliated entities in tax havens. Chapter 7 describes such anti-
avoidance provisions. Countries can also obtain help in controlling transfer
pricing abuses through cooperation with their tax treaty partners, particularly
their close neighbors. However, despite determined efforts by national tax
authorities and increased cooperation with their trading partners, transfer
pricing continues to be a serious ongoing problem.

6.9 TREATY ASPECTS OF TRANSFER PRICING

The provisions of the OECD and UN Model Treaties do not deal with the
problem of transfer pricing in any detailed way. Article 9(1) of both Model
Treaties authorizes an adjustment to the profits of an enterprise that is
associated with another enterprise if “conditions are made or imposed
between the two enterprises which differ from those which would be made
between independent enterprises.” Literally, therefore, Article 9(1) focuses
on the terms and conditions of related-party transactions, rather than just on
the prices charged in specific transactions.
Article 9 is entitled “Associated Enterprises.” It applies if an enterprise
of one contracting state participates directly or indirectly in the management,
control, or capital of another enterprise in the other contracting state or if the
same persons participate directly or indirectly in the management, control, or
capital of two or more enterprises in the contracting states. Beyond this
wording, Article 9 does not define “associated enterprises”; as a result,
whether enterprises are associated must be determined in accordance with
domestic law. Most countries apply their transfer pricing rules to transactions
between enterprises where one enterprise controls the other, or where they are
both controlled by another person and control is considered to be the
ownership of more than 50 percent of the shares or interests in an entity.
Article 9(1) of both the OECD and UN Model Treaties provides that the
profits that would have accrued to an enterprise, but did not because of non-
arm’s-length transactions with associated enterprises, “may be included” in
the profits of the enterprise. Thus, read literally (and unlike other treaty
provisions), Article 9(1) is not worded as a mandatory provision (“shall be
included”) that the contracting states must apply. Some commentators argue
that, as a result of the permissive wording of Article 9(1), contracting states
are entitled to tax resident enterprises on profits that are more or less than
their profits in accordance with the arm’s-length standard. For example, if
ACo, a resident of Country A, charges its subsidiary, BCo, resident in
Country B, 100 for goods or services that have an arm’s-length price of only
75, Country A is not required by Article 9(1) to unilaterally reduce the price
to 75. Conversely, if the arm’s-length price of the goods or services is 125,
Country A is not required by Article 9(1) to increase the price to 125,
although in most circumstances it will want to do so.
On this view, Article 9(1) is merely a statement of principle about the
determination of the profits of associated enterprises in accordance with the
arm’s-length standard. However, Article 9(2) of both the OECD and UN
Model Treaties requires a country to make adjustments to the transfer prices
used to compute taxable income of their taxpayers if those prices have been
adjusted by the other contracting state in accordance with the arm’s-length
standard.
Assume, on the facts of the previous example, that Country A adjusts
the price at which ACo sells its manufactured goods to its foreign affiliate,
BCo, from 60 (the actual sales price) to 90 (the price that Country A
considers that arm’s-length parties would have charged). Therefore, Country
A will increase ACo’s taxable income by 30. If Country B concurs with
Country A’s determination of the proper transfer price, it should allow BCo
to increase its actual cost of acquiring the goods by 30 (60 + 30 = 90) and
reduce its taxable income accordingly, to 60. An adjustment of a transfer
price used by one taxpayer to take into account an adjustment made by
another country to the transfer price used by an affiliated taxpayer is referred
to as a “corresponding adjustment”.
It is arguable whether profit-based transfer pricing methods, such as the
profit-split method and TNMM, are consistent with the language of Article
9(1). The reference to profits in Article 9(1) could be a reference to all
profits, or just the profits from particular transactions or types of business. In
any event, by endorsing the profit-based methods, the OECD Transfer
Pricing Guidelines clarify that those methods are acceptable in certain
circumstances under Article 9.
Conflicts between countries over transfer prices are commonplace,
despite the fact that all countries have agreed in their bilateral tax treaties to
adhere to the arm’s-length standard. Most tax treaties provide that an
enterprise that is subject to double taxation because of inconsistent transfer
prices may seek redress through the mutual agreement procedure. Under that
procedure, the competent authorities are required to try and deal with the
taxpayer’s complaint, but they are not generally obligated to resolve it. A few
newer treaties include a procedure for binding arbitration. For a discussion of
the mutual agreement procedure for resolving treaty disputes and arbitration,
see Chapter 8, sections 8.8.3 and 9.5, respectively.

6.10 TAX POLICY CONSIDERATIONS: FORMULARY

APPORTIONMENT AND THE FUTURE OF THE
ARM’S-LENGTH METHOD
To be blunt, the current transfer pricing rules based on the arm’s-length
standard do not work effectively, despite decades of practical experience and
refinement. Although there are many reasons for the failure of arm’s-length
transfer pricing rules, the most fundamental reason is that the underlying
assumption – that the separate parts of a multinational enterprise can be
considered to behave as if they were independent – is patently false.
Multinational enterprises exist because there are economic and financial
advantages to the economic integration of their activities. However, transfer
pricing in accordance with the arm’s-length standard ignores the economic
integration of a multinational enterprise’s activities, but respects its legal
structure and the legal form of its intercorporate transactions.
The arm’s-length standard has received considerable criticism from
academic commentators, from taxpayers directly affected by the standard,
and from tax administrators. Taxpayers complain that it often imposes
unreasonable burdens of proof on them, that it presents them with problems
of double taxation not resolved by the competent authority mechanism of tax
treaties, and that frequently it is not followed by the tax authorities during
audits. The tax authorities complain that the arm’s-length standard allows
considerable undertaxation of taxpayers engaged in cross-border transactions,
that it encourages taxpayers to take aggressive positions on their tax returns
in the hope of avoiding detection or of striking a favorable bargain on audit,
and that it is extremely time-consuming and expensive to enforce. Some
academics contend that, in some cases, the arm’s-length method necessarily
produces improper results because it cannot account for the profits that
related corporations typically enjoy from conducting an integrated business.
All of these criticisms are valid.
Although arm’s-length transfer pricing is generally acknowledged to be
seriously and irreparably flawed, it has one very important advantage –
namely, that it is widely accepted. The transitional costs in moving to a
different, even if better, system would likely be enormous. Moreover, it is not
clear that there is a better system. Some commentators argue that the profits
of multinational groups should be allocated among the members of the group
based on profit-split methods. In fact, as noted above, the OECD Transfer
Pricing Guidelines currently recognize the use of profit splits in certain
circumstances and, in practice, it would appear that they are used increasingly
to resolve transfer pricing disputes. The difficulty with profit-split methods is
that it is unclear on what basis the profits of a multinational enterprise should
be allocated to the countries in which it does business.
The alternative to the arm’s-length approach that is most frequently
advanced is a global formulary apportionment system. In a formulary
apportionment system, affiliated entities engaged in a common enterprise are
taxed as if they were a single entity. The worldwide income of the enterprise
is attributed by a predetermined formula among all the countries where the
enterprise engages in meaningful economic activity. Assuming that all
countries could agree on the use of this system and could also agree on a
reasonably uniform definition of taxable income, multinational corporations
would be taxable once, and only once, on their worldwide income.
For example, in the case of a multinational enterprise engaged in the
manufacture and sale of goods, an apportionment formula might allocate
some fraction – perhaps one-half – of the income of the enterprise among the
countries in accordance with the sales in those countries. The remaining part
of the income would be apportioned among the countries where the
manufacturing is conducted, with the allocation based on the total
manufacturing assets or the payroll of the enterprise, or some combination of
these two factors. Little or no income would be apportioned to any group
entity in a tax haven unless sales or manufacturing activities take place in that
country.
There are obviously many problems with the use of formulary
apportionment as a means of allocating profits among related corporations.
The arbitrariness of predetermined formulas makes it difficult to take into
account the particular circumstances of each multinational enterprise. It relies
heavily on access to foreign-based information. It almost guarantees that the
amount of profits attributed to each member of a multinational group will
differ, sometimes markedly, from the income shown on its books of account,
even if those books are kept in good faith and in accordance with approved
accounting methods. Substantial cooperation among governments is
necessary to solve these problems, and it is unlikely that countries would be
able to agree on a common apportionment formula.
 Nevertheless, formulary apportionment has some attractive features. A
well-designed system can eliminate the tax advantages of low-tax countries
without the need for complex, difficult-to-administer controlled foreign
corporation (CFC) rules. Formulary apportionment also avoids some of the
difficult audit problems that frequently arise under the arm’s-length approach.
Unlike the arm’s-length approach, it does not require separate agreement on
the source of gross income and deductions in order to avoid double taxation
because source rules are implicitly incorporated into the apportionment
formulas. Although it is sometimes argued that formulary apportionment
would be better for developing countries than arm’s-length transfer pricing,
this claim is questionable. The use of factors such as sales, property, and
employees or payroll in the formula are unlikely to result in the allocation of
substantial income to developing countries.
In comparing the formulary apportionment method to the arm’s-length
method, it is useful to contrast the treatment of income from intangible
property under the two approaches. Under formulary apportionment, all
income, including income derived from intangibles, would be apportioned to
the countries in which goods are produced and sold. In contrast, under the
arm’s-length method, such income is allocated to the entity owning the
intangible property. Ownership rights within a corporate group, however,
have little economic significance. As a result, under the arm’s-length method,
it is frequently possible for a multinational enterprise to avoid tax on income
from intangibles by shifting ownership of the intangibles to an affiliated
corporation in a low-tax country.
Formulary apportionment has an undeservedly bad reputation, largely
for political reasons. A sensible discussion of that method and the alternatives
to it must go beyond labels and preconceived ideas. Moreover, the arm’s-
length standard and formulary apportionment should not be seen as polar
extremes – instead, they should be viewed as part of a continuum of methods
ranging from CUPs to predetermined formulas.
A formulary apportionment system uses an arm’s-length approach in
some circumstances, and the arm’s-length approach sometimes uses formulas
(profit splits). Recent refinements in the arm’s-length approach rely
increasingly on formulas, and that trend seems to be gaining international
acceptance. Consequently, it is sometimes unclear where the arm’s-length
principle ceases and formulary apportionment begins. Applying pejorative
and misleading labels to either approach is counterproductive.
 Despite its deficiencies, the arm’s-length standard is likely to continue to
be the internationally accepted approach for resolving transfer pricing issues,
except in special circumstances. As the earlier discussion of pricing
methodologies indicates, however, the arm’s-length standard is vague and has
been interpreted to accommodate pricing methodologies, such as the profit-
split method and TNMM, that seem closer to formulary apportionment than
to an arm’s-length approach.
Formulary apportionment is used in some federal countries (e.g., Canada
and the US) to allocate the income of an entity among the subnational
governments. It has been proposed for internal use within the North America
Free Trade Agreement and the European Union. For several years, the
European Union has been exploring the possibility of adopting some type of
formulary apportionment to deal with the complex problems that Member
States encounter in determining the amount of income derived by
corporations from activities occurring within their borders. However,
agreement on a common consolidated tax base has been elusive.
The OECD has approved the use of formulas for apportioning the
income of corporate groups engaged in global trading, banking, and
insurance. Therefore, current transfer pricing rules are a mix of arm’s-length
methods and formulary apportionment methods. Given the deficiencies of
both the arm’s-length method and formulary apportionment, as well as the
difficulty of changing the existing system, it seems likely that transfer pricing
will continue to be a complex mixture of rules that, despite continuous
refinement, will never deal with the underlying problem in a satisfactory
manner.
CHAPTER 7
Anti-avoidance Measures

7.1 INTRODUCTION
International transactions provide many opportunities for the avoidance of
tax. In this context, tax avoidance must be distinguished from tax evasion,
which is illegal and usually involves the intentional nondisclosure of income
or fraud. Tax avoidance is difficult to define precisely, but generally means
transactions or arrangements entered into by a taxpayer in order to minimize
the amount of tax payable in a lawful manner.
The ways of avoiding tax through international transactions are far too
numerous to itemize. The following examples, however, illustrate the range
of possibilities:

– A taxpayer can shift his or her residence from one country to another
country that levies lower or no taxes.
– A taxpayer can divert domestic source income to a controlled foreign
entity, such as a trust or a corporation, established in a tax haven.
– A taxpayer can establish a tax haven subsidiary to earn foreign
source income or to receive dividends from subsidiaries in other
foreign countries.
– If advantageous treaties exist, a taxpayer can route dividends,
interest, royalties, and other amounts through subsidiaries established
in foreign countries in order to reduce the amount of withholding tax
on such amounts.

Not surprisingly, most countries have anti-avoidance rules to deal with

certain types of international tax avoidance, and some countries still have
exchange controls to regulate foreign investments and transactions by
residents. Although these controls can be effective in preventing international
tax avoidance, over the past several decades countries have abandoned
exchange controls in favor of the free movement of capital.
Countries that do not use exchange controls employ a wide variety of
tax measures to combat international tax avoidance. Some important
examples of such measures are described briefly below, with references in
some instances to more detailed treatment elsewhere in the Primer.
Anti-avoidance rules and doctrines. Many countries have judicial anti-
avoidance doctrines or statutory anti-avoidance rules under which
transactions may be disregarded for income tax purposes. These doctrines
and rules apply generally to tax avoidance transactions or arrangements,
including international transactions. Judicial anti-avoidance doctrines include
the sham transaction, substance-over-form, business purpose, step transaction
and abuse of law doctrines. For example, the existence of a holding company
established in a tax haven may be disregarded as a sham if it does not engage
in any genuine commercial activities. Statutory anti-avoidance rules include
specific and general anti-avoidance rules. Specific anti-avoidance rules
include transfer pricing rules, thin capitalization and earnings-stripping rules,
controlled foreign corporation (CFC) rules, and foreign investment fund
rules. General anti-avoidance rules (often referred to by the acronym
“GAAR”) are intended to be sufficiently broad to deal with most or all types
of abusive tax avoidance, including international tax avoidance transactions.
Typically, GAARs apply when the principal purpose or one of the principal
purposes of a transaction or a series of transactions is to avoid tax and the
transaction abuses, frustrates, or defeats, or is inconsistent with, the object
and purpose of the relevant tax legislation.
Special tax haven provisions. Some countries have specific provisions
designed to deal with particular tax haven abuses. For example, Germany
imposes a special tax on persons who move their domicile to a tax haven.
Other countries disallow the deduction of interest and royalty payments or
payments for services made to a tax haven entity unless the taxpayer
establishes that the transactions are genuine; in other words, the onus of proof
is placed on the taxpayer to justify such deductions.
Transfer pricing rules. Most countries have intercompany or transfer
pricing rules to prevent related taxpayers from carrying out transactions at
artificially high or low prices in order to shift income and expenses from one
country to another. It is arguable whether these rules are properly classified
as international anti-avoidance rules or whether they are just part of a
country’s basic tax system. Transfer pricing rules are discussed in detail in
Chapter 6.
CFC rules. Several countries have adopted CFC rules to prevent the
diversion of passive and certain other income to, and the accumulation of
such income in, a CFC established in a tax haven. These rules are discussed
in section 7.3 below. Some countries have similar rules with respect to
foreign trusts.
Foreign investment fund rules. Several countries have adopted foreign
investment fund rules to prevent the deferral of domestic tax by residents
investing in foreign mutual funds, unit trusts, or similar entities. These rules
are discussed in section 7.4 below.
Anti-treaty shopping rules. Several countries insist on the inclusion of
provisions in their tax treaties and/or in their domestic legislation to prevent
treaty shopping. Treaty shopping typically involves the establishment of a
legal entity in a country by nonresidents in order to obtain the benefits of the
country’s tax treaties. Treaty shopping is discussed in Chapter 8, section
8.8.2.2.
Thin capitalization and earnings-stripping rules. Several countries have
adopted thin capitalization and earnings-stripping rules to limit the
deduction of interest by resident corporations and other legal entities. Thin
capitalization rules are intended to prevent nonresident shareholders of
resident corporations from using excessive debt capital to extract corporate
profits in the form of deductible interest rather than non-deductible dividends.
Earnings-stripping rules are more broadly targeted at resident corporations
and other entities that claim disproportionately large interest deductions.
These rules are discussed in section 7.2 below.
Taxation of gains on transfers of property abroad and on expatriation.
When appreciated property – property with an accrued gain – is transferred to
a related nonresident, some countries deem the property to have been sold for
its fair market value so that the accrued gain is subject to tax. Otherwise,
domestic tax on the gain might be avoided entirely. Further, some countries
impose tax on accrued gains when a taxpayer ceases to be resident or for a
temporary period after a taxpayer ceases to be resident; such exit or departure
taxes and trailing taxes are discussed in Chapter 3, sections 3.4.1 and 3.4.2.
Back-to-back arrangements. Back-to-back arrangements are
commonly used as a tax planning device to obtain tax benefits that would not
otherwise be available to a taxpayer directly. For example, a country may
have rules dealing with related-party transactions; these rules can sometimes
be avoided by inserting an arm’s-length intermediary between the related
parties. Similarly, the benefits of a country’s treaties, such as reductions in
withholding taxes, may be inappropriately obtained through back-to-back
arrangements. Such arrangements are particularly common with respect to
financial transactions, since funds can be funneled through an arm’s-length
financial institution with relative ease. For example, assume that Country A
exempts interest payments by corporations resident in Country A to arm’s-
length nonresidents from its withholding tax on interest. If ACo, a
corporation resident in Country A, pays interest to BCo, a related corporation
resident in Country B, the interest would be subject to Country A’s
withholding tax. However, if BCo puts funds on deposit with a financial
institution that deals at arm’s length with both ACo and BCo and the
financial institution loans an equivalent amount to ACo, Country A’s
withholding tax would not apply to the interest payments by ACo to the
financial institution.
Hybrid entities and hybrid financial instruments. A “hybrid”
arrangement refers to situations in which two countries treat entities,
transactions, or arrangements differently and the different treatment is
exploited to produce tax benefits. For example, if one country treats preferred
shares issued by a resident corporation in accordance with their legal form as
shares on which dividends are paid, but another country treats the shares as
debt on which is interest is paid, this inconsistent treatment can be exploited
to produce tax savings. If the country in which the corporation is resident
treats the payments on the shares as interest, the payments will be deductible
and reduce that country’s tax base. If the country in which the recipient of the
payments is resident treats the payments as dividends, it may exempt those
dividends from tax as a result of its participation exemption. Hybrid
arrangements are discussed in more detail in Chapter 9, section 9.3.

7.2 RESTRICTIONS ON THE DEDUCTION OF

INTEREST: THIN CAPITALIZATION AND
EARNINGS-STRIPPING RULES

7.2.1 Introduction
When a resident corporation pays interest to nonresidents, the interest is
usually deductible by the payer in computing income unless there are special
rules to the contrary. The interest payments may be subject to withholding
tax, but the rate of withholding tax may be substantially reduced or
completely eliminated pursuant to an applicable tax treaty. The nonresident
lender may or may not be subject to tax on the interest in its country of
residence. If the nonresident lender is also the controlling shareholder of the
resident corporation, the nonresident lender/shareholder will usually have a
choice of financing its subsidiary with debt or equity and extract the profits of
the subsidiary by receiving either dividends or interest.
Unlike interest, dividends paid by a resident corporation generally are
not deductible. Accordingly, income earned by a resident corporation and
distributed to its shareholders is subject to two levels of tax – corporate tax
when the income is earned by the corporation, and shareholder tax when the
income is distributed to the shareholders as a dividend. If the shareholder is a
nonresident, the shareholder tax is usually imposed as a withholding tax.
In contrast, income earned by a resident corporation and distributed in
the form of interest to a nonresident lender who is also a shareholder of the
corporation is subject to only one level of tax. Because the interest is
deductible by the corporation, usually the only source country tax is the
withholding tax on the interest payment to the nonresident, and many
countries have reduced or eliminated their withholding taxes on interest,
either unilaterally or under their tax treaties. The advantage of paying interest
to nonresident shareholders compared to paying dividends constitutes an
inherent bias in favor of debt financing of resident corporations by
nonresident investors. This bias is illustrated in the following example.
NCo, a nonresident corporation, owns all the shares of RCo, a resident
corporation. RCo requires capital of one million to finance its business
activities. To provide that capital, NCo can either subscribe for one million in
additional shares of RCo, or it can loan RCo one million (or some
combination of debt and equity). RCo earns income, before the payment of
interest or dividends, of 100,000 and distributes its entire after-tax income as
a dividend. The arm’s-length interest rate payable on loans is 10 percent, and
the applicable rates of withholding tax are 5 percent on dividends and 10
percent on interest. A comparison of the tax results of advancing funds by
way of debt and equity are set out in Table 7.1.

Table 7.1 Relative Advantages of Debt and Equity Finance

Debt Equity
Corporate income before 100,000 100,000
payment of interest or
dividends
Deduction of interest 100,000 not applicable
Taxable income nil 100,000
Corporate tax (40%) Nil 40,000
Dividends not applicable 60,000
Withholding tax (10%, 5%) 10,000 3,000
Total tax 10,000 43,000

As this example illustrates, financing a resident corporation with debt is

considerably more effective in reducing the source country tax than financing
with equity. The major reason is that interest is deductible, whereas dividends
are not deductible. In addition, a resident corporation can repay a loan at any
time without triggering tax, whereas it may not be able to repay equity
investments (redeem shares or reduce capital) without triggering a taxable
dividend.
In response to the bias in favor of debt compared with equity, several
countries have adopted restrictions on the deduction of interest paid to
nonresidents, or on the deduction of interest more generally. Under “thin
capitalization” rules, the deduction for interest paid by a resident corporation
to a nonresident controlling shareholder is denied to the extent that interest
deductions claimed by the corporation are considered to be excessive. Under
these rules, interest is considered to be excessive to the extent that the
corporation’s debt relative to its equity exceeds a fixed debt:equity ratio
(often 1.5:1 or 2:1). The term “thin capitalization” is apt because the rules
apply only when a corporation’s equity capital is small in relation to its debt.
Under earnings-stripping rules, interest is considered to be excessive if it
exceeds a financial formula based on the earnings of the corporation (often
25–30 percent of earnings before the deduction of interest, taxes, depreciation
and amortization, or “EBITDA”).
The problem of deductible interest payments that erode a country’s tax
base relates primarily to payments to nonresidents. Interest payments to
residents are not generally problematic because the residents receiving the
payments are usually taxable on those payments. However, EU countries are
prohibited from discriminating against residents of other EU countries, and
the European Court of Justice has ruled that thin capitalization rules that are
applicable only to interest payments to nonresidents are invalid insofar as
they apply to residents of other EU countries. Consequently, some European
countries have revised their thin capitalization rues so that they apply to all
interest payments by resident corporations, including such payments to
residents. Other countries have adopted earnings-stripping rules that apply to
all interest payments by resident corporations irrespective of the residence of
the recipient.
Some countries try to deal with the problem of excessive interest
deductions by adopting statutory thin capitalization or earnings-stripping
rules; others rely on administrative guidelines or practices. Still others have
applied transfer pricing or GAARs. The statutory rules of the various
countries differ considerably. In some countries, thin capitalization rules are
seen as specific transfer pricing rules for interest that are limited to interest
payments to related or non-arm’s-length parties. In other countries, the rules
are targeted at interest payments that are viewed as disguised dividends: in
other words, debt held by nonresident shareholders with a substantial interest
in a resident corporation. Other countries consider the rules to be aimed at
interest payments generally.
Although most countries’ thin capitalization rules are targeted at certain
interest payments – rather than all interest payments – to nonresidents, it must
be recognized that all deductible interest payments by residents to
nonresidents reduce or erode a country’s tax base. Nevertheless, not all base-
eroding payments are objectionable; many deductible payments by residents
to nonresidents, including interest, represent legitimate income-earning
expenses.

7.2.2 The Structural Features of Thin Capitalization and

Earnings-Stripping Rules
Typically, thin capitalization and earnings-stripping rules have most of the
following structural features.
Nonresident lenders. Thin capitalization and earnings-stripping rules
generally apply only to interest paid to nonresidents who own a significant
percentage of the shares of a resident corporation. The level of share
ownership varies from a substantial interest in the shares (10–25 percent) to
control (more than 50 percent of the shares) of the resident corporation.
However, some countries, such as Australia, also apply their rules to resident
corporations that use debt to finance foreign investment (so-called outbound
thin capitalization rules). Moreover, as noted above, several European
countries apply their rules to interest paid to both resident and nonresident
lenders.
Domestic entities. The thin capitalization rules of most countries apply
only to resident corporations. However, the stripping of profits through the
payment of excessive interest to related persons may also arise with respect to
partnerships and trusts and branches (PEs) of nonresident corporations. As a
result, countries are increasingly extending the application of their thin
capitalization rules to these entities.
Determination of excessive interest. Generally, thin capitalization and
earnings-stripping rules apply only to certain “excessive” interest paid to
nonresidents by resident corporations. Countries use a variety of different
approaches to determine what constitutes excessive interest; there is no
international consensus on this issue. The most common approach is the use
of a fixed debt:equity ratio, under which only interest on a corporation’s debt
that is artificially large in relation to its equity – in effect, debt that is
disguised equity – is not deductible. An alternative approach, recommended
by the OECD for tax treaties, attempts to characterize debt and equity by
reference to all the facts and circumstances, including the debt:equity ratio of
the resident corporation. According to the OECD, this approach is consistent
with the arm’s-length standard used for transfer pricing generally and avoids
the inflexibility and arbitrariness of applying a fixed debt:equity ratio.
Under the earnings-stripping rules used by the United States (US) and
several European countries, excessive interest is determined by reference to
the relationship between a corporation’s interest expenses and its income. A
corporation is generally not entitled to deduct interest paid to certain
nonresident shareholders to the extent that the interest exceeds a percentage
of its income. The US approach is a combination of an earnings-stripping
rule, under which interest in excess of 50 percent of a corporation’s income is
not deductible, and a thin capitalization rule, under which corporations that
have a debt:equity ratio of no greater than 1.5:1 are not subject to the
earnings-stripping rule. Under the German earnings-stripping rules, the
deduction of interest by Germanresident corporations that are part of a
corporate group is denied if the interest exceeds 30 percent of EBITDA,
unless the German corporation is excessively leveraged compared to the
group as a whole.
The OECD’s BEPS Action 4: Limiting Base Erosion Through Interest
Payments and Other Financial Payments recommends that countries should
restrict interest deductions based on the net interest expense of the worldwide
group as a percentage of the group’s earnings (EBITDA). Thus, interest
deductions of any resident corporation would be limited by reference to the
earnings of the group as a whole rather than by an arbitrary debt:equity ratio.
One of the difficulties with this approach is that it requires the tax authorities
to have information about the interest expenses and earnings of the
worldwide group.
One significant difference between the determination of excessive
interest on the basis of earnings or a fixed debt:equity ratio is that earnings
are sensitive to fluctuations in interest rates, whereas a fixed debt:equity rule
is not. Thus, although a corporation may be better able to carry additional
debt if interest rates decline, this fact is irrelevant under a fixed debt:equity
ratio. However, under an earnings approach, taxpayers have an incentive to
reduce their debt during periods of rising interest rates in order to avoid
restrictions on the deduction of interest.
Computation of a debt:equity ratio. A debt:equity ratio for purposes of
thin capitalization rules can be established either:

– as an arbitrary ratio, computed on a consolidated basis, ignoring any

intercompany debt and equity; or
– by reference to the average debt:equity ratio for all resident
corporations or all resident corporations engaged in a particular
industrial or commercial sector.

Most countries seem to use an arbitrary debt:equity ratio of 1.5:1 to 3:1,

sometimes with a higher ratio for financial institutions. The calculation of
debt and equity as components of the ratio necessitates many subsidiary tax
policy decisions. For example, should all debt held by nonresidents be taken
into account, or just debt held by substantial nonresident shareholders?
Should equity include contributed surplus or only share capital and retained
earnings? How should hybrid securities such as preferred shares be
classified? Should debt that is guaranteed by a nonresident shareholder be
taken into account? Should a corporation’s gross debt be reduced by any cash
on hand, especially since such cash may be earning interest income? A
similar issue arises under an earnings-stripping rule: Should the limitation on
interest deductions be based on a corporation’s gross or net interest?
Consequences. The effect of the application of the thin capitalization or
earnings-stripping rules is generally that excessive interest is not deductible.
In some countries, this excessive interest is treated as a dividend. In other
countries, excessive interest that is not deductible in one year can be carried
forward and deducted in a subsequent year, assuming that it is not subject to
the limitation on the deduction of interest in that year.
Tax authorities must be aware that their thin capitalization rules can be
avoided if taxpayers channel their intercompany loans through back-to-back
arrangements with international banks and other financial intermediaries.
Assume, for example, that FCo, organized in Country F, lends money to B,
an unrelated bank, which then relends the money to ACo, organized in
Country A. FCo and ACo are members of an affiliated group of companies.
The loan from B to ACo may not be subject to Country A’s thin
capitalization rules because it appears to be an arm’s-length loan. Therefore,
some countries’ thin capitalization rules contain provisions that attempt to
prevent the use of back-to-back loans and similar tax avoidance devices.

7.3 CFC RULES

7.3.1 Introduction
As noted in Chapter 3, section 3.3.1, one of the most effective ways for
residents to avoid tax on their worldwide income is the use of CFCs and other
legal entities to earn foreign source income. Domestic tax on foreign source
income can easily be deferred or avoided completely by establishing a
foreign corporation or other legal entity, such as a trust, to earn the income.
Because the foreign corporation or trust is generally considered to be a
separate taxable entity and not resident in the country where its controlling
shareholders or beneficiaries are resident, those shareholders or beneficiaries
are not taxable when the income is earned by the foreign corporation or trust.
When distributions from the corporation or trust are paid, the
shareholders or beneficiaries may be taxable by the residence country.
However, many countries exempt dividends from foreign corporations from
residence country tax if the dividends are received on shares owned by a
resident corporation with a substantial interest in the foreign corporation.
Even if the distributions are taxable, the residence country tax is postponed
until distributions are received, which may be several years after the foreign
entity earns the income. Thus, earning income through a foreign entity may
result in the deferral or complete avoidance of residence country tax. The
benefit is greatest when the foreign tax on the income of the foreign
corporation or trust is low or nil. Therefore, the problem arises primarily from
the establishment of CFCs or trusts in tax havens.
The problem of tax avoidance and deferral through the use of controlled
foreign entities is most pronounced with respect to passive investment
income because such income can be easily diverted to or accumulated in an
offshore entity in a tax haven. For example, assume that a corporation
resident in Country A earns interest income of 1,000 from bonds and the tax
rate in Country A is 40 percent. If the corporation establishes a wholly owned
subsidiary in a tax haven that does not impose tax, it can defer tax of 400 by
transferring the bonds to the subsidiary. The interest income derived by the
subsidiary may not be subject to Country A’s withholding tax, either because
the interest is not sourced in Country A or because the interest is exempt
from withholding tax. However, even if the interest is subject to Country A’s
withholding tax, the corporation can defer Country A’s tax to the extent of
the difference between the corporate tax rate and the withholding tax rate
(which may be substantial). The tax benefit from the transfer of the bonds to
the subsidiary will be even greater if Country A provides a participation
exemption for dividends received from foreign corporations.
Where a residence country imposes tax on distributions from foreign
corporations and other entities, residence country tax is deferred, but not
avoided entirely. When a foreign subsidiary distributes dividends to its
resident parent corporation or when the parent corporation disposes of its
shares in the foreign subsidiary, the residence country will presumably tax
the distribution or gain. Therefore, the benefit of deferral in any particular
case depends on the difference between the domestic and foreign tax rates,
the rate of return on the deferred taxes, and the period of deferral. Under
standard present value calculations, indefinite deferral is nearly equivalent to
exemption.
Several countries have adopted detailed statutory rules to prevent or
restrict the use of CFCs to defer or avoid domestic tax. The US was the first
country to adopt CFC rules (Subpart F) in 1962; the rules were based on
similar rules (the foreign personal holding company rules), adopted in 1937,
that were targeted at the use of foreign corporations by individuals. The
adoption of the Subpart F rules was very controversial. The rules that were
finally enacted represented a compromise between the original proposal to
eliminate deferral for all income of CFCs and the arguments of US-based
multinationals that deferral should be eliminated only with respect to clearly
passive income. Subpart F remains controversial today. For many years, US
multinationals have argued that the Subpart F rules are broader and tougher
than the CFC rules of other countries and that they put US multinationals at a
competitive disadvantage. In January 2001, the US Treasury issued a report
on Subpart F (The Deferral of Income Earned Through U.S. Controlled
Foreign Corporations: A Policy Study, December 2000), which concluded
that the basic policy of Subpart F was appropriate and there was no
convincing evidence that the international competitiveness of US
multinationals was adversely affected by the rules.
Since the US adopted Subpart F in 1962, several other capital-exporting
countries have enacted CFC rules to protect their tax base. As of 2015, over
thirty countries had enacted CFC rules. This is a well-established trend that
will likely continue.
The basic pattern of CFC legislation is similar in all countries. Resident
shareholders that control, or have a substantial interest in, a foreign
corporation established in a no-tax or low-tax country are subject to residence
country tax currently on their proportionate share of all or some of the
income of the foreign corporation, whether or not the income is actually
distributed to them. If a foreign corporation is engaged in legitimate
commercial activities offshore, however, the CFC rules do not generally
apply to the income generated by those activities. For example, assume that
ACo, a resident of Country A, owns all the shares of a CFC resident in a low-
tax country. The CFC earns passive income of 1,000 and pays tax of 100 on
that income to its country of residence. The CFC does not distribute any of its
after-tax income to ACo. If Country A has CFC rules, ACo would be subject
to tax by Country A on the CFC’s income of 1,000 (despite the fact that ACo
has not received any distribution of that income from the CFC) and would
receive credit for the tax paid by the CFC of 100. Thus, if Country A imposes
tax at a rate of 40 percent, ACo would pay tax of 300 (400 less a foreign tax
credit of 100), which is exactly the amount of tax that ACo would have paid
if it had earned the income directly.
The basic structure of CFC legislation reflects two competing policies.
First, there is a desire to prevent tax avoidance and to advance the traditional
goals of fairness and economic efficiency discussed in Chapter 1, section 1.3.
At the same time, countries generally do not want to interfere unreasonably in
the ability of resident corporations to compete in foreign markets. In every
country with CFC measures, there is a balancing of these two policies,
although the balance is struck differently in every country. Other than Brazil,
no country eliminates entirely the benefits from the use of CFCs, and most
countries limit the application of their CFC rules to CFCs established in low-
tax countries and to passive income earned by CFCs. In contrast, the
Brazilian CFC rules apply to all of the income, active and passive, of all
CFCs in which Brazilian residents own 20 percent or more of the shares,
irrespective of the country in which the CFCs are resident. The original New
Zealand CFC rules also differed markedly from other countries’ CFC rules
and came close to eliminating deferral entirely; they applied to all the
income, active and passive, of all CFCs controlled by New Zealand residents,
except those established in seven listed countries. However, the New Zealand
CFC rules were revised in 2010 to exclude active business income from the
application of the CFC rules.

7.3.2 Structural Features of CFC Rules

Although CFC rules vary considerably, several fundamental structural
aspects of the rules are the same in most countries. These aspects of the
taxation of CFCs are discussed below.

7.3.2.1 Definition of a CFC

With a few exceptions, countries restrict the scope of their CFC rules to
income derived by entities (1) that are nonresident, (2) that are corporations
or similar entitles taxed separately from their owners, and (3) that are
controlled by domestic shareholders or in which domestic shareholders have
a substantial interest. Entities (such as partnerships) that are taxable on a
conduit or flow-through basis are not within the scope of the CFC rules if the
resident partners of a foreign partnership are subject to residence country tax
on their share of the partnership’s income. The status of an entity as a
nonresident is established in accordance with the residence country’s normal
residence rules for legal entities (place of incorporation or place of
management, or both). The residence rules for corporations and legal entities
are discussed in Chapter 2, section 2.2.2.
Although it may seem strange, some countries, such as France, apply
their CFC rules to foreign branches or PEs. The extension of CFC rules to
foreign branches or PEs is necessary where a country exempts income earned
through a foreign branch or PE and the exempt income includes passive
income that, if earned by a CFC, would be subject to the country’s CFC rules.
The application of the CFC rules to foreign branches or PEs is not necessary
if the exemption for income earned through foreign branches or PEs is
limited to active business income (i.e., income that would not be subject to
the CFC rules if it were earned by a CFC).
Not surprisingly, given the title of “controlled foreign corporation”
rules, most CFC legislation applies only to foreign corporations that are
controlled by certain domestic shareholders. In general, control means the
ownership of more than 50 percent of the outstanding voting shares. Some
countries extend the concept of control to include ownership of shares having
a value equal to more than 50 percent of the total value of the outstanding
shares of the corporation. Other countries have rules that presume residents to
control a foreign corporation in certain circumstances even if they own less
than 50 percent of the voting shares. For example, the Australian and New
Zealand CFC rules deem a resident to control a foreign corporation if the
resident owns 40 percent or more of the voting shares of the foreign
corporation and no nonresident person has voting control of the corporation.
Only a few countries have adopted a de facto control test as a
supplement to the basic de jure control test. Under a de facto control test, a
resident taxpayer is considered to control a foreign corporation if, based on
all the facts and circumstances, the taxpayer has the means to control the
affairs of the corporation even where it does not have voting control. For
example, a taxpayer that owns 20 percent of the shares of a corporation may
have de facto control of the corporation if the rest of the shares are widely
held. A de facto control test involves considerable uncertainty for taxpayers
and is also difficult for the tax authorities to apply.
The rationale for the control requirement is fairness. It would be unfair
to tax resident shareholders on the undistributed income of a foreign
corporation if they do not have sufficient legal or actual power or influence
over the foreign corporation to determine the activities it engages in (i.e.,
whether it is subject to the CFC rules) and to require it to distribute its
income.
A few countries, such as Brazil, Denmark, and Portugal, have rejected
the complexity and the limitations of a control test. They apply their CFC
rules to foreign corporations in which residents have a substantial (20 percent
in the case of Brazil; 25 percent for Portugal and Denmark) ownership
interest. Until 2004, France applied its CFC rules to French residents owning
10 percent or more of the shares of a CFC.
Control for purposes of CFC rules includes indirect control. The CFC
rules cannot be avoided by having the shares of a tax haven corporation
owned by another foreign corporation that is controlled by residents. For
example, if a resident owns 60 percent of the voting shares of ACo, which in
turn owns more than 50 percent of the voting shares of a second foreign
corporation, BCo, BCo is considered to be a CFC of the resident. Indirect
control is usually determined by multiplying a taxpayer’s interest in one
corporation by the corporation’s interest in other corporations, and so on. For
example, if ACo owns 40 percent of the shares of BCo and BCo owns 30
percent of the shares of CCo, ACo is considered to own 12 percent of CCo.
However, if one corporation controls another corporation, that corporation
should be considered to own all the shares of any other corporations owned
by it. For example, if ACo owns 51 percent of the shares of BCo and BCo
owns 51 percent of the shares of CCo, ACo would be considered to own all
the shares of CCo owned by BCo, rather than just 26.01 percent (51 percent
51 percent). Therefore, ACo would be considered to control BCo and CCo
(and any corporations controlled by CCo).
Most countries also have constructive ownership rules to prevent
taxpayers from avoiding the CFC rules by fragmenting the ownership of
shares among related persons. For example, if one resident corporation owns
40 percent and another resident corporation owns 20 percent of the voting
shares of a foreign corporation, the foreign corporation will be a CFC of both
resident corporations if they are related because, for example, they are both
wholly owned subsidiaries of another resident corporation. Whether persons
are related for this purpose is determined under the country’s domestic law.
In some countries, control must be concentrated in a small number of
resident shareholders in order for the CFC rules to apply. For example,
Australia, Canada, and New Zealand require that control of a foreign
corporation must be concentrated in five or fewer resident shareholders.
Under the US Subpart F rules, only US shareholders owning at least 10
percent of the shares of the foreign corporation are counted in determining
whether the foreign corporation is a CFC. In other countries, such as Norway
and Germany, even foreign corporations that are widely held by resident
shareholders are considered to be CFCs.
The concentrated-ownership requirement is related to the rationale for a
control test. Whenever the shares of a foreign corporation are widely held by
resident shareholders, those shareholders are unlikely to be able to exercise
sufficient power over the corporation to determine its income-earning
activities or require it to make distributions. A concentrated-ownership
requirement requires constructive ownership rules and perhaps anti-
avoidance rules.

7.3.2.2 Designated Jurisdiction or Global Approach

The primary focus of CFC rules is tax haven entities. As a result, the CFC
rules of most countries are limited to CFCs located in countries that are
defined and designated to be tax havens (the “designated jurisdiction
approach”). However, a few countries, such as Brazil, Canada, and the US,
apply their CFC rules to certain specified categories of income earned or
received by a CFC, regardless of whether the CFC is resident in a tax haven
or a high-tax country (the “global approach”).
Under the designated jurisdiction approach, the residence of a foreign
corporation in a designated tax haven is crucial to the application of the CFC
rules. Because all foreign corporations established in countries that are not
designated as tax havens are exempt from the CFC rules, the compliance and
administrative burden of the rules is reduced compared to the global
approach. Under the global approach, the residence of the foreign corporation
is irrelevant. The theory behind this approach is that all countries, even
generally high-tax countries, have aspects of their tax system that permit the
earning of preferentially or low-taxed income. Under the designated
jurisdiction approach, the legislation usually provides a general definition of
what constitutes a tax haven; the tax authorities then supplement that
definition by issuing a list of countries that are regarded as tax haven
countries, or that are not regarded as tax havens. However, some countries
simply use a list of tax haven or non-tax haven countries without any
statutory definition of what constitutes a tax haven.
The general definition of a tax haven is invariably based on a
comparison of the taxes levied in the foreign country and in the residence
country. If the foreign country actually levies taxes at approximately the same
rates as the residence country, the foreign country should not be considered to
be a tax haven because it cannot be used to defer or avoid a significant
amount of residence country tax. The comparison of domestic and foreign tax
rates can be based on:

– nominal tax rates;

– effective tax rates; or
– the actual foreign tax paid by a particular CFC.

The use of nominal tax rates to identify tax havens is problematic

because it ignores generous deductions, exemptions, credits, or allowances
that may be afforded by a foreign country. The use of effective tax rates is
also problematic, because effective tax rates are difficult to determine and
would have to be determined annually for every country in which a CFC of a
resident corporation is resident. Moreover, just because a country has high
effective tax rates does not mean that a particular CFC resident in that
country may not be subject to low foreign taxes. A few countries use the
effective-tax-rate approach. Most countries, however, focus on the actual
foreign tax paid by a CFC.
The comparison of the actual foreign tax paid by a CFC and the notional
domestic tax that the CFC would have paid as a resident corporation is the
theoretically correct approach because it focuses on the situation of each
particular CFC. However, this approach imposes onerous compliance
burdens on taxpayers because the income of a CFC must be recomputed in
accordance with residence country rules in order to determine the amount of
the notional domestic tax.
The specific relationship between the foreign tax and the residence
country tax varies considerably. Some countries define a tax haven as a
country whose tax rate is less than 55 percent (Sweden), 60 percent (Finland),
66 2/3 percent (France and Norway) or 75 percent (Spain and the United
Kingdom (UK)) of the residence country rate. Other countries define a tax
haven simply by reference to the foreign rate. For example, Germany and
Japan define a tax haven as a country that levies tax of less than 25 percent;
Korea uses 15 percent. Even a small difference between the foreign and
domestic tax rates may be sufficient to induce resident taxpayers to shift
passive income (which is easily shifted) to a CFC in a foreign country.
As a result of the difficulties that arise in defining a tax haven by
comparing its tax rate to the domestic tax rate – in particular, uncertainty for
both taxpayers and tax authorities – most countries that use a designated
jurisdiction approach have supplemented their definition of a tax haven with
a list of tax haven countries, or non-tax haven countries, or both. The list may
be either legislative (included in the legislation making up the CFC rules) or
administrative (issued by the tax authorities). Such a list is intended to
provide taxpayers and tax officials with concrete guidance. The lists vary
widely: some are determinative (legally binding) of a country’s status as a tax
haven or a non-tax haven, while others merely establish rebuttable
presumptions. The more sophisticated lists recognize that a country that
generally is a high-tax country may nevertheless impose little or no tax on
certain types of income or entities. Consequently, such countries may be
placed on a non-tax haven list, subject to certain exceptions for preferentially
taxed income or entities. As a result, CFCs that do not qualify for those
countries’ low-tax regimes will be exempt from the CFC rules.
The global approach is more precise than the designated jurisdiction
approach. As mentioned above, under the global approach, the country of
residence of the CFC is irrelevant and therefore, every transaction engaged in
by all CFCs of all resident taxpayers must be examined in order to determine
the nature of the income from the transaction. If the corporation has
“tainted” income, as discussed below, the income is attributed to the
domestic shareholders of the corporation and is subject to tax in their hands,
with a credit for any foreign taxes on the income. Consequently, the CFC
rules potentially apply even to CFCs in high-tax countries if they earn tainted
income. In contrast, although the designated jurisdiction approach is not as
precise, it minimizes the compliance and administrative burdens of the CFC
rules.

7.3.2.3 Definition and Computation of Attributable Income

Some countries employ what may be called an entity approach in taxing the
income of a CFC to its domestic shareholders. Under this approach, the CFC
rules usually provide an exemption for certain CFCs that are engaged
primarily in genuine business activities. This exemption is discussed in
section 7.3.2.4 below. If a CFC does not qualify for any of the exemptions,
all its income is attributable to its domestic shareholders. If, however, the
CFC qualifies for the exemption, none of its income, even its passive income,
is attributable to its domestic shareholders. This all-or-nothing result is the
essential characteristic of the entity approach.
In contrast, other countries follow a transactional approach, under
which only certain types of income (referred to as “tainted income”) derived
by a CFC are subject to attribution. Under a transactional approach, each
transaction entered into by a CFC must be analyzed to determine whether it
produces tainted or other income and, for this purpose, tainted income is
determined by applying residence country tax rules. Although the entity
approach is less precise than the transactional approach, it minimizes the
compliance and administrative burden of CFC rules. Some countries use a
hybrid approach that combines elements of the transactional and entity
approaches. For example, some countries, such as Australia, New Zealand,
and the US, use a transactional approach but provide an exemption for CFCs
whose tainted income is less than a specified percentage of its total income.
Tainted income usually consists of passive investment income and base
company income. Passive income consists of dividends, interest, rents,
royalties, and capital gains. All countries with CFC rules consider passive
income to be tainted income, although they define passive income differently.
Perhaps the most difficult issue in defining passive income for purposes of
CFC rules is identifying situations in which passive income should be
classified as active business income. For example, interest earned by a
genuine financial institution is generally considered to be active business
income and therefore exempt from CFC rules. Similar issues arise with
respect to rents and royalties.
The term “base company income” is used to refer to any income, other
than passive income, that is considered to be tainted income for purposes of
CFC rules. The definition of base company income is often quite complex
and the scope of the definition for purposes of various countries’ CFC rules
varies considerably.
In general, there are three major components of base company income:

(1) Income derived by a CFC from the country in which its controlling
shareholders are resident. If such income is not taxable by the
residence country, that country’s tax base is eroded. Many
countries consider the erosion of their tax base by a CFC in this
 manner to be inappropriate, especially since in many situations the
income could be earned by the parent of the CFC directly.
(2) Income derived by a CFC from transactions with related parties.
The treatment of income from related-party transactions as tainted
income is usually intended to bolster a country’s transfer pricing
rules. Transfer pricing rules are intended to prevent the diversion of
income to related foreign corporations through the non-arm’s-
length pricing of sales, services, and other transactions. (See
Chapter 6.) These rules are notoriously difficult to enforce. By
treating such income as tainted income for purposes of CFC rules,
countries can avoid the necessity of applying their transfer pricing
rules.
(3) Income derived by a CFC from transactions outside the country in
which it is resident. The rationale for treating income from
transactions outside the CFC’s local market as tainted income
relates to considerations of international competitiveness. Income
from local market transactions is usually exempt because the
current imposition of residence country tax on such income of a
CFC would adversely affect the ability of the multinational
enterprise of which the CFC is part to compete in that country.
Where, however, the CFC derives income outside its local market,
the deferral of residence country tax is not necessary for the CFC to
compete in its local market. Moreover, where a CFC does business
in its country of residence, there are probably good commercial
reasons for it to be established there.

These three categories of base company income are not mutually

exclusive. Thus, for example, some countries limit the definition of base
company income to income derived from related-party transactions outside a
CFC’s country of residence or to income from the country in which the
controlling shareholders are resident as a result of related-party transactions.
One key aspect of the definition of tainted income concerns income
from intercompany transactions between CFCs. For example, significant tax
savings can be achieved if a CFC that is resident in a high-tax country pays
interest, royalties, or other deductible amounts to a CFC resident in a low-tax
country. In the absence of special rules, interest, royalties, and other amounts
received by a CFC would likely be considered to be passive income subject
to the CFC rules. However, many countries have adopted special rules to
exclude such intercompany payments from the scope of tainted income.
Thus, in general, multinational enterprises can establish group finance
companies or holding companies for intellectual property in order to reduce
foreign taxes without becoming subject to the CFC rules.
Any tainted income of a CFC that is attributable to its domestic
shareholders should be computed in accordance with domestic tax rules and
in domestic currency. This obligation presents many difficulties because of
differences between foreign and domestic tax laws. Generally, a CFC is not
allowed to consolidate its profits and losses with the profits and losses of
other CFCs of the same domestic shareholders.

7.3.2.4 Nature and Scope of Exemptions

Countries may provide a variety of exemptions that limit the scope of their
CFC rules. The most important of these exemptions are described below. The
exemptions vary depending on whether a country uses an entity or
transactional approach, as described above. All countries with CFC
legislation provide at least some of these exemptions.
Exemption for genuine business activities or active business income. An
exemption is usually granted, expressly or implicitly, for CFCs engaged
primarily or exclusively in genuine business activities or for active business
income earned by CFCs. Countries that use a transactional approach, as
described above, tax only the tainted income of a CFC. Inherent in this
approach is the exemption of active business income, since such income is
not considered to be tainted income. Other countries use an entity approach,
under which each CFC is tested and either all or none of its income is
attributed to its domestic shareholders. Under the entity approach, an
exemption is invariably provided for CFCs engaged primarily or almost
exclusively in genuine business activities. This exemption is generally
available only if: (1) the CFC is engaged in certain defined active businesses
or is not engaged in investment activities; (2) it has a substantial presence in
the foreign country; and (3) less than a certain percentage (usually 50
percent) of its income is tainted income (generally passive income and base
company income).
Only the CFC rules of Sweden and Brazil do not distinguish between
active business income and other income. Under their CFC rules, all the
income of a CFC that is not exempt is attributable to its domestic
shareholders. However, Sweden provides broad exemptions from the CFC
rules; virtually all CFCs established in countries with which Sweden has a tax
treaty, as well as other high-tax countries, are exempt.
Distribution exemption. To the extent that a CFC distributes dividends
out of its current profits to its resident shareholders that are subject to
residence country tax, there is arguably no need to apply the CFC rules, since
residence country tax is neither avoided nor deferred. Despite the theoretical
justification for a distribution exemption, no country currently provides such
an exemption from the CFC rules. Part of the reason for the absence of a
distribution exemption is that many countries exempt dividends from CFCs;
another reason is that distribution exemptions are surprisingly complex and
some countries that had initially adopted such exemptions later eliminated
them.
De minimis exemption. A de minimis exemption is frequently granted for
CFCs whose income (or tainted income) does not exceed a minimum amount.
De minimis exemptions vary widely, and several countries do not provide any
such exemption. The Canadian exemption is available only if the CFC’s
tainted income is CAN 5,000 (approximately USD 3,750 as of late 2015 – a
meaninglessly small amount) or less. Other countries provide sizeable de
minimis exemptions, although they are difficult to justify on tax policy
grounds. De minimis exemptions appear to be largely motivated by political
considerations in order to allay fears that CFC rules will impose significant
compliance costs on taxpayers in respect of relatively small amounts of
income subject to tax.
Other exemptions. A few countries provide an exemption for CFCs that
are not used for the purpose of avoiding or reducing tax (a “motive”
exemption). Although a motive exemption is perhaps consistent with the anti-
avoidance purpose of CFC legislation, it gives the tax authorities
considerable discretion. It may be a simple way of limiting the scope of the
rules without the legislative complexity necessitated by more specific
exceptions.
The UK provides an exemption for CFCs that earn low profit margins.

7.3.2.5 Resident Taxpayers Subject to Tax

In most countries, both individual and corporate shareholders are subject to
tax under the CFC rules; in a few countries, the rules apply only to resident
corporations. There does not appear to be any good reason why the rules
should not apply to individuals.
In most countries, the undistributed income of a CFC is attributed to
resident shareholders who own shares in the corporation at the end of its
taxation year. This approach may appear to be unfair because it taxes
shareholders on their pro rata share of the CFC’s income for the entire year
despite the possibility that they may have owned their shares for only part of
the year. However, this approach is much simpler, although less precise, than
determining a taxpayer’s share of the income of a CFC for part of a year. In
addition, once the end-of-the-year rule is well understood, persons acquiring
shares of a CFC can be expected to take the rule into account in setting the
purchase price of the shares.
In most countries, resident shareholders of a CFC are not taxable on
their share of the undistributed income of the foreign corporation unless they
meet a minimum share ownership requirement (usually 10 percent). The
reason for this exemption for shareholders with small investments in a CFC is
that they may not have sufficient influence over the foreign corporation to
require it to distribute its income or obtain access to the information
necessary to compute their share of the income. Nevertheless, these small
shareholders may be counted in determining whether a foreign corporation is
controlled by resident shareholders.

7.3.2.6 Relief Provisions

Typically, countries provide some relief from double taxation that may
otherwise occur from the operation of their CFC rules. Because the basic
taxing mechanism under CFC rules is to tax the resident shareholders of a
CFC on their share of its undistributed income, the possibility of double
taxation arises where the CFC’s income is subject to foreign tax and where a
shareholder receives dividends from a CFC or disposes of its shares in the
corporation. Most countries provide relief for foreign taxes and subsequent
dividends out of previously taxed income of a CFC, but few countries
provide relief for capital gains from the disposal of shares of a CFC that
reflect previously taxed income of the CFC. These double taxation issues are
illustrated in the following example.
PCo, a resident of Country P, owns all the shares of SCo, a resident of
Country S. In 2016, SCo earns passive income of 1,000 in Country S and
pays tax of 100 to Country S on its income. Under Country P’s CFC rules,
PCo is taxable on SCo’s income of 1,000 at a rate of 40 percent (or 400) and
qualifies for a foreign tax credit of 100 for the tax paid by SCo to Country S
of 100. In 2018, SCo pays a dividend to PCo of 900. Since PCo has already
been taxed on the income out of which the dividend was paid, the dividend
should be exempt from Country P tax if it is otherwise taxable under the laws
of Country P. If Country S imposes withholding tax of 10 percent on the
dividend, Country P should also provide relief for that withholding tax,
perhaps by allowing it to be carried back and claimed as a foreign tax credit
in 2016 or by allowing it to be claimed against any tax on CFC income for
2018 or future years.
If PCo sells the shares of SCo in 2018 without receiving a dividend, the
proceeds of sale will presumably reflect SCo’s after-tax income for 2016 of
900. Since PCo has already paid tax to Country P on that amount, it should
not be required to pay tax on the capital gain realized on the sale of the shares
of SCo. If no relief is provided by Country P in this situation, PCo is likely to
consider having SCo pay a dividend of 900 (which will probably be exempt
from tax by Country P, as explained above) to reduce the capital gain on the
sale of the shares.
Losses of a CFC are not generally attributable to its resident
shareholders. Most countries permit such losses to be carried forward and
deducted in computing the attributable income of the CFC in future years.
Most countries’ CFC rules do not provide specific relief from double
taxation resulting from the application of the CFC rules of two or more
countries. For example, assume that ACo, resident in Country A, owns all the
shares of BCo, resident in Country B. BCo owns all the shares of a
corporation that is established in a tax haven and earns passive income. If
Country A and Country B both have CFC rules, the passive income of the tax
haven corporation may be subject to tax to ACo by Country A and to BCo by
Country B. Although it may appear that Country A should give credit for the
tax levied by Country B pursuant to its CFC rules, Country A may take the
position that the passive income of the tax haven corporation was shifted
from Country A and should be taxable in Country A. Some countries provide
specific relief, by way of deduction or credit, for foreign taxes levied
pursuant to another country’s CFC rules. Other countries provide no relief,
although relief might be available under the mutual agreement procedure of
an applicable tax treaty. This double-tax problem is becoming more serious
as more countries adopt CFC rules.

7.3.3 Tax Treaties and CFC Rules

The relationship between tax treaties and CFC rules is controversial. In
Finland, France, Japan, Sweden, and the UK taxpayers have challenged the
application of CFC rules to foreign subsidiaries as a violation of an
applicable tax treaty. The taxpayers have argued that the business profits
article of the typical tax treaty (Article 7 of the OECD and UN Model
Treaties) provides that a country (the country with CFC rules) cannot impose
tax on the business profits of a corporation resident in the other country (even
if controlled by residents of the first country) except to the extent that the
corporation has a PE in the first country and the profits are attributable to the
PE. In response, the tax authorities have argued that under CFC rules, tax is
imposed on the resident shareholders of the foreign corporation, not the CFC,
and nothing in a tax treaty prevents a country from taxing its own residents.
The strength of these arguments varies depending on the particular
country’s CFC rules and the specific provision of the treaty. As noted above,
most countries’ CFC rules do not apply to active business income; moreover,
several countries do not apply their CFC rules to CFCs resident in treaty
countries. As a result, the potential conflict between tax treaties and CFC
rules is limited. The cases have been decided uniformly in favor of the tax
authorities, except in France, where the highest French court held that Article
7 of the France-Switzerland treaty prevented the application of the French
CFC rules to a Swiss subsidiary of a French corporation.
The Commentary on Article 1 of the OECD Model Treaty was revised
in 2003 to clarify that, according to the OECD, there is no conflict between
CFC rules and tax treaties; therefore, tax treaties do not prevent the
application of CFC rules. Further, the revisions to the Commentary clarified
that it is not necessary for countries with CFC rules to put an explicit
provision in their treaties allowing the application of CFC rules. A few
countries – Belgium, Luxembourg, the Netherlands, and Switzerland – have
registered their disagreement with this aspect of the Commentary. The
Commentary does caution countries not to apply their CFC rules to
companies resident in treaty countries that are subject to tax in those
countries comparable to the tax imposed by the resident country.
7.4 NONRESIDENT TRUSTS
Trusts are legal relationships under which the legal ownership and
management of property is separated from its beneficial ownership. Trusts
originated under English law as part of the common law and are recognized
under the laws of most common law countries. Some civil law countries have
adopted legislation to allow the establishment of trusts or trust-like
relationships. Typically, a trust involves a settlor (the person who establishes
the trust and transfers (settles) property to the trust) and a trustee (the person
who has legal ownership and management of the property for the benefit of
one or more beneficiaries (the persons who are the beneficial owners of the
property)). A trust is a particularly flexible arrangement because the settlor of
the trust may also be a trustee and a beneficiary. Trusts may be either
discretionary or nondiscretionary. Under a nondiscretionary trust, the
interests of the beneficiaries are fixed and cannot be altered without the
amendment of the trust. In contrast, under a discretionary trust, the trustee has
discretion with respect to the amount of income or capital payable to any
particular beneficiary.
The flexibility of trusts makes them difficult to tax, and the difficulty is
magnified with respect to nonresident trusts. In many countries that recognize
trusts, they are taxed as entities, at least to the extent that they accumulate
their income.
Beneficiaries are generally taxable on trust income that is distributed to
them, but not on distributions of the capital of the trust, which generally
includes the after-tax income of the trust (in other words, the income earned
or received by a trust in a year is usually added to the trust’s capital if it is not
distributed in the year).
If a resident of Country A establishes a trust in a tax haven (and many
tax havens, especially former UK colonies, have adopted flexible trust
legislation to facilitate this practice) for the benefit of family members who
are also resident in Country A, in the absence of special rules, Country A
may be unable to tax the income of the trust unless the beneficiaries receive
current distributions out of the trust’s income. The trust itself is not a resident
of Country A and therefore is not taxable by Country A except to the extent
that it derives income from Country A. In most cases, any income earned by
the trust in a year will be accumulated in the trust and distributed to the
beneficiaries only in subsequent years as tax-free capital distributions or after
they have ceased to be resident in Country A.
To prevent this type of tax avoidance, some countries have adopted
special rules that attempt to impose tax if a resident transfers property to a
nonresident trust with resident beneficiaries. Taxpayers may attempt to avoid
these rules by establishing foreign trusts with a recognized international
charity as the only named beneficiary, but with a power in the trustee or a
protector (usually a trusted friend or adviser who is a nonresident) to add new
beneficiaries to the trust at any time. Alternatively, some tax havens allow the
establishment of purpose trusts, which do not require any named
beneficiaries. In response, some countries have extended their nonresident
trust rules to tax the resident settlor (any resident who transfers property to a
nonresident trust) on the income of the trust. This measure may be considered
to be draconian because legally the settlor has no right to obtain any funds
from the trust. However, it is intended to stop residents from transferring
funds to nonresident trusts in the first place, in recognition of the difficulty
that countries have in taxing the income of such trusts, either by taxing the
trust or its resident beneficiaries.

7.5 FOREIGN INVESTMENT FUNDS

As discussed in section 7.3.2.1 above, CFC legislation generally applies only
to foreign corporations that are controlled by resident shareholders, and in
some countries, only to foreign corporations that are controlled by a small
group of resident shareholders. Moreover, the CFC rules of several countries
apply only to resident shareholders that own a minimum percentage (usually
5-10 percent) of the shares of the foreign corporation. As a result, it is
relatively easy for foreign investment companies, mutual funds, or unit trusts
to be established in low-tax countries without being subject to the CFC rules.
Such foreign investment funds allow resident taxpayers to defer domestic tax
on their passive investment income. Foreign investment funds may also
permit taxpayers to convert what would otherwise be ordinary income into
capital gains on the disposition of interests in the fund.
Several countries have enacted detailed legislation to prevent the
deferral of domestic tax through the use of foreign investment funds. For
some countries, the purpose of these rules is to prevent the avoidance of CFC
legislation. For some other countries, the foreign investment fund rules have
a much broader purpose: they are intended to eliminate the benefit of deferral
for all investments in passive foreign corporations and other entities that are
not subject to the CFC rules.
When thinking about foreign investment fund rules, it is useful to
compare the tax consequences of three alternative investments: (1) an
investment in a foreign investment fund; (2) an investment in a domestic
investment fund; and (3) a direct foreign investment (e.g., purchase of a
foreign bond or rental real property located offshore). The essential
difference between the tax consequences for a resident investing in a
domestic investment fund as compared to a foreign investment fund is that
residence country tax is deferred with respect to a foreign fund until the
resident receives distributions or disposes of the interest in the fund. In
contrast, domestic tax is customarily imposed on the income derived by a
domestic investment fund.
The benefit of an investment in a foreign investment fund compared to a
direct foreign investment is that the income from the fund can be effectively
converted into capital gains if the fund accumulates its income. This
conversion of investment income into capital gains may also occur with
respect to investments in shares of resident corporations. However, resident
corporations are subject to current corporate tax in the residence country on
their income, whereas foreign corporations are not.
As a result, the tax systems of many countries contain an incentive for
resident individuals to invest in foreign corporations as compared to resident
corporations whenever (1) foreign taxes on the foreign corporation’s income
are less than the domestic taxes on the equivalent amount of income of a
resident corporation, and (2) the foreign corporation accumulates at least part
of its income. The incentive is greatest where the foreign corporation is based
in a tax haven and accumulates all of its income.
Countries use several different approaches to deal with investments in
foreign investment funds, and in certain circumstances, some countries use
more than one method. The methods are described briefly below.
CFC rules. In some countries, such as Germany, the foreign investment
fund rules form part of the CFC rules. The CFC rules apply to small investors
in foreign corporations and other entities controlled by residents if the entities
earn primarily passive income.
Purpose test. Some countries, such as Canada, have a purpose-based
specific anti-avoidance rule to deal with residents that own interests in a
foreign investment fund. Thus, residents who hold an interest in a foreign
investment fund are taxable on imputed income if one of the primary
purposes for the acquisition or holding of the interest in the fund is to avoid
tax.
Mark-to-market method. A mark-to-market method is essentially an
accrual-based capital gains tax under which any increase or decrease in the
value of a resident taxpayer’s interest in a foreign fund must be included in
computing the taxpayer’s income for each year. For example, if a taxpayer’s
interest in a foreign investment fund has a value of 300 at the start of a year
and 500 at the end of the year, the taxpayer would be subject to tax on a gain
of 200. If the value of the interest declined to 200 at the end of the following
year, the taxpayer would have a loss of 300.
The mark-to-market method is easy to apply if the foreign investment
fund is actively traded on a stock exchange or if the fund provides
information on the current value of interests in the fund (usually for the
purpose of redeeming investors’ interests). In other circumstances, it may be
quite difficult to value interests in a foreign investment fund except when
they are sold.
Imputed income approach. Under an imputed income or deemed rate of
return approach, the resident taxpayer is considered to have earned income on
the amount invested in the offshore fund at a specified rate, irrespective of the
actual income earned by the fund. For example, if the specified rate of return
is 10 percent, an individual who invests 10,000 in the fund would be taxable
on deemed income of 1,000. Any deemed income for a year would then be
added to the cost of the interest in the fund. Thus, assuming no distributions
from the fund are made, the individual would be taxable on deemed income
of 1,100 (10 percent of 11,000) for the following year.
The advantage of the imputed income method is that it is simple to apply
and minimizes the compliance burden on taxpayers and the administrative
burden on the tax authorities because it is unnecessary for them to obtain
specific information about the income of the foreign investment fund.
However, the imputed income method may result in the under- or over-
taxation of investors.
Deemed distribution approach. Under this approach, resident
shareholders are subject to tax on their pro rata share of the income of the
foreign fund regardless of whether the income of the fund is distributed. This
approach is the same as the method of taxation under CFC rules: it requires
taxpayers to have access to sufficient information in order to compute their
share of the foreign investment fund’s income. As a result, it is sometimes
limited to taxpayers who own a substantial interest in the offshore entity.
Deferral charge approach. Under this approach, residence country tax is
not imposed until distributions are received or gains are realized. However, at
that time an interest charge is imposed to eliminate the benefits of deferral
that the taxpayer has enjoyed.
Most countries that have foreign investment fund rules generally apply
their rules only to foreign entities that earn primarily passive income or
whose assets consist primarily of passive assets such as marketable securities.
Exemptions are often provided for foreign entities that are principally
engaged in an active business or that distribute virtually all of their income
currently. The distinction between active and passive income or assets is a
difficult one to make in a satisfactory manner and usually requires complex
rules. Most countries provide double-tax relief for foreign taxes, actual
distributions, and capital gains on the disposition of ownership rights in the
fund. These relief mechanisms are similar to those provided in comparable
circumstances for income derived through CFCs (see section 7.3.2.6 above).
CHAPTER 8
An Introduction to Tax Treaties

8.1 INTRODUCTION
Bilateral tax treaties are an important feature of the international tax
landscape that serve as a bridge between the tax systems of the contracting
states. Over 3,000 bilateral income tax treaties are currently in effect, and the
number is growing. Most bilateral tax treaties are based in large part on the
OECD Model Treaty or the UN Model Treaty. The UN Model Treaty is
substantially similar to the OECD Model Treaty, but includes some
additional and different provisions that permit source countries to impose
more tax than is permitted by the OECD Model Treaty. Both of these models
are discussed below.
Sections 8.2 through 8.6 below provide an overview of several of the
most important general aspects of tax treaties, including the legal nature of
tax treaties, their relationship with domestic law, their objectives, and the
interpretation of treaties. The main features of the influential OECD and UN
Model Treaties are summarized in section 8.7 in order to give readers a basic
understanding of the provisions a typical tax treaty. Some special topics,
including treaty abuse, nondiscrimination, resolution of disputes, and
administrative cooperation, are examined in section 8.8.
Although this chapter focuses exclusively on income tax treaties, several
other types of treaties deal with tax issues. For example, countries that
impose estate or inheritance taxes may have treaties to eliminate double
taxation with respect to those taxes. In addition, as of June 30, 2015 over
ninety countries had signed the Multilateral Convention on Mutual
Assistance in Tax Matters, sponsored by the OECD and the Council of
Europe, which entered into force in 1995 and was significantly revised in
2011. This Convention deals with tax administration issues such as exchange
of information, assistance in the collection of taxes, and dispute resolution. In
addition, there are many types of treaties that deal primarily with non-tax
matters but also include tax provisions. These non-tax treaties include air
transportation agreements and trade and investment treaties; most of these
agreements contain carve-out provisions indicating that any income tax issues
will be dealt with exclusively under the income tax treaty between the
countries. The General Agreement on Tariffs and Trade (GATT), as
renegotiated in 1994, and the General Agreement on Trade in Services
(GATS), both of which were consolidated as part of the Agreement
Establishing the World Trade Organization in 1994, contain some important
provisions relating to income taxation, primarily designed to prevent the use
of income tax provisions as disguised trade barriers or export incentives.
An important recent development is the proliferation of Tax Information
Exchange Agreements (TIEAs), typically between high-tax countries and
low- or no-tax countries with which the high-tax countries would not
otherwise have a comprehensive income tax treaty. In general, TIEAs require
the low- or no-tax countries to exchange information on the same basis as
provided in Article 26 of OECD and UN Model Treaties.
Income tax treaties are invariably bilateral, rather than multilateral.
Although proposals have been made from time to time for a multilateral
income tax treaty, to date multilateral agreements have been limited to
administrative issues. Countries seem to prefer customized agreements with
each treaty partner that take into account the cross-border trade and
investment flows between them and their income tax systems. However,
trade and investment treaties, such as the GATT and the GATS, are
multilateral agreements, and there is no legal impediment to a multilateral
income tax treaty. In fact, a multilateral agreement is a much more efficient
method of revising the vast network of bilateral treaties than renegotiating
each treaty. In this regard, in BEPS Action 15: Developing a Multilateral
Instrument to Modify Bilateral Tax Treaties, the OECD has recently proposed
producing a multilateral treaty to implement the changes to tax treaties that
were recommended as part of the BEPS project (see section 8.8.2.3 below).

8.2 LEGAL NATURE AND EFFECT OF TAX TREATIES

8.2.1 Vienna Convention on the Law of Treaties

All treaties, including tax treaties, are governed by the Vienna Convention on
the Law of Treaties (“Vienna Convention”), which was concluded on May
23, 1969 and entered into force on January 27, 1980. Although the Vienna
Convention has not been signed by some countries (most notably, the United
States (US)), it is generally considered to be binding on all nations because
its provisions are a codification of the principles of customary international
law dealing with treaties. All nations are considered to be subject to the
principles of customary international law.
Treaties are agreements between sovereign nations. According to Article
2 of the Vienna Convention:
A treaty is an international agreement (in one or more instruments, whatever called)
concluded between states and governed by international law.

Thus, it does not matter that some tax treaties are called “conventions”
and others are called “agreements.”
Tax treaties confer rights and impose obligations on the parties or
signatories to the treaty, called the contracting states. In most countries, they
do not confer rights on citizens or residents of the two states unless and until
the provisions of the treaty have been incorporated into the domestic laws of
the contracting states in some manner. The methods for incorporating treaties
into domestic law vary from country to country. In most Commonwealth
countries, tax treaties are usually incorporated into domestic law by means of
domestic legislation. In other countries, tax treaties are self-executing; they
become part of domestic law once they are concluded and ratified by the
responsible government officials. In other countries, treaties are subject to a
special legislative process; for example, in the US, tax treaties entered into by
the executive branch (the President) must receive the advice and consent of
the US Senate before they become effective.
Article 26 of the Vienna Convention contains the pacta sunt servanda
principle, under which treaties are binding on the contracting states and must
be performed by them in good faith. Such a fundamental principle is self-
evident. Treaties are binding agreements between sovereign states and must
be respected by them – countries are unlikely to be interested in entering into
treaties with countries that do not adhere to their obligations. Unfortunately,
although the pacta sunt servanda principle is essential for treaties to operate
as intended, there have been instances where countries have not respected the
provisions of their tax treaties.
Reciprocity is a fundamental underlying principle of tax treaties,
although its precise meaning is unclear. The provisions of almost all bilateral
tax treaties are reciprocal. For example, if the rate of tax on dividends,
interest, and royalties under a treaty is limited to 10 or 15 percent, that rate
invariably applies equally to payments of these amounts by residents of one
contracting state to residents of the other contracting state and by residents of
the other state to residents of the first state. Moreover, the reciprocal
limitation of the rates of tax imposed on dividends, interest, and royalties by
the contracting states applies notwithstanding that the flows of these
payments between the two contracting states may be unequal. The application
of the principle of reciprocity is especially difficult with respect to provisions
such as exchange of information and assistance in the collection of tax. These
provisions impose potentially costly obligations on states. Although the
provisions apply in the same manner in both states, does the principle of
reciprocity require that both states make reasonably equal use of the
provisions, or is it acceptable for one state to make disproportionate or even
exclusive use of the provisions?

8.2.2 The Relationship between Tax Treaties and Domestic

Law
The relationship between tax treaties and domestic tax legislation is much
more complex than many commentators and tax professionals realize. Many
of them think that the relationship consists of nothing more than the basic
principle that a treaty prevails in the event of a conflict between the
provisions of domestic law and the treaty. In fact, the relationship between
tax treaties and domestic law is complex, involving the following issues:

(1) the effect of tax treaties on subnational taxes;

(2) whether tax treaties limit domestic tax, allocate taxing rights, or
impose tax;
(3) the limited impact of tax treaties in that they do not displace
domestic law entirely;
(4) the incorporation of meanings of terms in domestic law into tax
treaties;
(5) the incorporation of tax treaty terms into domestic law; and
(6) domestic laws that override the provisions of tax treaties.

These issues are discussed briefly below.

In general, tax treaties apply to all income taxes imposed by the
contracting states, including taxes imposed by provincial (state), local, and
other subnational governments. In some federal states, however, the central
government is prevented by the constitution or established tradition from
entering into tax treaties that limit the taxing powers of their subnational
governments. Accordingly, the tax treaties of some federal states, such as
Canada and the US, apply only to national taxes. In such circumstances, a
subnational government may impose taxes that contravene the provisions of
an applicable tax treaty despite the fact that the central government could not
impose similar taxes.
In general, tax treaties do not impose tax, nor do they allocate taxing
rights between the contracting states – the right to tax is derived from the
domestic law of a state. Tax treaties limit the taxes otherwise imposed by a
state; in effect, they are primarily relieving in nature. France and several
African countries that follow the French practice are notable exceptions in
this regard because taxes may be imposed pursuant to treaty provisions even
where they are not imposed under domestic law. In other words, these
countries impose tax on amounts that a treaty allows them to tax, despite the
fact that those amounts may not be taxable under their domestic law. In
contrast, for most countries, if an amount is not taxable under domestic law,
that is the end of the matter; it is unnecessary to refer to the treaty.
The provisions of tax treaties do not displace the provisions of domestic
law entirely. For example, a person who is considered to be a resident of both
Country A and Country B under their domestic laws may be deemed to be a
resident of Country A pursuant to the tie-breaker rule in the treaty between
Country A and Country B. Article 4(2) (Resident) of both the OECD Model
and UN Model Treaties provides a series of tie-breaker rules to make a dual-
resident individual a resident of only one country for purposes of the treaty.
However, although the individual may be considered to be a resident of
Country A for purposes of the treaty, the individual will remain a resident of
Country B under its domestic law for all purposes not affected by the treaty.
If, for example, the individual makes payments of dividends, interest, or
royalties to nonresidents, the person may be subject to any withholding
obligations imposed by Country B on such payments made by residents of
Country B.
Many tax treaty provisions include explicit references to the meaning of
terms under domestic law. As a result, the meanings of terms in a tax treaty
are determined by reference to the meanings of those terms under the
domestic law of the contracting states. For example, under Article 6 (Income
from Immovable Property) of both the OECD and UN Model Treaties,
income from immovable or real property located in a country is taxable by
that country. For this purpose, the term “immovable property” is defined in
Article 6(2) to have the meaning that it has under the domestic law of the
country in which the property is located. In addition, Article 3(2) (General
Definitions), which is discussed below, provides that any undefined terms in
a treaty should be given the meaning that they have under the law of the
country applying the treaty. Conversely, in some countries where the
domestic law uses terms that are also used in the treaty, the meaning of those
terms for purposes of domestic law may be interpreted in accordance with the
meaning of the terms for purposes of the treaty. In effect, in these
circumstances, treaty meanings are incorporated into domestic law.
As noted above, as a general rule, the provisions of tax treaties prevail in
the event of a conflict with the provisions of domestic law. In some countries,
this principle is enshrined in the constitution; in other countries, it is enacted
as a statutory rule; in yet other countries, treaties prevail over other laws
because they are considered to be special (lex specialis). In the US, the basic
rule for resolving conflicts between statutes and treaties is that the later-in-
time prevails. Except in countries that give constitutional priority to treaties,
it is possible for countries to adopt legislation that takes priority over their tax
treaties. Such legislation is often referred to as a treaty override. For
example, some countries have passed legislation to modify or overturn the
interpretation of a tax treaty by its domestic courts, perhaps on the basis that
the court decisions are inconsistent with the Commentary on the OECD or
UN Model Treaties or the intentions of the contracting states. Such
legislation, adopted in good faith, may not violate a country’s obligations
under its tax treaties. A country contemplating a treaty override may consult
with its treaty partners in advance to demonstrate good faith and prevent
misunderstandings.
Occasionally, some countries have included treaty overrides in
legislation to prevent taxpayers from arguing in court that the countries’ tax
treaties prevent the application of the legislation. This type of treaty override
is very controversial. Tax treaties are solemn obligations that should not be
disregarded except in extraordinary circumstances. If a country becomes
dissatisfied with the provisions of a tax treaty, the appropriate remedy is the
renegotiation or termination of the treaty. At the same time, countries must
have the ability to amend the provisions of their domestic tax legislation to
keep it current and to clarify interpretive difficulties concerning the
relationship between the treaty and domestic law.

8.3 THE OECD AND UN MODEL TAX TREATIES

There are two influential model tax treaties – the OECD Model Tax
Convention on Income and on Capital (available at www.oecd.org) and the
United Nations Model Double Taxation Convention between Developed and
Developing Countries (available at
www.un.org/esa/ffd/documents/UN_Model_2011_Update.pdf), the most
recent versions of which are 2014 and 2011 respectively. Some countries
have their own model tax treaties, which are often not published but are
provided to other countries for the purpose of negotiating tax treaties. The US
has published its model treaty, available at www.irs.gov. The UN Model
Treaty and the various country models are broadly similar to the OECD
Model Treaty and can be viewed as modified versions of the OECD Model
Treaty rather than as separate models.
The OECD Model Treaty has a long history, beginning with early
diplomatic treaties of the nineteenth century. The limited objective of those
treaties was to ensure that diplomats of one country working in another
would not be discriminated against; they were extended to cover income
taxation once income taxes began to be widely adopted in the early part of
the twentieth century. After the First World War, the League of Nations
commenced work on the development of a model treaty dealing exclusively
with income tax issues. This work culminated in draft model conventions in
1943 and 1946. These conventions were not unanimously accepted, and the
work of creating an acceptable model treaty was taken over by the OECD.
Currently, the OECD has thirty-four members, consisting of most of the
major industrialized countries. Membership in the OECD has recently been
extended to Chile, Estonia, Israel, and Slovenia. It seems likely that several
other countries, such as Colombia, Costa Rica, Latvia, and Lithuania will
become members in the near future.
The OECD Model Treaty was first published, in draft form, in 1963. It
was revised in 1977 and again in 1992. In 1992, the OECD decided that the
Model Treaty should be ambulatory, with more frequent, periodic updates
rather than complete revisions at less frequent intervals. Consequently, since
1992 the OECD Model Treaty has been revised nine times: in 1994, 1995,
1997, 2000, 2002, 2005, 2008, 2010 and, most recently, 2014. The
Committee on Fiscal Affairs, which consists of senior tax officials from the
member countries, has responsibility for the Model Treaty as well as other
aspects of international tax cooperation. The Committee on Fiscal Affairs
operates through the Centre for Tax Policy and Administration, which was
created in early 2001; it has a large staff that deals with various aspects of
international taxation, including tax treaties, and oversees several Working
Parties. The Working Parties consist of delegates from the member countries;
Working Party No. 1 is responsible for the Model Treaty and examines issues
related to the treaty on an ongoing basis.
The work of the United Nations on a model treaty commenced in 1968
with the establishment by the UN Economic and Social Council (ECOSOC)
of the UN Ad Hoc Group of Experts on Tax Treaties between Developed and
Developing Countries. The Group of Experts produced a Manual for the
Negotiation of Bilateral Tax treaties between Developed and Developing
Countries, which led to the publication of the UN Model Taxation
Convention between Developed and Developing Countries in 1980. The UN
Model Treaty was revised in 2001 and again in 2011. In 2004, the Group of
Experts became the Committee of Experts on International Cooperation in
Tax Matters. In addition to maintaining the UN Model Treaty and its
Commentary, the Committee has published several useful reference works
dealing with international tax, including a manual on transfer pricing for
developing countries and a handbook on the administration of tax treaties.
The members of the Committee are tax officials nominated by their
governments and appointed by the Secretary-General of the UN. They serve
for four-year terms in their individual capacity rather than as representatives
of their governments; however, the practical reality is that many members of
the Committee usually adopt positions that are consistent with those of their
governments. A majority of the members of the Committee are from
developing countries and countries with economies in transition. The
members of the Committee are tax specialists and include several treaty
negotiators.
The UN Model Treaty follows the pattern set by the OECD Model
Treaty, and many of its provisions are identical, or nearly identical, to those
of the OECD Model Treaty. The chief difference between the two models is
that the UN Model Treaty imposes fewer restrictions on the taxes that may be
imposed by developing countries. For example, the UN Model Treaty does
not contain specific limitations on withholding tax rates on dividends,
interest, and royalties imposed by source countries; instead, the withholding
rate levels are left to bilateral negotiations between the contracting states.
Similarly, as discussed in section 8.7.3.2 below, the UN Model Treaty allows
source countries to tax more cross-border business profits than the OECD
Model by lowering the threshold for a PE.
A detailed Commentary, organized on an article-by-article basis,
accompanies both the OECD and the UN Model Treaties. In particular, the
OECD Commentary has become increasingly important with respect to the
interpretation and application of tax treaties, including treaties between
countries that are not members of the OECD. To take into account the
positions of some non-member states, the OECD opened up the Commentary
in 1997 for non-member countries to register their positions on the provisions
of the OECD Model Treaty and its Commentary. As of 2014, thirty-three
countries have done so.
In general, and in comparison to the UN Model Treaty, the OECD
Model Treaty favors capital-exporting (residence) countries over capital-
importing (source) countries. Often it eliminates or mitigates double taxation
by requiring the source country to give up some or all of its taxing rights on
certain categories of income earned by residents of the other treaty country.
This aspect of the OECD Model is appropriate if the flow of trade and
investment between the two countries is reasonably equal and the residence
country taxes any income that the source country exempts from tax. The
following example illustrates this point.
Country A and Country B are both developed countries contemplating a
tax treaty. Both countries tax their residents on a worldwide basis, and both
provide their residents with a foreign tax credit for withholding taxes paid
with respect to foreign source income. Country A has one taxpayer, Taxpayer
A, who earns royalties of 1,000 from Country B. Country B likewise has one
taxpayer, Taxpayer B, who earns royalties of 1,000 from Country A. Absent
a treaty, Country B would impose a 15 percent withholding tax on royalties
paid to Taxpayer A, and Country A would do the same with respect to
royalties paid to Taxpayer B. Country B would allow Taxpayer B to claim a
foreign tax credit for the withholding taxes paid to Country A, and Country A
would allow a credit for the comparable taxes paid by Taxpayer A to Country
B. If the two countries agree in their tax treaty to reduce withholding at
source on royalties to a rate of zero for residents of the other contracting
state, each country would thereby lose source-tax revenue of 150 (1,000 ×
0.15). They would recoup the lost source-tax revenue, however, by collecting
150 in additional tax from their own residents.
Two important points may be drawn from the above example. First, a
country that gives up the domestic tax that it imposes on income derived from
its territory cannot expect to recoup the lost revenue from an expansion of its
residence jurisdiction if it uses an exemption system to relieve international
double taxation.
Second, a trade off of domestic tax based on the source of income for
tax, which is based on the residence of the taxpayer, is likely to be
unfavorable for a country that is a net importer of capital. Of course,
investment flows between two contracting states are never as exact as in the
above example; some deviations from strict reciprocity are acceptable,
especially for countries with a network of tax treaties. If a group of countries
enters into bilateral treaties based on the OECD Model Treaty, what is
important is that the investment flows within that group be roughly in
balance. In such circumstances, a country that loses revenue under some of
its treaties can expect to recoup the revenue under other treaties.
Developing countries are net capital importers, and many of them use
the exemption method for granting double taxation relief to their resident
taxpayers. Consequently, developing countries entering into a tax treaty with
a developed country would not benefit from the trade-off of taxation based on
the source of income for taxation, based on the residence of taxpayers,
contained in the OECD Model Treaty. As noted above, because of the
shortcomings of the OECD Model Treaty, developing countries devised their
own model treaty under the auspices of the UN.
The success of the OECD Model Treaty has been remarkable – virtually
all existing bilateral tax treaties are based on it. As noted above, even the UN
Model Treaty and US Model Treaty follow the basic pattern of the OECD
Model Treaty. This wide acceptance of the OECD Model Treaty, and the
resulting standardization of many international tax rules, has been an
important factor in reducing international double taxation and facilitating
international trade and investment.
Nevertheless, the OECD Model Treaty has several deficiencies. For
example, some provisions are intentionally vague in order to disguise
disagreements among OECD member countries. Also, many important
aspects of international tax, such as foreign-currency gains and losses,
sophisticated financial arrangements, cross-border reorganizations, and
corporate integration methods, are not dealt with in the Model Treaty at all.
Moreover, in some ways the OECD Model Treaty is a victim of its own
success. Changing the articles of the OECD Model Treaty to correct flaws
and respond to new developments is extremely difficult. One source of this
difficulty is that countries can adopt revisions of the OECD Model Treaty
only by renegotiating their existing treaties, which is time-consuming,
especially for countries with large treaty networks.
Another source of difficulty is the OECD tradition of amending the
Model Treaty only with the unanimous consent of all OECD members. The
practical significance of the unanimity rule is diminished because member
countries that disagree with any aspect of the Model Treaty can register a
reservation to the particular provision of the Model Treaty. These
reservations are found in the Commentary on the Model Treaty. A
reservation indicates that the country does not intend to adopt the particular
provision of the OECD Model Treaty in its tax treaties. Most countries have
entered reservations on some aspects of the Model Treaty. For example,
fourteen member countries have entered reservations on Article 12 dealing
with royalties by asserting their intention to levy withholding taxes on
royalties.
The Commentary also contains observations by particular countries on
specific aspects of the Commentary. Observations are usually used to indicate
that a particular country’s interpretation of a provision of the Model Treaty or
a part of the Commentary differs from the interpretation of the majority of the
member countries. A country making an observation does not reject the
particular article of the Model Treaty. The purpose of the observation is to
indicate that the country will include the provision in its treaties, but will
interpret and apply the provision in a manner different from the view
expressed in the Commentary. A country is not expected to enter an
observation if the Commentary sets out alternative positions and the country
adopts one of those positions.
The Commentary on the OECD Model Treaty is much easier to change
than the articles of the Model Treaty itself. According to the Commentary,
the views expressed in the Commentary, subject to any reservations or
observations, should be taken into account in interpreting and applying all
treaties between Member States, even those treaties entered into before the
revisions of the Commentary. This practice of applying revisions of the
Commentary to previously existing tax treaties raises some interesting
questions of interpretation that are dealt with in section 8.6 below.

8.4 THE PROCESS OF NEGOTIATING AND REVISING

TAX TREATIES
The negotiation of a tax treaty typically begins with initial contacts between
the countries. Usually, a country will consider many factors, including the
level of trade and investment with another country, in deciding whether to
enter into negotiations for a tax treaty with that country. Once the countries
have decided to negotiate, they exchange their model treaties (or their most
recent tax treaties, if they do not have a model treaty) and schedule face-to-
face negotiations.
Traditionally, treaties are negotiated in two rounds, one in each country.
During the first round of negotiations, the negotiating teams agree on a
particular text – usually the OECD Model or UN Model Treaty – to use as the
basis for the negotiations. After presentations by both sides about their
domestic tax systems, the negotiations proceed on an article-by-article basis.
Aspects of the text that cannot be agreed on are usually placed in square
brackets, to be dealt with later. Once the wording of an article is agreed on,
the parties initial it. Once all the articles have been agreed on by the treaty
negotiators, arrangements are made for the treaty to be signed by an
authorized official (often an ambassador or government official). After
signature, each state must ratify the treaty in accordance with its own
ratification procedures. The treaty is concluded when the countries exchange
instruments of ratification. The treaty enters into force in accordance with the
specific rules in the treaty (Article 30 (Entry into Force) of the OECD Model
Treaty and Article 29 (Entry into Force) of the UN Model Treaty).
Thus, the process for the adoption of a tax treaty involves several
separate steps or stages: signature, ratification, conclusion, and entry into
force. Each of these steps has a special meaning and particular consequences.
Once a treaty has been adopted, it may be modified in minor or major
ways by the mutual consent of the contracting states. It is common for a tax
treaty to be amended by the parties entering into a Protocol to the treaty.
Under the provisions of the Vienna Convention as discussed in section 8.2.1
above, an agreement designated as a Protocol is simply a treaty under a
different name. Thus, as described above, it must be ratified under the rules
applicable to treaties before it becomes effective.
Tax treaties require updating just like domestic tax laws. In practice, the
amendment process is often exceedingly slow and difficult – it is not
uncommon for a Protocol to take as long to negotiate as a treaty. Often, once
one aspect of a treaty is opened up for renegotiation, other aspects of the
treaty become negotiable. Consequently, if a country has a broad network of
tax treaties – some have tax treaties with over 100 countries – the
renegotiation of the entire network could take decades.
To a limited degree, tax treaties may be updated without a formal
amendment procedure through the interpretive process. For example, as
discussed in section 8.3 above, the OECD Commentary is frequently updated
to clarify the meaning of the articles of the treaty, and the OECD takes the
position that the revisions to the Commentary should be applied to tax
treaties that were entered into before those revisions were made (see
paragraphs 33-36 of the Introduction to the OECD Model Treaty). In
addition, Article 25 (Mutual Agreement Procedure) of both the OECD and
UN Model Treaties authorizes the competent authorities of the two states to
resolve issues of interpretation. The general rules for interpreting treaties are
discussed below in section 8.6.

8.5 OBJECTIVES OF TAX TREATIES

The fundamental objective of tax treaties, broadly stated, is to facilitate cross-
border trade and investment by eliminating the tax impediments to these
cross-border flows. This broad objective is supplemented by several more
specific operational objectives.
From the perspective of taxpayers, the most important operational
objective of bilateral tax treaties is the elimination of double taxation. Several
provisions of the typical bilateral tax treaty are directed at the achievement of
this goal. For example, tax treaties contain tie-breaker rules (Article 4(2) and
(3) of the OECD and UN Model Treaties) to deem a taxpayer who is
otherwise resident in both countries to be a resident of only one of the
countries. They also limit or eliminate the source country tax on certain types
of income (Articles 10, 11 and 12 of the OECD and UN Model Treaties) and
require residence countries to provide relief for source country taxes, either
by way of a foreign tax credit or an exemption for the foreign source income
(Article 23 of the OECD and UN Model Treaties). The mechanisms for
granting relief from double taxation are discussed in Chapter 4, section 4.3.
In the mid- twentieth century, the focus of tax treaties was almost
exclusively on solving the problem of double taxation. This focus was
reflected in the title of the 1963 and 1977 OECD Model Treaties:
“Convention between (State A) and (State B) for the avoidance of double
taxation with respect to taxes on income and on capital.” At that time,
multinational enterprises were facing risks of substantial double taxation, few
countries provided unilateral relief for double taxation, and widespread treaty
networks were just starting to develop. Treaty solutions to most of the major
double-tax problems were worked out in the 1950s and early 1960s, however,
and they are now routinely accepted by states when they enter into tax
treaties.
The historical emphasis on the elimination of double taxation should not
obscure the fact that most tax treaties have another equally important
operational objective – the prevention of tax evasion and avoidance. This
objective is the converse of the elimination of double taxation. Tax treaties
are intended to eliminate double taxation of cross-border income, but are not
intended to facilitate double non-taxation. The underlying assumption of tax
treaties is that cross-border income should be taxed, but should be taxed only
once.
Originally, the OECD and UN Model Treaties included a preamble
stating that the treaty was intended to eliminate double taxation and prevent
fiscal evasion. The meaning of the term “fiscal evasion” was unclear; some
countries interpreted it broadly to include tax avoidance while others, such as
Switzerland, restricted the term to criminal tax evasion. The references in the
preamble to the avoidance of double taxation and prevention of fiscal evasion
were eliminated from the OECD and UN Model Treaties and moved to a
footnote in 1992 and 2001 respectively. However, in 2003 the Commentary
on the OECD Model Treaty was revised to include an explicit statement that
“[I]t is also a purpose of tax conventions to prevent tax avoidance and
evasion” (paragraph 7 of the Commentary on Article 1). Although the
Commentary on the UN Model Treaty does not reproduce this sentence from
the OECD Commentary, it is reasonably clear from the UN Commentary on
Article 1 that one of the important purposes of the treaty is to prevent tax
avoidance and the improper use of the treaty.
The increasing focus on the use of tax treaties to facilitate tax avoidance
led the OECD, as part of the BEPS project, to recommend changes to the title
and preamble of the OECD Model Treaty to refer explicitly to the prevention
of tax avoidance (see BEPS Action 6: Preventing the Granting of Treaty
Benefits in Inappropriate Circumstances, available at www.oecd.org and
discussed in section 8.8.2.3). The proposed title of the OECD Model Treaty
will be “Convention between (State A) and (State B) for the elimination of
double taxation with respect to taxes on income and on capital and the
prevention of tax evasion and avoidance.” The proposed preamble will also
include explicit references to double non-taxation and treaty shopping:
Intending to conclude a Convention for the elimination of double taxation with respect to
taxes on income and on capital without creating opportunities for non-taxation or reduced
taxation through tax evasion or avoidance (including through treaty-shopping arrangements
aimed at obtaining reliefs provided in this Convention for the indirect benefit of residents of
third States).

Although the elimination of tax avoidance and evasion is an explicit

objective of most tax treaties, few provisions in those treaties are designed to
achieve that objective. Both the OECD and UN Model Treaties contain
provisions for the exchange of information (Article 25) and assistance in the
collection of taxes (Article 27). These provisions give the contracting states
two important tools to prevent tax avoidance and evasion; however, neither
Model Treaty currently contains any general anti-avoidance rule, and both
Model Treaties contain few specific anti-avoidance rules. The OECD BEPS
Action 6 proposes to add a detailed anti-treaty shopping rule similar to the
limitation-on-benefits provision included in US tax treaties; it also proposes
to add a general anti-abuse provision to the OECD Model, under which treaty
benefits would be denied if one of the principal purposes of a transaction or
arrangement was to avoid tax, unless the treaty benefits are in accordance
with the object and purpose of the treaty. See sections 8.8.2.1 and 8.8.2.2 for
a discussion of treaty abuse and treaty shopping respectively.
In addition to the two principal operational objectives of tax treaties,
there are several other ancillary objectives. One ancillary objective, which is
discussed in section 8.8.1, below, is the elimination of discrimination against
foreign nationals and nonresidents. Most countries entering into tax treaties
want to ensure that their residents are treated the same as residents of the
other contracting state, and certainly no worse than residents of any third
state. Other ancillary objectives, discussed in section 8.8.4 below, are the
exchange of information between the contracting states and assistance in the
collection of taxes. As noted above, allowing countries to obtain information
about the income-earning activities of taxpayers is an important tool in
combating tax avoidance and, more generally, in ensuring that the provisions
of the treaty are applied properly. Similarly, requiring countries to provide
assistance in collecting the taxes imposed by their treaty partners can prevent
taxpayers from avoiding and evading tax by moving to another country or
hiding assets or funds in another country. Finally, most contracting states
provide a mechanism in their treaties for resolving disputes with respect to
the application of the treaty and, in particular, transfer pricing disputes.
Dispute-resolution mechanisms are discussed in section 8.8.3, below.
An important effect of tax treaties – and an implicit purpose – is to
provide certainty for taxpayers with respect to the tax consequences of cross-
border investment. Investors like certainty. Tax treaties have an average life
of approximately fifteen years. As a result, nonresident investors know that,
despite changes that will inevitably be made to the tax laws of the source
country, the basic limitations in the treaty on the source country’s right to tax
will prevent future changes from affecting those limitations. For example, if
Company A, a resident of Country A, makes an investment in the shares of
Company B, a resident Country B, Company A knows that the limit provided
in the treaty between Country A and Country B on the rate of withholding tax
imposed by Country B on dividends will continue to apply even if Country B
increases the rate of withholding tax on dividends under its domestic law.
Another important effect, and an implicit objective, of a tax treaty is the
allocation of tax revenues from cross-border activity between the two
contracting states. The provisions of tax treaties determine how much tax
revenue from the cross-border activity between the two states will be subject
to tax by each of those states. For example, if Country A agrees to include in
its treaty with Country B a provision similar to Article 12 of the OECD
Model Treaty dealing with royalties, any royalties paid by residents of
Country A to residents of Country B will be taxable exclusively by Country
B, and vice versa. If royalty flows between Country A and Country B are
relatively equal, the allocation of the tax revenues from those flows will also
be relatively equal. However, if the flows are disproportionately from
residents of Country A to Country B (as would usually be the case if Country
A is a developing country and Country B is a developed country), the tax
revenues would be allocated disproportionately to Country B. Country A
might attempt to avoid this result by insisting that Article 12 of the treaty
allow the source country to tax royalties paid by its residents to residents of
the other country at a limited rate of, say 15 percent. In this case, Country A
would derive tax revenues equal to 15 percent of any royalties paid by its
residents to residents of Country B, and Country B would derive tax revenues
equal to its tax rate on royalties derived by its residents from residents of
Country A, less 15 percent of those royalties.

8.6 INTERPRETATION OF TAX TREATIES

8.6.1 Introduction
In certain respects, the interpretation of tax treaties is similar to the
interpretation of domestic tax legislation. The meaning of the words of the
treaty, the context in which they are used, and the purpose of the treaty are
generally important factors in interpreting both treaties and domestic tax
legislation. As a result, it seems likely that a country’s tax authorities and its
courts would interpret tax treaties in the same manner as domestic tax
legislation. There are, however, several important differences between tax
treaties and domestic tax legislation:

(1) Because tax treaties are bilateral, questions of interpretation should

be resolved by reference to the intentions of both states.
(2) Tax treaties are addressed to both the governments and the
taxpayers of both countries, whereas domestic tax legislation has a
narrower scope.
(3) Tax treaties are often drafted using different terms from those used
in domestic legislation. For example, the OECD and UN Model
Treaties use the term “enterprise,” which is not used in the domestic
legislation of many countries.
(4) Unlike domestic tax legislation, tax treaties do not generally impose
tax; they limit the taxes imposed by the contracting states.
(5) The influential OECD Model Treaty and Commentary and the UN
Model Treaty and Commentary have no counterparts in the context
of domestic tax legislation.

These differences may suggest that tax treaties should be interpreted

differently from domestic tax legislation. However, the interpretation of any
language, including the provisions of tax treaties and domestic tax rules, is a
matter of judgment that cannot be reduced to mechanical rules.

8.6.2 The Interpretive Provisions of the Vienna Convention on

the Law of Treaties
The interpretation of tax treaties is governed by customary international law,
as embodied in the Vienna Convention on the Law of Treaties. The
interpretive rules of the Vienna Convention apply to all treaties, not just tax
treaties. As discussed in section 8.2.1 above, the provisions of the Vienna
Convention are binding on all nations because they represent a codification of
customary international law.
Article 31(1) of the Vienna Convention provides a basic rule for the
interpretation of treaties:
A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given
to the terms of the treaty in their context and in light of its object and purpose.

Under Article 31(2), the context of a treaty includes the text of the
treaty, any agreements between the parties made in connection with the
conclusion of the treaty, and any instrument made by one party and accepted
by the other party. For example, the US produces a technical explanation for
each of its tax treaties, and Canada publicly announced its acceptance of the
US technical explanation of the US-Canada treaty. Under Article 31(3),
subsequent agreements between the parties to the treaty and their subsequent
practice with respect to the interpretation of the treaty, and any applicable
rules of international law, must also be taken into account, together with the
context. Article 31(4) provides that a treaty term may have a special meaning
rather than its ordinary meaning if it is established that the parties so intend.
The Commentary on the OECD or UN Model Treaty may provide evidence
that a term is intended to have a special meaning.
The basic interpretive rule in Article 31(1) of the Vienna Convention
makes intuitive sense. Obviously, the first step in any interpretive exercise
must be to carefully consider the ordinary meaning of the words of the treaty.
And those words must be read in their context – the particular provision in
which the words are used and the treaty as a whole – because the meaning of
words is always dependent on the context in which they are used. It also
makes sense to interpret the terms of a treaty in light of the purpose of the
provision and the treaty as a whole because, obviously, the contracting states
are trying to accomplish something by entering into the treaty and agreeing
on its terms.
Although Article 31(1) of the Vienna Convention makes sense, it must
also be acknowledged that it is vague and does not provide any clear,
meaningful guidance for taxpayers, tax authorities, or courts about how to
interpret treaties. Most importantly, it does not indicate how much weight to
give to the ordinary meaning of the words, the context, and the purpose of the
relevant provisions of the treaty in any particular case. For example, if there
is a conflict between the ordinary meaning of the words and the purpose of
the relevant provision, Article 31(1) does not indicate how the conflict should
be resolved. Although most courts and commentators would take the position
that words with a relatively clear meaning should not be disregarded in order
to carry out an unexpressed, uncertain purpose, it is difficult to write an
interpretive rule as to how all the relevant factors should be weighed in any
particular case.
Under Article 32 of the Vienna Convention, other material, referred to
as supplementary means of interpretation, which includes the travaux
préparatoires (preparatory work) of the treaty, should be considered only to
confirm the meaning established pursuant to Article 31, or to establish the
meaning if applying Article 31 produces an ambiguous, obscure, absurd, or
unreasonable result.
Although the Commentaries on the OECD and UN Model Treaties are
very important for the interpretation of tax treaties, their legal status under the
provisions of the Vienna Convention is unclear. At first glance, they appear
to be supplementary means of interpretation under Article 32. If so, they are
relevant only to confirm the meaning otherwise established by the application
of the basic interpretive rule in Article 31, or to establish the meaning if the
meaning under Article 31 is ambiguous, obscure, absurd, or unreasonable.
The OECD does not intend for the Commentary to have such a limited role.
In the Introduction to the Commentary, it is stated that the Commentary “can
be & of great assistance in the application and interpretation of the
conventions and, in particular, in the settlement of any disputes” (paragraph
29). It is difficult, however, to justify including the Commentary as part of
the context of a treaty under Article 31 of the Vienna Convention, especially
if the treaty being interpreted was entered into before the Commentary was
revised, or if one of the contracting states is not a member of the OECD and
therefore had no part in the preparation of the Commentary.
 Although the status of the OECD Model Treaty and Commentary under
the Vienna Convention is a controversial topic among international tax
scholars, the issue appears to be of little practical significance. In treaty cases
from virtually all countries, the courts invariably give the OECD Model
Treaty and Commentary substantial weight.
The provisions of tax treaties should be interpreted in the same way in
both countries (the principle of common interpretation) because otherwise
income may be taxed twice, or not at all. Assume, for example, that Company
A, a resident of Country A, performs services in Country B for the benefit of
Company B, a resident of Country B. The services result in the creation of
some work product used by Company B. Company A receives a payment
from Company B that is characterized under the laws of Country B as
compensation for services performed in Country B. In contrast, Country A
characterizes the payment as a royalty for allowing Company B to use
Company A’s work product. Under the tax treaty between the two countries,
fees for personal services are taxable in the source state, but royalties are
taxable exclusively by the residence state. Under these circumstances,
Company A will be subject to double taxation unless the competent
authorities of the two countries can resolve the matter. Country B will impose
tax on Company A’s income in accordance with Article 7 of the treaty
(assuming that Company A has a PE in Country B); in contrast, Country A
will impose tax on the payments received by Company A as royalties under
Article 12 of the treaty. Country A may not provide any relief for the tax
imposed by Country B because Country B’s tax is not imposed on the
royalties.
Several countries have multiple official languages. When these countries
enter into tax treaties, there may be multiple official versions of the treaty in
different languages. Article 33 of the Vienna Convention provides that all
versions of tax treaties concluded in multiple languages are considered to be
equally authentic unless the provisions of the treaty specify that one version
is to govern in the event of a conflict. Some countries that conclude their tax
treaties in multiple languages, such as China, provide that the English-
language version of the treaty will prevail where the versions conflict.

8.6.3 The Interpretation of Undefined Terms in Accordance

with Domestic Law – Article 3(2)
In addition to the provisions of the Vienna Convention, tax treaties based on
the OECD and UN Model Treaties contain an internal rule of interpretation.
Article 3(2) of the OECD and UN Model Treaties provides as follows:
As regards the application of the Convention at any time by a contracting state, any term not
defined therein shall, unless the context requires otherwise, have the meaning that it has at
that time under the law of that State for the purposes of the taxes to which the Convention
applies, any meaning under the applicable tax laws of that State prevailing over a meaning
given to the term under other laws of that State.

In effect, Article 3(2) provides that undefined terms used in the treaty
have the meaning that they have under the domestic law of the country
applying the treaty unless the context requires otherwise. For this purpose, a
country applies a treaty when it takes any relevant action with respect to the
treaty, such as issuing a ruling or an assessment of tax.
The application of Article 3(2) involves a three-stage process:

(1) Does the treaty provide a definition of the term?

(2) If the treaty does not provide a definition of the term, what is its
domestic meaning?
(3) Does the context of the treaty require a meaning different from the
domestic meaning?

The first step is not as simple as it appears, in part because some

definitions in tax treaties are inclusive, while others are exclusive. For
example, Article 3(1)(a) defines a “person” to include an individual, a
company, and any other body of persons. In contrast, the definition of
“company” in Article 3(1)(b) is exclusive (“company means &”). Generally,
an inclusive definition means that the term has its ordinary meaning plus the
items that are specifically mentioned. Does Article 3(2) apply to determine
the ordinary meaning under domestic law of terms that are defined
inclusively, such as “person”? In my view, Article 3(2) should apply in these
circumstances so that the term “person” would include any legal entity that is
considered to be a person under the domestic law of the state applying the
treaty. A further difficulty is that definitions in a treaty often contain terms
that are undefined; for example, the terms “individual” and “body of persons”
in Article 3(1)(a) are not defined. Again, in my view, these terms should take
their meaning from domestic law by virtue of Article 3(2).
The determination of the meaning of a term under domestic law may
also be difficult. Article 3(2) provides that the meaning of an undefined term
under a country’s tax law prevails over the meaning under other domestic
laws. An undefined term, however, may have more than one meaning for
purposes of a country’s tax law. In this situation, the domestic meaning that is
most appropriate should be used for purposes of the treaty. Also, Article 3(2)
refers to the “meaning” of an undefined term, not to its definition, under
domestic law. A term may not be defined for purposes of a country’s tax law,
but it should have an ordinary meaning.
The final step in the application of Article 3(2) is to consider whether
the context of the treaty requires a term to be given a different meaning from
its meaning under domestic law. For this purpose, it is necessary to consider
alternative meanings for the term for purposes of the treaty and whether one
of these meanings is more appropriate in the context of the treaty than the
domestic law meaning. Matters that should be considered in this analysis
include:

– the ordinary meaning of the term as compared to the meaning under

the domestic tax law;
– the meaning of the term under the other country’s tax law;
– the purpose of the relevant provision of the treaty; and
– extrinsic material such as the Commentary on the OECD or UN
Model Treaty.

Some international tax scholars argue that in applying Article 3(2), if at

all possible, undefined terms should be given a meaning that is independent
of domestic law and that a domestic law meaning should be used only as a
last resort. Other scholars argue that Article 3(2) contains a preference for
domestic law meanings because those meanings are displaced by treaty
meanings only if “the context requires otherwise.” The use of the word
“requires,” they argue, places a substantial onus on those seeking to justify a
treaty meaning.
In my view, Article 3(2) does not establish any clear preference for
domestic law meanings or treaty meanings for undefined terms. Furthermore,
there are no strong arguments for establishing any residual presumption in
favor of either a domestic or treaty meaning of an undefined term. As noted
above, the meaning of undefined terms in a tax treaty should be determined
by reference to all the relevant information.
 Another important and controversial issue of interpretation in
connection with Article 3(2) of the OECD and UN Model Treaties is whether
a term has its meaning under domestic law at the time that the treaty was
entered into (the static approach) or its meaning under domestic law as
amended from time to time (the ambulatory approach). Article 3(2) of the
OECD Model Treaty was amended in 1995 to clarify that Article 3(2) should
be applied in accordance with the ambulatory approach. A similar
conforming amendment was made to the UN Model Treaty in 2001. The
ambulatory approach allows treaties to accommodate necessary changes to
domestic law without the need to renegotiate the treaty. As a result, the
ambulatory approach effectively permits a country to unilaterally amend its
tax treaty with another country by changing its domestic law. However, an
amendment to domestic law that significantly alters the bargain between the
two countries, and was not contemplated by both countries when the treaty
was negotiated, is equivalent to a treaty override.

8.7 SUMMARY OF THE PROVISIONS OF THE OECD

AND UN MODEL TREATIES

8.7.1 Introduction
This section describes the major provisions of the OECD and UN Model
Treaties. Section 8.7.2 describes the provisions that identify the parties to a
treaty and the persons whose tax obligations are affected by it, that establish
the scope of the treaty, and that govern its ratification, termination, and
amendment. Section 8.7.3 describes the treatment of various categories of
income under a typical tax treaty; these provisions are known as the
distributive rules of a treaty. Section 8.7.4 describes the rules dealing with
administrative matters and cooperation between the treaty partners.
Every tax treaty includes some provision for relieving or mitigating
double taxation. In the OECD and UN Model Treaties, relief from double
taxation is provided either by Article 23A (Exemption Method) or Article
23B (Credit Method). Methods of providing relief from double taxation are
discussed in Chapter 4.
To prevent tax avoidance through transfer pricing, the domestic tax laws
of most countries give the tax authorities the power to adjust prices in
transactions between related persons to reflect the prices that would have
prevailed if the transaction had taken place at arm’s length with an unrelated
person. Article 9 (Associated Enterprises) of the OECD and UN Model
Treaties provides that the contracting states are permitted (indeed, expected)
to adjust prices and recompute profits from related-party transactions in
accordance with this so-called arm’s-length standard. Transfer pricing and the
arm’s-length standard are discussed in Chapter 6.

8.7.2 Coverage, Scope, and Legal Effect

The two countries that enter into a bilateral income tax treaty are called the
“contracting states.” Under Article 1 (Personal Scope) of the OECD and UN
Model Treaties, the provisions of the treaty apply to persons who are
“residents of one or both of the contracting states.” Article 4 (Resident)
defines a “resident” of a contracting state for purposes of the treaty as a
person who is liable to tax under the domestic laws of that contracting state
on the basis of certain connecting factors, such as residence, domicile, place
of management, or other similar criterion. As discussed in Chapter 2, section
2.2.3, Article 4 also includes tie-breaker rules that prevent a person from
being a resident of both contracting states for purposes of the treaty. A
“person” is defined in Article 3 (General Definitions) to include “an
individual, a company and any other body of persons.” The OECD
Commentary indicates that a charitable foundation is a “person” within the
meaning of Article 3 – indeed, any legal entity that is recognized under the
laws of a contracting state is likely to be treated as a “person” for tax treaty
purposes. Although a partnership is probably a body of persons and therefore
a person for purposes of the OECD and UN Model Treaties, it may not be a
resident of a contracting state if the partners rather than the partnership are
liable to tax in that state.
Article 2 (Taxes Covered) of the OECD and UN Model Treaties
specifies that the treaty applies “to taxes on income and on capital imposed
on behalf of a contracting state or of its political subdivisions or local
authorities.” Some treaties do not extend coverage to subnational income
taxes. Despite any limitations in Article 2, Articles 24, 26 and 27 of the
OECD and UN Model Treaties dealing with nondiscrimination, exchange of
information, and assistance in the collection of taxes apply to all taxes
imposed by the contracting states and not just those taxes described in Article
2.
 The typical tax treaty expressly lists the national taxes (and sometimes
the subnational taxes) of the contracting states to which the treaty applies.
Each country’s personal income tax and corporate income tax are invariably
listed. Most treaties also provide that the treaty applies to amendments of the
listed taxes and to subsequently imposed taxes that are identical or
substantially similar to the listed taxes. Some treaties also list certain income
and capital taxes that are not to be covered by the treaty; for example, many
tax treaties exclude from their scope payroll and social security taxes
earmarked for government pensions.
According to Article 30 (Entry into Force) of the OECD Model Treaty
and Article 29 (Entry into Force) of the UN Model Treaties, tax treaties
become effective on ratification and the states should agree to exchange
instruments of ratification as soon as possible. Each contracting state has its
own internal procedures for ratifying treaties that must be satisfied. For
example, many countries provide that a treaty negotiated by the government
must receive legislative approval to be effective. Once these internal
procedures have been satisfied, the contracting states will exchange
instruments of ratification. Generally, tax treaties become effective on the
first day of the calendar year following the exchange of the instruments of
ratification with respect to provisions of the treaty that apply on the basis of
taxation years, such as Article 7 (Business Profits). Other provisions of the
treaty, such as the provisions dealing with withholding rates in the source
country, may take effect earlier or later than the rest of the treaty. In addition,
certain provisions of the OECD and UN Model Treaties may apply
retrospectively. For example, a request for information or for assistance in the
collection of taxes may relate to a year before the treaty entered into force.
Such a request is valid as long as it is made after the treaty becomes effective.
Although tax treaty partners contemplate that their relationship will last
indefinitely, their treaties provide for the termination of the treaty at the
request of either party. Under Article 31 (Termination) of the OECD Model
Treaty and Article 30 (Termination) of the UN Model Treaty, a contracting
state may unilaterally terminate a treaty as of the beginning of the next
calendar year by giving notice of termination to its treaty partner at least six
months before the end of the current year. The contracting states may
terminate a treaty at any time by mutual consent, although some treaties
provide that a new treaty must remain in effect for a minimum period after it
enters into force.
8.7.3 The Distributive Rules: Articles 6 through 21

8.7.3.1 Introduction

Articles 6 through 21 of the OECD and UN Model Treaties deal with the
treatment of various types of income, from broad categories such as business
profits to quite narrow categories such as directors’ fees and pensions. This
approach inevitably means that conflicts arise as to how amounts should be
categorized for purposes of the treaty. These conflicts are sometimes resolved
by definitions in the relevant articles. For example, Article 10(3) defines
dividends as income from shares or other rights “not being debt-claims” and
Article 11(3) defines interest as “income from debt-claims of every kind.”
Consequently, Articles 10 and 11 cannot both apply to the same amount – if
an amount is income from a debt-claim, it is interest and cannot be a
dividend. Sometimes the conflicts are resolved pursuant to specific
provisions in the treaty. For example, where Article 7 and another article both
apply, Article 7(4) of the OECD Model (Article 7(6) of the UN Model) gives
priority to the other article. Some conflicts are not resolved by specific rules.
Where an item of income is not covered by any of the specific articles
(Articles 6–20), it is dealt with in Article 21 (Other Income).
The wording of the distributive rules of the OECD and UN Model
Treaties is remarkably consistent. Where an article uses the words “shall be
taxable only” in one of the contracting states, it means that the other state is
precluded from taxing the relevant income. For example, under Article 8(1),
profits from the operation of ships or aircraft in international traffic “shall be
taxable only” in the country in which the enterprise has its place of effective
management. In contrast, where the words “may be taxed” in a contracting
state are used, it means that the relevant amount may be taxed by that
country; however, it does not mean that the amount is not taxable by the other
contracting state. In other words, the words “may be taxed” mean that the
contracting states are both entitled to tax the relevant amount under the treaty.
In these circumstances, the source country’s tax takes priority and Article 23
requires the residence country to provide relief from double taxation by
exempting the income from residence country tax or granting a credit for the
source country tax against the residence country tax.

8.7.3.2 Business Income

The OECD and UN Model Treaties distinguish between several types of
business income. For example, profits from immovable property, profits from
international shipping and air transportation, and profits from entertainment
and athletic activities are dealt with under Articles 6, 8, and 17 respectively.
Article 7 of the OECD and UN Model Treaties applies to business profits that
are not covered by a more specific article. Article 7 (Business Profits)
provides that “an enterprise of a contracting state” is exempt from tax on its
profits derived from business carried on in the other contracting state unless
the business is carried on through a permanent establishment (PE) located in
that other contracting state and the profits are attributable to the PE. This
limitation on a country’s source jurisdiction is discussed in Chapter 2. The
definition of a PE is provided in Article 5 (Permanent Establishment) and is
discussed below. Article 3 (General Definitions) defines “an enterprise of a
contracting state” as an enterprise carried on by a resident of a contracting
state.
If an enterprise of a contracting state has a PE in the other contracting
state, it is taxable by the other state only on the profits attributable to the PE.
Article 7(2) of the OECD and UN Model Treaties provides that the profits of
a PE should be determined on the assumption that the PE is a separate and
distinct entity dealing independently with the other parts of the enterprise of
which the PE is a part. The effect of these assumptions in Article 7(2) is that
the transfer pricing rules applicable to associated enterprises under Article 9
also apply, by analogy, for the purpose of determining the profits attributable
to a PE. The difficulties that arise in applying the arm’s-length principle to
determine the profits attributable to a PE are addressed in section 8.8.5
below.
Article 7 of the OECD Model Treaty does not use a so-called force-of-
attraction approach, under which all of a taxpayer’s income derived from a
country is subject to tax by that country if it has a PE in that country. Under
Article 7, if a taxpayer has a PE in a country, only the taxpayer’s profits from
the business that are attributable to the PE are subject to tax by that country.
Other income may be taxable under other articles of the treaty, but not under
Article 7.
Article 7(1) of the UN Model Treaty employs a limited force-of-
attraction principle in determining the income attributable to a PE. Under that
principle, if an enterprise has a PE in a contracting state, it is taxable by that
state not only on the profits attributable to the PE, but also on profits derived
from sales in that state of goods similar to those sold through the PE or from
business activities in that state similar to the activities conducted through the
PE. Although this limited force-of-attraction rule in the UN Model Treaty
may appear to broaden the scope of Article 7 and reduce the opportunities for
tax avoidance, the rule is easily avoided if profits unrelated to a PE are earned
by a related nonresident entity.
The Definition of a Permanent Establishment
Under Article 5(1) of the OECD and UN Model Treaties, a PE is defined as
“a fixed place of business through which the business of an enterprise is
wholly or partly carried on.” This language is used in essentially identical
form in almost all tax treaties.
For an enterprise to have a “fixed place of business” in a contracting
state, it must operate at a specific geographical location, and its activities at
that location must continue for more than a temporary period (generally for
more than six months). Thus, a taxpayer that does business at various
locations in a country for an aggregate of more than six months does not have
a fixed place of business PE in that country. The place where equipment,
such as an oil pumping machine, is used can constitute a fixed place of
business even if that machine is unattended by human agents of the
enterprise. However, some countries take the position that human
intervention is necessary for a place of business to constitute a PE.
For a place of business to be “fixed” in a geographical sense, it must
have both geographical and commercial coherence. For example, a
marketplace can be the fixed place of business of an enterprise if the
enterprise operates within that marketplace on a regular basis, even though it
may use a different stall from time to time, because the marketplace has both
geographical and commercial coherence. In contrast, if an interior designer
provides services for a client in the client’s office occupying a floor in a large
office building for four months and then provides services for a different
client with an office on a different floor in the same building for another four
months, the office building cannot be considered to be the designer’s PE.
Although the building has geographical coherence (it is a fixed place), the
offices where the designer works do not have commercial coherence because
they are leased by different clients. Similarly, if the designer does work in
two different buildings owned by the same person, those buildings are not the
designer’s PE, since the buildings do not have geographical coherence (they
are different fixed places) although they do have commercial coherence.
However, multiple buildings may have geographical coherence if they are
part of a campus or office complex that constitutes a single location.
It is immaterial whether an enterprise rents or owns its premises in
determining whether the premises constitute a PE, as long as the place is at its
disposal (see paragraph 4 of the OECD Commentary). This concept of a
place being at the taxpayer’s disposal is problematic. It is clear that the
taxpayer does need to have a legal right to use the place; on the other hand, it
appears that the mere use of a place for more than six months is not sufficient
to constitute a PE. For example, the OECD Commentary indicates that, if a
salesperson employed by an enterprise visits a client’s office every day to
take orders, the client’s office is not at the disposal of the enterprise and is
not a PE of the enterprise. The OECD has proposed to amend the OECD
Commentary to clarify that a taxpayer must have effective power over a place
for it to be at the taxpayer’s disposal (see OECD Discussion Draft, OECD,
Discussion Draft, Interpretation and Application of Article 5 (Permanent
Establishment) of the OECD Model Tax Convention (OECD 2011),
www.oecd.org/dataoecd/23/7/48836726.pdf) and OECD, Revised Proposals
concerning the Interpretation and Application of Article 5 (Permanent
Establishment) of the OECD Model Tax Convention (OECD 2012), available
at www.oecd.org/ctp/treaties/PermanentEstablishment.pdf).
Both the OECD and UN Model Treaties provide that a building site or
construction or installation project constitutes a PE if the project continues
for at least twelve months in the case of the OECD Model Treaty (Article
5(3)) and six months in the case of the UN Model Treaty (Article 5(3)(a)).
Both provisions apply to assembly and supervisory activities conducted in
connection with a building or assembly site. These activities are explicitly
included in Article 5(3)(a) of the UN Model Treaty, but not in the OECD
Model Treaty, where they are dealt with in the Commentary (paragraphs 17
and 20 of the Commentary on Article 5).
Developing countries typically adopt the six-month period in the UN
Model Treaty (or an even shorter minimum period for construction sites) in
their tax treaties; for example, the minimum period in the India-US treaty is
four months. A few treaties between developed countries extend the
minimum period beyond one year. The Japan-US treaty, for example, has a
twenty-four month period.
The construction site provisions in the OECD and UN Model Treaties
are commonly misunderstood as deeming provisions. They are properly
interpreted as an additional condition that must be met in order for a
construction site to constitute a PE. In other words, a construction site must
satisfy the conditions of a fixed place of business under Article 5(1) and must
also last for at least twelve months (OECD) or six months (UN). Thus, a
project that involves construction activities at different locations in a country
for an aggregate of more than twelve months is not a PE if the activities at
each location do not last for at least twelve months. Each place at which
construction occurs must be treated as a separate place unless the places have
geographical and commercial coherence, as discussed above.
If the definition of a PE were limited to fixed places of business, it
would be too narrow and would not apply to many types of businesses that do
not need to be carried on through a fixed place of business. Consequently,
both the OECD and UN Model Treaties extend the definition of a PE to
include certain dependent agents acting on behalf of an enterprise. Under
Article 5(5) of both Model Treaties, a resident of one contracting state is
deemed to have a PE in the other contracting state if a person (often referred
to as a dependent agent) has and habitually exercises on behalf of the resident
an authority to conclude contracts that are binding on the resident. The agent
must have not only legal authority to conclude contracts “in the name of the
enterprise” but it must also do so habitually, which means regularly or
repeatedly. The phrase “in the name of the enterprise” is not intended to have
a literal meaning; it is sufficient if the contracts are legally binding on the
enterprise.
The deemed PE rule in Article 5(5) does not apply if the person acting
on behalf of an enterprise is an agent of independent status who is acting in
the ordinary course of business (Article 5(6) of the OECD Model Treaty and
Article 5(7) of the UN Model Treaty). Whether a person is an independent
agent depends on all the facts and circumstances, although the Commentary
on Article 5(6) of the OECD Model indicates that a person must be both
legally and economically independent. Thus, an employee is invariably a
dependent agent and a person who acts exclusively for one enterprise is likely
to be considered a dependent agent in most circumstances.
The agency rule in the UN Model Treaty is more expansive than the
OECD rule; it extends to dependent agents that habitually maintain a stock of
goods from which they make deliveries on behalf of an enterprise. In
addition, under Article 5(7) of the UN Model Treaty, an agent cannot be
independent if the agent acts exclusively or almost exclusively for one
enterprise and the commercial and financial relations between them are not at
arm’s length.
Many multinational corporations have used commissionaire
arrangements to avoid having a PE in a country. A commissionaire
arrangement is a legal relationship recognized under the civil law, under
which one person enters into contracts in the name of or on behalf of another
person, but those contracts are not legally binding on that other person.
Therefore, multinational corporations can structure their affairs so that a
group company in a low-tax country sells its products through another group
company in a high-tax country to customers in that country as a
commissionaire for the low-tax group company. The commissionaire does
not own the goods and the contracts it enters into with customers are not
legally binding on the group company that owns the goods. The result is that
the group company acting as a commissionaire earns only a small profit from
its activities; that profit is taxable in the high-tax country in which it is
resident. However, the group company in the low-tax country earns most of
the profit from the sale of the products, and that company is not taxable in the
high-tax country because it does not have a PE there.
Dell computers used this type of commissionaire arrangement to avoid
tax in several high-tax European countries. A Dell group company
established in Ireland sold computers to another group company established
in Norway that resold the computers to customers in Norway. The Norwegian
Supreme Court held that Dell Ireland did not have a PE in Norway (see Dell
Products v. The State, December 2, 2011 (Tax East), HR-2011-02245-A
(Case No. 2011.755), Tax Treaty Case Law IBFD). The French Conseil
d’Etat reached the same result in the Zimmer case, (see Societe Zimmer
Limited, March 31, 2010, Case No. 304715 and No. 308525, Tax Treaty Case
Law IBFD.)
The OECD’s revised discussion draft on BEPS Action 7: Preventing the
Artificial Avoidance of PE Status, May 2015, available at
www.oecd.org/tax/treaties/revised-discussion-draft-beps-action-7-pe-
status.pdf, proposes to amend Articles 5(5) and (6) to prevent the use of
commissionaire arrangements to avoid PE status. Article 5(5) will be
amended to deem an enterprise of one contracting state to have a PE in the
other state if a person habitually concludes contracts or negotiates the
material elements of contracts on behalf of the enterprise and if the contracts
are in the name of the enterprise, for the transfer of the ownership or use of
property owned or leased by the enterprise, or for the provision of services by
the enterprise. Thus, a typical commissionaire arrangement would be covered
by this provision because the commissionaire negotiates the material
elements of the contracts for the sale of the goods owned by the principal
even though those contracts are not legally binding on the principal. Article
5(6) will be amended to provide that a person who deals exclusively or
almost exclusively with one or more enterprises with which it is connected
will not be considered to be an independent agent. Thus, commissionaires
will not be able to argue that they are independent agents.
The deemed agency PE rule in Article 5(5) of both Model Treaties does
not apply if an agent’s activities are limited to the exempt activities (generally
preparatory or auxiliary activities) listed in Article 5(4), which is discussed
below.
Under Article 5(6) of the UN Model Treaty, an enterprise engaged in the
sale of insurance in a contracting state is deemed to have a PE in that state if
it collects premiums in that state or ensures risks located in that state. This
rule does not apply, however, if these activities are conducted by an
independent agent acting in the ordinary course of business.
Article 5(3)(b) of the UN Model Treaty provides that an enterprise is
deemed to have a PE in a contracting state if it performs services in that state
through employees or other personnel for a period of at least 183 days in any
twelve-month period with respect to the same or a connected project.
Whether projects are connected must be determined based on all the facts and
circumstances. The OECD Model Treaty has no comparable provision,
although the OECD Commentary contains an alternative services PE
provision that countries may adopt (see paragraph 42.23 of the OECD
Commentary on Article 5).
Under the UN Model Treaty, income from the performance of
independent personal services is taxable under Article 14 and not under
Article 7. This approach was also followed under the OECD Model Treaty
until 2000, when Article 14 was deleted. The taxation of independent
personal services is discussed in section 8.7.3.3 below.
Article 5(4) of the OECD and UN Model Treaties provides an
exemption from the definition of a PE for fixed places of business that are
used exclusively for certain preparatory or auxiliary activities. Under both
models, a fixed place of business of an enterprise used solely for the
following activities is deemed not to be a PE:
– the storage or display of goods owned by the enterprise;
– the maintenance of a stock of goods owned by the enterprise for
storage or display or for processing by another enterprise;
– purchasing goods or collecting information for the enterprise; and
– other activities of a preparatory or auxiliary character.
In addition, a fixed place of business used solely for a combination of
the activities set out above is deemed not to be a PE as long as the overall
activity is preparatory or auxiliary. Article 5(4) of the OECD Model also
applies to a fixed place of business used solely for the delivery of goods
owned by an enterprise. Therefore, if an enterprise owns or rents a warehouse
in a country that it uses to store goods owned by it and to deliver those goods
to customers, the warehouse would not be a PE of the enterprise. In contrast,
under the UN Model Treaty, the warehouse would be a PE because delivery
of goods is not an exempt activity. (Note that under both Model Treaties, if a
warehouse is used to store goods owned by other enterprises, it will be a PE.)
The OECD’s BEPS Action 7: Preventing the Artificial Avoidance of PE
Status (October 31, 2014), available at www.oecd.org/ctp/treaties/action-7-
pe-staus-public-discussion-draft.pdf proposes that Article 5(4) should be
revised to ensure that the exemption in that article is available for the listed
activities only if they are truly preparatory or auxiliary. Thus, a large
warehouse owned by an enterprise and used to store goods that are sold
through online shopping websites would be considered to be a PE. An
alternative would be to delete the exemptions for delivery of goods,
purchasing, and collecting information.
According to Article 5(7) of the OECD Model Treaty and Article 5(8) of
the UN Model Treaty, a subsidiary resident in a country or carrying on
business in a country does not constitute a PE of its parent company in that
country simply because the parent controls it. Similarly, a parent company is
not a PE of its subsidiary. Thus, a company resident in Country A that owns a
subsidiary resident or carrying on business in Country B does not have a PE
in Country B simply because it controls the subsidiary. However, the parent
company might have a PE in Country B if its subsidiary habitually enters into
contracts binding on its parent or if facilities owned or leased by the
subsidiary are at the disposal of the parent company and used by it for more
than six months. The Commentary on both the OECD and UN Model
Treaties indicates that, with respect to a multinational group, the
determination whether a PE exists must be made separately for each company
in the group; just because one group company has a PE in a country does not
mean that any other group company has a PE in that country.
International Shipping and Air Transportation
Under Article 8 of the OECD Model Treaty and Article 8 (Alternative A) of
the UN Model Treaty, business profits derived by an enterprise resident in
one contracting state from the operation of ships or aircraft in international
traffic are taxable exclusively by the state in which the enterprise has its place
of effective management. Some treaties assign the exclusive right to tax to
the country of residence of the enterprise in order to avoid uncertainty about
the meaning of the place of effective management of an enterprise. The
definition of “international traffic” (Article 3(1)(e)) for purposes of Article 8
is extremely broad and includes all transport other than transport of goods or
persons solely between places in a country. Thus, Article 8 does not permit a
country to tax the profits derived by an enterprise whose place of effective
management is located in the other contracting state from taking goods or
passengers on board in the country or unloading goods or passengers in the
country. For example, an airline with its place of effective management in
Country A that starts flights outside Country B, stops in Country B to drop
off passengers and take on passengers, and completes the flights outside
Country B would not be subject to tax by Country B. However, if the airline
operates a flight that stops in Country B to pick up passengers and then stops
at another location in Country B, Country B is allowed to tax the profits from
the portion of the flight that takes place solely in Country B.
Article 8 (Alternative B) of the UN Model Treaty permits the source
country to tax income derived from shipping (but not air transportation)
activities if such activities are “more than casual.” According to the
Commentary, more than casual means any planned trip to a country to pick
up goods or passengers.
Income from Entertainment and Athletic Activities
Under Article 17 of both the OECD and UN Model Treaties, income derived
by an entertainer or athlete resident in one contracting state from
entertainment or athletic activities performed in the other state are taxable in
that other state without any threshold requirement or any limitations. Thus, a
country is entitled to tax an entertainer or athlete who is present in the
country for only a short period on the gross amount received by the
entertainer or athlete without any limit on the rate of tax.
Article 17 provides a sharp contrast with Article 7, under which a source
country is entitled to tax a nonresident on business profits only if the
nonresident has a PE in the source country and only if the net profits are
attributable to the PE. It is difficult to justify Article 17 on any principled
basis. Some entertainers and athletes can make large sums of money in a
relatively short time, and countries want their share of that money. However,
other taxpayers, such as consultants and celebrities, who can also earn large
sums in a relatively short time, are not subject to tax if they earn income in
another country without any PE or fixed base in that country. Also, many
entertainers and athletes earn only modest amounts from their entertainment
and athletic activities; nevertheless, under the terms of Article 17, they are
subject to tax without any limitations by the country in which their activities
are exercised.
Article 17(2) contains an anti-avoidance rule that allows a country to tax
income from entertainment or athletic activities occurring in the country even
if the income is assigned by the entertainer or athlete to another person. For
example, an entertainer might perform in a country as an employee of a
personal services corporation so that most of the income from the
performance is derived by the corporation rather than the entertainer. Article
17(2) allows the country to tax both the entertainer and the corporation.
Leasing income
Rental income derived from leasing equipment is taxable in accordance with
Article 7 of the OECD Model Treaty. Therefore, such income is subject to
tax by a country only if the taxpayer has a PE in the country and the income
is attributable to the PE. Until 1992, such rental income was included in the
definition of royalties for purposes of Article 12, so that the source country
was precluded from taxing the income unless the taxpayer had a PE in the
source country and the equipment was effectively connected with the PE (in
which case Article 7 applied). However, Article 12 of the UN Model Treaty,
which permits taxation of royalties by the source country, continues to treat
income from equipment rentals as royalties. As a result, under the OECD
Model Treaty, income from equipment leasing is not taxable by the source
country unless the taxpayer has a PE in the source country and the income is
attributable to the PE. In contrast, under the UN Model Treaty, income from
equipment leasing is subject to tax by the source country at a limited rate on
the gross rental payments under Article 12 even if the taxpayer does not have
a PE in the source country. If the taxpayer does have a PE in the source
country and the leasing equipment is effectively connected with the PE, then
Article 7 applies.
Rent derived from immovable property situated in a country is taxable
by that country in accordance with Article 6 of the OECD and UN Model
Treaties irrespective of whether the rent is derived from a business or the
taxpayer has a PE in that country. For example, income derived from renting
an apartment building would be taxable in the contracting state where the
building is located. Article 6 is discussed in section 8.7.3.4 below.
Many difficult issues arise in determining whether an enterprise has a
PE in a contracting state as a result of engaging in electronic commerce in
that State. Those issues are addressed in Chapter 9, section 9.6.

8.7.3.3 Employment and Personal Services Income

Many provisions of the OECD and UN Model Treaties deal with a wide
variety of income from services. The provisions differ significantly, and
therefore it is important to distinguish between the various types of services.
For example, the basic rule for the taxation of income from employment is
Article 15. However, some types of income from employment – such as
income from entertainment and athletic activities, directors' fees,
remuneration of top-level managerial officials, pensions, and income from
government service – are subject to special rules.
It is frequently necessary to distinguish between employment income
and independent personal services income in order to determine whether the
rules for the treatment of employment income under Article 15, or the rules
for the treatment of independent personal services income under Article 14
(or Article 7 of the UN Model Treaty) apply in a particular case. This
distinction is important because the rules differ significantly. For example, an
individual resident in one contracting state who is employed by an employer
resident in the other contracting state is taxable by that other state on any
income derived from employment exercised in that state. In contrast, if the
individual is an independent contractor, the individual is taxable by the other
contracting state only if the individual has a regularly available fixed base or
PE in that state or stays in that state for more than 183 days.
Under Article 14 (Independent Personal Services) of the UN Model
Treaty, a resident of a contracting state who performs “professional services
or other activities of an independent nature” in the other contracting state is
not taxable in that State unless he or she has a “fixed base” in the state that is
regularly available or stays in the state for more than 183 days in any twelve-
month period. The term “professional services” includes the services of
physicians, lawyers, engineers, architects, dentists, and accountants, as well
as independent scientific, literary, artistic, educational, and teaching
activities. Article 14 was included in the OECD Model Treaty until it was
removed in 2000. As a result of this change in the OECD Model Treaty,
individuals and companies engaged in the performance of independent
personal services in a contracting state are taxable in that state only if they
have a PE in that state and their income is attributable to the PE.
The concept of a fixed base contained in Article 14 of the UN Model
Treaty and the pre-2000 version of the OECD Model Treaty is intended to be
equivalent to the concept of a fixed place of business in the definition of a PE
in Article 5. However, the deemed agency PE rules and the exception for
preparatory and auxiliary activities in Article 5 are not applicable in
determining whether an enterprise has a fixed base for purposes of Article 14.
Income from employment performed in a country may be taxable in the
country under Article 15 (Dependent Personal Services) of the OECD and
UN Model Treaties whether or not the employee has a fixed base in the
country. However, such income is exempt from tax in the source country if
an employee is paid by a nonresident employer without a PE in the source
country and the employee is present in the source country for not more than
183 days in any twelve-month period.
The limitations on source country taxation of professionals and
employees do not generally apply to entertainers and athletes (see Article 17
of the OECD and UN Model Treaties). Nor do they apply to nonresidents
receiving fees as corporate directors of resident corporations (see Article 16
of the OECD and UN Model Treaties) or remuneration as top-level managers
of resident corporations (see Article 16 of the UN Model Treaty). Under
Article 16, it is immaterial whether the income of the directors or top-level
officials of a company arises from services performed in the contracting state
in which the company is resident.
 With certain exceptions, individuals performing employment services
for the government of a contracting state are taxable only by that state (see
Article 19 of the OECD and UN Model Treaties). Diplomats and consular
officials who work in a foreign country as members of their government’s
diplomatic missions are exempt from tax under special agreements or under
the rules of international law. A tax treaty would not affect such exemptions
(see Article 27 of the OECD and UN Model Treaties).
Under Article 18 of the OECD Model Treaty, individuals receiving
pensions on account of past employment are generally taxable only by the
contracting state in which they are resident. In contrast, the UN Model Treaty
provides some limited scope for taxation by the country where the payer of
the pension is resident. Government pensions generally are taxable by the
contracting state making the pension payment unless the individual receiving
the pension is both a resident and a national of the other contracting state (see
Article 19(2) of the OECD and UN Model Treaties).
Students and certain business apprentices or trainees who visit a
contracting state for educational or training purposes are generally not taxable
in that contracting state on payments for their maintenance, education, or
training received from persons resident in the other state (see Article 20 of
the OECD and UN Model Treaties). Some tax treaties also provide reciprocal
exemptions for visiting professors and teachers.

8.7.3.4 Income and Gains from Immovable Property

Most countries want to retain the right to tax income derived from the sale
and rental of immovable property and from the extraction of natural resources
located within their territory. Reflecting this consensus view, Article 6 of the
OECD and UN Model Treaties allows the country of source to tax income
derived from “immovable property” situated in the country. The meaning of
the term “immovable property” is determined in accordance with the law of
the country in which the property is situated; the term includes income from
agriculture, forestry, mineral deposits, and other natural resources. Article 13
of both Model Treaties allows gains from the disposition of immovable
property to be taxed by the source country.
Because the source country is entitled to tax both the income derived
from immovable property and gains from the disposition of such property, it
does not generally matter for purposes of the Model Treaties whether a gain
from the disposition of immovable property is characterized as income or
capital gain; this characterization issue is left to domestic law. The same
approach is used under the OECD and UN Model Treaties for income and
gains from property other than immovable property.
Article 13(4) of the OECD and UN Model Treaties provides that a
source country is entitled to tax gains from the disposition of shares of a
company or an interest in a partnership or other entity if the value of the
company, partnership, or other entity is derived primarily from immovable
property situated in the country. That provision is intended to prevent a
taxpayer from avoiding source country taxation on gains derived from
immovable property by transferring the property to a controlled corporation,
partnership, or other entity and then disposing of the interests in the
corporation, partnership, or entity in a transaction that would otherwise be
exempt from source taxation under the tax treaty. Under Article 13(5) of the
UN Model Treaty, the country in which a company is resident is entitled to
tax gains from the disposal of a substantial interest in the company.

8.7.3.5 Reduced Withholding Rates on Certain Investment

Income
A major objective of most tax treaties is to provide for reduced rates of
withholding tax levied by the source country on dividends, interest, and
royalties paid to residents of the other contracting state. The goal of these
reduced rates is to provide for some sharing of the tax revenue between the
source country and the residence country.
The OECD Model Treaty provides that the withholding taxes impoln the
table below.

Table 8.1 Maximum Withholding Rates Endorsed by OECD Model Treaty

Dividends Dividends Interest Royalties

Paid to Paid to Other
Corporations Persons
with a
Substantial
Interest
Maximum 5% 15% 10% 0%
Rate
Source: OECD Model Treaty, Article 10 (Dividends), Article 11 (Interest), and Article 12 (Royalties).

The maximum rates proposed in the OECD Model Treaty, especially the
zero rate on royalties, are unacceptable to most developing countries and to
many developed countries. The UN Model Treaty does not provide any
specific limits on withholding rates, leaving those limits to be negotiated by
the contracting states. Most tax treaties with developing countries allow
maximum withholding rates that are substantially in excess of the rates
provided in the OECD Model Treaty; it is uncommon, for example, for
developing countries to agree to a maximum withholding rate on royalties of
lower than 15 percent.
Many tax treaties provide for a more complicated set of maximum
withholding rates than the simple pattern proposed in the OECD Model
Treaty. For example, it is common for tax treaties to impose separate
limitations on the withholding rates applicable to industrial royalties,
royalties paid with respect to copyrights of literary works, and royalties paid
for the showing of motion picture films.
The rules for the taxation of dividends, interest, and royalties under
Articles 10, 11, and 12 of the OECD and UN Model Treaties respectively
take priority over the rules for the taxation of business profits in Article 7.
For example, under Article 11, interest paid by a resident of one contracting
state to a resident of the other contracting state is taxable by the first state
even if the interest forms part of the business profits of the resident of the
other state. However, if the resident of the other state has a PE in the first
state and the debt-claim in respect of which the interest is paid is effectively
connected with the PE, the interest is taxable by the first state in accordance
with Article 7 rather than Article 11 (see Articles 7(4) and 11(4) of the OECD
Model Treaty and Articles 7(6) and 11(4) of the UN Model Treaty). The rules
in Articles 10(4), 11(4), and 12(4) of the OECD and UN Model Treaties are
known as “throwback rules” because, in the first instance, Article 7 applies to
dividends, interest, and royalties that constitute business profits; Article 7
then gives priority to Articles 10, 11, or 12, but those articles then make
Article 7 applicable once again.

8.7.3.6 Other Types of Income

Most tax treaties do not impose limits on the rights of the contracting states to
tax income, other than those types of income discussed above, derived by
their residents. Article 13 of the OECD and UN Model Treaties generally
provides that capital gains, other than gains from the disposal of the assets of
a PE in the source country, immovable property situated in the source
country, ships and aircraft used in international traffic, and interests in
corporations, partnerships, and other entities the value of which is derived
primarily from immovable property situated in the country, are taxable
exclusively by the residence country.
The residual rule contained in Article 21 (Other Income) of the OECD
Model Treaty similarly provides that items of income not dealt with in other
articles of the treaty are taxable exclusively by the residence country. In
contrast, Article 21 of the UN Model Treaty allows the source country to tax
other income that arises in the source country. Article 21 of the OECD Model
is important with respect to income derived from financial instruments. A tax
treaty following the OECD Model Treaty precludes taxation at source of
income items that may resemble various traditional types of income, but
which are modified by contractual arrangements to constitute a type of
income that is not mentioned in the treaty.

8.7.4 Administrative Cooperation

Several provisions in the OECD and UN Model Treaties are designed to
promote administrative cooperation between the contracting states. Article 24
(Non-Discrimination) of the OECD and UN Model Treaties requires each
contracting state not to discriminate unfairly against the residents and
nationals of the other contracting state. Although nondiscrimination is a
worthy objective, it is not easily attained. Issues arising under the
nondiscrimination article are addressed in section 8.8.1 below.
Article 25 (Mutual Agreement Procedure) of the OECD and UN Model
Treaties establishes a mechanism for resolving disputes that arise from the
interaction of the tax systems of the contracting states or from the operation
of the treaty itself. The mutual agreement procedure, including arbitration of
tax disputes, is discussed in section 8.8.3 below.
Virtually all tax treaties provide for some cooperation between the
contracting states in the administration of the tax treaty and their domestic tax
systems. Article 26 (Exchange of Information) of the OECD and UN Model
Treaties provides for the exchange of “such information as is foreseeably
relevant for carrying out the provisions of this Convention or of the domestic
laws of the contracting states concerning taxes covered by the Convention.”
Article 27 (Assistance in the Collection of Taxes) is a recent addition to both
the OECD Model and the UN Model Treaty, which provides for mutual
assistance in the collection or enforcement of taxes. Exchange of information
and mutual assistance are dealt with in section 8.8.4 below.

8.8 SPECIAL TREATY ISSUES

8.8.1 Nondiscrimination
In general, there are no significant legal restrictions on a country’s
jurisdiction to tax, and consequently, a country could consider taxing
nonresidents more harshly than residents. In fact, however, most countries
generally treat nonresidents in the same way as, or better than, residents for
tax purposes. Probably the most important constraints on the unequal
treatment of nonresidents are the possibility of retaliation by other countries
and the need to attract investment by nonresidents.
The most important type of legal protection against discrimination for
tax purposes is the nondiscrimination article of bilateral tax treaties. The
nondiscrimination provisions of the GATT, the GATS, and other trade and
investment treaties generally provide that tax discrimination is to be dealt
with in accordance with bilateral tax treaties. Article 24 of the OECD and UN
Model Treaties prohibits the contracting states from imposing tax
consequences on the citizens or residents of the treaty partner that are less
favorable or more adverse than the tax consequences imposed on their own
citizens or residents. The treaties do not define discrimination or
nondiscrimination. In general, however, discrimination means distinguishing
between persons adversely on grounds that are unreasonable, irrelevant, or
arbitrary. Conversely, nondiscrimination means equal (functionally
equivalent) or neutral treatment. In any nondiscrimination case, the crucial
issue is to determine the precise situations that are to be compared.
Article 24 of the OECD and UN Model Treaties prohibits discrimination
against foreign nationals and nonresidents in several respects:

(1) Article 24(1) prohibits discrimination on the basis of nationality.

 Because most countries do not tax individuals on the basis of
nationality (the US is a major exception), this provision is primarily
important with respect to legal entities.
(2) Article 24(3) prohibits discrimination against nonresidents carrying
on business in a country through a PE. Such nonresidents must be
treated no less favorably than residents of the treaty country
engaged in similar activities.
(3) Article 24(4) requires countries to allow the deduction of amounts
paid by residents of a contracting state to residents of the other
contracting state on the same basis as amounts paid to residents of
the first state. In effect, this provision provides protection against
discrimination indirectly because the direct beneficiaries of the legal
protection are domestic enterprises. Thin capitalization rules, which
are discussed in Chapter 7, section 7.2, are widely considered to
violate this aspect of a nondiscrimination article.
(4) Article 24(5) ensures that corporations, partnerships, and other
entities resident in a contracting state whose capital is owned or
controlled by residents of the other contracting state must be treated
no less favorably than enterprises owned or controlled by residents.
As with Article 24(4), the protection against discrimination is
provided directly to resident entities and only indirectly to the
nonresident owners of those entities.

Article 24 of the OECD and UN Model Treaties provides important but

limited protection against tax discrimination. Some countries, such as
Canada, refuse to treat nonresidents the same as residents (national
treatment) for tax purposes. Instead, they agree in their tax treaties to
provide most-favored-nation treatment to the residents of a particular treaty
country. Most-favored-nation treatment ensures that the residents of a treaty
country are treated the same as residents of other foreign countries. It does
not guarantee that they will be treated the same or as well as resident
taxpayers.

8.8.2 Treaty Abuse

8.8.2.1 Introduction
In general, tax treaties limit the taxes imposed by the contracting states. Not
surprisingly, therefore, tax treaties have been widely used by taxpayers to
avoid tax. This section examines the ways in which countries protect
themselves from the abuse or improper use of their tax treaties. Section
8.8.2.2 deals with one particular aspect of treaty abuse, the problem of treaty
shopping. Section 8.8.2.3 describes the OECD BEPS Action 6 proposals for
dealing with treaty abuse.
As discussed in section 8.5 above, the two most obvious purposes of tax
treaties based on the OECD and UN Model Treaties are the elimination of
double taxation and the prevention of tax avoidance and evasion. Although
the Model Treaties contain several explicit provisions aimed at eliminating or
preventing double taxation, they have few provisions dealing with tax
avoidance. Article 9 dealing with transfer pricing, Article 13(4) (allowing
source countries to tax gains from the disposal of interests in land-rich
enterprises), and Article 17(2) (allowing source countries to tax income from
entertainment and athletic activities that accrue to a person other than the
entertainer or athlete) are the only specific anti-abuse rules found in the
Model Treaties. Nevertheless, the Commentary on Article 1 of both Model
Treaties deals reasonably comprehensively with the topic of treaty abuse.
The Commentary on Article 1 of the OECD Model Treaty was revised
in 2003 pursuant to the OECD’s Harmful Tax Competition Project, which is
described in Chapter 9, section 9.2.2. The 2003 Commentary states explicitly
that “[I]t is also a purpose of tax conventions to prevent tax avoidance and
evasion.” Until that time, the only explicit statement found in the Treaty or
the Commentary was that the purpose of tax treaties was to eliminate double
taxation and prevent fiscal evasion (preventing tax avoidance was not
mentioned). Thus, taxpayers were able to argue that tax treaties could not be
interpreted to prevent tax avoidance because there was no strong evidence
that the prevention of tax avoidance was one of the purposes of tax treaties.
In addition, the 2003 Commentary on the OECD Model Treaty dealt
with the interpretation of the provisions of tax treaties to prevent the granting
of treaty benefits in abusive cases, and the relationship between domestic
anti-avoidance rules and tax treaties. With respect to the first issue, the
Commentary (paragraph 9.5) adopted, in effect, a general anti-abuse rule in
the guise of the following “guiding principle”:
A guiding principle is that the benefits of a double taxation convention should not be available
where a main purpose for entering into certain transactions or arrangements was to secure a
 more favourable tax position and obtaining that more favourable treatment in these
circumstances would be contrary to the object and purpose of the relevant provisions.

With respect to the second issue, the Commentary (paragraph 22.1)

indicates explicitly that domestic anti-avoidance rules are used to determine
the underlying facts on which tax liability is based and, as such, they are not
dealt with in tax treaties and are not affected by tax treaties. In general,
therefore, there is no conflict between domestic anti-avoidance rules and the
provisions of tax treaties, with the result that tax treaties do not prevent the
application of domestic anti-avoidance rules.
Although the main points of the Commentary on Article 1 of the OECD
Model Treaty are reasonably clear, as summarized above, that Commentary is
somewhat disorganized and makes some distinctions between different types
of domestic anti-avoidance rules, such as substance-over-form rules and CFC
rules, that are difficult to justify. In contrast, the Commentary on Article 1 of
the United Nations Model, which was revised as part of the 2011 update of
that Model, provides a better organized and more comprehensive discussion
of treaty abuse.
The UN Commentary on Article 1 (paragraph 10) identifies and
discusses the following seven approaches for dealing with treaty abuse:
– specific anti-avoidance rules in domestic law;
– general anti-avoidance rules in domestic law;
– judicial anti-abuse rules;
– specific anti-avoidance rules in tax treaties;
– general anti-avoidance rules in tax treaties; and
– the interpretation of treaty provisions to prevent abuse.
The Commentary explains that, although tax treaties generally prevail
over domestic law in the event of a conflict between them, conflicts between
domestic anti-avoidance rules and tax treaties can often be avoided.
Sometimes the provisions of the treaty (Article 9 with respect to transfer
pricing is an example) may explicitly allow the application of domestic anti-
avoidance rules; sometimes the treaty may depend on the application of
domestic law, including domestic anti-avoidance rules; and sometimes the
interpretation of the treaty will result in the denial of the benefits of the treaty
consistent with the denial of such benefits under domestic anti-avoidance
rules.
The UN Commentary on Article 1 also endorses the OECD guiding
principle (quoted above) as to what constitutes an abuse of a treaty. Further, it
indicates that, for countries that prefer not to rely on a non-binding statement
of the guiding principle in the Commentary, the guiding principle could form
the basis for a general anti-abuse rule to be included in the treaty. One
difficulty with the inclusion of such a general anti-abuse rule is that it may
create a negative implication that treaties without such a rule cannot be
interpreted to prevent treaty abuse. This is particularly problematic for
countries that have a large number of tax treaties because of the time needed
to renegotiate all the treaties. The OECD BEPS project has proposed to deal
with this difficulty through the negotiation of a multilateral treaty, which
would amend existing bilateral treaties with respect to the BEPS
recommendations for changes to the OECD Model Treaty. The OECD BEPS
Action 6 proposals with respect to treaty abuse are discussed in section
8.8.2.3.
Finally, the UN Commentary on Article 1 provides several useful
examples to illustrate the wide variety of potential treaty abuses and the rules
necessary to deal with them.

8.8.2.2 Treaty Shopping

Only residents and (in some cases) nationals of a contracting state are entitled
to benefits under an income tax treaty. Article 1 of the OECD and UN Model
Treaties provides that a treaty applies to persons who are residents of one or
both of the contracting states. Taxpayers who are not residents or nationals of
a contracting state have frequently sought to obtain the benefits of a tax treaty
by organizing a corporation or other legal entity in one of the contracting
states to serve as a conduit for income earned in the other contracting state.
This practice is commonly referred to as treaty shopping. Although a
taxpayer may engage in treaty shopping to obtain any treaty benefit not
otherwise available, most treaty shopping involves attempts by taxpayers to
obtain reduced withholding rates on dividends, interest, and royalties. Treaty
shopping is just one form of treaty abuse.
One classic form of treaty shopping involves the use of an unrelated
financial intermediary located in a treaty country to make investments for
taxpayers who are not themselves eligible for treaty benefits. For example,
assume that T is a resident of Country TH, a tax haven jurisdiction that does
not have a tax treaty with Country A. However, Country A has a tax treaty
with Country B, under which Country A reduces its normal withholding tax
rate from 30 percent to zero on interest paid to residents of Country B. T
invests 1 million with BCo, an independent financial intermediary that is
resident in Country B. BCo uses the 1 million to purchase a bond issued by
ACo, an unrelated corporation resident in Country A. ACo pays BCo 100,000
of interest on the bond. BCo claims that the 100,000 is exempt from County
A’s withholding tax under the treaty with Country B. BCo pays 100,000 to T,
minus some commission, as a return on T’s original investment.
This example utilizes what is commonly referred to as a back-to-back
arrangement to minimize taxes. BCo, the financial intermediary, avoids tax in
Country A under the tax treaty with Country B and avoids paying significant
tax in Country B because the 100,000 of interest received from ACo is offset
by the deduction of the amount paid to T.
Another classic form of treaty shopping involves the use of a controlled
corporation organized in a treaty country. For example, assume that T, in the
example above, organizes a wholly owned corporation, CCo, in Country C. T
subscribes 2 million for shares of CCo, and CCo uses that money to purchase
shares of stock in various companies resident in Country A and listed on
Country A’s stock exchange. CCo receives dividends of 400,000 on the
shares. Country A has a tax treaty with Country C that reduces Country A’s
withholding tax on dividends paid to residents of Country C from 30 percent
to 15 percent. As a resident of Country C, CCo claims the benefit of the
treaty to reduce its tax otherwise payable on the 400,000 of dividends from
120,000 to 60,000. Assuming that CCo is exempt from tax in Country C
because Country C does not tax foreign dividends under its domestic tax law,
this simple type of treaty shopping results in a tax saving of 60,000.
The international tax community has been slow to take action to curtail
treaty shopping. The OECD and UN Model Treaties do not contain any
specific provisions designed to combat the types of treaty shopping abuses
illustrated in the above examples. Indeed, some countries apparently have
concluded that tolerance of treaty shopping is in their national interest. The
US has been the leading proponent of aggressive international action to
curtail treaty shopping. The problem of treaty shopping is exacerbated for
countries, like the US, that enter into tax treaties with countries that have
widely varying withholding rates because taxpayers have an incentive to take
advantage of the most favorable treaty.
All recent tax treaties entered into by the US include a “limitation-on-
benefits” article intended to curtail treaty shopping. The basic policy of the
limitation-on-benefits article is to deny treaty benefits to a corporation that is
resident in one of the contracting states, but is in effect serving as a conduit
for residents of some third country. One way of thinking about a limitation-
on-benefits provision is to consider it an attempt to limit treaty benefits to
genuine residents of a contracting state.
The specific provisions of the limitation-on-benefits article contained in
US tax treaties vary from treaty to treaty. However, the limitation-on-benefits
provision in the US Model Treaty gives a good idea of the general US
approach. In general, a corporation resident in a contracting state is not
denied treaty benefits if it derives income from the active conduct of a trade
or business (other than the business of making investments) in the other
contracting state.
A corporation that fails to meet this active-business test must satisfy
both of the following conditions to qualify for treaty benefits:

(1) the corporation’s gross income must not be used in substantial part
to pay interest, royalties, or other liabilities to persons not entitled to
treaty benefits; and
(2) over 50 percent of the shares of the corporation (determined by
reference to both the voting rights attached to the shares and the
value of the shares) must be owned, directly or indirectly, by certain
qualified persons – typically, individuals who are residents of one
of the contracting states.

The first of these conditions is intended to combat treaty shopping of the

type illustrated in the first example above. The second condition addresses
treaty shopping of the type illustrated in the second example.
The limitation-on-benefits article contained in US tax treaties is
invariably more complex than the discussion above might suggest. Much of
the complexity involves the definition of a qualified person. In general, the
list of qualified persons includes individuals resident in a contracting state,
US citizens, publicly traded companies (and certain of their subsidiaries),
charitable foundations, and the contracting states themselves (including their
political subdivisions and local authorities).

8.8.2.3 OECD BEPS Action 6: Preventing Treaty Abuse

The OECD issued a Discussion Draft on BEPS Action 6: Preventing the
Granting of Treaty Benefits in Inappropriate Circumstances in March 2014
and a Revised Discussion Draft in May 2015. It is anticipated that the
proposals will be finalized by the end of 2015 and the changes incorporated
into the next update of the OECD Model Treaty. The Discussion Draft
presents a comprehensive set of proposals to deal with treaty abuse, ranging
from amendments to the title and preamble of the OECD Model to the
addition of a general anti-abuse rule to the Model. Although the Commentary
on Article 1 (paragraph 7) was revised in 2003 to establish that one of the
purposes of tax treaties is to prevent tax avoidance, the OECD proposes to
amend the title to the OECD Model Treaty to include specific references to
the elimination of double taxation and the prevention of tax evasion and
avoidance.
In addition to the revised title, the preamble to the OECD Model Treaty
will include a statement that the intentions of the contracting states are to
eliminate double taxation and prevent tax evasion and avoidance, including
tax avoidance “through treaty shopping arrangements aimed at obtaining
reliefs provided in this Convention for the indirect benefit of residents of
third States” (see section 8.5 above for the full preamble). The intention
behind these changes to the title and preamble is to elevate the prevention of
tax avoidance to a main purpose of tax treaties, along with the elimination of
double taxation, so that national courts will take this purpose into account in
interpreting the provisions of actual bilateral treaties that adopt the OECD
recommendations.
The changes to the title and preamble will be reinforced by revisions to
the Introduction to the OECD Model Treaty. Most importantly, the
Introduction will state that the elimination of double taxation and the
prevention of tax evasion and avoidance are the main purposes of the OECD
Model Treaty. Currently, the Introduction indicates that the elimination of
double taxation is the main purpose of the OECD Model Treaty.
The Discussion Draft makes a conceptual distinction between the
circumvention of the provisions of a treaty itself and the use of a treaty to
circumvent domestic law. The most obvious example of the first type of tax
avoidance is treaty shopping. The
Discussion Draft proposes to deal with treaty shopping by including a
new article (Entitlement to Treaty Benefits) containing a detailed US-style
limitation-on-benefits provision (see section 8.8.2.2 above), supplemented by
a general anti-abuse rule.
The detailed limitation-on-benefits provision denies treaty benefits
based on the legal nature, ownership and control, and activities of a resident
of a contracting state. The inclusion of a derivative benefits provision in the
limitation on benefits is still under consideration. A derivative provision
would allow an entity to qualify for treaty benefits to the extent that the
owners of the interests in the entity would be entitled to equivalent or more
favorable benefits if they received the payments or income directly.
Since the proposed limitation-on-benefits provision would not cover all
transactions involving treaty shopping (e.g., conduit financing arrangements),
the Discussion Draft recommends that a general anti-abuse rule should be
added to the OECD Model Treaty as follows:
Notwithstanding the other provisions of this Convention, a benefit under this Convention shall
not be granted in respect of an item if it is reasonable to conclude, having regard to all
relevant facts and circumstances, that obtaining that benefit was one of the main purposes of
any arrangement or transaction that resulted directly or indirectly in that benefit, unless it is
established that granting that benefit in these circumstances would be in accordance with the
object and purpose of the relevant provisions of this Convention.

In effect, the guiding principle in paragraph 9.5 of the Commentary on

Article 1, discussed in section 8.8.2.1 above, would be incorporated into the
Model itself. The inclusion of such a general anti-abuse rule in actual
bilateral tax treaties will have significant impact. Although the Commentary
is considered to be of persuasive value, it is clearly not binding on domestic
courts; however, a rule contained in the treaty itself cannot be easily ignored.
The proposed general anti-abuse rule consists of a two-part test. First,
one of the main purposes of a transaction or arrangement must be to obtain
treaty benefits and second, obtaining treaty benefits in the particular
circumstances must be contrary to the object and purpose of the relevant
provisions of the treaty. If the purpose test is met, the onus is on the taxpayer
to establish that obtaining the treaty benefits would be in accordance with the
purpose of the treaty.
The Discussion Draft also proposes to add several specific anti-abuse
rules to the OECD Model Treaty, as follows:
– A holding-period requirement will be added for purposes of the
application of the 5 percent tax rate on dividends in Article 10(2)
and the taxation of capital gains from the alienation of shares in
land-rich companies under Article 13(4). For the latter purpose, the
 alternative provision in the Commentary, which extends the
provision to interests in other entities such as partnerships and trusts,
will be incorporated into Article 13. The length of the holding period
has yet to be determined.
– The tie-breaker rule for entities based on the place of effective
management in Article 4(3) will be replaced by a determination by
the competent authorities on a case-by-case basis. If the competent
authorities cannot agree, the entity will not be entitled to any treaty
benefits.
– Treaty benefits will be denied with respect to income attributable to a
permanent establishment (PE) in a third state. The type of rule that
would be used to accomplish this result has not been decided. The
Discussion Draft suggests a possible approach based on the rule in
some US treaties, under which treaty benefits are denied where the
income is subject to a preferential rate of tax (taking into account the
taxes in both the residence country and the PE country) that is less
than 60 percent of the corporate tax rate in the residence country.
The Discussion Draft proposes to clarify the Commentary on Article 1
with respect to the relationship between tax treaties and specific anti-
avoidance rules in domestic law by adopting the approach used in the
Commentary on Article 1 of the UN Model Treaty (2011). The OECD
Commentary is confusing because portions of it were adopted at different
times in response to different concerns. As discussed in section 8.8.2.1, the
UN Commentary on Article 1 is much clearer than the OECD Commentary
in this regard.
The final proposal in the Discussion Draft is the addition of a saving
clause to the OECD Model to prevent residents of a country from relying on
the provisions of a tax treaty to avoid residence country tax. The saving
clause is a standard feature of US tax treaties and provides that, subject to
certain exceptions, the treaty does not affect the taxation by a state of its own
residents. The exceptions to the saving clause are the benefits provided under
Articles 7(3), 9(2), 19, 20, 23, 24, 25, and 28.
The addition of a saving clause to the OECD Model is a good idea in
principle because tax treaties are not generally intended to prevent a country
from taxing its residents, and it will put other countries on the same footing
as the US. Because of the saving clause, the US does not need to worry about
applying its transfer pricing rules, CFC rules, or other anti-avoidance rules to
its residents. In my view, it does not make sense for the US to be able to
apply its domestic anti-avoidance rules without regard to its tax treaties, but
not for other countries to do so, except to the extent permitted by the
Commentary on Article 1, by specific provisions in their treaties, or by virtue
of a treaty override.
In summary, the Discussion Draft presents a fairly comprehensive list of
proposed changes to the OECD Model to deal with treaty abuse. The list is
not completely comprehensive yet because it will be supplemented by other
proposals stemming from other BEPS actions. For example, a new provision,
Article 1(2), will be added to the OECD Model Treaty, indicating that income
derived through a transparent entity will be considered to be income of a
resident of a contracting state for purposes of the treaty only to the extent that
it is treated as income of a resident of that state for purposes of taxation by
that state.
It will take a long time for a country with numerous tax treaties to
renegotiate all its treaties to include the proposed general anti-abuse rule and
the other proposed changes. In the meantime, if the changes are included in
only some of a country’s tax treaties, a negative implication may arise that
treaties without those provisions cannot be interpreted to prevent treaty
abuse. Therefore, the key issue with respect to the OECD proposals on treaty
abuse is that their effectiveness depends on the conclusion of a multilateral
treaty. Although the Discussion Draft on treaty abuse does not make any
reference to BEPS Action 15: Developing a Multilateral Instrument to
Modify Bilateral Tax Treaties, if OECD member countries (and possibly
other countries as well) can agree to a multilateral treaty that provides for the
amendment of all their existing treaties to incorporate the changes aimed at
preventing treaty abuse, the changes could be implemented relatively quickly.
Therefore, a multilateral treaty is critical to the effective implementation
of the OECD proposals dealing with treaty abuse, but a multilateral tax treaty
has been an elusive dream since the beginning. Negotiations for the
multilateral treaty are planned to commence early in 2016.

8.8.3 Resolution of Disputes

Most tax treaties provide a mutual agreement procedure for resolving
disputes that arise under the treaty. A person resident in a contracting state
who believes that the actions of one or both contracting states will cause the
payment of tax not in accordance with the treaty may request relief from the
“competent authority” of the state of which the person is a resident. The
competent authority is typically a senior official in the country’s tax
department who is responsible for international tax matters. The competent
authority will make a determination whether the taxpayer’s request appears
justified and, if so, will attempt to provide an appropriate remedy. If the
competent authority does not have the power to resolve a dispute on its own,
it may attempt to resolve the dispute through consultations with the
competent authority of the other contracting state.
The dispute-resolution mechanism of the OECD and UN Model Treaties
is set forth in Article 25 (Mutual Agreement Procedure). This article provides
that the competent authorities “shall endeavor” to resolve matters referred to
them. Thus, it is notable that the competent authorities are not required to
reach an agreement, even if the result is that a taxpayer is subject to double
taxation. For this reason, mandatory arbitration was added to Article 25 of the
OECD Model Treaty in 2008 for the resolution of issues that the competent
authorities are unable to agree on within two years. Arbitration is discussed in
agree on within two years. Arbitration is discussed in Chapter 9, section 9.5.
Although a taxpayer is entitled to make its case to the competent
authority of its country of residence, it is not allowed to participate directly in
the consultative procedure between the competent authorities of the two
contracting states. However, a few treaties contemplate that the taxpayer is
entitled to present its case independently to both competent authorities.
A variety of disputes may be referred to the competent authorities. Some
of these disputes involve the proper interpretation of treaty language, while
others involve disputes over the facts on which a taxpayer’s tax liability is
based. The most common and complex disputes referred to the competent
authorities involve the proper application of the arm’s-length standard to
transfer prices in cross-border transactions. These disputes are sometimes
difficult to resolve for a variety of reasons, including the large amounts of tax
revenue frequently at stake. Article 9(2) of the OECD and UN Model
Treaties provides that if one country adjusts the transfer prices used by
related corporations in accordance with the arm’s-length standard (and the
other country agrees with the adjustment), the other country must make a
corresponding adjustment to the profits of the related corporation in order to
avoid double taxation. Transfer pricing is discussed in Chapter 6.
 Under the domestic laws of many countries, multinational companies
may request formal advance approval from the tax authorities of their
methodology for establishing prices in their inter-group transactions. An
administrative ruling entered into under this procedure is commonly referred
to as an Advance Pricing Agreement (APA). In many situations,
multinational enterprises prefer, if possible, to use the competent authority
procedure to arrange for the joint issuance of an APA by several of the
countries in which they do business.

8.8.4 Administrative Cooperation

The tax authorities of a country often experience difficulty in obtaining
information concerning the foreign activities of residents and verifying that
the information is correct. In the past, this difficulty was exacerbated by bank
secrecy laws in many countries and the unwillingness of tax havens to
exchange information with high-tax countries.
Article 26 (Exchange of Information) of the OECD and UN Model
Treaties provides for an exchange of “such information as is foreseeably
relevant for carrying out the provisions of this Convention or to the
administration and enforcement of the domestic laws of the contracting states
concerning taxes of every kind and description.” Article 26 of the OECD
Model Treaty was revised in 2005 to change the standard for exchanging
information from “necessary” to “foreseeably relevant” and to require the
exchange of information with respect to all taxes imposed by the contracting
states, and not just the taxes covered by the treaty. These changes were
intended to clarify the meaning of the Article, but not to change its substance.
Under the OECD and UN Model Treaties, an exchange of information
may take place as a result of a specific request from a treaty partner, through
an arrangement for an automatic exchange of information, or by the initiative
of a contracting state acting spontaneously. Information requested by one
state must be provided by the other state despite the fact that the information
may not be necessary or relevant for the purposes of the other state’s own
taxes (i.e., the other state may have no domestic interest in the information).
Information obtained by the tax department of a contracting state under
an exchange-of-information article must be kept confidential, although
release of the information in court proceedings is generally allowed. Under
the exchange-of-information article, a contracting state is not obligated to
carry out administrative procedures on behalf of its treaty partner that are
contrary to its own laws or practices, to supply information that is not
obtainable under its domestic laws or in the normal course of administration
of both states, or that would result in the disclosure of trade secrets or similar
information. An escape clause generally allows a contracting state not to
provide requested information if its disclosure would be “contrary to public
policy.” However, under Article 26(5), a country cannot refuse to provide
information solely because the information is held by a financial institution, a
nominee or agent, or because it relates to ownership interests in a corporation
or other person. Since most countries have been required to eliminate their
bank secrecy laws, this provision is not as important as it would otherwise be.
Exchanges of information between tax authorities can take place only
pursuant to international agreements between countries; therefore, in the
absence of a bilateral tax treaty, it is usually impossible for the tax authorities
to share information. Before this century, it was impossible for high-tax
countries to get information from tax havens or low-tax countries with which
they did not have a bilateral tax treaty. As a result, the practice developed for
TIEAs to be concluded on a bilateral basis between OECD member countries
and low-tax countries with which there was no need for a comprehensive
income tax treaty. The US has been the leader in this regard; several OECD
member countries have also concluded or are negotiating TIEAs with low-tax
countries. TIEAs provide for countries to exchange information on a basis
similar to Article 26 of the OECD Model Treaty.
As part of the OECD’s Harmful Tax Competition Project in the late
1990s, the OECD proposed that tax havens should be required to obtain
information about the beneficial ownership of companies and other entities
formed under their laws and to require the companies to maintain financial
accounts in accordance with generally accepted accounting standards and
make those accounts available for the regulatory or tax authorities. In 2002,
the OECD issued a Model Agreement on Exchange of Information on Tax
Matters. Article 26 of the OECD Model Treaty was revised in 2005 to
override any bank secrecy or other confidentiality laws of the country
requested to provide information, and to delete the necessity for any domestic
tax interest in the requested information. In 2011, Article 26 of the UN Model
Treaty was revised to conform to Article 26 of the OECD Model, although
Article 26 of the UN Model Treaty is broader in certain respects. For
example, Article 26(1) includes the statement that information that is helpful
in preventing tax avoidance or evasion shall be exchanged. In addition,
Article 26(6), which authorizes the competent authorities to establish
procedures for the exchange of information, has no counterpart in Article 26
of the OECD Model Treaty.
In 2001, the OECD established a Global Forum on Taxation to discuss
exchange-of-information issues with non-member countries. This Global
Forum, which is now known as the Global Forum on Transparency and
Exchange of Information, has been very successful in ensuring that
international standards for the exchange of information are implemented
effectively. The Global Forum, which currently consists of 127 countries,
engages in peer review exercises of both the legal and administrative capacity
of countries to exchange information and of their actual performance in
exchanging information. Countries are given ratings (which are available to
the public) based on their compliance with the international standard for
exchange of information.
Until recently, the international standard for exchanges of information
between tax authorities has required only exchanges on request. In other
words, one country was required to provide information only if the tax
authorities of the other country specifically requested that information.
However, in 2014 the OECD formulated a new Standard for Automatic
Exchange of Financial Information in Tax Matters, (available at
www.oecd.org) which provides for certain financial information (e.g.,
information about dividends, interest, proceeds of sale of financial products,
and balances of certain accounts) obtained from a country’s financial
institutions to be provided automatically (i.e., without the necessity for a
request by the tax authorities of another country) to the tax authorities of
other countries on an annual basis. Over ninety countries have agreed to this
new standard. In addition, over sixty countries have signed a Multilateral
Competent Authority Agreement (available at ww.oecd.org) to implement
automatic exchanges of information.
Often, the tax authorities of a country will audit the international affairs
of taxpayers in addition to requesting information from other countries.
Under international custom, however, the tax officials of one country cannot
visit another country for the purpose of auditing a taxpayer’s records unless
invited to do so by the foreign government. Some governments consider it
inappropriate for tax officials to make such visits without also obtaining the
concurrence of the taxpayer. Several countries have addressed this problem
by conducting joint audit programs, under which a particular taxpayer (and
its affiliates) is audited by the tax authorities of both countries.
Once the tax authorities of a country have conducted an audit and
assessed a tax deficiency against a taxpayer, they must collect any taxes
owing. Tax authorities often encounter severe difficulties in enforcing tax
liability in another country. Under the domestic law of most countries, the tax
judgments of a foreign country generally are not enforceable, in accordance
with the “revenue rule.” Article 27 (Assistance in the Collection of Taxes) of
the OECD and UN Model Treaties overcomes the limitations of the revenue
rule by requiring each country to provide assistance in the collection of the
other country’s taxes. Like Articles 24 (Nondiscrimination) and 26
(Exchange of Information), Article 27 is not limited to the taxes covered by
the tax treaty, but extends to all taxes imposed by the contracting states.
A request for assistance must be accepted by the requested state if the
taxpayer cannot resist the collection of the taxes under the laws of the
requesting state. In addition, the requested state must collect the taxes of the
requesting state as if those taxes were its own. However, it is not required to
provide assistance unless the requesting state has exhausted all the measures
for the collection or conservancy of the taxes available under its domestic law
and administrative practices, or the administrative burden to collect the taxes
is disproportionate to the benefit to the requesting state. A taxpayer is not
entitled to contest the existence, validity, or amount of the taxes owing in the
courts or administrative bodies of the requested state. In providing assistance
in the collection of the taxes owing to the other state, a state is not required to
take any measures that are inconsistent with its own laws or administrative
practices or contrary to public policy.

8.8.5 Attribution of Profits to Permanent Establishments

As discussed above in Chapter 5, section 5.8.1.2, an entity resident in one
country may engage in substantial business activities in another country
through a branch or PE in that country or through an entity (such as a
subsidiary) established in that country. Unlike a subsidiary, a branch or PE is
not a legal entity and cannot take actions on its own. The property and
activities of a branch or PE are actually the property and activities of the
entity of which it is a part.
When an enterprise resident in one country is engaged in business
activities through a branch or PE in another country, it is necessary for both
countries to determine the amount of income of the branch or PE. The source
country requires this information in order to determine the amount of the
nonresident enterprise’s profits subject to source country tax. The country in
which the enterprise is resident requires the information in order to provide
relief from double taxation – by way of exemption or a foreign tax credit – of
the profits earned by the enterprise through the foreign branch or PE. The
domestic rules used by countries to compute the profits earned by a
nonresident vary considerably, as discussed in Chapter 5, section 5.8.1.1.
Under Article 7(1) of both the OECD and UN Model Treaties, a resident
of one country is taxable with respect to business profits by the other country
only if it carries on business in the other country through a PE located in that
country. Under the OECD Model Treaty, a resident carrying on business
through a PE in the other country is taxable only on the profits attributable to
the PE. Under the UN Model Treaty, such a resident is also taxable on profits
from sales of goods similar to those sold through the PE and from other
business activities similar to those carried out through the PE. This limited
force-of-attraction rule in the UN Model Treaty is not very important because
it is so easy to avoid.
Article 7(2) of both Model Treaties requires the profits attributable to a
PE to be determined on the assumption that the PE is a separate enterprise
dealing independently with the rest of the enterprise of which it is a part. (The
rest of the enterprise means the head office of the enterprise and any other
PEs of the enterprise.) This assumption is generally considered to make the
transfer pricing rules of Article 9 applicable in determining the profits
attributable to a PE.
Article 7 of the OECD Model Treaty was substantially revised in 2010
pursuant to a decade-long project, which culminated in 2008 with a report,
Attribution of Profits to Permanent Establishments, and a new chapter
dealing with PEs in the OECD’s Transfer Pricing Guidelines. This new
chapter was developed largely by economists working through Working
Party 6, which did not take into account any of the constraints imposed by the
wording of the existing Article 7 of the treaty. In effect, Working Party 6
took the separate-entity assumption in Article 7(2) to its logical conclusion;
thus, a PE was required to be treated as a separate entity for all purposes in
computing its profits. This new approach required a complete overhaul of the
wording of Article 7.
 Article 7(2) of the OECD Model Treaty was revised to provide that the
profits attributable to a PE must be computed as if the PE were a separate
entity “taking into account the functions performed, assets used and risks
assumed by the enterprise through the permanent establishment and through
the other parts of the enterprise.” Paragraph 3 through 6 of Article 7,
discussed below, were deleted and a new corresponding adjustment similar to
Article 9(2) was added to Article 7(3). Article 7(3) requires a contracting
state to make an adjustment to the profits of an enterprise if the other
contracting state makes an adjustment in accordance with Article 7 to the
profits of a PE of the enterprise in that other state.
The changes to Article 7 have been controversial. Seven OECD member
countries have entered reservations on the new Article, indicating that they
reject the new Article 7 and intend to adhere to the former version of the
article. In addition, the UN Committee of Experts refused to adopt the
OECD’s new version of Article 7 in the 2011 revision of the UN Model
Treaty.
Serious conceptual and practical difficulties arise in applying the
transfer pricing rules to PEs, as described in Chapter 6. The transfer pricing
rules in Article 9 apply to transactions between related persons. They do not
apply to PEs because a PE is not a person; it is merely part of a legal
enterprise, and transactions do not take place between parts of the same
enterprise. As a legal matter, a transfer requires a change in ownership from
one person to another, and a PE cannot own property. What is often described
metaphorically as a transfer of property between the head office and a PE of
an enterprise or between two PEs of the same enterprise is merely a change in
the use or location of property owned by that corporation, and not a genuine
transfer.
Therefore, in order to apply transfer pricing rules to PEs, it is necessary
to treat a PE as if it were a separate legal entity and to construct some
hypothetical transactions between the PE and the rest of the enterprise of
which the PE is a part. Assume, for example, that ACo, resident in Country
A, manufactures goods in Country A and sells those goods through a sales
outlet in Country B. To apportion the income of ACo between the two
countries by reference to the transfer pricing rules, ACo’s sales activities
through its PE in Country B would be treated as if they were carried on by a
subsidiary corporation resident in Country B (e.g., BCo). ACo would be
treated as if it had made a transfer of goods, either by sale or consignment, to
the assumed BCo. The transfer pricing rules would then be applied to that
notional transfer. An assumption would have to be made as to whether BCo
was operating in Country B as an independent distributor or as an agent of
ACo because the income earned by distributors and agents in the marketplace
might not be the same.
Under Article 7(2) of the OECD Model Treaty, the determination of the
profits attributable to a PE is a two-step process. The first step involves a
functional and factual analysis of the PE in order to determine the functions
performed by the PE, the economic ownership of assets by the PE, the risks
assumed by the PE, the capital of the PE, and the hypothetical “dealings”
between the PE and the other parts of the enterprise. The second step
involves the application of the transfer pricing rules, by analogy, to those
dealings in order to establish an arm’s-length price.
Because of the necessity to invent dealings between a PE and another
part of the enterprise, the application of the OECD’s transfer pricing rules to
PEs is even more difficult, uncertain, and less reliable than their application
to transactions between associated enterprises. As noted above, several
countries have rejected the OECD’s new approach, and many countries will
be unable to apply those rules effectively.
Most existing tax treaties contain a business-profits article that is similar
to the former version of Article 7 of the OECD Model Treaty and the current
version of Article 7 of the UN Model Treaty. As noted above, under Article
7(2), the profits of a PE are required to be computed on the assumption that a
PE is a separate entity dealing independently with the rest of the enterprise of
which it is a part. Thus, this version of Article 7(2) is not significantly
different from the new OECD version of Article 7(2). However, the
Commentary on the former version did not take the separate-entity
assumption to its logical conclusion; instead, it provided a series of ad hoc
practical rules for computing the profits attributable to a PE. Sometimes the
Commentary required the PE to be treated as a separate entity; for example, if
a PE transferred assets to the head office of the enterprise, the profits of the
PE were computed as if the PE had sold the assets for their fair market value
at the time of the transfer and the head office had acquired the assets for the
same amount at the same time. In contrast, sometimes the Commentary
rejected treating the PE as a separate entity; for example, notional payments
of interest or royalties were not deductible in computing the profits of a PE
under the former version of Article 7 (except in the case of financial
institutions), whereas they are deductible under the new OECD Article 7.
Consider the following example, which illustrates the different
approaches under the two versions of Article 7. ACo is an enterprise resident
in Country A that commences to carry on business through a PE in Country
B. ACo borrows 200,000 with interest at 10 percent, which it transfers to the
PE to finance the establishment of the business carried on in Country B. ACo
also advances an additional 1 million to the PE to finance the PE’s business.
Under the new Article 7, it would be necessary to determine how much debt
and equity a separate entity would have if it performed the functions of the
PE, owned the assets of the PE, and assumed the risks of the PE. Thus, if
such a separate entity would have debt equal to twice its equity, the PE would
be assumed to have 400,000 of equity and 800,000 of debt; thus, interest at
10 percent (assuming that 10 percent is an arm’s-length interest rate) would
be deductible in computing the profits attributable to the PE. In contrast,
under the former version of Article 7 of the OECD Model Treaty and the
current version of Article 7 of the UN Model Treaty (see Article 7(3), which
explicitly denies any deduction for notional payments of interest or royalties),
only interest on 200,000 (the actual interest expense incurred by the ACo in
respect of debt used for the purposes of the PE) would be deductible in
computing the profits of the PE.
Allowing the deduction of notional payments of interest and royalties is
problematic. Where actual payments of interest and royalties are made and
those payments are deductible in computing the profits of a PE, under Article
11 or 12 of the Model Treaties, the source country is entitled to impose
withholding tax on the payments. However, notional payments of interest and
royalties are not subject to source country withholding tax.
International banks, insurance companies, and other financial services
companies often operate their global businesses through branches. Often the
reason for using a branch is to satisfy capital reserve requirements imposed in
many countries to protect investors and customers. Under the Commentary
on former Article 7(3) of the OECD Model Treaty (paragraph 49) and the
Commentary on Article 7 (3) of the UN Model Treaty (paragraph 41), a
financial institution is allowed to deduct notional interest payments in
computing the profits of a PE. According to the Commentary, the special
treatment of banks and other financial institutions is appropriate “in view of
the fact that making and receiving advances is closely related to the ordinary
business of such enterprises.”
 Article 7(3) of the UN Model Treaty (and former Article 7(3) of the
OECD Model Treaty) provides that expenses incurred by an enterprise for the
purposes of a PE must be allowed as deductions in computing the profits of
the PE irrespective of where the expenses are incurred and whether they are
incurred wholly on behalf of the PE. Thus, the PE country cannot deny the
deduction of expenses because they are incurred outside the source country or
because only a portion of the expenses relates to the PE. For example, head
office expenses, such as accounting and legal expenses, which are incurred
by the head office on behalf of a PE, must be allowed as deductions in
computing the profits of the PE. However, it must be emphasized that the
deductibility of expenses is a matter for domestic law. Therefore, for
example, if only a portion of entertainment expenses is deductible under the
domestic law of the source country, the source country is not required by
Article 7 to allow the full amount to be deductible in computing the profits of
the PE.
Article 7(4) of the UN Model Treaty (and former Article 7(4) of the
OECD Model Treaty) provides that the profits of a PE may be computed in
accordance with a formulary apportionment of the profits of the enterprise as
a whole as long as that has been the customary practice of the country in
which the PE is located. However, any such formulary apportionment must
produce a result that is consistent with the principles of Article 7; in other
words, it must be consistent with the profits that the PE would be expected to
make if it were a separate entity.
Article 7(5) of the UN Model Treaty (and former Article 7(5) of the
OECD Model Treaty) provides that the profits of a PE must be determined on
a consistent basis from year to year unless there is some good and sufficient
justification for a change in the method for computing PE profits.
No country has developed detailed and comprehensive domestic rules
for extending transfer pricing rules to branches. In practice, most countries
compute the profits of a PE on the basis of the profits shown in the taxpayer’s
books of account and make ad hoc adjustments if those books do not produce
a reasonable result. However, since the books and records of a PE are within
the control of the enterprise, they may not be reliable, and the tax authorities
must scrutinize them carefully to ensure that they accurately reflect the
profits of the PE.
CHAPTER 9
Emerging Issues

9.1 INTRODUCTION
This chapter deals with several important recent developments in
international tax. Although many of these developments have been
mentioned in previous chapters, they are dealt with in detail here for
convenience. There is no overarching theme to the topics dealt with in this
chapter other than the fact that they have become increasingly important in
recent years.
The issues discussed in this chapter are wide-ranging. They include
international aspects of domestic law and tax treaties, substantive issues, and
administrative issues such as exchange of information and arbitration.
The prominence of the issues discussed here is largely attributable to the
work of the OECD and the United Nations (UN). The OECD has become a
dominant player with respect to international tax issues, as the recent G-
20/OECD BEPS project shows. In recent years, the UN, through the
Committee of Experts and its Capacity Development Unit, has also started to
exert an increasingly important influence in the international tax arena. For
example, the proposal to add a new article to the UN Model Treaty dealing
with fees for technical services, discussed in section 9.4 below, represents a
potentially important step in the evolution of tax treaties.

9.2 BASE EROSION AND PROFIT SHIFTING (BEPS)

9.2.1 Introduction
Since 2012, the OECD’s BEPS project has dominated the international tax
agenda. As discussed below in section 9.2.3, the scope of the project is huge,
comprising fifteen action items, and the time frame for its completion – the
end of 2015 – is unrealistically tight. The project involves not only OECD
member countries, but also G20 and developing countries. Many of the action
items are discussed elsewhere in this Primer. This section presents an
overview of the BEPS project and its implications for the international tax
system. It is important to understand that the problems with the international
tax system targeted by the BEPS project are not new. Many of them involve
fundamental structural features of the allocation of taxing rights between
source and residence countries that countries and international organizations,
such as the OECD and the UN, have struggled with for decades. As
background for understanding the BEPS project, section 9.2.2 discusses the
OECD’s first attempt to deal with some of the fundamental problems of the
international tax system – the harmful tax competition initiative of the late
1990s.

9.2.2 The 1998 OECD Harmful Tax Competition Report

Over the last fifty years, the proliferation and increased use of tax havens and
preferential regimes in otherwise high-tax countries has been staggering.
Many developments have contributed to this phenomenon, including:

– the elimination of exchange controls and the liberalization of cross-

border trade and investment;
– improved communications, transportation, and financial services;
– the globalization of tax advisory firms;
– the adoption of flexible commercial regimes and strict bank secrecy
and confidentiality requirements by tax havens; and
– aggressive marketing.

Tax havens and countries with preferential regimes have become very
sophisticated in marketing their services to every geographical region of the
world and to every conceivable tax planning need. Although the competition
among them is fierce, the field remains crowded, with newer havens and
regimes continually being introduced to try to get a piece of the action.
This proliferation of low-tax regimes and their base-eroding effects are
what the OECD has called a “race to the bottom.” Concern about this race to
the bottom led to the OECD and European Union (EU) initiatives against
harmful tax competition in the late 1990s. After years of work and difficult
negotiations, the OECD issued a Report on Harmful Tax Competition in
April 1998. The EU adopted a Code of Conduct concerning harmful tax
competition in December 1997 (Commission of the European Communities,
A Package to Tackle Harmful Tax Competition in the European Union).
Although the two initiatives were complementary, the EU Code of Conduct
was considerably broader than the OECD Report because it was not limited
to geographically mobile activities (the OECD Report was limited to
geographically mobile activities, including financial services). The Report
recognized that harmful tax competition also occurs with respect to non-
mobile activities, such as manufacturing and personal savings, but these
activities were left for future action because they were considered more
difficult to deal with.
Although the OECD Report was targeted at “harmful tax competition,”
no definition of that expression was provided in the Report. This shortcoming
was understandable because the term is impossible to define precisely. The
use of the term “competition” was perhaps unfortunate because in an era of
global free trade and capitalism, it was an article of faith that all competition
is good. However, the essential concept inherent in the term “harmful tax
competition” is that there is a difference between the type of tax competition
that led to a worldwide lowering of tax rates and broadening of tax bases in
the late 1980s, and harmful tax competition, which involves the race to the
bottom described earlier. According to the OECD, although every country
has the sovereign right to determine its own tax policy, a country should not
enact policies intended to “poach” the tax base of other countries.
The OECD Report targeted both tax havens and harmful preferential tax
regimes, including such regimes in the tax systems of OECD member
countries. The Report also suggested a lengthy list of countermeasures that
countries might take on both a unilateral and a coordinated basis to deal with
tax havens and low-tax regimes. However, the OECD soon abandoned the
idea of sanctions against tax havens in favor of a strategy of dialogue and
cooperation.
The harmful tax competition initiative was significant because it
signaled the beginning of serious international cooperation in fighting tax
evasion and avoidance. However, the OECD’s action was very controversial.
It was accused by some commentators and several smaller tax haven
countries of being a cartel of rich countries dictating to small,
underdeveloped countries, and of trying to impose a particular type of tax
system – an income tax – and a minimum rate of tax. Other commentators,
however, have welcomed the OECD initiative as the next logical step for
countries to take to protect their domestic tax bases.
Any major new tax policy initiative is a political issue, and the harmful
tax competition project was no exception. In late 2000, the United States
(US), which until that time had been an enthusiastic supporter of the harmful
tax competition project, effectively compelled the OECD to refocus the
project almost exclusively on exchange of information. As a result, the
OECD committed to developing standardized exchange-of-information
provisions in both bilateral and multilateral formats, dealing with both
criminal and civil tax matters; its efforts have resulted in a much more
effective and efficient system with respect to exchange of information.
Exchange of information is discussed in Chapter 8, section 8.8.4.
Although the original goal of the OECD’s project was changed from
eliminating harmful tax competition to the more modest goal of effective
exchange of information, the project has produced some significant results.
For example, the Commentary on Article 1 of the OECD Model Treaty was
revised extensively in 2001 to clarify the relationship between tax treaties and
domestic anti-avoidance rules. This issue is discussed in Chapter 8, section
8.8.2. Also, some of the most obvious preferential tax regimes in OECD
member countries have been eliminated.

9.2.3 The G20/ BEPS Project

In retrospect, it was only a question of time before pressures on countries’ tax
revenues led to additional efforts to protect their domestic tax bases from the
tax planning strategies of multinational enterprises; these efforts gained
momentum during the financial crisis of 2008. In 2012, the OECD launched
its project against BEPS. The project received a shot in the arm when
aggressive tax-planning techniques used by several US multinationals,
including Apple, Amazon, Starbucks, and Google, came under intense
criticism by European and American politicians.
The BEPS initiative was a natural outgrowth of the OECD’s work on
exchange of information as a tool to combat international tax avoidance,
although the impetus for the project was the final declaration of the meeting
of the G20 finance ministers in June 2012, which emphasized “the need to
prevent base erosion and profit shifting.” In February 2013, the OECD
responded to the G20’s concerns by issuing a short note, “Addressing Base
Erosion and Profit Shifting,” that identified several areas for action and
deadlines for the implementation of responses. The G20 finance ministers
meeting in February 2013 welcomed the OECD report and strongly
supported the initiative in the following terms:
We are determined to develop measures to address base erosion and profit shifting, take
necessary collective actions and look forward to the comprehensive action plan the OECD
will present to us in July.

On July 19, 2013 the OECD released a detailed Action Plan on Base
Erosion and Profit Shifting (OECD Action Plan). This Action Plan sets out
an ambitious agenda with tight deadlines; it consists of the following actions:

(1) Develop rules to allow countries to impose direct and indirect

taxation on electronic commerce (the digital economy) (electronic
commerce is dealt with in section 9.6).
(2) Develop treaty provisions and recommendations for domestic rules
to deal with hybrid mismatch arrangements involving the use of
hybrid entities and hybrid financial instruments (the BEPS
proposals with respect to hybrid entities and hybrid financial
instruments are discussed in section 9.3).
(3) Develop proposals to strengthen controlled foreign corporation
rules (the BEPS proposals with respect to CFC rules are discussed
in Chapter 7, section 7.3).
(4) Develop recommendations to deal with base erosion through
interest and other financing expenses (the BEPS proposals with
respect to interest deductions are discussed in Chapter 7, section
7.2).
(5) Develop more effective countermeasures for harmful tax practices.
(6) Develop treaty anti-abuse provisions and recommendations for the
adoption of domestic anti-abuse rules (the BEPS proposals with
respect to treaty abuse, including treaty shopping, are discussed in
Chapter 8, section 8.8.2.3).
(7) Revise the definition of permanent establishment (PE) in the OECD
Model Treaty to prevent the use of commissionaire and other
arrangements to avoid PE status (the BEPS proposals with respect
to the definition of a PE are discussed in Chapter 8, section
8.7.3.2).
(8) Revise transfer pricing rules to ensure that transfer pricing cannot
be used for base erosion and profit-shifting purposes (the BEPS
proposals with respect to transfer pricing are discussed in Chapter
6).
 (9) Develop methodologies for the collection and analysis of data
concerning BEPS and for the evaluation of the effectiveness of
measures to counteract BEPS.
(10) Develop measures to require the disclosure of aggressive tax
planning arrangements.
(11) Improve transfer pricing documentation (see Chapter 6, section
6.8).
(12) Improve the treaty process for the resolution of disputes, including
arbitration (the resolution of tax disputes is discussed in Chapter 8,
section 8.8.3, and arbitration is discussed below in section 9.5).
(13) Develop a multilateral treaty to implement BEPS countermeasures
and to amend bilateral treaties (the multilateral treaty is discussed
in Chapter 8, section 8.8.2.3).

The OECD’s BEPS Action Plan is obviously an ambitious one, even

taking into account the fact that the OECD had been working on some of the
issues, such as hybrid mismatch arrangements and transfer pricing, before the
announcement of the BEPS project. It is even more ambitious considering the
timeframe for the delivery of the OECD’s recommendations. Only the work
on the transfer pricing aspects of financial instruments, part of the work on
harmful tax practices, and the development of a multilateral treaty was
scheduled to take more than two years. With respect to the other issues,
OECD recommendations were due to be completed by September 2014 or
September 2015. The OECD has adhered to this timetable, although it has
acknowledged that work on many of the issues will necessarily continue after
2015. The tight timing of the BEPS project emphasizes the importance of the
work to the OECD and the G20; however, it also raises some concerns about
the quality of the work produced under such tight time constraints.
At this stage, several general comments about the BEPS project seem
appropriate. First, the project has gained widespread support from the G20,
which includes Brazil, Russia, India, China, and South Africa (the so-called
BRICS), as well as from developing countries. The work is supported by the
UN, the World Bank, and the International Monetary Fund. One crucial
question is whether the widespread support for the BEPS project in principle
will continue when the project calls for coordinated action by these nations
on specific issues.
Second, coordinated action by the member countries of the OECD, the
BRICS, and developing countries is essential for the success of any action to
combat aggressive tax planning by multinational enterprises, since
multinationals operate and engage in tax planning on a worldwide basis.
Unilateral action by countries, even countries with the largest economies,
such as the US, is likely either to prove ineffective or to inflict serious
damage on the domestic economy.
Third, the bad publicity generated by the attacks of politicians in Europe
and the US against multinationals has generated a public perception that
multinational corporations are engaged in tax avoidance activities that are
probably illegal, and certainly immoral – multinationals appear to have been
put on the defensive.
Fourth, the problems of BEPS – (incidentally, the terms “base erosion”
and “profit shifting” are synonyms; they do not describe different problems)
– are not new. International tax planning has been a standard part of business
practice for decades, and inevitably, governments have responded with
various types of rules to protect their tax bases. Therefore, it is important to
see the current situation as just the most recent manifestation of the tension
between multinational corporations and national tax authorities. In the past,
however, tax authorities responded to international tax avoidance primarily
with unilateral anti-avoidance measures in their domestic law or in their
bilateral tax treaties. In contrast, multinational enterprises engage in tax
planning on a worldwide basis.
As noted above, any effective response to the problem of international
tax avoidance requires coordinated action by national tax authorities. It
remains to be seen whether such coordinated action is feasible. On the one
hand, the failure of the OECD’s previous attempt to deal with harmful tax
competition and the inability of countries to see beyond their narrow self-
interest makes one skeptical about the BEPS project. On the other hand, the
support of the G20 and the apparent success of the OECD-led efforts to
improve exchange of tax information among all countries, including tax
havens, suggests that the situation may be different today.
Finally, the BEPS project represents an opportunity to take some
tentative first steps to redesign an international tax regime that has become a
bit tired. Thus, the BEPS project should be viewed as simply the most recent
and most ambitious episode in a long-term effort to make the international
tax system less vulnerable to international tax planning by multinational
enterprises. Even if only a few of the BEPS action items are successfully
implemented, the effects will be significant.
The tax policy considerations underlying the OECD’s BEPS initiative
are superficially clear. Aggressive international tax avoidance by
multinational enterprises has several harmful consequences for national tax
systems:

(1) It reduces government tax revenues and increases the cost to

governments of ensuring compliance by multinational enterprises
with the tax rules.
(2) It undermines the perceived integrity of the tax system and may
have a deleterious effect on tax compliance generally.
(3) It undermines the fairness of national tax systems because
taxpayers other than multinational corporations must bear a greater
share of the tax burden.
(4) Small- and medium-sized enterprises may not be able to take
advantage of the same international tax planning opportunities as
multinational enterprises and may therefore be placed at a
competitive disadvantage.
(5) Finally, distortions in the location of investment may result to the
extent that opportunities for aggressive tax planning are greater in
the international context than in the domestic context.

However, the tax policy analysis of BEPS for any particular country is
much more subtle and difficult than the foregoing list of consequences may
suggest. The tax systems of many capital-exporting countries contain features
designed to facilitate the international competitiveness of their resident
multinational corporations. Over the past two or three decades, these
countries have consistently taken domestic measures to enhance the
competitive position of their resident multinational corporations, or at the
least have avoided taking measures that would place their multinationals at a
competitive disadvantage. Thus, many countries seem likely to have a two-
faced attitude to BEPS. On the one hand, they want to protect their domestic
tax bases from the aggressive tax-planning strategies of foreign-based
multinationals. On the other hand, they have no interest in preventing their
own resident multinationals from eroding the tax base of other countries (i.e.,
making them pay more foreign tax).
From any particular country’s perspective, therefore, the ideal result
from the BEPS project would be that other countries take action to require
their multinational corporations to pay more foreign tax, while it does little or
nothing with respect to its resident multinationals so that they gain a
competitive advantage. In simple terms, this explains why coordinated action
is so important and so difficult. Part of the difficulty relates to the widely
varying interests of OECD member countries. For example, net capital-
importing countries have different interests from net capital-exporting
countries. In addition, some OECD members, such as Belgium, Ireland,
Luxembourg, and the Netherlands, have significant interests both from a
public perspective (tax revenue, investment, employment) and a private-
sector perspective (professional firms and financial institutions) in facilitating
international tax planning by multinational enterprises, and therefore have an
interest in maintaining the status quo.

9.3 HYBRID ARRANGEMENTS

9.3.1 What Is a Hybrid Arrangement?

The term “hybrid arrangement” is generally used to describe situations in
which two countries take different and inconsistent positions with respect to
the tax treatment of some aspect of an arrangement. This inconsistent
treatment may result in either beneficial or harmful consequences for a
taxpayer. For example, inconsistent positions on transfer prices may result in
double taxation, whereas inconsistent positions on the character of certain
payments may result in a deduction in one country and no taxation in the
other country. Of course, well-advised taxpayers, such as multinational
corporations, will plan to avoid the harmful consequences of hybrid
arrangements and use them to generate tax benefits.
Hybrid arrangements, as broadly defined above, include all the
fundamental features of an income tax system: the persons subject to tax, the
type of activities giving rise to income (employment, business, and
investment), and the types of payments relevant for the computation of net
income. However, the most important types of hybrid arrangements are
hybrid entities and hybrid financial instruments.
Hybrid arrangements are often used as substitutes for tax planning
arrangements involving tax haven entities. For example, tax haven entities are
often used as intermediaries to receive deductible payments from an entity in
a high-tax source country and then pay those amounts in a non-taxable form
to a related entity in a high-tax residence country. A hybrid arrangement can
achieve the same result without the use of a tax haven. For example, if an
entity in a high-tax source country, HE (the hybrid entity) is treated as a flow-
through or transparent entity, that country may treat deductible payments to
HE as payments made to the owner of the entity, ACo, which might be
resident in a high-tax country. The payments to HE might be exempt from
source country tax, either under its domestic law or under the provisions of a
tax treaty. The payments also might not be taxable by the residence country
because it treats HE as a separate taxable entity. Therefore, as far as the
residence country is concerned, the payments are not received by ACo.

9.3.2 Hybrid Entities

A hybrid entity is a legal relationship that is treated as a separate taxable
entity in one jurisdiction and as a transparent or flow-through entity in
another jurisdiction. For example, under the laws of Country A, it may be
possible to establish a form of business organization in which the members
own interests and have limited liability. Country A may treat this
organization as a partnership for tax purposes, with the result that the
members or partners are taxable on their shares of the income of the
organization. On the other hand, under the tax laws of Country B, the
organization may be characterized as a corporation – as a legal entity separate
from its members or shareholders – with the result that the organization itself
is subject to tax on its income. Accordingly, if one or more of the members of
the organization is resident in Country B, the different treatment of the
organization in the two countries creates many tax planning opportunities.
Hybrid entities may take many different shapes and forms. Whether or
not a particular entity is a hybrid entity depends on the domestic laws of the
countries involved and, in particular, how they characterize entities for tax
purposes. For example, trusts and other similar fiduciary relationships may be
hybrids because they are treated as entities by some common law countries
but are ignored by some civil law countries. Hybrid entities may also include
special entities or arrangements, such as corporations limited by guarantee,
that are recognized under the laws of some tax havens. The “owners” of the
value of the company (the guarantors) have rights and obligations pursuant to
a contract, but have no rights to vote or receive dividends. The shares of the
company are owned by shareholders who have the right to elect the directors
but only limited rights to receive dividends. Dividends can be paid to persons
who have no relationship with the company (i.e., persons related to the
guarantor). The shareholders are typically trust companies that provide
wealth management and estate planning services. Several tax havens have
enacted legislation allowing the establishment of these types of entities,
which have many of the characteristics of inter vivos trusts.
Several countries recognize silent partnerships, which are essentially
contractual arrangements. The silent partners contribute assets to a managing
partner in consideration for a share of the profits from a business. Silent
partnerships are not treated as entities; the managing partner owns the assets
transferred by the silent partners. The payments made to the silent partners
are usually deductible in computing the income of the managing partner,
although they may be subject to withholding tax. If the country in which a
silent partner is resident treats the silent partnership as a partnership, and if it
taxes business income on a territorial basis, then there will be no tax, except
possibly withholding tax, in either country.
In a 1998 case in the United Kingdom, Memec PLC v. IRC [1998] STC
754 (Court of Appeal), Memec, a UK company, owned shares of a German
company, which in turn owned shares of two German operating companies.
The operating companies paid German trade taxes that were not creditable
against UK tax when Memec received dividends from its top-tier German
subsidiary. Therefore, Memec entered into a silent partnership with the
German corporation and then claimed that it had received dividends as a
partner directly from the operating companies, so that the trade taxes were
creditable. The UK Court of Appeal held that the German silent partnership
was not a partnership under UK law and that the source of the income was
the contractual arrangements.
In the last fifteen to twenty years, tax planning opportunities through the
use of hybrid entities have proliferated as a result of the US check-the-box
rules. Before 1997, entities were classified as corporations or partnerships
under US tax law based on six factors, including limited liability, continuity
of life, centralized management, and free transferability of interests; tax
planners were able to manipulate these factors to achieve the desired status
for an entity. For example, the use of limited liability companies (LLCs) as
transparent investment vehicles became very popular. LLCs were created
under special state statutes as vehicles for tax shelters, providing limited
liability for investors but treated as transparent for US tax purposes.
In 1997, the US government adopted check-the-box rules, which made
the classification of many business entities elective for taxpayers. Instead of
manipulating the six factors to achieve the desired characterization, taxpayers
can simply elect to have an entity treated as a corporation, a partnership, or a
disregarded entity (if the entity has only one member) by checking a box on a
prescribed form. A disregarded entity is treated as a sole proprietorship if the
single owner is an individual, or as a branch if the single owner is a
corporation. The election can be made with respect to LLCs, partnerships,
joint ventures, branches, and other business entities, and it can be made with
respect to entities created under foreign laws. Once made, an election cannot
be altered for five years. The election cannot be made for certain entities that
are clearly corporations (so-called per se corporations).
The US check-the-box rules were motivated by a desire to simplify
domestic tax planning and administration. Perhaps inadvertently, they also
made the use of hybrid entities for investment into and out of the US much
more attractive and certain. For example, a double-dip financing arrangement
for the acquisition or expansion of a US business might be structured in the
following way.

ACo, a resident of Country A, proposes to acquire all the shares of

USCo, a US corporation engaged in an active business in the US. ACo forms
a hybrid entity (e.g., an LLC) under the laws of the US or a tax haven in
which ACo owns all the units or interests. Under the tax law of Country A,
the hybrid entity is treated as a corporation. However, an election is filed
under the US check-the-box rules to treat the hybrid entity as a disregarded or
transparent entity. ACo also establishes a wholly owned US subsidiary,
USHoldco, to acquire the shares of USCo. ACo borrows the purchase price
of the USCo shares from a bank in Country A and contributes the borrowed
funds to the hybrid entity. The hybrid entity loans the funds to USHoldco,
which uses them to acquire the shares of USCo.
The tax consequences of this type of arrangement are as follows:

– the interest paid by USHoldco is deductible in computing the income

of the US consolidated group consisting of USHoldco and USCo;
– the hybrid entity is not subject to US tax on the interest payments
received because it is disregarded (i.e., treated as transparent) for US
tax purposes;
– the interest payments from US Holdco to the hybrid entity are subject
to US withholding tax because they are considered to be paid directly
by US Holdco to ACo for US tax purposes;
– the hybrid entity is not subject to Country A tax because it is treated
as a nonresident corporation (assuming that the interest is not taxable
under Country A’s CFC rules); and
– the interest paid by ACo is deductible in computing its income for
purposes of Country A tax.

Originally, the rate of US withholding tax on the interest payments was

the reduced rate provided in the treaty between the US and Country A, which
in some cases was zero. However, in 1997 the US adopted rules to deny the
benefit of any reduced rate of withholding under a treaty if the other country
did not tax the payment because it treated the hybrid entity as a separate
taxable entity. Thus, the structure would no longer be effective with respect
to the US because the interest would be subject to a 30 percent US
withholding tax.
For several years, some taxpayers used what were referred to in the US
as “reverse hybrid” arrangements to avoid US withholding tax.
 On the facts of the previous example, a reverse hybrid would involve
ACo forming a hybrid entity under the laws of the US that elects to be treated
as a corporation for US tax purposes. The hybrid entity would borrow from a
US bank and use the borrowed funds to acquire the shares of USCo, a US-
resident corporation engaged in active business. This arrangement is referred
to as a reverse hybrid simply because, for US tax purposes, the hybrid entity
is treated as a corporation rather than as a partnership or a disregarded entity.
The tax consequences of this reverse hybrid arrangement are as follows:

– the interest paid to the bank is deductible for US tax purposes in

computing the income of the hybrid entity and the US consolidated
group, consisting of the hybrid entity and USCo;
– no US withholding tax is payable because the interest is paid to a US
bank;
– the interest paid to the bank is deductible in computing the income of
ACo and ASub for purposes of Country A tax because the hybrid
entity is treated as a partnership between ACo and ASub.

In 2001, the US adopted rules for reverse hybrid arrangements, under

which a hybrid entity is not entitled to deduct interest payments made to
related nonresident persons. Such payments are treated as dividends for US
tax purposes and US tax treaties. This dividend treatment applies only to the
extent that the hybrid entity receives dividends from related US corporations.
Tax planning with hybrid entities is not for the faint-hearted. It involves
the complex interplay of domestic and foreign tax rules, domestic and foreign
corporate and commercial law, and tax treaties. The rewards of such planning
may be substantial; however, there are also serious risks. For example, in the
first hybrid example dealt with above, if the hybrid entity is considered to be
resident in Country A or if the hybrid is considered to be a PE of ACo in the
US, the benefits of the arrangement will be nullified. Moreover, any US tax
on the payments to the hybrid entity will not be creditable against any
Country A tax payable by ACo pursuant to Country A’s CFC rules because
the US tax is payable by ACo, not by the hybrid entity.

9.3.3 Hybrid Financial Instruments

In general, a hybrid financial instrument is a financial instrument that is
characterized differently by two counties. For example, an instrument with
characteristics of both debt and equity may be treated as debt by one country
but as equity (a share) by another country. Like hybrid entities, these hybrid
financial instruments are widely used for tax planning purposes. If a source
country treats an instrument issued by a corporation as debt, it will usually
characterize the payments on the debt as interest and allow a deduction for
those payments. However, the country in which the recipient of the payments
is resident may characterize the instrument as a share of the corporate issuer
and treat the payments on the share as tax-exempt dividends.
As an example of a hybrid financial instrument that results in a
deduction in both the residence and source countries, consider the following
sale and repurchase (“repo”) arrangement. ACo, a resident of Country A,
owns all the shares of BCo, a corporation resident in Country B. In turn, BCo
owns all the common and preferred shares of Cco, which is also resident in
Country B. ACo borrows money to acquire the preferred shares of CCo from
BCo. At the same time, ACo and BCo enter into an agreement under which
BCo agrees to repurchase the preferred shares of CCo in five years at a fixed
price.
 Country A treats these transactions in accordance with their legal form.
Thus, ACo is entitled to deduct its interest payments, but, assuming that
Country A has a participation exemption for dividends from foreign
corporations, ACo is not taxable on any dividends received on the preferred
shares of CCo. In contrast, Country B treats the arrangement in accordance
with its economic substance as a five-year loan by ACo to BCo, with the
preferred shares of CCo held by ACo as security. Thus, Country B allows
BCo to deduct any dividends paid by CCo on the preferred shares as interest.
In the result, the arrangement gives rise to interest deductions in both Country
A and Country B with no offsetting income inclusion.

9.3.4 OECD BEPS Proposals for Hybrids

The OECD’s BEPS Action 2: Neutralise the Effects of Hybrid Mismatch
Arrangements (Recommendations for Domestic Law) and (Treaty Issues),
March 19, 2014, is targeted only at hybrid entities and hybrid financial
instruments and transfers rather than all types of hybrid arrangements. More
specifically, the OECD proposals are concerned with hybrids that result in a
deduction in one country without any income inclusion in the other country,
or that result in a deduction in both countries. The OECD proposals are not
aimed at timing differences or differences in the value of payments with
respect to financial instruments that are attributable to foreign-currency gains
and losses.
 With respect to hybrid financial instruments, the OECD proposals focus
on differences in the characterization of payments (e.g., interest or dividends)
on a financial instrument or the underlying property (e.g., debt or shares). For
this purpose, a financial instrument is intended to be defined under domestic
law, which in some countries may be based on International Financial
Reporting Standards (IFRS) or other accounting rules.
With respect to hybrid entities, the OECD proposals target payments
made by a hybrid entity where the two countries involved view the payment
as being made by different entities, or where one country does not consider a
payment to have been made at all. As an example of the second type of
hybrid mismatch, consider a situation in which ACo, resident in Country A,
loans funds to a wholly owned entity, BCo, resident in Country B. Country B
treats BCo as a taxable entity and allows a deduction for the interest paid to
ACo on the loan. However, Country A treats BCo as transparent and
considers ACo to have a branch or PE in Country B; as a result, Country A
does not recognize any payment of interest by BCo to ACo and may exempt
ACo’s foreign branch profits from Country A tax.
The OECD proposals also target “reverse hybrids,” which involve
payments received by hybrid entities, and “imported mismatches,” which are
hybrid arrangements created under the laws of two countries and imported
into a third country. As an example of an imported hybrid mismatch
arrangement, consider the following example. BCo, resident in Country B,
issues a hybrid financial instrument to its parent, ACo, resident in Country A.
The payments by BCo to ACo are deductible by BCo in Country B but are
not taxable by Country A to ACo. BCo then loans funds to an unrelated
company, CCo, in Country C. The interest payments on the loan are
deductible by CCo in Country C and are included in BCo’s income, offsetting
the deductible payments by BCo to ACo. In effect, the benefits of the hybrid
arrangement between Country A and Country B are shifted to a third country,
Country C.
To counter the beneficial tax results of the different types of hybrid
arrangements described above, the OECD recommends a series of specific
rules rather than a comprehensive general response. The rules are very
complex and will be difficult for the tax authorities of many countries,
especially developing countries, to apply.
The OECD recommendations consist of domestic rules to be adopted by
countries unilaterally and domestic rules that, although adopted unilaterally,
are intended to be coordinated with the rules of other countries. For this
purpose, one country is considered to be the primary country to take action
against the targeted hybrid arrangement, and the other country is considered
to be the secondary country whose anti-hybrid rules should apply only if the
primary country does not deal with the hybrid arrangement.
In the case of hybrid arrangements involving deductions in one country
but no income inclusion in another country, the country in which the payer is
resident is the primary country. However, with respect to hybrid financial
instruments, the OECD recommends that the country in which the recipient is
resident (the secondary country) should deny any deduction or exemption for
payments that are deductible by the payer. In either case, in order to apply the
hybrid arrangement rules properly, it will be necessary for countries to obtain
detailed information about the treatment of payments on hybrid financial
instruments in another country. This is not typically something about which
the tax authorities of most countries have much experience.
In the case of hybrid entity payments involving double deductions, the
primary country is the country in which the recipient of the payment is
resident, and it is expected to deny a deduction for the payments. A similar
approach applies to hybrid entity payments involving a deduction in one
country but no income inclusion in the other country, reverse hybrids, and
imported mismatches. In these situations, the primary country is the country
in which the recipient of the payment is resident. Thus, the country in which
the payer is resident is expected to deny a deduction for the payment, or
require the payment to be included in income if the other country allows a
deduction. If, however, the primary country does not do so, the secondary
country is expected to deny any deduction for the payment.

9.4 FEES FOR MANAGEMENT, TECHNICAL, AND

CONSULTING SERVICES

9.4.1 Introduction
As discussed in Chapters 2 and 5, most countries tax nonresidents on their
domestic source income. Some countries tax nonresidents on all their
domestic source income; other countries tax nonresidents on their domestic
source business income only if a minimum threshold, such as a PE or a
minimum period of physical presence, is met. Under the provisions of the
OECD and UN Model Treaties, a source country is generally entitled to tax
the profits of a business carried on by a resident of the residence country only
if the business is carried on through a PE located in the source country, and
only to the extent that the profits are attributable to the PE.
As a result of these provisions of domestic law and tax treaties,
taxpayers, especially multinational enterprises, can structure their business
operations to erode the tax bases of source countries through payments for
technical, management, and consulting services, as shown in the following
example. ACo, a corporation resident in Country A, is the parent corporation
of a multinational group of companies involved in the hotel business. The
group has a hotel in Country B that is owned by BCo, a wholly owned
subsidiary of ACo, which is resident in Country B, a high-tax country; BCo is
taxable by Country B on its profits from the operation of the hotel. In
computing its profits, BCo is entitled to deduct expenses incurred to earn its
income. The expenses incurred by BCo include management fees and
consulting fees paid to CCo, another wholly owned subsidiary of ACo, that is
resident in Country C, a low-tax country.
This structure may result in substantial tax savings for the multinational
enterprise, essentially because the management and consulting fees are
deductible against Country B’s tax base and are taxable at a low rate in
Country C. The deductions for the fees claimed by BCo are, of course,
subject to Country B’s transfer pricing rules since the fees are paid to a
related party. However, the tax savings are not dependent on the payment of
amounts that are more or less than the arm’s-length amount; they are
available even if the fees are what arm’s-length parties would have paid.
Also, CCo would not be taxable by Country B on the fees because Country C
does not impose tax on CCo’s fees under its domestic law, or because, under
the tax treaty between Country B and Country C, Country B is entitled to tax
business profits derived by a resident of Country C only if the resident carries
on business through a PE in Country B. CCo will be careful to avoid creating
a PE in Country B.
Some developing countries have special rules in their domestic law for
taxing nonresidents on fees for management, technical, and consulting
services derived by nonresidents. Under these rules, withholding tax is
imposed on the gross amount of payments by residents to nonresidents for
management, technical, and consulting services whether or not the services
are performed in the country. If the nonresident service provider performs
services in the source country, the income is clearly derived from the source
country, in accordance with the conventional notion that the source of income
from services is where the services are performed. If, however, the services
are provided outside the source country, it is quite controversial to consider
the income from the services to be derived in the country in which the
services are used or consumed.
One basic difficulty with these rules is that the types of services to
which the rules apply are often not defined precisely. Some countries
distinguish between technical assistance, which generally involves a transfer
of know-how or technical expertise (analogous to the transfer of the right to
use intellectual property), and technical services, which involve the
application of specialized knowledge or skill. The definition of technical
services is similarly problematic under the provisions of tax treaties, as
discussed below.
Even if the provisions of a developing country’s domestic law impose
tax on income from technical services earned by a nonresident, the provisions
of an applicable tax treaty may limit that tax, as discussed in the next section.

9.4.2 The Taxation of Income from Management, Technical,

and Consulting Services under Tax Treaties
This section provides a brief review of the provisions of the OECD and UN
Model Treaties that are potentially applicable to income from management,
technical, and consulting services (referred to here for convenience simply as
“technical services”) and an overview of the provisions dealing with income
from such services that some developing countries have included in their tax
treaties.
Neither the OECD nor the UN Model Treaty currently contains any
specific provisions dealing with income from technical services provided by
a resident of one state in the other contracting state or to customers in the
other contracting state. In general, income from business services is covered
by Article 7 of the OECD Model Treaty or Article 7 or 14 of the UN Model
Treaty. Under Article 7(1) of both Model Treaties, a country is entitled to tax
a nonresident’s business profits only if the nonresident carries on business in
the country through a PE. A PE is defined in Article 5 to be a fixed place of
business that must generally last for a minimum period of six months. Under
Article 5(3)(b) of the UN Model Treaty, a nonresident is deemed to have a
PE in the source country if the nonresident furnishes services in the source
country for more than 183 days in any twelve-month period in respect of the
same or a connected project. The fixed-place-of-business rule can be easily
avoided by some nonresident service providers by limiting the time spent in
the source country, by not working at any one place for more than six
months, by splitting large contracts into several smaller ones, or by having
projects carried out by related parties.
Under Article 14 of the UN Model Treaty (Article 14 was deleted from
the OECD Model Treaty in 2000), income derived by a nonresident
individual from professional or other independent services performed in the
source country is taxable by the other state only if the income is attributable
to a fixed base in that country that is regularly available to the nonresident, or
if the nonresident is present in the source country for 183 days or more in any
twelve-month period. It is generally accepted that the source country must tax
income under Article 7 or 14 on a net basis. The distinction between
technical services and professional and business services that involve
technical expertise is unclear. For example, engineering services would often
be considered to be technical services; however, the independent activities of
engineers are included in the definition of “professional services” for
purposes of Article 14 of the UN Model Treaty. Thus, income from
engineering services, or at least those performed by individuals, would be
taxable by a source country only if the engineer has a fixed base in that
country or stays in that country for 183 days or more in any twelve-month
period.
Article 12 of both the OECD and UN Model Treaties dealing with
royalties does not apply to fees for management, technical, or consulting
services because the definition of royalties in Article 12(3) is limited to
payments for the use of, or the right to use, intellectual property, equipment,
or information.
The erosion of the source country’s tax base by payments for technical
services, and the inability of the source country to tax such payments, has led
some countries to add specific provisions to their tax treaties to allow them to
tax payments for technical services on a gross basis. A 2011 survey by the
International Bureau for Fiscal Documentation (IBFD) found that 134 of the
1,600 tax treaties concluded between 1997 and 2011 contained a separate
article dealing with fees for management, technical, and consulting services.
Under these special articles, income from technical services are, in effect,
treated like royalties. Some countries, such as India, extend Article 12 dealing
with royalties to include payments for services that are ancillary or subsidiary
to the application of intellectual property, or that make available technical
knowledge, skill, know-how or processes, or that involve the development of
a technical plan or design. The Indian approach is unsatisfactory because it
muddles the distinction between royalties for the transfer of the right to use
intellectual property and fees for services, which do not involve any transfer
of know-how or property by a service provider to a client.
Typically, the special articles in tax treaties dealing with fees for
technical services limit source country tax to fees for such services “arising”
in the source country, which usually means that the services must be
performed in the source country. Moreover, the expression typically used in
these provisions – “managerial, technical or consultancy services” – is not
usually defined.

9.4.3 Proposed Article on Fees for Management, Technical,

and Consulting Services in the UN Model Treaty
The UN Committee of Experts has been working on the taxation of income
from services for several years, and in 2013 the Committee decided in
principle to add a new article to the UN Model Treaty dealing with fees for
management, technical, and consulting services. The new article and
Commentary are likely to be added to the UN Model Treaty at the time of its
next update, which is scheduled for 2016 or 2017. Under the new article, fees
for management, technical, and consulting services will be taxable by a
contracting state on a gross basis by way of withholding if the fees arise in
that state. Fees for management, technical, and consulting services will be
considered to arise in a contracting state if the payer is a resident of that state,
or is a nonresident with a PE in that state and the fees are deductible in
computing the profits attributable to the PE.
In effect, under the new article, fees for management, technical and
consulting services will be taxable by a state on a gross basis irrespective of
where the services are provided and without any threshold such as a PE or
fixed base. This contrasts sharply with the treatment of such fees under the
existing provisions of both the OECD and the UN Model Treaties (taxation
on a net basis only if the services are performed in the source country through
a PE or fixed base). Not surprisingly, the new article is controversial and has
been vigorously opposed by several developed countries. It remains to be
seen whether developing countries will be successful in negotiating the
inclusion of the new article in their tax treaties, especially treaties with
developed countries.

9.5 ARBITRATION
As discussed in Chapter 8, section 8.8.3, tax treaties typically contain an
article providing for a mutual agreement procedure (MAP) to resolve
disputes between the contracting states with respect to the application and
interpretation of the treaty. Under Article 25(2) of the OECD and UN Model
Treaties, the competent authorities of the two states are required to endeavor
to resolve disputes submitted by a taxpayer; however, they are not obliged to
do so. Therefore, in some cases, the competent authorities may not be able to
reach an agreement, with the result that a taxpayer may be subject to
unrelieved double taxation. Moreover, there are no time limits for the
resolution of disputes by the competent authorities, and mutual agreement
cases have been known to drag on for many years.
Disputes between taxpayers and tax authorities concerning relief from
double taxation and other issues have arisen more frequently as international
trade and investment have become more sophisticated, tax treaties have
proliferated, and countries have become more aggressive in enforcing their
transfer pricing rules. Consequently, the need for a more efficient and certain
process for resolving tax disputes became obvious. Pressure from
multinationals and a few influential developed countries led the OECD to
include a compulsory arbitration provision in Article 25(5) of the OECD
Model Treaty in 2008; Article 25 of the UN Model was amended in 2011 to
add an optional arbitration provision. Because arbitration has been added to
the OECD and UN Models so recently, it has been included in only a few
bilateral tax treaties to date, and those treaties are invariably between
developed OECD member countries. As a result, information about the actual
experience of countries with arbitration is scarce, and at best anecdotal.
Arbitration forms part of the MAP provided in Article 25 of both Model
Treaties. Arbitration is not available independently of a MAP, and is not
available if the competent authorities of the contracting states agree that
taxation has been imposed in accordance with the treaty; it is available only
where the competent authorities have not been able to reach an agreement on
one or more issues. Arbitration is not available for an entire MAP case, since
the resolution of a case as a whole is the function of the MAP. In effect,
arbitration is used to resolve certain issues within the MAP on which the
competent authorities are unable to agree. Because arbitration is a part of a
MAP, it is subject to any and all of the limitations on the MAP.
Article 25(5) of the OECD Model Treaty provides that a taxpayer can
request that any unresolved issues in a case submitted for a MAP under
Article 25(1) be resolved by arbitration if the competent authorities have not
been able to resolve the case within two years. However, arbitration is not
available if the unresolved issues have already been decided by a domestic
court or administrative tribunal. The decision of the arbitrators is binding on
the competent authorities (unless it is rejected by the taxpayer), and must be
implemented irrespective of any time limits in domestic law. The competent
authorities are authorized to settle the details of the arbitration process by
way of a mutual agreement. A sample mutual agreement on arbitration is
included in an annex to the OECD Commentary on Article 25.
Article 25 of the UN Model Treaty contains two alternative versions of
the MAP, only one of which provides for arbitration. Under paragraph 5 of
Article 25 (alternative B), unresolved issues in a MAP case under Article
25(1) that have not been resolved within three years can be submitted for
arbitration at the request of one of the competent authorities. The taxpayer is
entitled to be notified of the request.
The major differences between arbitration under the UN and OECD
Model Treaties are as follows:

– arbitration is an alternative under the UN Model Treaty;

– arbitration under the UN Model Treaty is available at the request of
the competent authorities rather than at the request of the taxpayer, as
is the case under the OECD Model Treaty;
– issues can be submitted for arbitration under the UN Model Treaty
only if they have not been resolved by the competent authorities
within three years, rather than the two-year period under the OECD
Model Treaty; and
– under the UN Model Treaty, the competent authorities have six
months after the arbitration decision has been made to reach a
different resolution of the issues.
 Apart from these differences, the arbitration process under the UN
Model Treaty is the same as under the OECD Model Treaty, as described
above. However, the Commentary on the UN Model Treaty also provides for
an alternative voluntary arbitration process, under which both competent
authorities must agree to submit cases to arbitration on a case-by-case basis.
According to the OECD, the purpose of arbitration is to enhance the
effectiveness of the MAP by providing a dispute-resolution mechanism for
issues about which the competent authorities cannot agree. However, from a
taxpayer’s perspective, arbitration has the effect of and, arguably at least, the
purpose of, forcing the resolution of issues submitted to arbitration in a
manner that is binding on the competent authorities. Typically, in a MAP
without binding arbitration where the competent authorities cannot agree, the
result is unrelieved double taxation. Therefore, forcing the competent
authorities to agree through arbitration provides taxpayers with certainty and
will often eliminate double taxation, to the benefit of taxpayers.
To the extent that issues are submitted to arbitration, the competent
authorities give up control over the resolution of these issues to independent
arbitrators, who may not fully appreciate the significance of the issue for a
country’s tax system. Rather than have the issue decided through arbitration,
the competent authorities are effectively nudged toward resolving the issue
themselves. Thus, arbitration imposes discipline on the competent authorities
to resolve cases subject to a MAP that would otherwise be lacking.
Under both the OECD and UN Model Treaties, arbitration is available
only if a person has presented a case for a MAP under Article 25(1) to the
competent authority of the state of which the person is resident. The case
must involve actions by one or both of the contracting states that have
resulted (or will result) in taxation contrary to the provisions of the treaty,
and it must be presented to the competent authority within two or three years
of the first notification to the person of the actions. MAP cases under Article
25(3) that involve the interpretation or application of the treaty or the
elimination of double taxation not provided for in the treaty do not qualify for
arbitration, although contracting states have the option of extending
arbitration to those issues.
Although arbitration under the OECD Model Treaty is initiated by a
taxpayer, once it begins, it is a state-to-state process controlled by the
competent authorities. In contrast, under the UN Model Treaty, arbitration is
initiated at the request of one of the competent authorities. Although the
initiation of arbitration by one of the competent authorities seems to be
significantly different from taxpayer-initiated arbitration, the competent
authority requesting arbitration would likely have consulted with the taxpayer
involved beforehand, and in most cases would accede to the taxpayer’s
wishes concerning recourse to arbitration.
Under the arbitration provisions in both the OECD and UN Model
Treaties, arbitration is not available for issues where a domestic court or
administrative tribunal of either contracting state has already rendered a
decision on those issues. In most countries, the competent authorities would
not be able to implement an arbitration decision that is contrary to a domestic
decision. This limitation is consistent with the similar limitation on the MAP
generally.
Arbitration is available only for cases presented by a person for a MAP
under Article 25(1) where the actions of one or both of the contracting states
have “resulted for the person in taxation not in accordance with the
provisions of this Convention.” Therefore, for example, issues involving
exchange of information and assistance in collection would not qualify for
arbitration because they do not involve taxation contrary to the treaty. Most
issues submitted for arbitration are likely to involve the distributive
provisions of a treaty (Article 6 through 21), especially Article 9 dealing with
transfer pricing disputes. Some countries limit arbitration to specified
provisions of the treaty, such as Article 4 (residence), Article 5 (PEs), Article
7 (attribution of profits to PEs), Article 9 (transfer pricing), and Article 12
(royalties); others limit arbitration to questions of fact. Under several US tax
treaties, arbitration may be rejected for an issue if both competent authorities
agree; however, there are no standards or criteria for the denial of access to
arbitration.
Like the MAP generally, arbitration is not intended to displace domestic
law remedies or to force a taxpayer to choose between the two dispute
resolution mechanisms. To avoid duplication, domestic remedies are usually
suspended until the arbitration is complete. Once the arbitration and MAP are
completed, the taxpayer has the choice of accepting the decision of the
competent authorities or rejecting it and pursuing domestic remedies. If a
taxpayer rejects the result of a MAP involving arbitration, the process might
be viewed as a waste of time and resources. Since arbitration is invoked at the
option of the taxpayer, arguably the taxpayer should be bound by the decision
of the arbitrators. However, it would be unfair (and possibly a violation of
human rights) to deprive taxpayers of their access to domestic courts,
although, according to the OECD, it is rare for taxpayers to reject a mutual
agreement in order to have recourse to domestic courts. Alternatively, the
taxpayer may be required to waive access to domestic remedies as a condition
for arbitration; the competent authorities will usually require the taxpayer to
do so as a condition for the implementation of a MAP involving arbitration.
Until countries develop sufficient experience with arbitration, especially
mandatory arbitration, to feel comfortable with its operation as a mechanism
for resolving disputes, they are unlikely to expand the availability of
arbitration beyond what is currently provided in the OECD and UN Model
Treaties. In fact, it might be expected that countries will limit the types of
issues qualifying for arbitration even more narrowly.
The detailed procedures for arbitration under a tax treaty are usually set
out in a memorandum of understanding between the competent authorities. In
general, there are two basic types of arbitration. One type is a quasi-judicial
process in which the arbitration panel receives submissions, hears arguments,
and provides a written decision with reasons, which is not limited to the
positions put forward by the competent authorities. Decisions may or may not
be published publicly, depending on whether the competent authorities want
to establish a body of decisions that might have precedential value. Under so-
called “baseball” or last-best-offer arbitration, the arbitration panel is limited
to choosing between the positions taken by the competent authorities and is
not required or allowed to provide written reasons for its decision. Baseball
arbitration is used by countries that are concerned about compromising their
sovereignty and want to limit arbitration as much as possible. It is thought to
be less expensive, faster, and more effective in forcing the competent
authorities to adopt reasonable positions and resolve cases without the need
for arbitration.
Typically, an arbitration procedure involves the appointment of a panel
consisting of three arbitrators, one appointed by each competent authority,
and the chair, chosen by the other two arbitrators, often from a list
preapproved by the competent authorities. The arbitrators are often former
judges, government officials, or internationally recognized tax experts. The
panel receives written (and sometimes oral) submissions from the competent
authorities, and possibly from the taxpayer, depending on the procedures
agreed to by the competent authorities.
The OECD BEPS Action 14: Make Dispute Resolution Mechanisms
More Effective, December 18, 2014, acknowledges the absence of a clear
obligation on the competent authorities to resolve MAP cases and the lack of
consensus on mandatory arbitration. Therefore, Action 14 recommends that
complementary steps be taken to improve access to the MAP and to make the
MAP more efficient and effective in order to ensure that disputes are resolved
once a MAP has been initiated. In general, Action 14 recommends that
minimum standards for MAP cases should be established, and also that a
monitoring process should be established to ensure that participating
countries adhere to the minimum standard. With respect to arbitration, the
OECD suggests that the Commentary on the OECD Model Treaty could be
revised to allow countries to narrow the scope of issues subject to arbitration,
clarify the relationship between arbitration decisions and domestic law
remedies, and promote the use of most-favored- nation provisions to facilitate
the adoption of arbitration.

9.6 THE DIGITAL ECONOMY – ELECTRONIC

COMMERCE

9.6.1 Introduction
One of the major technological innovations of the late twentieth century was
the development and widespread use of the Internet for conducting and
facilitating various personal and business activities. The Internet was initially
developed by some US research universities operating under grants from the
US government. It was designed to facilitate communication among
researchers and to serve as a secure communication system in the event of a
major war. As the Internet expanded in scope and reach, its commercial
implications were discovered and exploited by a tremendous variety of large
and small vendors, leading to further expansion and development. Commerce
has clearly driven the development of the Internet.
In the second edition of this Primer, the discussion in this section
focused on commercial activities conducted over the Internet (“electronic
commerce”) such as online sales and services. The term “electronic
commerce” suggests that such commerce can be distinguished from other,
more traditional forms of commerce. In the last decade, however, digital
activities have permeated so many aspects of commercial life that it is now
customary to refer to the digital economy.
 The digital economy is the product of information and communication
technologies featuring global connectivity on a 24/7 basis through a variety
of mobile devices such as tablets, smart phones, and wearable devices. Key
features of the digital economy include the conversion of tangible goods into
digital products, global connectivity through mobile devices, and the
gathering, analysis, and commercialization of data collected from customers.
A common example of the conversion of tangible goods into digital products
is the increasing replacement of traditional books, magazines, and
newspapers by e-publications. The growth in 3D printing has the potential to
turn traditional manufactured goods into intangibles that are licensed to
customers who perform the manufacturing themselves. The connectivity of
the digital economy means that businesses can access customers wherever
they are and at all times (as long as they have a mobile device and access to
the Internet) and that businesses can locate their personnel and information
technology wherever they want. For this reason, the digital economy is often
said to be borderless. Consumers play a more active role in the digital
economy in a variety of ways: creating content (e.g., YouTube), using multi-
sided platforms (e.g., Google, eBay, and Amazon) and providing a source of
big data.
Examples of the digital economy include:

– the collection of personal data by businesses from customers that is

used to tailor marketing activities to them on a personalized basis;
– the use of logistics to track the movement of vehicles and cargo
globally;
– 3D printing and cloud computing;
– the use of social networks for accessing news and entertainment;
– the delivery of online educational materials and training;
– the diagnosis and treatment of health care patients from remote
locations;
– virtual currencies such as Bitcoin; and
– the sharing economy, such as the taxi service Uber and the
accommodation sharing service Airbnb.

The digital economy has resulted in the development of several new

models for doing business in addition to electronic commerce. Different types
of payment services and new forms of online advertising have been
developed. Many types of business activities take place on networked
platforms; online app stores have flourished with the development of digital
goods and services; cloud computing has arisen to provide a variety of
services (infrastructure-as-a-service, platform-as-a-service, software-as-a-
service, content-as-a-service, and data-as-a-service) through a network of
computers storing software, data and other resources that are available to
customers for a fee. The sources of revenue from these new business models
include sales of digital goods and services, advertising, subscriptions, sales of
data, and fees for transactions.

9.6.2 Tax Challenges Posed by the Digital Economy

In general, the digital economy presents three major challenges for the
current international tax system. First, as noted above, the digital economy is
borderless; it permits businesses to be conducted globally and remotely. In
the digital economy, businesses can engage with customers in a country
without the need for any physical presence – assets or personnel – in that
country. Second, the digital economy presents many difficult issues of
characterization with respect to new sources of revenue. Third, although data
have become an important source of value in the digital economy, it is
difficult for tax systems to capture the income from such data. Each of these
issues is discussed further below.
The tax issues raised by the digital economy are not new. For example,
for many years mail-order businesses have been able to sell to customers in
countries without any physical presence there. It is the scale of the digital
economy that threatens existing sources of tax revenue.
In general, nonresidents are taxable on income derived from the digital
economy under the same domestic tax rules applicable to income derived
from other types of activities. As discussed in Chapter 5, many countries tax
nonresidents on business profits only if the nonresidents meet some minimum
threshold based on physical presence in a country. Similarly, under most tax
treaties, a resident of one contracting state is not taxable on business profits
derived from the other contracting state unless the resident has a PE in the
other state and the profits are attributable to the PE. A PE is defined to be a
fixed place of business or a dependent agent with authority to conclude
contracts, and does not include a place or an agent that is used to carry out
only preparatory or auxiliary activities.
 Therefore, often the first issue for countries that wish to tax income from
the digital economy is whether the nonresident enterprise deriving the income
meets the minimum domestic threshold for source country taxation and, if it
does, whether it has a PE in the source country for the purposes of an
applicable tax treaty. The Commentary on Article 5 of both the OECD and
UN Model Treaties indicates that the following activities will not generally
give rise to a PE:

– The sale of goods and services and other activities in a country

through a website on a computer located in that country. According
to the OECD and UN Commentary on Article 5, a PE requires an
enterprise to have some physical presence (although not necessarily a
human presence) in a country, and a website is intangible. Although a
few countries (Chile, Greece, India, and Portugal) have indicated that
they disagree with the position taken in the Commentaries, that
position seems to represent an international consensus. Although an
argument can be made that a website that performs the same
functions as a bricks-and-mortar store should be treated as PE, there
would be insurmountable problems in applying and enforcing such a
rule.
– The use of computer equipment, such as a server, in a country to
store or provide access to electronic files or otherwise to assist a
taxpayer in conducting its business operations. The Commentary
recognizes that the place where computer equipment is kept could be
considered to be a PE if the place meets the requirements of Article
5. In most cases, however, such a place will not meet these
requirements because the place is not at the disposal of the
nonresident enterprise or the activities carried on by the server are
preparatory or auxiliary. In any event, any country that takes the
position that a computer server constitutes a PE would likely find that
any servers in the country would be moved to more hospitable
jurisdictions.
– The use in a country of an agent, such as an internet service provider
(ISP), to provide access to the Internet or to host a web site. Under
Article 5 of both Model Treaties, an independent agent acting in the
ordinary course of its business does not create a PE for its principal.
In most cases, therefore, an ISP would be considered to be an
 independent agent because it is merely providing Internet access to
the public and is roughly comparable to a telephone company
providing telephone service to its customers. This position, which is
reflected in the Commentary on Article 5 of both the OECD and UN
Model Treaties, does not appear to be controversial.

The characterization of income derived from digital transactions is

important in applying the provisions of a tax treaty. For example, income
characterized as business profits is taxable by a source country in accordance
with Article 7 of the OECD and UN Model Treaties only if the enterprise has
a PE in that country and the income is attributable to the PE. Business profits
are taxable on a net basis after the deduction of the costs of earning the
income. As discussed above, the OECD and UN Commentaries suggest that a
PE does not include a website that functions as a virtual office. As a result,
most income from digital activities characterized as business profits would
not be taxable in the source country.
Income characterized as royalties is taxable under Article 12 of the
OECD and UN Model Treaties. Article 12 of the OECD Model Treaty
provides that royalty income is exempt from tax in the source country. In
contrast, the UN Model Treaty allows a source country to tax royalties at a
rate to be agreed on by the contracting states.
In many cases, income from digital transactions is easy to characterize.
For example, income derived from sales over the Internet of tangible goods
would be business income, and income from providing access over the
Internet to a database containing proprietary information would be royalties.
Some types of income from the digital economy, however, present problems
of characterization. The Commentary on Article 12 of both the OECD and
UN Model Treaties takes the position that the consideration for the
acquisition of computer software over the Internet may be treated as royalties
or as business profits, depending on the facts. If the transferee acquires rights
to use the software in a way that would constitute an infringement of
copyright in the absence of a license to use the software, the consideration
should be treated as a royalty. However, if the transferee acquires only
limited rights necessary to operate the program, (e.g., to copy a program onto
the user’s computer), the consideration should be treated as business profits,
despite the legal form of the transaction as a license. However, some
countries take the position that payments for software are royalties. The
OECD and UN Commentaries also extend the principles applicable to the
treatment of computer software to other digital products.
In summary, the tax challenges posed by the digital economy are like the
weather: everybody likes to talk about it, but nobody does anything about it.
The OECD BEPS Action 1: Addressing the Tax Challenges of the Digital
Economy, is no exception. The OECD Report presents an excellent
description of the digital economy and the risks it poses for the international
tax system. However, concrete recommendations for coordinated action to
deal with digital transactions are absent from the Report. The reason is that
any meaningful response to the challenges posed by the digital economy
requires fundamental changes to the rules of the international tax system that
allocate tax revenues between source and residence countries. Although the
Report raises the possibility of adopting a significant-presence threshold
(rather than the existing PE threshold) for cross-border sales of digital goods
and services and a withholding tax on payments for digital goods and
services, it does not suggest that these options will be pursued except as
aspects of ongoing work on the digital economy. Instead, the Report
expresses the view (the hope?) that other aspects of the BEPS project, such as
the work on Action 7: Preventing the Artificial Avoidance of PE Status, may
limit the base-eroding effects of the digital economy.
Glossary of International Tax Terms

183-day rule A rule used by many countries under which an individual is

deemed to be a resident if the individual is present in the country for 183 days
or more in the aggregate in any twelve-month period.
Advance pricing agreement (APA) An agreement between a multinational
enterprise and the tax authorities of one or more countries approving the
transfer pricing method to be used by the enterprise in future tax years.
Arm’s-length method, (standard, principle, or approach) The
establishment of transfer prices in transactions between related parties based
on the prices charged (or sometimes the profits derived) in similar
transactions between unrelated parties.
Arm’s-length price A price set on a transfer of goods, services, or intangible
property between related persons that corresponds to the price that would be
set in a similar transfer between unrelated persons.
Back-to-back arrangements An arrangement involving a transaction, such
as a loan or lease, by a person to an intermediary, followed by a similar
transaction by the intermediary to another person who may be related to the
first person. A back-to-back arrangement is almost always used as a tax
avoidance device.
Base company income Business income derived by a controlled foreign
corporation from the country in which its controlling shareholders are
resident or from transactions with related parties that occur outside the
country in which the foreign corporation is established.
Base erosion and profit shifting (BEPS) A project launched by the
G20/OECD in 2012 to deal with aggressive tax planning by multinational
enterprises that result in the reduction of the tax bases of high-tax countries.
Beneficiary or beneficiaries of a trust The persons, individuals, or legal
entities who beneficially own property held in trust for them by a trustee or
trustees, who have legal title to the property.
Branch A business carried on by a taxpayer, usually through an office or
other fixed place of business. A branch is not separately incorporated. A
foreign branch is a branch located outside the taxpayer’s country of
residence.
Branch tax A tax imposed by a country on the profits of a branch of a
nonresident enterprise that are not reinvested in the country; the tax is
intended to perform the same function as a withholding tax on dividends paid
by a resident subsidiary to its foreign parent corporation.
Capital-export neutrality The situation that exists where resident investors
bear the same tax burden whether they invest at home or abroad.
Capital-import neutrality The situation that exists where residents investing
in a source country bear the same tax burden as other investors in that
country.
Capital-ownership neutrality The situation that exists where the owners of
investments bear the same tax burden irrespective of their countries of
residence.
Check-the-box rules US tax rules that, under most circumstances, allow
taxpayers to choose to have an entity (other than certain per se corporations)
treated as a corporation, partnership, branch, or disregarded entity for tax
purposes.
Commentary The Commentary on the OECD Model Treaty or on the UN
Model Treaty, which explains how the provisions of the Model Treaty should
be interpreted and applied.
Commissionaire arrangements A legal relationship recognized under the
law of civil law countries under which one person, the commissionaire, enters
into contracts for the sale of goods owned by another person that are legally
binding on the commissionaire, but not on that other person.
Comparable profit method (CPM) A transfer pricing method used by the
United States that is based on a comparison of the profits earned from similar
businesses. CPM is similar to the transactional net margin method (TNMM)
authorized by the OECD Transfer Pricing Guidelines.
Competent authority An official of a treaty country who is responsible for
the resolution of disputes and issues of interpretation arising under a tax
treaty.
Consolidation (consolidated) Rules that permit related corporations -
typically a parent corporation and its subsidiaries - to aggregate their income
and losses, thereby allowing the losses of one affiliated corporation to offset
the profits of another corporation.
Contracting states The countries that are parties to a tax treaty.
Controlled foreign corporation (CFC) rules Rules that require the passive
income and certain other tainted income of foreign corporations controlled by
resident shareholders to be included in the income of those shareholders,
whether or not such income is distributed.
Corresponding adjustment A modification by one country to a transfer
price used by a taxpayer to take account of a modification made by another
country to the transfer price used by a related taxpayer.
Cost-contribution arrangements A contractual arrangement under which
the prospective users of intangible property jointly develop and share in the
costs of developing that property and in ownership rights to that property.
Such arrangements are called “cost-sharing arrangements” in the United
States.
Credit method Foreign taxes paid by a resident of a country are credited
against the residence country’s tax on the resident’s foreign source income.
Cross-border transactions Transactions that have potential tax
consequences in more than one country.
Deduction method Foreign taxes paid by a resident of a country are
deductible in computing the resident’s taxable income in the residence
country.
Deferral The practice of subjecting to taxation the profits derived from
foreign investment through foreign corporations only when the profits are
repatriated to the country of residence of the investor.
Designated jurisdiction approach Under this approach, a country identifies
certain countries as low-tax countries for purposes of applying its CFC rules
to CFCs resident in those countries.
Direct investment An equity investment in a company that is likely to
provide the investor with substantial influence in the management of the
company. An ownership interest of over 50 percent of the outstanding shares
of a company is always classified as a direct investment. Many countries treat
an ownership interest of 10 percent or more as a direct investment.
Double-dip lease A leasing arrangement under which the lessor and the
lessee are both able to claim the tax benefits of ownership of the leased
property in their country of residence because the countries characterize the
transaction differently.
Dual-resident taxpayer A taxpayer who is a tax resident of two or more
countries for the same tax year.
Earnings-stripping rules Rules under which the interest deductions claimed
by resident enterprises are limited by reference to a percentage of some
measure of their income, such as EBITDA.
Entity approach An approach for applying CFC rules under which either all,
or none, of a CFC’s income is taxable to its resident shareholders
Exemption method Exemption from domestic tax of some or all foreign
source income derived by residents.
Exemption with progression A method for relieving double taxation under
which foreign source income is exempt from tax but is taken into account in
determining the rate of tax applicable to other income.
Exit or departure tax A tax imposed by a country on the accrued income
and capital gains of a person giving up residence in that country.
Force-of-attraction principle The taxation by a country of a nonresident
with a PE in that country on all the income derived in that country by the
nonresident, and not just the income attributable to the PE.
Foreign affiliate A foreign corporation in which a domestic taxpayer has a
significant direct or indirect ownership interest (usually 10 percent or more of
the shares).
Foreign investment fund rules Rules designed to tax residents on their
interest in an offshore or foreign investment fund.
Foreign tax credit A provision that permits domestic tax otherwise payable
to be reduced by foreign tax paid on foreign source income.
Formulary apportionment A method for allocating the profits or losses of a
multinational enterprise among the countries in which it operates in
accordance with a formula based on factors such as sales, assets, and payroll.
Global approach An approach for applying CFC rules under which the CFC
rules apply to CFCs irrespective of the country in which they are resident.
Gross-up A notional amount added to the amount of a dividend received by a
shareholder, usually equal to the portion of the tax paid by the corporation
that is attributable to the dividend, with the result that the dividend plus the
gross-up amount equals the before-tax income of the corporation out of
which the dividend was paid.
Hybrid entity An entity that is treated as a separate taxable entity (usually as
a corporation) in one country and as a transparent or flow-through entity
(often as a partnership) in another country.
Hybrid financial instrument A financial product or instrument that is
characterized in one way (e.g., debt) for purposes of one country’s tax system
but in another way (e.g., equity or shares) for purposes of another country’s
tax system.
Indirect foreign tax credit A foreign tax credit allowed to a domestic
taxpayer when it receives a dividend from a foreign corporation for the
underlying foreign corporate taxes paid by the foreign corporation on the
income out of which the dividend is paid.
Inter-nation equity A concept requiring a fair sharing of the tax revenue
derived from taxation of transnational income between capital-importing and
capital-exporting countries.
International double taxation The imposition of income tax by two or more
countries on the same income of the same taxpayer for the same taxable
period.
Inward-bound (in-bound) transaction A transaction in which a nonresident
of a country invests capital or other resources in the country.
Limited liability company (LLC) An entity that is treated as transparent
under the tax laws of a country but which provides the investors in the entity
with limited liability under the general laws of that country.
Limitation-on-benefits article A provision included in tax treaties that is
intended to prevent the benefits of the treaty from being granted to residents
of third states and to limit the benefit of the treaty to persons who are genuine
residents of one of the contracting states.
Most-favored-nation treatment The treatment by one country of the
residents or citizens of another country not less favorably than the treatment
of the residents or citizens of any other country (but not its own residents or
citizens).
Mutual agreement procedure A process provided under tax treaties for the
resolution of disputes concerning the interpretation and application of the
treaty by the competent authorities of the contracting states.
National treatment The treatment of nonresidents or foreigners by a country
not less favorably than the treatment of its own residents or citizens.
Neutrality A tax is neutral if its imposition does not alter the economic
decisions that would be taken in its absence.
Nondiscrimination A generally accepted notion that a country should tax
nonresidents, foreigners, and foreign-owned domestic corporations in a
manner that is the same as or functionally equivalent to the treatment of
residents, citizens, or domestically owned corporations in similar
circumstances.
Nonresident A person who does not have sufficient connections with a
country to be liable to tax there on worldwide income and who is taxable
only on income from sources in that country.
Organisation for Economic Co-operation and Development (OECD) An
organization of the major developed countries of the world, with its
headquarters in Paris.
OECD Model Treaty A model income tax treaty sponsored by the OECD on
which virtually all bilateral income tax treaties are patterned.
Foreign investment fund A unit trust or mutual fund established in a foreign
country, often a tax haven, to make passive investments in stocks, bonds, and
other investment assets. The investors are generally residents of high-tax
countries, and the fund usually accumulates its income.
Outward-bound (out-bound) transaction A transaction in which a resident
of a country invests capital or other resources outside the country.
Parent corporation A corporation that controls another corporation (referred
to as a subsidiary).
Participation exemption A tax regime under which dividends received from
foreign corporations by a resident corporation are exempt from residence
country tax if the resident corporation owns at least some minimum
percentage of the shares of the foreign corporation.
Permanent establishment (PE) A concept used to determine if an enterprise
has sufficient connections with a country to subject it to tax on its income
attributable to the PE.
Place-of-incorporation test A rule under which a corporation is considered
to be a tax resident of the country in which it is incorporated.
Place-of-management test A rule under which a corporation is considered to
be a tax resident of the country in which it is controlled or managed (usually
where the board of directors meets and exercises control over the affairs of
the corporation).
Portfolio investment An equity or debt investment in a company that does
not provide the investor with substantial influence in the management of the
company. An equity ownership interest of less than 10 percent of the
outstanding shares of a company is typically classified as a portfolio
investment.
Profit-split method A method for the allocation of the worldwide profits of a
multinational enterprise among its members in various countries in
proportion to their contributions to the earning of the profits.
Residence country The country in which a person is resident for income tax
purposes.
Residence jurisdiction A principle of taxation under which all income
accruing to residents of a country (worldwide income), regardless of its
source, is subject to tax by that country.
Resident A person who has sufficiently close connections to a country to be
liable to tax there on worldwide income.
Saving clause A provision included in all US tax treaties that preserves the
right of the United States, subject to certain exceptions, to tax its residents
and its citizens as if the treaty had not been entered into.
Settlor of a trust A person who establishes a trust and usually transfers
property to the trust.
Soak-up taxes Taxes levied by a country on nonresidents solely with the
intention of those taxes being claimed by the nonresidents as credits against
the taxes payable in their country of residence.
Source country The country where a foreign investment is located or where
income arises.
Source jurisdiction A principle of taxation under which residents and
nonresidents alike are taxed on income from economic activity within a
particular country.
Subsidiary A corporation that is directly owned or controlled by another
corporation. A foreign subsidiary of a corporation is a corporation resident
outside the country of residence of the controlling corporation.
Tainted income Income of a controlled foreign corporation that is taxed to
the resident shareholders of the corporation when earned by the foreign
corporation rather than when distributed. Generally, tainted income consists
of passive investment income and base company income.
Tax avoidance The deferral, avoidance, or reduction of tax by lawful means.
Tax evasion The reduction of tax by illegal means, usually involving
fraudulent nondisclosure or willful deceit.
Tax havens Countries that subject income (or some forms of income) or
entities (or certain entities) to low or no taxation.
Tax incentives or preferences General terms used to describe tax
exemptions, deductions, credits, rate reductions, or other concessions
designed to attract foreign investment or otherwise to affect economic
behavior.
Tax sparing The allowing of a credit for the amount of foreign taxes that
were not paid because of a tax incentive or holiday in the foreign country.
Territorial basis A method of taxation under a country imposes income tax
only on income derived in that country (whether by residents or
nonresidents).
Thin capitalization rules Restrictions on the deductibility of interest
payments made by corporations with excessive debt-to-equity ratios to their
substantial nonresident shareholders.
Tie-breaker rules Rules in tax treaties that establish the residence of a dual-
resident taxpayer in one country for treaty purposes.
Trailing taxes Taxes imposed by a country after a person ceases to be a
resident of that country that would not usually be imposed on nonresidents.
Transactional approach An approach for applying CFC rules under which
only the tainted income (passive income and base company income) of a
CFC is taxable to its resident shareholders.
Transactional net margin method (TNMM) A method for determining
transfer prices in transactions between related parties, typically based on the
ratio of profits earned by parties engaged in similar activities to some
economic indicator, such as invested capital or gross receipts.
Transfer pricing rules Rules that limit the ability of related parties to set
prices on transfers of property or services that are different from the prices
that would be set in similar transfers involving unrelated parties.
Treaty override A term used (often in a negative sense) to describe a
country’s domestic legislation that takes priority over the provisions of a
country’s tax treaties.
Treaty shopping The use of a tax treaty by a person who is not resident in
either of the treaty countries, usually through the use of a conduit entity
resident in one of the countries.
Trust An arrangement allowed under the laws of common law jurisdictions
for the holding of property by a person (trustee) transferred from a person
(settlor) for the benefit of other persons (beneficiaries).
Trustee The individual or legal entity that has legal title to property and the
power to manage that property for the benefit of other persons (beneficiaries).
UN Model Treaty A model income tax treaty sponsored by the United
Nations that is based on the OECD Model Treaty, with some modifications
made to reflect the interests of developing countries.
Withholding tax A tax levied by the source country at a flat rate on the gross
amount of dividends, royalties, interest, or other payments made by residents
to nonresidents. The tax is collected and paid to the government by the
resident payer.
Worldwide basis A method of taxation under which a country taxes its
residents on both their domestic source income and their foreign source
income (i.e., their worldwide income).
Index
A

Accounting methods
Accounting rules
Accrual taxation
Accrued gains
Active business income
Active business test
Adjustment, corresponding
Administrative aspects of taxing nonresidents
exchanges of information
obtaining information about foreign source income
reporting requirements
Administrative cooperation, treaty contents
Advance Pricing Agreement (APA)
Agents
Agreement Establishing the World Trade Organization
Alienation of immovable property
Allocation formula, pro rata
Allocation of expenses
Amendment of treaties
Ancillary treaty objectives
Anti-avoidance measures
controlled foreign corporation (CFC) rules
foreign investment funds
rules and doctrines
thin capitalization rules
Anti-treaty shopping rule
APA. See Advance Pricing Agreement (APA)
Apportionment
expense allocation
formulary
income
Apprentices
Arbitrage, cross-border
Arbitration
arbitration under the OECD Model
arbitration under the UN Model
baseball arbitration
Arm’s-length method
cost-contribution arrangements
debt: equity ratio and
future of
PE income determination
sales of tangible personal property
sharing of corporate resources
treaty provisions
Athletes
Audits
Australia
Average effective tax rates
Avoidance of taxes. See Anti-avoidance measures; Tax avoidance; Tax
havens

Back-to-back arrangements
Banks
Base erosion and profit shifting, OECD project
Action plan
Action 1 (digital economy)
Action 2 (hybrids)
Action 3 (controlled foreign corporation (CFC) rules)
Action 4 (interest deductions)
Action 6 (treaty shopping and treaty abuse)
Action 7 (permanent establishments)
Action 8 (intangibles)
Action 13 (country-by-country reporting and disclosure requirements)
Action 14 (dispute resolution and arbitration)
Action 15 (multilateral treaty)
Beggar-thy-neighbor policies
Belgium
Body of persons
Branches
hybrid entities
income determination
permanent establishment rules
Brazil
Building or construction site
Burden of proof, arm’s length method
Business
active versus passive
exemptions
fixed place of
source of income
Business income

Canada
exemptions
formulary apportionment use in
Capital-export neutrality
Capital gains
double-tax relief
passive income
relief provisions
source jurisdiction
Capital-import neutrality
Capital outflows
Check-the-box rules
China
Classical method of corporate taxation
Code of Conduct, European Union
Collection of taxes, information exchange
Commentary on OECD Model Treaty
Commentary on UN Model Treaty
Commission of the European Communities
Commodity markets
Comparability analysis
Comparable Profit Method (CPM)
Comparable uncontrolled price (CUP) method
Competent authorities
Competitiveness of domestic economy
Computer software
Conduit companies
Confidentiality
Conflicts. See also Dispute resolution
Constitutions, treaties and
Consultancy services
Consumption taxes
Controlled foreign corporation (CFC) rules
anti-avoidance measures
attributable income, definition and computation of
avoidance of taxes
base company income
control test
definition of CFC
designated jurisdiction or global approach
domestic taxpayers subject to tax
exemptions
foreign tax rates
general
global approach
inactive income
indirect control
losses, relief provisions
motive exemption
nominal tax rates
passive income
passive investment income
relief provisions
tainted income
tax haven countermeasures
 tax haven definition
Convention on Mutual Administrative Assistance in Tax Matters
Cooperation
Copyright
Corporate reorganizations
Corporate resource sharing
Corporate tax
Corporations limited by guarantee
Corresponding adjustment
Cost-contribution arrangements
Cost of goods sold
Cost-plus method
Costs of financing
Council of Europe
Credit lines, sharing of corporate resources
Credit method
comparison of exemption method with
double taxation relief
general rules
indirect credit
treaty aspects
types of limitations
Credits
indirect foreign tax
Cross-border tax arbitrage
Cross-border transactions

183-day period
Debt: equity ratio
Debt financing
Deduction method
Deductions
allocation of expenses
apportionment formulas
formulary apportionment and
 hybrid entities
source rules
treaty provisions
treaty shopping
Deemed distribution approach
Deferral charge approach
Deferral of taxes
attributable income
CFC rules
de minimis exemption
foreign investment funds
resident tax on foreign source income
Deferred income, expense allocation
Definitions
international tax
treaty interpretation
Delivery and transport of goods
Denmark
Developing countries, OECD Model Treaty
Digital economy
access to data through websites
characterization of income from digital transactions
sales or services through websites
use of internet service provider
Direct investment
Disclosure laws
Discrimination. See also Nondiscrimination
Dispute resolution
fair dealing and cooperation provisions of treaties
pricing methodology
treaties
Disregarded transactions
Distinction between business and other income
Dividends
controlled foreign corporation (CFC) rules
exemptions
expense allocation
 gross-up procedure
indirect foreign tax credit
passive income
source jurisdiction
treaty issues
treaty shopping
Documentation requirements, transfer pricing
Domestic economy
competitiveness of
of tax havens
Domestic law. See Legislation
Double-dip financing
Double-dip leases
Double taxation
allocation of expenses
defined
mechanisms of
credit method
deduction method
exemption method
property transfers
tax sparing
transfer pricing issues
treaties
Dual residence
double taxation relief
jurisdiction

Earnings stripping
Economic competitiveness
Education and training
Electronic commerce
Emerging issues
harmful tax competition
countermeasures
 identifying
recent developments
hybrid entities
defining
types of
Employment services
Enforcement, administrative assistance
Enterprise, treaty language
Entertainers
Equity, debt disguised as
Equity finance
Estate taxes
Estonia
European Union
administrative assistance
Code of Conduct concerning tax havens
Company Taxation in the Internal Market
formulary apportionment use
Package to Tackle Harmful Tax Competition in the European Union, A
treaty shopping issues
Evasion, fiscal
Exemption method
comparison with credit method
double taxation relief
tax sparing and
Exemptions
controlled foreign corporation (CFC) rules
de minimis
dividend
expense deduction
professional services income
with progression
shareholders
Exit or departure taxes
Expense allocation

F
Fair dealing, treaty contents
Fair market value
Federalism, fiscal
Federal states
Fees for management, technical, and consulting services
proposed article in UN Model
Fees for personal service
Financial intermediaries
Financial services industries
Financing costs
Finland
Fiscal evasion, treaty objectives
Fiscal federalism
Fixed place of business
Force-of-attraction principle
Foreign affiliates
Foreign corporations. See Controlled foreign corporation rules
Foreign currency gains and losses. See Foreign exchange gains and losses
Foreign distributors
Foreign exchange gains and losses hedging
Foreign investment fund rules
Foreign sales
Foreign-source income. See also Income
double taxation relief mechanisms
tax havens, domestic income diversion
Foreign subsidiaries
Foreign tax credits
direct
indirect
limitations
country-by-country
item-by-item
overall
Foreign tax laws
Formulary apportionment
France
G

GATS. See General Agreement on Trade in Services (GATS)

GATT. See General Agreement on Tariffs and Trade (GATT)
General Agreement on Tariffs and Trade (GATT)
General Agreement on Trade in Services (GATS)
Germany
Globalization
Goals of tax rules
Gross income
Gross-up amounts
Guarantee, corporations limited by
Guidelines, transfer pricing

Harmful preferential tax regimes

Harmful tax competition
countermeasures
identifying
recent developments
Head office expenses
Hong Kong
Hybrid arrangements
hybrid entities
branch and subsidiary
check-the-box rules
defining
types of
hybrid financial instruments
repo
reverse hybrid
Hybrid securities

Imputed income approach

Income
active versus passive
branch, determination of
derived by nonresidents
branch taxes
capital gains
income from employment
income from immovable property
investment income
diversion of
source of
business
employment and personal services
investment
tainted
treaty contents
business income
employment and personal services income
investment income
treaty coverage, scope, and legal effect
Income-earning activities, tax havens
Incorporation, place-of-incorporation test
Independent contractors
Indirect foreign tax credit (indirect credit)
Individual
residence, defining
treaty terminology
Industrial royalties
Information exchange
tax havens
treaty provisions
Intangible property transfer
Intellectual property
Interest
allocation of expenses
deduction of
 differences between model treaties
excessive
hybrid entities
passive income
thin capitalization rules
treaty shopping
Interest income
source jurisdiction
treaty provisions
Intermediaries, financial
International banks
International cooperation. See Cooperation
International double taxation. See also Double taxation
Inter-nation equity
Internet sales
Inventory accounting methods
Investment
direct
offshore investment funds
outward-versus inward-bound
portfolio
Investment income
differences between model treaties
thin capitalization rules
treaty contents
treaty shopping
withholding tax rates
Inward-bound transactions
Israel
Item-by-item limitation

Japan
Jurisdiction
CFC rules
conflicts over, and double taxation
 defining residence
individuals
legal entities
treaty issues related to
differences between model treaties
OECD Model Treaty
source of income
business
employment and personal services
investment

Language, treaty
Law. See Legislation
League of Nations
Leases
Legal department expenses
Legal documents, administrative assistance
Legal entities
planning considerations
residence, defining
Legislation
anti-avoidance measures
and CFC rules
foreign investment funds
information exchange
intersection of domestic with foreign tax law
profit determination
treaties and
fair dealing and cooperation provisions of
interpretations of
terminology
Legislative approval of treaties
Licenses
Liechtenstein
Limitada
Limitation on benefits (LOB)
Limited liability companies (LLCs)
Loans
corporate resource sharing
thin capitalization rule avoidance
LOB. See Limitation on benefits (LOB)
Luxembourg

Mail-order sales
Management
permanent establishment, treaty definitions
place-of-management test
Manufacturing
Marginal tax rates
Mark-to-market method
Maximum withholding rates
Memec
Minimization of taxes
Model treaties
Most-favored-nation treatment
Movable property, rental income from
Multinational Convention on Mutual Administrative Assistance in Tax
Matters
Mutual agreement procedure
Mutual consent, treaty modification by
Mutual funds

National self-interest
National taxes
National treatment
Netherlands
Neutrality, capital-export and capital-import
New product development costs
New Zealand
Nondiscrimination
fair dealing and cooperation
treaty contents
Nonresident lenders, thin capitalization rules
Nonresidents
administrative aspects
dividend taxation
election to pay on a net basis
excessive taxation of
nexus or jurisdictional basis for source country tax
role of tax adviser
source of income rules
thresholds for source country tax
withholding taxes
final
provisional
Norway
Notional transactions

Office expenses
Organisation for Economic Co-operation and Development (OECD). See also
specific taxation topics
Centre for Tax Policy and Administration
Committee on Fiscal Affairs
excessive interest determination
harmful tax competition policies
Report on Harmful Tax Competition
transfer pricing publications
Organisation for Economic Co-operation and Development (OECD) Model
Treaty
ambulatory approach
business income
Commentary on
coverage, scope, and legal effect
interpretation of
 success of
Outward-bound transactions
Ownership rights disposition
Ownership, tax credit claim requirements

Parent corporation
Partial exemption methods
Participation exemption
Partnerships
double taxation
silent
treaty coverage, scope, and legal effect
Passive income
Passive investment funds
Penalties, transfer pricing documentation requirements
Pensions
Permanent establishment (PEs)
business income
personal services income
profit determination
transfer pricing rules
treaty provisions
Personal property sales
Personal services income
permanent establishment
sources of income
Person, treaty terminology
Place-of-incorporation test
Place-of-management test
Place-of-organization test
Planning, role of tax adviser in
Portfolio investment
Portugal
Preferential tax regimes, harmful
Preferred shares
Present value calculations
Price, arm’s-length
Pricing, Advance Pricing Agreement (APA)
Professional services. See also Personal services income
Professionals, source jurisdiction
Professors
Profits
determination of
formulary apportionment
stripping
Profit-split method
Progression, exemption with
Property sales/transfers
intangible property
planning considerations
source jurisdiction
tangible personal property
transfer pricing methods
taxation of accrued gain
Proprietary information
Pro rata share
Protocols to treaties
Provincial governments

Qualified person, definition of

Ratification of treaties
Related-party transaction
Relief mechanisms. See also Double taxation
Rents
Reorganization of corporations
Report on Harmful Tax Competition
Resale price method
Residence
 avoidance of taxes
defining
individuals
legal entities
treaty issues related to
terminology
Residence changes
Residence jurisdiction
Residence-of-the-shareholders test
Residence-residence conflicts
Residence-source conflicts
Residence taxation of investment income
Resident corporations
nondiscrimination and
thin capitalization rules
Resolution of disputes. See Dispute resolution
Resource sharing, corporate resources
Role of tax adviser
Royalties
differences between model treaties
passive income
sharing of corporate resources
source jurisdiction
treaty provisions
treaty shopping
Russia

Sale of property. See Property sales/transfers

Sales taxes
Saving clause
Separate-baskets approach
Services
corporate resource sharing
income from
Shareholders
 controlled foreign corporation (CFC) definition
dividend exemption systems
domestic taxpayers subject to tax
relief provisions
residence-of-the-shareholders test
Shareholder tax
Shares
exemptions
preferred
voting
Sharing of corporate resources
Shipping
Ships
Silent partnerships
Soak-up taxes
Software
Source country
Source jurisdiction. See also Jurisdiction
defined
OECD Model Treaty
tax avoidance possibilities
Source of income
Source rules,
Source-source conflicts
Source taxation of investment income
South Africa
South Korea
Sovereign power,
Sovereign states
Spain
Special tax haven provisions
Standard inventory accounting methods
State government
States, Contracting
Stockholders. See Shareholders
Storage facilities
Stripping of profits/ earnings
Students
Subnational jurisdictions
formulary apportionment use
treaty effect
Subpart F rules
Subsidiaries
hybrid entities
permanent establishment definition
planning considerations
tax haven
Sweden
Switzerland

Tax arbitrage, cross-border

Tax avoidance. See also Tax havens
role of tax adviser
transfer pricing (see Transfer pricing)
treaties and
Tax evasion
Tax havens
anti-avoidance measures
avoidance of taxes
defining
formulary apportionment
harmful tax competition
information gathering
jurisdiction issues
planning considerations
role of tax adviser
tainted income
Tax incentives
Tax Information Exchange Agreements (TIEAs)
Tax jurisdiction. See Jurisdiction
Tax sparing
Tax treaties. See Treaties
Teachers
Technical fees. See Fees for management, technical, and consulting services
Telecommunication income
Termination, treaty
Terminology, treaty interpretation
Territorial basis of jurisdiction
Territorial taxation
Tests of residency
individuals
legal entities
Thin capitalization rules
Threshold requirements
carrying on business
engaged in a trade or business
fixed base
immovable property
permanent establishment
physical presence
TIEAs. See Tax Information Exchange Agreements (TIEAs)
Tie-breaker rules
TNMM. See Transactional net margin method (TNMM)
Tracing approach
Trade barriers
Trade secrets
Trainees
Transactional net margin method (TNMM)
Transfer of property. See Property sales/transfers
Transfer pricing
anti-avoidance measures
arm’s-length method
cost-contribution arrangements
determining income of branch or permanent establishment of corporation
documentation requirements
future of
sales of tangible personal property
sharing of corporate resources
treaty aspects of transfer pricing methods
 formulary apportionment and future of arm’s-length method
tainted income
tax sparing and
Transport
Travaux préparatoires
Treaties
business income
contents of
coverage, scope, and legal effect
double taxation relief
employment and personal services income
fair dealing and cooperation
formulary apportionment use
investment income
other types of
overview
interpretation of
legal nature and effect of
model treaties
objectives of treaties
revisions and overrides
reduced withholding rates on certain investment income
residence
schedular basis
special treaty issues
administrative cooperation
nondiscrimination
resolution of disputes
treaty shopping
tax haven countermeasures
tax sparing
transfer pricing methods
Treaty shopping
anti-avoidance measures
investment income
treaty contents
Trust income

United Kingdom
United Nations Committee of Experts
United Nations Model Treaty
business income
differences between model treaties
interpretation of
personal services providers
termination provisions
United Nations Practical Manual on Transfer Pricing for Developing
Countries
United States
controlled foreign corporation (CFC) rules
earnings-stripping rule
exemptions
harmful tax competition policies
tax sparing
transfer pricing rule enforcement
treaty effects
treaty shopping
United States-Canada treaty

Vienna Convention on the Law of Treaties

Voting shares

Warehouses
Wealth tax
Withholding taxes
coverage, scope, and legal effect
differences between model treaties
investment income
 source jurisdiction
treaty issues
treaty shopping
World Trade Organization, Agreement Establishing
Worldwide taxation

Zero rate of withholding