Corporate tax avoidance
Submission 75
February 6, 2015
Senate Standing Committees on Economics
PO Box 6100
Parliament House
Canberra ACT 2600
E-mail: [email protected]
Senate Submission to the Senate Economics References Committee: Inquiry into
Corporate Tax Avoidance and Minimisation
by Ross McClure* and Roman Lanis*
Ross McClure is a PhD candidate at the University of Technology Sydney and Dr. Roman
Lanis (corresponding author) is Associate Professor at the University of Technology Sydney.
The submission is partly based Mr McClure's draft PhD thesis.
Corporate tax avoidance
Submission 75
Executive Summary
The purpose of this submission is to rigorously analyse the second term of reference which
addresses the following comment: any need for greater transparency to deter tax avoidance
and provide assurance that all companies are complying fully with Australia‘s tax laws.
In order to evaluate the adequacy of corporate tax disclosures and the associated measures
that rely on those disclosures, it is necessary to understand what constitutes tax avoidance and
the methods, transactions and activities involved in carrying out tax avoidance schemes. Only
then can one assess the ability of the measure to gauge a theoretical construct and specific
activities.
The definition of tax avoidance employed by any relevant party will not only provide the
scope of activities, but will also define the methods used to measure them. Some definitions
are quite narrow and have sharp boundaries, while others are broader with more organic
borders. It is difficult to pinpoint a single definition that embraces both the broad concept of
tax avoidance as well as capturing the smaller details of the transactions involved. It is
ultimately the question at hand that will determine the definition.
The research and question should establish the tax activities and behaviours to be examined.
These activities determine the definition to be used and guide the choice of methodology
employed to measure their effect. This analytical framework attempts to link the transaction
level activities with both the various definitions of tax avoidance, and with the proxies that
attempt to measure tax avoidance. The definition should match the activities being examined
and be consistent with the measures employed. Outside factors such as disclosure rules, tax
policies, enforcement action and regulatory response, influence this framework and modify
Page 1
Corporate tax avoidance
Submission 75
the definitions, while other factors such as tax havens, ‘noise’ from earnings management,
and unrecognised tax benefits can have an impact on both the definition and the measurement
of tax avoidance. Therefore by providing an analytical framework which includes the
definition of tax avoidance, the specific activities which are associated with tax avoidances
and the measures of tax avoidance will enable conclusions to be made about the need for
greater transparency to deter tax avoidance and provide assurance that all companies are
complying fully with Australia‘s tax laws. More importantly, if there is a need for more
transparency by way of additional disclosures we shall specify in more detail which
disclosures can increase the transparency of corporate tax payment data to deter tax
avoidance and provide assurance that all companies are complying fully with Australia‘s tax
laws.
Page 2
Corporate tax avoidance
Submission 75
Recommendations for Greater Disclosure and Transparency

Require companies to at the minimum disclose, by way of the financial reporting regime
(Australian accounting standards), foreign tax paid, local tax paid and a reconciliation
between the effective and statutory tax rates;

More preferably to require companies to disclose, on a country-by-country basis, profit
and tax paid information as an extension of the segment reporting regime within the
Australian accounting standards; and

Require companies, by way of the financial reporting regime (accounting standards), to
disclose unreported tax benefits.
Page 3
Corporate tax avoidance
Submission 75
TableofContents
Table of Contents ............................................................................................................... 4 1. 2. Definitions.......................................................................................................................... 7 1.1 Accounting-based definitions .................................................................................... 10 1.2 Economic-based definitions ...................................................................................... 19 1.3 Legal-based definitions ............................................................................................. 22 1.4 Summary and conclusions ......................................................................................... 27 Methods used for tax avoidance ...................................................................................... 29 2.1 Incentives and opportunities...................................................................................... 30 2.2 Specific Methods used in Tax Avoidance ................................................................. 34 2.2.1 Transfer Pricing ................................................................................................. 34 2.2.2 Thin Capitalisation ............................................................................................. 39 2.2.3 Real Estate Investment Trusts (REIT’s) ............................................................ 41 2.3 Activities related to U.S. multinationals operating in Australia. .............................. 48 2.3.1 Subpart F ............................................................................................................ 48 2.3.2 Alternating Short-Term Loans ........................................................................... 51 Page 4
Corporate tax avoidance
Submission 75
2.3.3 Tax Treaties ....................................................................................................... 52 2.3.4 Double Dutch with Irish Sandwich .................................................................... 53 2.4 3. Conclusions ............................................................................................................... 56 Measuring tax avoidance ................................................................................................. 57 3.1 Effective Tax Rates (ETR’s) ..................................................................................... 58 3.1.1 Zimmerman (1983) ............................................................................................ 59 3.1.2 U.S. Congress, Joint Committee on Taxation (1984) ........................................ 61 3.1.3 Citizens for Tax Justice (1985) .......................................................................... 62 3.1.4 Shevlin (1987).................................................................................................... 64 3.1.5 Manzon and Plesko (1984) ................................................................................ 65 3.1.6 Dyreng, Hanlon and Maydew (2008) ................................................................ 66 3.1.7 Chen, Chen, Cheng and Shevlin (2010)............................................................. 68 3.1.8 Conclusions ........................................................................................................ 69 3.2 Book-Tax Gaps or Differences (BTG’s) ................................................................... 71 3.2.1 Desai (2003) ....................................................................................................... 71 3.2.2 Conclusion ......................................................................................................... 72 Page 5
Corporate tax avoidance
Submission 75
3.3 Regression-based Measures ...................................................................................... 73 3.3.1 Desai and Dharmapala (2006) ........................................................................... 73 3.3.2 Frank, Lynch and Rego (2009) .......................................................................... 74 3.4 Characteristics-based Measures ................................................................................ 77 3.4.1 Transfer Pricing Index (TPRICE) ...................................................................... 77 3.4.2 Maximum Allowable Debt (MAD) Ratio .......................................................... 78 3.5 Conclusions ............................................................................................................... 80 4. Recommendations and Conclusions ................................................................................ 81 5. References ........................................................................................................................ 87 Page 6
Corporate tax avoidance
Submission 75
1. Definitions
Defining tax avoidance or tax aggressiveness1 is fundamental to determining the set or group
of activities and behaviours that are to be investigated, as well as to the choice of
methodology in measuring their effect. The definition and measurement of tax avoidance by
empirical researchers has been a principal concern in the business literature (Hanlon and
Heitzman 2010; Dunbar 2010; Lisowsky, Robinson and Schmidt 2013). In their discussion
on measuring tax avoidance, Hanlon and Heitzman (2010) made the point “that not all
measures are appropriate to all research questions.” (p.139) The definition of tax avoidance
that is adopted is important to the strength of the research process, as it links the research
question and the tax avoidant methods, to the appropriate proxies and measures.
A major difficulty in studying tax avoidance is finding a generally accepted definition
(Hanlon and Heitzman 2010; Dunbar 2010; Lisowsky, Robinson and Schmidt 2013), as it
depends on matters of degree, scope and judgement (Lisowsky 2010). These degrees, scope
and judgements have arisen from numerous sources and have significantly influenced the
debate on a suitable definition. The accounting profession and the firms themselves have
developed concepts and parameters for decision making purposes. Tax researchers in both
the Economics and Finance disciplines who investigate similar questions to accounting
researchers have also developed their own sets of concepts. So have tax administrators and
legislators, whose thinking is represented in the tax laws, regulations and determinations, and
the courts who have developed further tests and definitions to interpret the tax legislation, and
1
While the terms “tax aggressiveness” and “tax avoidance” have separate meanings within academic research,
in reality the two terms are used interchangeably. In this situation, the use of the term “tax aggressiveness” is
also important to understanding this behaviour in tax research in accounting. The two terms will be discussed
separately in the Definitions Section, but the term “tax avoidance” will be used throughout the rest of the report.
Page 7
Corporate tax avoidance
Submission 75
to apply those facts in tax cases. These concepts and constructs, as to what constitutes tax
avoidance, are essential for understanding the complexity of the mechanisms and behaviour
of tax avoidance, and for analysing and measuring their effect. However, the measures are
only as reliable as what is observable in nature which is directly related to transparency in tax
disclosures.
Some tax research in accounting has defined tax aggressiveness and tax avoidance in very
broad terms (Hanlon and Heitzman 2010; Lopez et al. 1998; Dyreng 2008). This has been
influenced by previous tax research in both economics and finance.
Tax research in
accounting has been concerned with the magnitude, determinants and consequences of tax
aggressiveness, while economics research has focussed, amongst other things, on the tax
burden, and where the incidence of that tax burden is situated. Financial research has given
attention to the effect of taxes on firm value, expected returns and leverage (Hanlon and
Heitzman 2010).
A broad definition, and hence the use of broad measures, would be
appropriate to tax research in these areas. Recent tax accounting research appears to have
adopted the broad definitions of tax aggressiveness (Lanis and Richardson 2013; Lennox et
al. 2013; Donohoe and Knechel 2013; Gallemore, Maydew and Thornock 2013).
The use of a broad definition of tax aggressiveness overcomes some problems with the
definitions and usage of other terms used in tax research. Deliberations over terms such as
“tax avoidance”, “tax evasion”, “tax sheltering”, “tax planning”, and “tax minimisation”
becomes redundant or simply ignored. However, these terms can have specific meanings,
quite often based on legal or legislative definitions, and they are used throughout the tax
research literature from almost all disciplines. At other times, these terms are used almost
interchangeably with the term “tax aggressiveness”, especially “tax avoidance”. Nonetheless,
Page 8
Corporate tax avoidance
Submission 75
these terms do have meaning within tax research in accounting and there have been recent
attempts to categorise and characterise these terms within tax research in accounting
(Lisowsky, Robinson and Schmidt 2013, Hanlon and Heitzman 2010).
However, the use of a broad definition of tax aggressiveness is not without problems. A
definition should be able to designate those activities that are regarded as tax aggressive. The
problem is that the word “aggressive” is a relative concept and needs to be quantified. In this
sense, tax aggressiveness can be opposed to tax passiveness, a state where an agent takes no
explicit actions to reduce their taxes. In other words, anything that is not tax passive is tax
aggressive. A broad definition such as this would include all activities, regardless of the level
of aggressiveness involved, and while this would relieve the researcher from the necessity of
specifying and describing the mechanisms used to minimise tax, it is of little use if the
research topic was the operation of a specific tax avoidance scheme, or that the methods of
tax avoidance captured by the definition and the measure include benign activities that are
actively encouraged. If accounting research was interested in estimating the magnitude of the
tax avoidance involved, a broad definition would require a broad measure that captures most
activities, regardless of their “aggressiveness”. While this may be useful in ranking firms on
their level of “tax aggressiveness”, or putting them on a scale, it is not so useful in
determining the tax-reducing strategies being employed by these firms especially with respect
to legitimate reasons for the different tax levels between firms. Therefore, some of the
definitions and constructs from outside the accounting research are helpful in identifying
these other issues.
The need to encompass the definitions from other disciplines becomes obvious when a firm,
or firms, have been identified or accused of being tax aggressive. The first announcement by
Page 9
Corporate tax avoidance
Submission 75
the firm, or firms, or an industry representative, is nearly always that they fully comply with
all laws and pay all taxes required of them in all jurisdictions in which they operate (SMH, 10
Oct 2014). Whether an activity or scheme is legal or not is a fundamental issue in gauging
the level of tax aggressiveness. Borek, Frattarelli & Hart (2014) point out that “[l]itigation
has long revealed difficulties in designing, implementing and interpreting tax law in a manner
that allows taxpayers to claim intended benefits without encouraging abuse.” (p.2) The abuse
of tax mechanisms for unintended consequences is the type of activity and corporate
behaviour that interests and motivates much of the tax research in accounting. The research
question, and therefore the definition, needs to be clear about the activities and mechanisms
being examined, whether and when they are legal, and the way they are used to minimise the
tax liability of a firm.
Definitions link the research aims and questions to the types of tax avoidance methods being
employed by firms and to the measures required to analyse them. While accounting research
has not yet formulated a widely used and generally accepted definition of “tax
aggressiveness,” there are influences from other disciplines involved in this type of research.
An analysis of the contributions to the definition debate from accounting, economics, and
legal research is outlined in the following sub-sections.
1.1
Accounting-based definitions
While tax research in accounting may suffer from a lack of a generally accepted definition,
the efforts to clarify what is meant by the term “tax aggressiveness” have been worthwhile
and have produced a better understanding of the concept and therefore, the methods used to
Page 10
Corporate tax avoidance
Submission 75
measure it. Early attempts to define “tax aggressiveness” produced broad concepts that
captured all tax minimising activities and behaviours, regardless of their level of
“aggressiveness”. While this is suitable for some research, it does not provide a clear focus
on the activities of most interest. “Aggressiveness” is after all, a concept of degrees or levels.
A major difficulty in studying tax aggressiveness in accounting is finding a generally
accepted definition (Hanlon and Heitzman 2010; Dunbar 2010; Lisowsky, Robinson and
Schmidt 2013). For Dunbar et al. (2010), the “conceptual challenge in this stream of research
is the lack of a universal definition of tax aggressiveness.” (p.146) Hanlon and Heitzman
(2010) reiterate this assessment in their review of tax research that “there are no universally
accepted definitions of, or constructs for, tax avoidance or tax aggressiveness; the terms mean
different things to different people. … [This is] similar to financial accounting research on
earnings quality.” (p.137) Since Hanlon and Heitzman (2010) made their call for a better
definition and construct for tax aggressiveness, there has been numerous innovations and
improvements in defining tax aggressiveness, but defining and measuring the point at which
tax minimising activities become “aggressive” is still elusive.
Using a framework created by Dyreng et al. (2008), Taylor and Richardson (2014) define
corporate tax avoidance as “any transaction or event … that leads to a reduction in the
amount of corporate taxes paid.” (p.1, fn.2) While this has broad scope, the defining of
“aggressive” tax reduction methods as “structuring transactions or activities with one of the
principle objectives to reduce the amount of corporate taxes” (p.3), draws on narrower and
more specific legal definitions and judicial doctrines of economic substance and business
purpose. This also accords with Braithwaite’s (2005) “scheme or arrangement put in place
with the dominant purpose of avoiding tax” (p.16), and with the description given by the
Page 11
Corporate tax avoidance
Submission 75
former US Deputy Assistant Secretary of the Treasury for Tax Policy, Michael Graetz, who
said that these transactions are “deal[s] done by very smart people that, absent tax
considerations, would be very stupid.” (Herman 1999, p. A1)
The term “tax aggressiveness” entered the accounting research literature in the late 1990’s,
against a background of an ever widening gap between reported financial income and
reported tax income.2 When the term was originally used, it was in the literal sense that some
behaviour in relation to the reporting of tax income was seen as being very aggressive3 (eg.
Lopez et al. 1998).
Although Lopez et al. (1998) do not formally define “tax
aggressiveness”, it is described as “the propensity of a firm to engage in tax-minimising
behaviour”. (p. 38) Other researchers have used a number of terms, such as “tax avoidance”,
“tax minimisation” or “tax planning” to label these types of activities and behaviours, and in
some cases, added the descriptor, “aggressive,” or “abusive,” before them. In the U.S. the
term “tax sheltering” became synonymous with abuse of the tax system, while in Australia,
such activities were euphemistically known as “aggressive tax planning”.
In their definition, Lopez et al. (1998) did not put a benchmark on when the propensity to
engage in tax minimising behaviour became aggressive and left it open.
Their broad
definition of tax aggressiveness implies varying degrees of tax aggressiveness, and links the
quantitative character of “tax aggressiveness” to the quantitative concept of “propensity”.
Variations in the propensity or willingness of firms to engage in tax aggressive behaviours
2
This gap was identified by Desai (2003) who found that by 1998, the book-tax gap could no longer be
explained by the previously attributed determinants.
3
A similar nomenclature has been used with respect to “earnings management” being referred to as “aggressive
financial reporting” (Frank, Lynch and Rego, 2009)
Page 12
Corporate tax avoidance
Submission 75
and activities underlies one of the central themes of tax research in accounting, with
Weisbach’s (2002) under-sheltering puzzle4.
Lopez et al. (1998) use the term “tax
minimisation” to depict the overall group of activities that reduces the tax liability from a tax
passive level5. Other literature has termed this construct as “tax avoidance” (Hanlon and
Heitzman 2010; Dunbar et al. 2010; Lisowsky, Robinson and Schmidt 2013), “tax planning”
(Lanis and Richardson 2013), or “tax aggressiveness” (Chen et al 2010; Frank, Lynch and
Rego 2009).
Lisowsky, Robinson and Schmidt (2013) provide a taxonomy of terms,
concepts and measures used in tax research in accounting, and while this may improve the
use of terminology in tax research discourse and bring more clarity to the notion of “tax
aggressiveness”, it does not to help distinguish the point at which tax avoidance, tax
minimisation, or tax planning activities and behaviours become “aggressive”.
Many of the definitions of “tax aggressiveness” that have been proffered are very broad,
capturing all tax-reducing activities and transactions, but being unhelpful in either, directing
attention to particular activities and tax avoidance schemes, or for the selection of relevant
and appropriate measures. Even though Frank, Lynch and Rego (2009) base their definition
of “aggressive tax reporting” around the notion of uncertain tax reporting positions, overall
their definition is quite broad incorporating the “downward manipulation of taxable income
through tax planning that may or may not be considered fraudulent tax evasion” (p.468).
While using a similar definition, Chen et al. (2010) drop the term, “manipulation” for the
4
Weisbach (2002) was interested to understand the causes or determinants of the variation in the level of tax
minimising activities between firms, industries, and even geographical locations. The lack of obvious reasons
led to the “puzzle” moniker.
5
This is the level that a firm’s tax liability would be if it did not consider taxation when making business
decisions.
Page 13
Corporate tax avoidance
Submission 75
more wide-ranging and less value-laden term, “management”, and define tax aggressiveness
as “the downward management of taxable income through tax-planning activities” (p. 42).
“Tax planning” as a whole is a fairly broad concept, incorporating a large group of activities,
and it is difficult to distinguish those tax planning activities that become downward
manipulation or management of taxable income, from those that remain “acceptable”. In
their review of tax research, Hanlon and Heitzman (2010) also “define tax avoidance broadly
as the reduction of explicit taxes” 6 (p.137), a definition that “reflects all transactions that
have any effect on the firm’s explicit tax liability” (p.137). While this seemed as if each
definition was trying to out-do its predecessors with new dimensions of depth and breadth,
these broad definitions finally set out a base level that encompasses everything as a starting
point. These definitions established the condition of the null hypothesis, where there is no tax
aggressiveness, no tax avoidance, no tax planning, etc. It is a point where the tax response
of firms would be a random walk. The question then becomes: at what point from the
null hypothesis do tax reducing activities become aggressive?
Using the concept of first-order, second-order or marginal importance, Lisowsky (2010)
constructs a more specific definition of “tax aggressiveness”. In this scheme, tax shelter
activity, which is prohibited by the IRS, would be of a first-order importance, as the tax
benefit is the main, if not the only, reason for undertaking the activity. On the other hand,
depreciation or stock options are only of second-order or marginal importance to the capital
allocation or management incentive objectives. Lisowsky (2010) conceptualises a hierarchy
6
Implicit taxes are costs, such as the lower rates of return on tax-favoured investments that firms incur when
adopting behaviour favoured, and encouraged, by the tax system. An example is local Government bonds where
the interest is tax-free but the rate of return on capital is lower. Tax research in accounting has largely ignored
implicit taxes, presumably because they are so hard to measure or estimate. See Jennings, Weaver & Mayhew
(2012) for a recent summary of the (very scant) research.
Page 14
Corporate tax avoidance
Submission 75
of mechanisms, and a definition that would identify and allow for different levels of “tax
aggressiveness.” Frank, Lynch and Rego (2009) go further and state that from an empirical
perspective, firms are considered to have “aggressive financial (tax) reporting if they have
high discretionary pre-tax accruals (permanent book-tax differences)” (p. 468), thereby
identifying a measurable boundary at which tax avoidance or tax-reducing activities become
“aggressive”. Frank, Lynch and Rego (2009) also devise a measure that attempts to capture
those activities that lie beyond the boundary.
Attempting to narrow the range of activities and behaviours that were included in broad
definitions of “tax aggressiveness,” Frischmann et al. (2008) define it as the act of “engaging
in significant tax positions with relatively weak supporting facts.” (p.265) Dunbar et al.
(2010) also express the view that the tax aggressive activities of interest to researchers are
those activities “towards the end of a continuum of tax avoidance activities that range from
legitimate tax planning to investments in abusive tax shelters” (p.18). While these studies
identify a criterion to rank tax aggressiveness and the start of a benchmark for what
constitutes “tax aggressiveness”, the criterion is difficult to measure and the benchmark is
merely an arrow pointed in the right direction.
In their review of tax research, Hanlon and Heitzman (2010) build on Dunbar et al. (2010)
and designate “tax aggressiveness” as a broad continuum of activities from benign behaviours
that were envisioned by tax policies at one end, to outright tax fraud and evasion at the other.
In this framework, tax avoidance covers a “spectrum” of tax planning activities with
outcomes that range from certain to uncertain tax positions. This incorporates Frischmann et
al.’s (2008) use of the level of uncertainty about the outcomes of tax positions that have been
adopted by managers, to determine the positions or rankings of firms and activities on the
Page 15
Corporate tax avoidance
Submission 75
continuum. Uncertain positions are those that are based on a weak set of facts and are
unlikely to be sustained during an audit (Rego and Wilson, 2012). The more uncertain a
firm’s tax position, the more aggressive and risky the behaviour, and the further the firm sits
along the tax aggressiveness continuum. Rego & Wilson (2012) expanded the spectrum of
activities to “include[s] standard tax planning practices that do not necessarily violate income
tax rules, as well as fraudulent tax avoidance, including tax shelter transactions considered
abusive by the IRS and the Treasury Department.” (p.1) To further refine the types of
activities and transactions along the “spectrum”, Richardson, Taylor and Lanis (2013) include
transactions ranging from those that may be legal, through those that fall into a grey area that
requires tax authority or judicial adjudication, to those that may involve outright and obvious
illegal activities. These attempts at grading tax aggressive activities have entrenched the
“continuum” view and also support the incorporation of judicial influence into its definition,
as to when tax-reducing activities become aggressive.
Based on a test from U.S. Financial Instrument No. 487 (FIN 48) that recognises and
measures the uncertainty of a firm’s ability to sustain its tax positions, Lisowsky, Robinson
and Schmidt (2013) provide a conceptual framework to classify and categorise, both the
terminology associated with research into tax aggressiveness, and the various measures from
the empirical tax literature, that are used to capture it. Lisowsky, Robinson and Schmidt
(2013) use the term “tax avoidance” to cover all activities that reduce tax from the level of
the null hypothesis. This range of activities forms a tax avoidance continuum from least to
most aggressive tax positions. In this framework, “tax aggressiveness” is conceptualised as a
7
[INSERT FIN48 SUMMARY]
Page 16
Corporate tax avoidance
Submission 75
sub-set of tax avoidance, “in which the underlying positions likely have weak legal support”
(pp.589-590).
“Tax sheltering” is categorised as the most extreme sub-set of tax
aggressiveness.
A “tax shelter” is a generic term used primarily in the U.S. that covers many tax minimising
activities. In Australia, this type of activity is euphemistically referred to as “aggressive tax
planning” (Braithewaite, 2005). Tax shelters are transactions or arrangements that generate
tax losses, without incurring economic losses or the risk of losses8 (U.S. Treasury, 1999).
Under U.S. judicial principles, these arrangements are illegal if they do not exhibit "economic
substance" or "business purpose", or they are created with the sole purpose of evading tax
(Lisowsky, 2010). In Australia, Part IVA of the ITAA 369 provides a similar prohibition of
such schemes based on the principle of “sole or dominant purpose”.10 Part IVA contains a
number of tests for the objective purpose of a scheme, arrangement or transaction, including
the amount of economic risk involved. These schemes are usually “sold” as a package to
companies by a purveyor such as an accounting or legal firm. KPMG has come under
scrutiny in the U.S. Senate for the part it has played in creating and selling numerous tax
shelter schemes, and PwC was recently named in association with aggressive tax
minimisation schemes in Luxembourg.
In attempting to locate tax sheltering within the accounting definitions, Gallemore, Maydew
and Thornock (2013), posited that tax shelters “tend to be at the extreme end of the tax
8
This does not include risk of losses from the scheme being detected by tax authorities.
9
Commonwealth of Australia, Income Tax Assessment Act, 1936
Federal Commissioner of Taxation v Spotless Services Limited and Anor [1996] HCA 34; (1996) 186 CLR
404
10
Page 17
Corporate tax avoidance
Submission 75
avoidance spectrum.” (p.8) However, as Bankman (1999) noted, “a tax shelter can be
defined as a product whose useful life is apt to end soon after it is discovered by Treasury”
(Wilson 2009, pp. 971-972). This makes it is difficult to codify against such schemes and has
led to the development of judicial doctrines and enactment of GAAR’s to combat these
schemes. The term, “tax shelter” does not refer to a specific method of tax minimisation, it
refers to a class of activities or schemes that are particularly tax aggressive and do not appear
to be the intended consequences of the tax legislation.
To determine the ranking of firms along the tax avoidance continuum, Lisowsky, Robinson
and Schmidt (2013) use the uncertainty of tax positions that “arise when firm managers and
their advisors apply ambiguous tax law to complicated facts and circumstances.” ( p.589) In
order to demarcate the boundary where the sub-set of “tax sheltering” begins, Lisowsky,
Robinson and Schmidt (2013) use the definition from Frank, Lynch and Rego (2009),
whereby “tax sheltering” firms are those with “high discretionary pre-tax accruals
(permanent book-tax differences)” (Frank, Lynch and Rego 2009, p. 468). Without some
guidance on what is regarded as “high”, this definition provides a fairly porous border. It is
undefined as to where along the tax avoidance continuum, the boundary of “tax
aggressiveness” is situated. While much research is only interested in the means, rankings
and variances of the levels of tax aggressiveness, it is important to know where in the
rankings, how far from the mean, and at what level of variance does “tax aggressiveness”
begin. Although a firm maybe ranked in the bottom quartile of all firms, it may still be
avoiding tax in terms of the methods being employed or the transactions it has entered into.
The most common current tax minimisation practice is transfer pricing. In their study of
transfer pricing in Australia, Richardson, Taylor and Lanis (2013) define Aggressive transfer
Page 18
Corporate tax avoidance
Submission 75
pricing as activity is reflected by extensive non-arm’s length transactions between related
parties.” (p.137)
Despite the fascination with the concept of the “continuum” that has taken place since Hanlon
and Heitzman (2010) first coined it in their review of tax research, not all researchers share
the same view of tax aggressiveness. For instance, Balakrishnan, Blouin, and Guay (2011)
define tax aggressiveness as “paying an unusually low amount of tax given a firm’s industry
and size.” (p.3) However, this definition also suffers from subjectivity in deciding when low
becomes unusual, but it does identify the firms and industries of interest.
Rego and Wilson (2012) discuss the difficulty in developing theoretical explanations and
definitions of tax aggressiveness when there is an incomplete understanding of why some
firms are more tax aggressive than others. In these circumstances, the choice of proxies to
measure tax aggressiveness needs to be informed by the definitions. Some definitions may
require legal concepts and others may require costs and benefits to be estimated. These
aspects of the definitions come from economics and legal tax research.
1.2
Economic-based definitions
Most of tax research in economics has been based around the concepts of tax burden and tax
incidence, with little economic research examining “tax aggressiveness.” When economics
research does enter this field, much of the research examines tax evasion, rather than “tax
aggressiveness” or “tax avoidance”. This research defines “tax evasion” as the “wilful
understatement” of taxable income. The focus is principally on individual taxpayers rather
than corporations, and employs theories from risk-based decisions under conditions of
Page 19
Corporate tax avoidance
Submission 75
uncertainty, to analyse the activities and behaviours involved in tax evasion. Despite this,
economics has developed a comprehensive analysis of the costs and benefits of tax
avoidance, and opened up the important issue of implicit taxes. This has provided a better
perspective of the incentives for tax avoidance, particularly for studies into the characteristics
of tax aggressive firms and corporate governance issues.
Early economic theories focused on tax evasion rather than tax avoidance (Allingham and
Sandmo 1972; Srinivasan 1973; Greenberg 1984). These studies analysed tax evasion in
terms of decision-making under conditions of uncertainty: the benefit to be obtained by
evading taxes versus the risk and subsequent penalties of being detected (Allingham and
Sandmo 1972).
The issue was analysed under a “classical” economic framework of a
maximisation problem for the tax payer in which tax policy is exogenously given (Greenberg
1984). By adopting the viewpoint of the tax authorities, Landsberger and Meilijson (1982)
provided evidence that tax policy is not endogenous, as taxpayers entering into certain
activities in order to avoid taxes, alter their chances of being audited by the tax authorities.
As a whole, the concentration of tax research in economics is on tax evasion, as opposed to
tax avoidance or tax aggressiveness, resulting in definitions based on legal and judicial
principles.
Weisbach (2002) introduced the concept of marginal cost to the study of tax evasion in
economics, finding that it can induce a distortionary response by taxpayers. The distortionary
effect reduces the efficiency of the tax system, and of the business purpose and economic
substance doctrines used to control tax avoidance (Weisbach 2002). However, Slemrod
(2007) shows that the dividing line between illegal tax evasion and legal tax avoidance is
blurry. Under U.S. law, tax evasion refers to a case in which a person or corporation, through
Page 20
Corporate tax avoidance
Submission 75
commission of a fraud, unlawfully pays less tax than the law mandates. Tax evasion is a
criminal offense under U.S. federal statutes, and subjects a person convicted of the offense to
a prison sentence, a fine, or both. Under that tax law, an overt act is necessary to give rise to
the crime of tax evasion, and therefore, the government must show “wilfulness and an
affirmative act intended to mislead” (Slemrod 2007, p.26).
However, some tax
understatement is inadvertent error, due to ignorance or confusion about the tax law,
including the overpayment of taxes in some cases. Although the theoretical models of this
issue generally refer to the wilful understatement of the tax liability, empirical analyses
cannot precisely identify the taxpayers’ intent, and therefore, cannot precisely separate the
wilful from the inadvertent. Nor can they precisely distinguish legal from illegal activities in
complicated areas of tax law.
The difficulty of identifying wilful tax noncompliance is reflected in the varying terms used
by tax research in economics, such as “evasion,” “noncompliance,” “misreporting,” and the
“tax gap.”
As with tax research in accounting, economics has difficulty defining “tax
aggressiveness” for empirical purposes. However, they both have strong links to judicial
determinations on tax matters. Economic research requires a judicial definition to ascertain
those activities and behaviours that are regarded as “wilful non-compliance”, while
accounting research also requires a legal determination of the level of uncertainty regarding
tax positions taken by a firm.
Page 21
Corporate tax avoidance
Submission 75
1.3
Legal-based definitions
Accounting-based definitions involve the concept of uncertain tax positions, whereas legal
definitions of “tax aggressiveness” are mostly concerned with what is illegal tax minimisation
or tax evasion, as opposed to legal tax avoidance, or tax planning, activities, and are based, in
part, on legislative measures that have been introduced to combat abuse of the tax system.
An uncertain tax position is only uncertain until it has been judicially determined. In 1999,
Bankman wrote: “[A] tax shelter can be defined as a product whose useful life is apt to end
soon after it is discovered by Treasury” (Wilson 2009, pp. 971-972). Judicial definitions are
therefore important to understanding the level of uncertainty of tax positions adopted by
corporations. Based on the concepts of economic substance and business purpose, these
definitions have evolved through the judicial system in interpreting tax legislation, as well as
through changes to the legislation.
In a 2008 study into tax intermediaries11, the Organisation for Economic Co-operation and
Development (OECD) used the perspective of the tax authorities to define “aggressive tax
planning” as tax revenue-reducing behaviour unforeseen by legislators and lacking in
transparency. The tax authorities are concerned about “the risk that the tax legislation will be
misused to achieve results which were not foreseen by the legislators” (OECD 2008, p.87).
They are also concerned that a taxpayer with a favourable tax return will not openly disclose
in the tax return, the uncertainty of significant matters being in accordance with the law. In
1999, Bankmann (1999) used a similar conception in defining tax shelter activities as “taxmotivated transactions based on a literal interpretation of government regulations inconsistent
11
Tax advisors, lawyers, accountants and banks providing tax advice to corporations and individual taxpayers.
Page 22
Corporate tax avoidance
Submission 75
with the original intent of the legislation.” (Wilson 2009, p.972) However, the fact that the
tax revenue consequences of a transaction are not what were expected by the tax authorities
does not mean they were not as expected by the legislature when they enacted the legislation.
Furthermore, it is up to the courts to interpret the legislation and to determine whether tax
planning is “acceptable”. For Freedman, Loomer and Vella (2009), this raises questions
about parliamentary intentions, the role of legislators, the courts and the administration, and
also the rights and duties of taxpayers and their advisers.
According to Freedman, Loomer and Vella (2009), large corporate taxpayers, in most
instances, want their tax planning to be legal, efficient and appropriate, to avoid reputational
risks, and to ensure that it satisfies all reporting and compliance requirements. This will not
always concur with the views of the tax authorities regarding the degree of risk relating to
particular transactions or the interpretation of legislation. All tax jurisdictions have difficulty
in drawing the boundary at which “tax planning”, or “acceptable” tax behaviour, becomes
“unacceptable”, “aggressive tax planning” or “tax avoidance” (Freedman, Loomer and Vella
2009). This gives rise to the uncertainty of the tax positions taken by firms.
It can be seen from tax research in both accounting and economics that “tax aggressiveness”
can encompass both legal and illegal activities.
When examining the issue of tax
aggressiveness from a judicial standpoint, the issue is whether the transactions, or scheme of
arrangements, are legal or not. Part IVA of the Australian Income Tax Assessment Act, 1936
(ITAA 36) is a general anti-avoidance provision or rule (GAAR). It applies to tax benefits
obtained through schemes or arrangements that are not covered by specific anti-avoidance
Page 23
Corporate tax avoidance
Submission 75
rules elsewhere in the tax legislation.12 It was introduced in 1981 to counter the rampant tax
avoidance industry which had proven itself largely immune to specific anti-avoidance
measures (Murphy, 2000). According to Murphy (2000) the intent of the legislation was to
target arrangements whereby a taxpayer could enter into an artificial transaction without any
commercial basis solely for the purpose of reducing their tax liabiliy and without suffering
the risk of a commensurate fiscal cost. However, Part IVA was subsequently applied to
arrangements that were quite clearly commercial in nature but had been structured to obtain
the maximum tax benefits.
As with the accounting and economic definitions of tax
aggressiveness, the question for the courts was “how far can a commercial transaction be
manipulated to obtain a taxation advantage without attracting [Part IVA]” (Murphy 2000,
p.198).
Section 177F of ITAA 36 allows the Commissioner of Taxation to cancel income tax benefits
derived from a scheme that’s sole or dominant purpose is to enable the taxpayer to obtain the
tax benefit. Although the legislation does not define the term “sole or dominant purpose”,
Section 177D(2) sets out eight criteria that need to be considered in determining the purpose
of a scheme. These include the manner in which the scheme was entered into, its form and
substance, the timing of the scheme, the results to be achieved, and the economic
consequences of entering into the scheme. These criteria constituted the tests for the “sole or
dominant purpose” doctrine that was articulated by the High Court in FCT v Spotless
[1996]13.
The judgement stated that “in its ordinary meaning, dominant indicates that
Tax legislation in this context includes both the Income Tax Assessment Act, 1936 and the Income Tax
Assessment Act, 1997.
12
Federal Commissioner of Taxation v Spotless Services Limited and Anor [1996] HCA 34; (1996) 186 CLR
404
13
Page 24
Corporate tax avoidance
Submission 75
purpose which was the ruling, prevailing, or most influential purpose.” (p.416) The test to
determine the dominant purpose of a scheme is an objective test based on what the purpose of
a reasonable person would be under the eight criteria outlined in Section 177D(2), and it
ignores the stated motivations of the taxpayer. The judgement in Spotless made it clear that
the presence of a rational commercial decision “does not determine the answer to the question
whether, within the meaning of Part IVA, a person entered into or carried out a 'scheme' for
the 'dominant purpose' of enabling the taxpayer to obtain a 'tax benefit.” (p. 416) In fact,
Callinan J in FCT v Hart [2004)14 made a distinction between the objectives of the taxpayer
that may be “entirely irreproachable and proper” … [and the] … “means adopted to achieve
these results”. (p.96) The other judges in the Hart case held that the tax benefits may be
cancelled even if the scheme “also advances a wider commercial objective.” (p.16) These
judicial doctrines accord with the business purpose and economic substance doctrines applied
by the U.S. judiciary, as the U.S does not have any GAAR in their tax code.
Braithwaite (2005) adopts the notion of dominant purpose as the basis of his definition of
aggressive tax planning in Australia. He defines aggressive tax planning as “a scheme or
arrangement put in place with the dominant purpose of avoiding tax.” (p.16) Lanis and
Richardson (2013) endorse this approach by defining corporate tax aggressiveness as “a
scheme or arrangement put in place with the sole or dominant purpose of avoiding tax which
is not within the spirit of the law” (p.75). The issues of dominant purpose and economic
consequences underline the judicial approach to judgements regarding the legality of
arrangements and schemes to reduce tax.
14
FC of T v. Hart & Anor [2004] HCA 26
Page 25
Corporate tax avoidance
Submission 75
With such strong anti-avoidance measures developed and put in place, it would be expected
that not many tax shelters could still exist. However, there are two main methods used by
corporations to circumvent the tax legislation including the GAAR’s. They are transfer
pricing and profit shifting by debt loading. These two methods of aggressive tax planning are
covered in the tax legislation but are difficult to apply in some situations.
The transfer pricing rules are set out in Sub-division 815-B of the ITAA 97. They were
updated with effect from 1 July 2013 to bring them into line with more recent OECD transfer
pricing guidance. The transfer pricing rules are based on the arm’s length principle and only
apply to cross border dealings. The parties to a transaction do not need to be related for the
rules to apply. There is a seven year limit for the Commissioner to make adjustments for
abuse of transfer pricing, as well as provisions for actual transactions to be disregarded and
substituted with hypothetical arm’s length transactions.
There are new documentation
requirements as the rules are self-assessed with specific provisions for the interaction with the
thin-capitalisation rules as well as increased penalty provisions.
The arm’s length principle, or standard, is a common feature of transfer-pricing rules in most
jurisdictions. It is based on the price that would be agreed between two unrelated parties
dealing at arm’s length such as would occur in a transparent and open market, e.g.
commodities or the stock market. The problem for the arm’s length principle or standard is
when trying to establish an arm’s length price for intangible assets and products such as
intellectual property rights, patents, licenses and copyright. There is no market for these
products and the rights can be easily transferred between jurisdictions. Apple uses this
method to transfer profits to its subsidiaries in Ireland where is has negotiated a tax rate of
just 2.5% on its profits (U.S Senate, 2013).
It uses a cost-sharing agreement with its
Page 26
Corporate tax avoidance
Submission 75
counterpart in the Americas to transfer the rights for the rest of the world to Ireland. Another
Irish subsidiary purchases its products from manufacturers in China and elsewhere and
invoices all sales outside the Americas to there. The invoice includes the license fee to the
other subsidiary and is the cost of the goods sold in Australia. The cost of the goods sold is
usually very close to the price of the goods in Australia, minus a small amount for local
marketing and distribution costs. The cost of the goods sold is a legitimate deduction against
sales revenues for tax purposes.
Thin capitalisation is based on the tax deductibility of interest payments. The rules for thin
capitalisation are found in Division 820 of the ITAA 97, and attempt to put a limit on the
level of debt and interest a company can claim in deductions on its tax. The rules do not
apply to companies that operate on a purely domestic basis and also use the arm’s length
principle to set rates of interest that can be claimed between entities. However, in the recent
Chevron case, a company merger and restructure reduced the paid up capital of Chevron
Australia Pty Ltd from over $3 billion to only $29 million with the shortfall made up by
USD2.5 billion loan from another subsidiary. A similar situation has occurred with the U.Kbased William Hill bookmaker in its takeover of Sportingbet and Tom Waterhouse. The new
entity has had its capital reduced and replaced with debt from a Gibralta-based subsidiary.
As changes in capital structure are common in mergers and takeovers, it is hard for the ATO
to mount a case against such activities using the current rules.
1.4
Summary and conclusions
The development of a definition of tax aggressiveness may not have provided a definite
position on when tax planning or tax minimisation activities become aggressive along the tax
avoidance continuum. But it has provided a taxonomy of terms and a structure to assist with
Page 27
Corporate tax avoidance
Submission 75
understanding the overall system and how the definitions relate to, and interact with, the
mechanisms and measures of “tax aggressiveness.”
The following sections provide an
overview of the major forms of tax avoidance undertaken by corporations, the measures that
have been developed to measure and analyse these activities, and locates these methods and
measures within the current definitions.
Page 28
Corporate tax avoidance
Submission 75
2. Methods used for tax avoidance
There is an old saying among tax professionals that “the poor evade and the rich avoid,”
meaning that the rich tend to reduce their taxes through legal “avoidance” measures such as
tax shelters, while those with lower incomes attempt more outright evasion (Slemrod, 2007).
The same could be applied to large versus small corporations and businesses. When large
corporations are faced with accusations of being tax aggressive, their first response is usually
that they fully comply with all statutory requirements, and pay the correct amount of tax
under all relevant tax laws and regulations. Therefore, the interest of this section is not to
examine whether these companies are doing anything illegal, but to identify the methods that
they are using to reduce their tax rates to below the statutory level, and to identify whether
the methods they are using contravene the intention of the tax laws. More importantly these
methods will be associated with the definitions presented in the previous section because as
standalone activities and transactions they may not be tax aggressive.
Little tax research in accounting has actually described the methods corporations use to
minimise their tax liabilities.
Commentators often use the terms “tax shelters”, or
“aggressive tax planning”, to describe some of these methods. They range from serious tax
reductions such as transfer pricing and profit shifting, to less serious arrangements such
accelerating depreciation and allowances, through to transactions that were actually
encouraged and anticipated by the regulators, such as research and development
expenditures. Currently, the main areas of concern over corporate tax avoidance in the
Australia are transfer pricing and thin capitalisation. This section begins by outlining the
incentives and opportunities that face corporate tax officials, including the costs and benefits,
and then describes the various means by which corporations have reduced their tax liability.
Page 29
Corporate tax avoidance
Submission 75
2.1
Incentives and opportunities
Corporations engaging in tax avoidance, as defined above, may benefit from increased cashflows and lower tax liabilities.
They also face the risk of substantial financial and
reputational costs to the firm, the individual executives involved, and even the board
members, if the methods they have utilised are publicly disclosed (Crocker and Slemrod
2005; Chen and Chu 2005). The decision to engage in tax aggressive behaviours requires an
analysis of the costs and benefits of the arrangement. According to Gergen (2002), the
factors a firm would use to evaluate whether to engage in an overly aggressive tax strategy
are; the tax savings, the cost of executing the strategy, the risk of detection by the tax
authorities, the probability of an adverse judicial decision if detected, and the size of any
penalty that may be imposed. However, there are also disincentives for firms, shareholders
(Desai and Dharmapala 2007) and individuals (Crocker and Slemrod 2005) to engage in and
transact these strategies.
In determining the costs, benefits and drivers of tax avoidance, two world views are put
forward with neither view dominating. One view of corporate tax avoidance focusses on the
incentives required by the individual managers of a corporation, acting as agents of the
shareholders, to engage in tax aggressive behaviours (Crocker and Slemrod 2005; Chen and
Chu 2005) and considers tax aggressive behaviour by the tax manager as a legitimate activity
to increase shareholder value. The other view of corporate tax avoidance places it in context
of the social costs to society as a whole (Shaviro 2001; Avi-Yonah 2009; Williams 2007) and
is based on a different understanding of the nature of the corporation. Under the societal
view, tax avoidance is seen as an illegitimate activity that has adverse effects on governments
and other taxpayers, such as shifting the tax burden.
Page 30
Corporate tax avoidance
Submission 75
In 1976, Jensen and Meckling argued that the key challenge when it came to corporate
governance was one of agency – how to get executives (the agents) to focus on maximizing
the wealth of the shareholders (the principals). Avi-Yonah (2006) describes this as the
“nexus of contracts” or “aggregate” view in which a corporation is seen as just an aggregate
of its individual members and shareholders rather than an entity in its own right. This view
precludes the influence of government in the legitimacy of the corporation. The “agency”
view has adopted the framework from Allingham and Sando’s (1972) deterrence model to
examine tax avoidance in the context of the contractual relationships between the
shareholders of the firm and the tax manager who possesses private information about the
extent of legally permissible tax minimisation strategies (Slemrod 2007). The basic model
under this view assumes that tax avoidance and any reduction in taxes paid to the government
is beneficial to shareholders through increased after tax profits and increased cash for the
payment of dividends. Under this model, the imposition of penalties on the individuals
within an organisation is more effective in limiting tax avoidance than penalties on a firm as a
whole, such as the increased responsibilities imposed on individuals under Sarbannes-Oxley.
According to Slemrod (2007) there is no empirical evidence that the level of non-compliance
by tax managers is affected by the size of the penalty. Of far more importance is the
probability, or risk, that the activity will be detected. The basic tenet of the “agency” view is
that to counteract this deterrence, incentives need to be tailored to the particular manager,
such as cash bonuses rather than equity-based bonuses for the tax manager (Chen and Chu
2005). Therefore, the story of tax avoidance involves no more than an amoral cost–benefit
calculation.
Page 31
Corporate tax avoidance
Submission 75
Under the societal view, corporate tax avoidance has society-wide implications, as opposed to
being just another operating cost of the corporation. If it was just an operating cost, the
objective of minimizing the amount of corporate tax paid would be an understandable one
and would involve few ethical, community or other stakeholder considerations on the part of
the corporation (Avi-Yonah, 2008). The question of whether a corporation pays its fair share
of taxes has a significant impact on the society in which it operates, and raises significant
public concern (Christensen and Murphy, 2004; Landolf, 2006; Williams, 2007; Sikka, 2010;
Lanis and Richardson, 2013). For instance, Frey (1997) differentiates between intrinsic
motivation, under which taxpayers comply with tax liabilities because of “civic virtue,” and
extrinsic motivation, in which they pay because of threat of punishment.
Frey (1997)
suggests that increasing extrinsic motivation can “crowd out” intrinsic motivation.
For
instance, Scholz and Lubell (2001) find that the level of co-operation in certain settings
declines significantly when penalties are introduced. A tax system that is seen as fair for all,
is expected to increase the level of voluntary tax compliance (Williams 2007; Avi-Yonah
2008). However, an experiment by Blumenthal, Christian and Slemrod (2001) “found that
moral suasion has hardly any effect on taxpayers’ compliance behavior” (Slemrod 2007,
p.40). Despite this, the moral and ethical influence over corporate behavior are better able to
explain Weisbach’s (2002) under-sheltering puzzle – why do some companies pursue
aggressive tax minimization and others do not.
The main benefits of corporate tax avoidance is the increase in cash and liquidity (Saveedra
2014), the increase in after-tax profits that are represented in firm performance metrics such
as earnings per share, and a reduced tax liability (Hanlon and Slemrod 2009). The reduced
effective tax rate produced by tax avoidance can also be seen as a positive signal to investors
Page 32
Corporate tax avoidance
Submission 75
(Chi, Pincus and Teoh, 2014; McGuire, Omer and Wilde, 2014; Inger, 2014), reducing the
cost of equity.
The cost of tax avoidance also includes the transaction costs to set up the tax planning
strategy, the risk of detection if the activities are illegal or in the “grey” area, and agency
costs such as the increased ability of mangers to extract rent for themselves (Desai and
Dharmapala, 2009) and the cost of incentives required to engage the tax manager in these
activities (Crocker and Slemrod, 2005; Chen and Chu, 2005).
There are further costs
involved for a company if the activity is detected and disallowed. While it may seem an
intuitive outcome, Gergen (2002) found empirical evidence that the risk of detection
increases as more firms engaging in the same strategy, and as the period a firm pursues the
strategy extends. The costs for a firm to comply with a tax audit if abusive tax planning is
detected, includes staff and mangers time, as well as disruptions from normal activities.
There is also the tax benefit that is disallowed along with back taxes, interest on the tax
deficiency and penalties imposed on both mangers and the firm. Furthermore, there are
reputational and political costs from being associated with tax avoidance, both for the firm
(Gallermore, Maydew & Thornock, 2013) and the individual managers (Chy, 2013).
Detection may also lead to an increased cost of equity capital (Hanlon and Slemrod, 2009).
If the only determinant of tax avoidance was the weighing up of the costs and benefits it
would most probably be much more widespread than it is and involve all firms. However,
different risk profiles in different firms, and amongst individuals, will lead to the variation in
tax avoidance observed in most research (Weisbach, 2002, Hanlon and Heitzman, 2010).
Page 33
Corporate tax avoidance
Submission 75
2.2
Specific Methods used in Tax Avoidance
If we take the broad definitions of tax avoidance from tax accounting research, then a broad
range of tax aggressive activities fit within that scope. Therefore, traditionally accounting
based definitions of tax avoidance attempt to implicitly include all manner of tax avoidance
strategies. Accounting research has shown that firms engage in all manner of tax avoidance
strategies, methods and arrangements, ranging from cross-border avoidance strategies such as
transfer pricing and thin capitalisation (Dyreng and Lindsey, 2009), to intangible holding
companies (Dyreng et al., 2013) and corporate inversions (Clausing, 2014). More recent
accounting definitions have targeted specific activities such as tax sheltering. Legal research
is even more specific in identifying particular schemes or arrangements which may
contravene the anti-avoidance provisions. The aim of this section is to link the various
business activities and transactions that are consistent with the definitions provided in the
section above.
2.2.1
Transfer Pricing
In a typical aggressive transfer pricing (TP) deal, a corporation produces or buys an asset or
product through a low-tax foreign subsidiary and then sells it to the parent company at an
above-market price. The transfer of the assets has not been done at arm’s length because the
price does not reflect the asset’s true value. Tax avoidance through transfer pricing has the
effect of subjecting a disproportional amount of the profit from the ultimate sale of the asset
to the subsidiary in the relatively low tax rate jurisdiction. Therefore, the parent company’s
taxable income is lower than it would be with arm’s length transfer pricing, and the firm has
less need for deductions from debt income.
This is consistent specifically with the
Richardson et al. (2013) definition of transfer pricing. The activities described above are also
Page 34
Corporate tax avoidance
Submission 75
consistent with definitions in Subsection 815 of the ITAA 97, and also within the general
definition of the tax avoidance continuum.
There are several ways in which assets or services are transferred between a parent and an
offshore affiliated entity: an outright sale of the asset; a licensing agreement where the
economic rights transferred to an affiliate in exchange for a licensing fee or royalty stream;
sale of services or a cost sharing agreement; and an agreement between related entities to
share the cost of developing an intangible asset, which typically includes a “buy-in” payment.
Of these approaches, “licensing and cost-sharing are among the most popular and
controversial.” Generally, legal ownership is not transferred; instead economic ownership of
certain specified rights to the property is transferred.
The use of debt to generate tax deductible interest payments is covered more fully under thin
capitalisation (discussed below). However, high leverage ratios have long been utilised due
to the deductibility of interest payments as opposed to dividends which are not deductible.
According to Graham and Tucker (2006), transfer pricing is also used as an indirect substitute
for this interest deduction because transfer pricing serves to reduce ordinary income for tax
purposes. Graham & Tucker (2006) found that, all else being equal, a firm engaged in
income shifting via transfer pricing uses less debt than other firms. This may be useful for
companies that have poor credit ratings or little available collateral to be able to use interest
to shield income.
There are other methods used within abusive transfer pricing, that assist with the transfer of
intellectual property rights and other intangible assets to low tax jurisdictions, such as
Page 35
Corporate tax avoidance
Submission 75
offshore intellectual property havens, royalty and other intellectual property payments
involving tax havens, and cost sharing agreements
2.2.1.1 Offshore Intellectual Property Haven
A variation of transfer pricing occurs when a multinational corporation (MNC) sells or
licenses the rights to intangible assets that were developed in a relatively high tax country, to
a subsidiary in a low-tax country. Intangible assets used for this purpose are items such as
copyright, patents, band names, and intellectual property. For example, an Australian-based
parent company may sell the license for the economic rights of its intellectual property, to a
subsidiary located in Jersey or Bermuda. Once the foreign subsidiary owns the rights, the
profits derived from the exploitation of the rights belong to the subsidiary, not the parent. The
license payment made by the subsidiary to its parent is taxable income, but the parent has an
incentive to set the price as low as possible. If the price paid is low compared to future
profits generated by the license rights, then less income is taxable to the parent, and on the
other hand, the subsidiary’s expenses are lower. Thus, the Australian-based parent has
successfully shifted taxable profits out of the Australia to Jersey or Bermuda, where no
corporate taxes apply. This is also fits within Richardson et al.’s (2013) specific definition of
transfer pricing.
The activities described above are also consistent with definitions in
Subsection 815 of the ITAA 97, and also within the general definition of the tax avoidance
continuum.
Multinational corporations have long had an incentive to house intellectual property abroad in
order to shelter the income derived from overseas sales (Wall Street Journal, 2002). For
example, an Australian multinational sells a patent to a newly formed Luxembourg
Page 36
Corporate tax avoidance
Submission 75
subsidiary, so that the royalties from any sales of products made outside Australia, flow to the
Luxembourg subsidiary, where they are subject to lower tax rates. The subsidiary usually
only pays a small portion of the cost to produce or buy the patent asset. While the price is
supposed to follow the “arm’s length” principle, this is a very subjective decision as there is
no active market for intellectual property to guide the price. While royalties that are collected
by the subsidiary are supposed be to be reported to the ATO, and subject to Australian
company tax. Offshore intellectual property haven (OIPH) deals are therefore akin to transfer
pricing in that they reduce revenue streams for the parent company.
Much of the focus on OIPH’s comes from the U.S. where the U.S. Senate has held inquiries
into large, high-technology firms such as Microsoft, Hewlett-Packard and Apple adopting
these strategies. The lack of large, high-technology companies should not preclude this being
an important tax minimisation issue in Australia. Firstly, Starbuck’s has been able to utilise
similar transactions and benefits with a brand name as the intellectual property.
This
confirms that many channels outside information technology are available to take advantage
of the loop-hole created by the current valuation of intra-corporate transactions involving
intangible assets.
The second issue is the reverse side of these transactions, whereby
Australia is in the “outside the U.S.” zone and all sales revenue from Australia, and the rest of
the world, is routed through a low-tax jurisdiction such as Ireland. The sales in Australia are
offset by large royalty and licence fees paid to the Irish affiliate to reduce the profit in
Australia and boost that in Ireland.
Page 37
Corporate tax avoidance
Submission 75
2.2.1.2 Royalty/Intellectual Property Payments (to tax havens)
This is the same as Offshore Intellectual Property Haven but concentrates on royalty
payments from the use of intellectual property. It is created using the same methods to
transfer intellectual property rights through transfer pricing or cost sharing agreements and is
based on the difficulty of valuing these assets. Valuing intangible assets at the time they are
transferred is complex, often because of the unique nature of the asset, which is frequently a
new invention without comparable prices, making it hard to know what an unrelated third
party would pay for a license.
2.2.1.3 Cost Sharing Agreements
A cost sharing agreement is an arrangement between two parties to share the cost of
developing an intangible asset, such as computer code, production methods, or patents and
assists with transfer pricing strategies. Such an arrangement is used to reduce or avoid taxes
on the transfer of those assets to a subsidiary in a low tax country. For example, if a parent
company wanted a foreign subsidiary to use one of its patents, tax authorities might consider
the transfer a taxable transaction. By establishing a cost sharing agreement, the parent
company and the subsidiary share in the cost of developing the patent, so that both are
entitled to use it as it is not transferred from one entity to the other. This practice is used to
overcome transfer pricing rules. As both parties have the same rights to the intangible asset,
no transfer takes place. For example, Apple uses a cost sharing agreement with its Irish
subsidiaries and has facilitated the shift of an estimated $74 billion worldwide between 2009
and 2012 (U.S. Senate, 2013).
Page 38
Corporate tax avoidance
Submission 75
2.2.2
Thin Capitalisation
Thin capitalization refers to a strategy by the firm to finance business operations and capital
structure primarily with debt capital rather than equity capital (Richardson et al., 1998;
Taylor & Tower, 2009; Taylor, Tower, & Van der Zahn, 2010). A company is said to be
thinly capitalised when its capital is made up of a much greater proportion of debt than
equity, i.e. its gearing, or leverage, is too high.
This excessive use of debt financing
compared to equity finance creates “thinly capitalized” structures of subsidiaries located in
higher tax jurisdictions. This constitutes an important international corporate tax avoidance
technique used by multinational firms (Shackelford & Shevlin, 2001; Shackelford, Slemrod,
& Sallee, 2007; Taylor & Richardson, 2013) These activities are consistent with the legal
definition of “sole and dominant purpose.”
Thin capitalisation is perceived to create problems for two classes of people:

Creditors bear the solvency risk of the company, which has to repay the bulk of its
capital with interest; and

Revenue authorities, who are concerned about abuse through excessive interest
deductions reducing taxable income.
The corporate laws in some countries actually allow for companies to be thinly capitalised.
However, the revenue authorities in those countries will often limit the amount of interest that
a company can claim as a tax deduction, particularly when it receives loans at noncommercial rates (e.g. from associated or connected parties). However, some countries
simply disallow interest deductions above a certain level, from all sources, when the
company is considered to be too highly geared under applicable tax regulations.
Page 39
Corporate tax avoidance
Submission 75
Some tax authorities limit the applicability of thin capitalisation rules to corporate groups
with foreign entities to avoid “tax leakage” to other jurisdictions.
The United States
“earnings stripping” rules are an example. Hong Kong protects tax revenue by prohibiting
payers from claiming tax deductions for interest paid to foreign entities, thus eliminating the
possibility of using thin capitalisation to shift income to a lower-tax jurisdiction.
Thin capitalisation rules determine how much of the interest paid on corporate debt is
allowable as a deduction for tax purposes. Such rules are of primary interest to private-equity
firms, which use significant amounts of debt to finance leveraged buyouts.
The rules for thin capitalisation in Australia are found in Division 820 of the ITAA 97. They
attempt to put a limit on the amount of interest a company can claim in deductions on its tax.
The rules do not apply to companies that operate on a purely domestic basis. The rules use
the arm’s length principle to set rates of interest that can be claimed between related entities.
However, in a recent case involving Chevron and the ATO, a company merger and
restructure reduced the paid up capital of Chevron Australia Pty Ltd from over $3 billion to
only $29 million with the shortfall made up by USD2.5 billion loan from another Chevron
subsidiary. The ATO is attempting to have the interest deductions disallowed. A similar
change to a company’s capital structure has occurred with the U.K-based William Hill
bookmaker in its acquisition of Sportingbet and Tom Waterhouse. The new entity has had its
capital reduced and replaced with debt from a Gibralta-based subsidiary. As changes in
capital structure are common in mergers and takeovers, it is hard for the ATO to mount a case
against such activities using the current rules.
Page 40
Corporate tax avoidance
Submission 75
2.2.2.1 Debt Loading
Debt loading is similar to thin capitalisation but the term usually describes a situation where a
company is heading for insolvency, leading to different outcomes. In these situations, there
is an incentive for managers and owners to load as much debt as possible onto such a
company including tax liabilities. The debt loading is usually accompanied by disbursements
of any remaining cash. The debts and liabilities will be written off by the insolvency when
there are insufficient funds remaining to cover them. This is basically an abuse of the limited
liability status that corporations enjoy and was a major component of the “bottom-of-theharbour” schemes that proliferated in Australia in the 1980’s. These schemes have been
closed through Directors now carrying personal liability for losses incurred from trading
while insolvent.
2.2.3
Real Estate Investment Trusts (REIT’s)
REIT’s are a special application of the general laws of trusts in Australia and are classified as
a sub-class of Managed Investment Trusts (MIT’s).
These are widely held collective
investment vehicles which allow individuals to pool together their capital to enable
investment in larger and more diversified assets than would otherwise be the case. They also
provide a relatively effortless method to invest in these sectors as the investment decisions
and everyday management of these trusts is undertaken by investment managers.
In general, REIT’s get favourable tax treatment in exchange for disbursing most or all of their
income to the unit-holders. In reality, a REIT is taxed the same as a corporation, but unlike
an ordinary corporation, a REIT can deduct dividends paid (distributions) to its shareholders
(unit-holder) in determining its taxable income. The net income of a trust estate is the taxable
Page 41
Corporate tax avoidance
Submission 75
income of the trust calculated as if the trustee is a resident taxpayer. The trust income may
differ from the net income. In general terms, a beneficiary is taxable on a share of the net
income of the trust estate. A beneficiary is entitled to a share of the trust income depending
on the nature of a beneficiary’s interest in the trust, the type of trust or whether there are
different classes of beneficiary with varying entitlements to trust income or capital. There is
confusion, even within the ATO, as to the definitions and concepts of trust income, present
entitlement or share of the trust income, leading to uncertainty as to how beneficiaries and
trustees are to be taxed. However, REIT’s generally attempt to make distributions of an
amount that reduces taxable income of the trust to zero. REIT’s are not consistent with any
specific definition but are within the continuum of activities that reduce tax liabilities.
The taxation treatment of distributions to beneficiaries is dependent on a number of factors
such as the individual circumstances of the beneficiary, the nature of their interest in the trust,
and the type of the distribution. Where MIT’s are established as unit trusts, unit holders
would normally not have a specific interest in any particular asset of the trust. Rather, a unit
is usually taken to represent a proportional interest in the whole of the trust property net of
any liabilities. Recent High Court and Federal Court decisions have raised some uncertainty
for MIT’s and investors in applying provisions which require a beneficiary to have a specific
interest in the trust property.
The Capital Gains Tax (CGT) provisions for MIT’s mean that acts of the trustee in relation to
the assets are ignored and the beneficiary is taken to have done the acts for CGT purposes.
CGT applies separately to the disposal of trust assets by the trustee and the disposal of a
beneficiary’s unit in the trust. Eligible beneficiaries offset their own capital losses against the
Page 42
Corporate tax avoidance
Submission 75
grossed up capital gain before applying the CGT discount to determine their individual
capital gains.
Certain widely held trusts such as MIT’s may be provided with more flexible rules in some
areas of tax law. For example, under the trust loss rules less onerous tests apply to determine
when losses can be recouped by the trust. Another example is where beneficiaries are able to
access special rules when determining whether they satisfy the 45 day holding period rule for
the purposes of being eligible to receive a share of the franking credits attributed to a
dividend distribution received by the trustee.
REIT’s were first developed in the U.S. in the 1960’s and were introduced in Australia in the
early 1970’s. In Australia, these investment structures were originally known as Listed
Property Trusts (LPT’s).
The name was changed to A-REIT’s (Australian Real Estate
Investment Trusts) in 2008 for international conformity.
REIT’s were introduced into
Canada and the U.K. as late as 2007 and the number of countries introducing them is
growing, particularly in Asia. The legislation that allows REIT’s is fairly uniform between
different countries. Some of the qualification criteria may differ but the flow-through nature
of distributions from REIT’s are similar in Australia, the U.S. Canada and the U.K. Some
jurisdictions, such as the U.S., impose a withholding tax of 30% on distributions to non–
resident unit-holders (in the U.K. it is 20%), but this is not imposed on A-REIT’s. In the
U.S., only 90% of REIT income has to be paid out as distributions.
The main qualification for REIT’s is that they must have “wide” ownership, which means
that the ownership must not be concentrated in the hands of just a few investors. The tax
treatment of the distributions when in the hands of the individual investors varies between
Page 43
Corporate tax avoidance
Submission 75
jurisdictions, however, most legislation allows corporations to invest in REIT’s, and the
distributions are treated as company income, subject to company taxes. In Australia, noncorporate unit-holders can access the capital gains tax (CGT) 50% discount on any capital
gains that are distributed by the REIT, if they have held the units for more than twelve
months. CGT for Australian superannuation funds can be as low as 10% if the assets were
owned for more than twelve months.
The main problem with REIT’s is that there is no disclosure of the amount of tax paid on
these earnings. There are no disclosures of the beneficiaries of these vehicles or the tax
treatment ultimately applied. REIT’s produce a favourable tax arrangement for non-resident
unit-holders from countries with a tax treaty as this can reduce the withholding tax on
distributions to only 15%. There is substantial evidence of REIT’s being advertised as
superior investment vehicles to overseas clients by accounting firms and investment advisors
which is consistent with the general definition of tax aggressiveness (Ashurst, 2013; PwC,
2013; Minter Ellison 2014).
2.2.3.1 Stapled Securities
Australia is relatively unique internationally in permitting business entities to issue stapled
securities. These are used by many A-REITS and infrastructure funds and at the end of 2011,
fifteen out of eighteen listed infrastructure funds and 28 out of 49 listed REITs had stapled
security structures. Stapled securities involve the stapling together of separate securities,
such as a share in a company and a unit in a trust, which cannot be traded separately. For
instance, prior to their most recent restructure, shares in the Westfield Group (WDC) were
stapled securities made up of shares in Westfield Holdings Ltd, units in the Westfield Retail
Page 44
Corporate tax avoidance
Submission 75
Trust and units in the Westfield American Trust. Stapling is relatively uncommon in the rest
of the world. In mid-2011 the Canadian authorities introduced new tax legislation to prevent
emerging use of stapled structures by REITs, and the USA has prevented such structures
since 1984.
Stapling of securities requires some form of contractual relationship between the entities
whose securities are being stapled. The approach used by WDC was for a contractual
operating agreement between the management company, Westfield Holdings Ltd, and the
Westfield trusts that underpins the stapling of the securities. The trusts own the physical
assets and lease them to Westfield Holdings Ltd for use in generating income.
The most obvious rationale for the stapling of securities is tax arbitrage, or taking advantage
of different tax treatments of income and allowable deductions in different jurisdictions. If
stapling reduces the total tax bill paid on income generated by a particular business activity,
stakeholders (other than government) benefit. The trust is a “pass-through” entity for tax
purposes and any income it receives is not subject to company tax as long as it is paid out to
unit holders. Consequently lease payments by the company reduce its taxable income and
company tax paid, and increase the income of the trust on which company tax is not paid.15
Stapled securities are not consistent with any specific definition of tax avoidance, but are
within the continuum of activities that are used to reduce tax liabilities.
15
Similar effects could be achieved by the trust making a loan to the company, which buys the assets, with
interest payments from the company to the trust reducing company tax paid. There are a number of complicating
factors in design of such arrangements including efficient use of depreciation tax shelters. The imputation tax
system used in Australia also complicates the analysis.
Page 45
Corporate tax avoidance
Submission 75
Other potential sources of value from stapled securities results from possible market
imperfections or investor behavioural biases. Constraints on companies paying dividends
when unprofitable may mean that trusts are better able to distribute available cash flow to
investors. Attaching franking credits to stapled securities which are more “debt-like” may
appeal to low tax rate investors who do not want increased equity exposure. The ASX
practice of measuring market capitalization by the aggregate value of stapled securities,
rather than the equity component alone, means that size of the business is overstated relative
to a similar entity with separate equity and debt. This can bring advantages where fund
managers invest primarily in larger capitalization stocks.
However, there is also the possibility that these structures are established by the sponsors, or
controlling interests of the business involved, to extract wealth from investors and
consolidate their control over the business. Virtually all such business structures are opaque
and complex, and result in governance arrangements which give investors limited control
rights. Sponsors of the business (such as investment banks and holding companies) can
generate fee income streams as the entity responsible for the trust, as well as from providing
external management and other services for the operating company, and may profit from
selling assets into the structure. The use of complex structures to reduce tax liabilities has
been found to accompany increased rent extraction by mangers (Desai and Dharmapala
2006).
Unfortunately investors, and advisors, do not appear to fully understand the nature of the cash
flow streams which they receive from such structures. Typically some part of that cash flow
will be a return of capital – but the total cash flow is generally described as a “yield”, causing
stapled securities to appear to be high (and stable) yield investments. This exploits a well-
Page 46
Corporate tax avoidance
Submission 75
known behavioural bias of investors who have an unwillingness to consume out of capital
rather than income. This is augmented by the unusual practice, accepted by the ATO, of
referring to the capital return component as “tax deferred income” – rather than as a return of
capital.
The rationale for preventing stapling in most other countries is straightforward – to prevent
tax arbitrage which reduces company tax collections. However, with the dividend imputation
tax system that operates in Australia, the payment of company tax is “washed out” by tax
(franking) credits attached to dividend payments and which reduces tax paid at the investor
level for Australian resident shareholders. Consequently, the dividend imputation tax system
means that the incentive for the Australian government to legislate against such structures is
much reduced. However, the main cost to the Australian government tax revenue arises from
distributions to foreign shareholders. Given the popularity of stapling in Australia, the
imputation tax system should also reduce the incentives for Australian business entities to
adopt such structures in order to avoid company tax, unless overseas investors are a
significant clientele, and therefore, some tax arbitrage is involved. It is possible that tax
revenue is adversely affected by the use of stapled securities. However, details on foreign
ownership of stapled securities are difficult to access. Only around 60 per cent of A-REITS
use stapled securities. The apparent ability of non-stapled A-REITS to successfully compete
suggests that stapling is not a necessary condition for efficient operations.
Page 47
Corporate tax avoidance
Submission 75
2.3
Activities related to U.S. multinationals operating in Australia.
Australian companies using tax avoidance methods outlined above have a direct effect on tax
revenues in Australia by transferring the tax base to a low tax or secrecy jurisdiction.
However, overseas companies operating in Australia, whether as a market for their goods, or
to engage in production operations, also use other methods to reduce their exposure to
Australian tax and the potential loss to the Australian tax revenues is significant. The
following analysis is based on the methods used by U.S. companies, even though MNC’s
from other countries also use many of these methods to shift profits out of Australia. These
methods are consistent with the general definitions of tax avoidance and tax sheltering.
2.3.1
Subpart F
Rather than being a method of tax minimisation, Subpart F is an attempt to prevent abuse of
U.S. corporations’ ability to defer foreign income. Subpart F of the U.S. Internal Revenue
Code was enacted to deter U.S. taxpayers from using controlled foreign corporations in tax
havens to accumulate earnings that should have accumulated to the parent company in the
United States. A substantial part of the design of Subpart F was to address some of the tax
avoidance techniques that are still being utilized by U.S. multinational corporations.
Subpart F applies to certain income of ‘controlled foreign corporations’ (CFC’s). A CFC is a
foreign corporation more than 50% of which, by vote or value, is owned by U.S. persons
owning a 10% or greater interest in the corporation by vote. ‘U.S. persons’ includes U.S.
citizens, residents, corporations, partnerships, trusts and estates. If a CFC has subpart F
income, “each U.S. shareholder must currently include its pro rata share of that income in
their gross income as a deemed dividend.” (U.S. Treasury, 2000, p. xii) Certain regulations
Page 48
Corporate tax avoidance
Submission 75
and temporary statutory changes such as permanent overseas reinvestment and the “checkthe-box” rules, have undercut the application of Subpart F, however, which is discussed
below.
2.3.1.1 Permanent Overseas Profit Reinvestment
If the profits from CFC’s are designated as being permanently reinvested overseas, they are
exempt from U.S. taxation. This was the compromise from the total abolition of overseas
profit deferral when Subpart F was introduced. U.S. tax code ASC 740 allows companies to
designate foreign earnings as “permanently reinvested earnings” (PRE’s) and therefore, delay
the recognition of the deferred tax liability that would arise from the U.S. repatriation tax.
This not only reduces future tax liabilities, but also increases reported after-tax income. This
leads to an incentive for U.S. MNC’s to delay the repatriation of overseas profits and has
substantially increased the overseas cash holdings of these entities (Edwards, Kravet and
Wilson, 2014). Recent estimates put the amount of PRE’s being held by U.S. MNC’s at
nearly $US2 trillion saving these corporations about $US550 billion in tax (CTJ, 2014;
Economic Policy Institute, 2014). ASC 740 was a watering-down of the restrictions on
deferral of overseas income that were introduced with Subpart F and assisted the continued
erosion of the U.S. tax base. However, there appears to be little political will within the U.S.
to change this arrangement as can also be seen with the introduction of the “check-the-box”
regulations.
2.3.1.2 Check-the-Box Regulations
The check-the-box regulations were issued by the US Treasury Department in 1997. These
rules allow a controlled foreign corporation (CFC) to be designated as a branch of the parent
Page 49
Corporate tax avoidance
Submission 75
company instead of a subsidiary for tax purposes. This allows the corporation to circumvent
the CFC and Subpart F regulations as they only apply to CFC’s.
This reduced the
effectiveness of the anti-deferral rules of Subpart F further and facilitates an increase in
offshore profit shifting. The check-the-box regulations enable a multinational corporation,
for tax purposes, to ignore the facts reported in its books – that it received passive income.
The regulations eliminated a multi-factor test that had been used to determine the proper
classification of an entity, in favour of a simple, elective "check-the-box” regime. Check-thebox was intended to eliminate the complexity and uncertainty inherent in the test, instead
allowing entities to simply select their preferred tax treatment. It was introduced as a means
to reduce “red tape”, and relieve the compliance burden of companies.
The regulations, however, had significant unintended consequences, and opened the door to a
host of tax avoidance schemes. Under Subpart F, passive income paid from one separate
legal entity to another separate legal entity – even if they were both within the same corporate
structure – was designated as passive income and immediately subject to U.S. tax. However,
with the implementation of the check-the-box regulations, a U.S. multinational corporation
could set up a CFC subsidiary in a tax haven and direct it to receive passive income, such as
interest, dividend, or royalty payments, from a lower tiered related CFC without incurring
Subpart F designated income. The check-the-box rule permitted this development, because it
enabled the multinational corporation to choose to have the lower-tiered CFC disregarded or
ignored for federal tax purposes. In other words, the lower tier CFC, although it is legally
still a separate entity, would be viewed as part of the CFC shell and not as a separate entity.
Therefore, for tax purposes, any passive income paid by the lower tier separate entity to the
Page 50
Corporate tax avoidance
Submission 75
higher tier CFC subsidiary would not be considered as a payment between two legally
separate entities and, thus, would not constitute Subpart F income.
The use of “check-the-box” rules is now widespread and, although it was introduced as a
temporary measure, it has subsequently been embedded in other legislation to extend it, and
attempts to remove it have been met by intensive lobbying by the corporate sector. The use
of the “permanently reinvested” and check-the-box” rules have effectively subverted the
intention of Subpart F, and only when foreign income is repatriated to the U.S. as dividends
does it become subject to U.S. tax.
2.3.1.3 CFC Look-Through Rules
The CFC “look-through” rules were enacted by Congress as a temporary measure in 2004
and made permanent in 2006. The outcome was to reduce the effectiveness of the antideferral rules of Subpart F and it has facilitated a further increase in offshore profit shifting.
If a foreign subsidiary of U.S. Corporation can trace passive income, which should be subject
to Subpart F tax, to the active income of the payer CFC, it can ignore that income for tax
purposes. This has led to the creation of transactions that strip earnings from high tax
countries to low tax environments. In their submission regarding the permanent extension of
the scheme in 2014, the Financial Accountability & Corporate Transparency Coalition said
that “[s]ubsidizing highly profitable corporations comes at the expense of ordinary American
taxpayers, who shoulder their cost through cuts to programs or higher taxes.” (p.1)
2.3.2
Alternating Short-Term Loans
Alternating short-term loans are used to overcome the problems faced by multinational
corporations with the taxation of overseas profits. The use of constantly renewed short-term
Page 51
Corporate tax avoidance
Submission 75
loans allows some corporations to sidestep the thin capitalisation rules and rules that deem
certain intra-company investments to be dividends and therefore liable for tax. In 2012, the
U.S. Senate heard that since 2008, Hewlett Packard Co. and General Electric Co. have used a
loophole in Section 956 of the U.S. tax code to facilitate billions of dollars in short-term
intercompany offshore loans to effectively repatriate untaxed foreign profits back to the
United States to run their U.S. operations, contrary to the intent of U.S. tax policy. There are
no reported cases of Australian companies using this strategy to repatriate profits held in lowtax jurisdictions, however, if they are set up to carefully comply with tax rules and reporting
standards, companies are not required to disclose these arrangements. These schemes also
demonstrate that many tax rules can be vanquished by creative methods.
2.3.3
Tax Treaties
Tax treaties, in their attempt to eliminate double taxation, have left loopholes for tax
arbitrage.
Tax treaties have been negotiated and enacted since the 1920's in order to
eliminate double taxation. However, overlaps between domestic tax systems still result in
some instances of double taxation and at the opposite side, there is no tax system applying to
some situations as a result of the differences between the tax rules and the effect of tax
treaties. These loopholes are used to give effect to tax avoidance schemes such as the Double
Irish/Dutch Sandwich. They are also useful for avoiding withholding taxes when receiving
distributions from trusts and other flow-through enterprises and other forms of tax arbitrage.
In the OECD 2013 BEPS Action Plan it states that "BEPS relates chiefly to instances where
the interaction of different tax rules leads to double non-taxation or less than single taxation.”
(OECD, 2013, p.10) These activities artificially segregate income from the activities and
jurisdictions that generate them, through arrangements that achieve no or low taxation. This
Page 52
Corporate tax avoidance
Submission 75
creates fears with tax policy, caused by gaps in the interactions between the different tax
systems, and in some cases because of the application of bilateral tax treaties, that “income
from cross-border activities may go untaxed anywhere, or be only unduly lowly taxed."
(OECD, 2023, p.10)
2.3.4
Double Dutch with Irish Sandwich
The double Irish with a Dutch sandwich technique is just one of a class of similar
international tax avoidance schemes. Each involves arranging transactions between
subsidiary companies to take advantage of the idiosyncrasies of varied national tax codes.
These techniques are most prominently used by technology companies because these firms
can easily shift large portions of profits to other countries by assigning intellectual property
rights to subsidiaries abroad.
The Double Dutch with Irish Sandwich scheme utilises an array of other tax avoidance
techniques to shift profits to low, or zero, tax jurisdictions.
It involves the use of a
combination of Irish and Dutch subsidiary companies to shift profits to low or no tax
jurisdictions. The double Irish with a Dutch sandwich technique involves sending profits first
through one Irish company, then to a Dutch company and finally to a second Irish company
headquartered in a tax haven. This technique has allowed certain corporations to dramatically
reduce their overall corporate tax rates. Use of the Dutch company allows for the avoidance
of withholding tax on funds being transferred out of the EU as the recipient is an Irish
registered company, but with headquarters in a tax haven. Ireland is one of the few countries
that allow companies to choose their own tax residency. It is this feature alone that allows
this technique to function.
Page 53
Corporate tax avoidance
Submission 75
The "double Irish" provision allows corporations with operations in Ireland to make royalty
payments for intellectual property to a separate Irish-registered subsidiary. The subsidiary,
though incorporated in Ireland, typically has its tax home in a country with no corporate
income tax. Usually, the company arranges for the rights to exploit intellectual property
outside the United States to be owned by an offshore company. This is achieved by entering
into a cost sharing agreement between the U.S. parent and the off-shore company in order to
subvert the terms of U.S. transfer pricing rules. The off-shore company continues to receive
all of the profits from exploitation of the rights outside the U.S., without paying U.S. tax on
the profits, unless and until they are remitted to the U.S.
It is called double Irish because it requires two Irish companies to complete the structure. One
of these companies is tax resident in a tax haven, such as the Cayman Islands or Bermuda.
Irish tax law provides that a company is tax resident where its central management and
control is located, not where it is incorporated, so that it is possible for the first Irish company
not to be tax resident in Ireland. This company is the offshore entity which owns the valuable
non US rights that are then licensed to a second Irish company (tax resident in Ireland) in
return for substantial royalties or other fees. The second Irish company receives income from
use of the asset in countries outside the US, but its taxable profits are low because the
royalties or fees paid to the first Irish company are deductible expenses. The remaining
profits are taxed at the Irish rate of 12.5%.
For companies whose ultimate ownership is located in the United States, the payments
between the two related Irish companies might be non-tax-deferrable and subject to current
taxation as Subpart F income under the Internal Revenue Service's Controlled Foreign
Corporation regulations if the structure is not set up properly. This is avoided by organizing
Page 54
Corporate tax avoidance
Submission 75
the second Irish company, as a fully owned subsidiary of the first Irish company and a tax
resident in a tax haven, making an entity classification election to be disregarded as a separate
entity from its owner (the first Irish company).
The payments between the two Irish
companies are then ignored for US tax purposes.
The addition of a Dutch sandwich to the double Irish scheme further reduces tax liabilities.
Ireland does not levy withholding tax on certain receipts from European Union member
States. Revenues from income of sales of the products shipped by the second Irish company
are first booked by a shell company in the Netherlands, taking advantage of generous tax
laws there. Overcoming the Irish tax system, the remaining profits are transferred directly to
the Cayman Islands or Bermuda. Thus if the two Irish holding companies are thought of as
"bread" and the Netherlands company as "cheese", this scheme is referred to as the "Dutch
sandwich". The Irish authorities never see the full revenues and hence cannot tax them, even
at the low Irish corporate tax rates. There are equivalent Luxembourgish and Swiss
sandwiches.
One example is Google, whose Dublin headquarters is its main hub outside the United States
and employs more than 2,500 people. A Dublin-based subsidiary for Google generates the
revenue, mostly from online advertising, and then pays it in royalties to a separate Google
unit in Ireland, which is resident in Bermuda for tax purposes.
In 2013, following pressure from the European Commission, the Irish government announced
that companies will no longer be able to incorporate in Ireland without also being tax resident
there, a measure intended to counter arrangements similar to the Double Irish. Michael
Noonan, Ireland's finance minister, recently told the Irish Parliament, "I am abolishing the
Page 55
Corporate tax avoidance
Submission 75
ability of companies to use the 'double Irish' by changing our residency rules to require all
companies registered in Ireland to also be tax resident." (NYT, 14/10/2014) These changes
are proposed to take effect in January 2015. However, there will be a phase-out period for
existing schemes that will continue until 2020.
As a result of the increased scrutiny of companies and their relationships with known tax
havens, companies such as Google, Oracle and FedEx are declaring fewer of their ongoing
offshore subsidiaries in their public financial filings, which have the effect of reducing
visibility of entities declared in known tax havens.
2.4
Conclusions
There is a wide variety of business activities and transactions that Australian companies and
MNC’s operating in Australia utilise to avoid paying tax in Australia. Absent an intention to
avoid tax, consistent with definitions presented in Section 1, these activities and transactions
can be a normal part of business operations. This makes it difficult to designate them as tax
avoidance activities without relating them back to the definitions of tax avoidance. However,
they all can be related back to either specific definitions or to general definitions. To place
them along the tax avoidance continuum requires specific measures that capture the business
purpose and the nature of these activities and transactions.
Page 56
Corporate tax avoidance
Submission 75
3. Measuring tax avoidance
The following section presents the various measures used in academic research to proxy for
tax avoidance. Most measures are accounting based, and therefore were developed within
accounting research, as all publically available tax data is disclosed in a company financial
reports (which includes the financial statements and notes to the financial statements). Each
proxy measure will be linked back to the tax avoidance definitions and methods discussed in
the previous sections. The adequacy of each measure, with respect to the definitions and
methods, will also be assessed.
The proxies for tax avoidance can be classified into four groups. Firstly, there are effective
tax rates (ETR’s) that compute a ratio of some measure of the tax liability of a firm to some
measure of its profits. The ratio can be compared to the statutory tax rate. The second group
of proxies are the book-tax gaps (BTG’s) that are sometimes also referred to as book-tax
differences (BTD’s). BTG’s estimate the taxable income by grossing up the measure of tax
liability, using the statutory tax rate, and subtracting it from the measure of the firm’s
accounting profits. This gives the magnitude of the difference or gap. In order to compare
BTG’s between firms, the gap amount is usually scaled by a measure of the firm’s size, such
as total assets, market value, or sales revenue. The third group of measures is based on some
of the other two groups but utilises regression analysis to identify particular categories of tax
avoidance.
The fourth group of proxies are characteristics-based measures that have
examined the effect of different company attributes on the level of tax avoidance. These
measures have been validated through examination of companies that have been detected
using known tax avoidance measures.
All proxies have problems due to inadequate
accounting disclosures and others due to the accounting treatment of some transactions.
Page 57
Corporate tax avoidance
Submission 75
3.1
Effective Tax Rates (ETR’s)
Effective tax rates (ETR’s) were probably the earliest measure of tax avoidance and have
been used extensively since firm-level tax and financial analysis using large amounts of data
that became available in the 1980’s allowed for widespread firm-level tax and financial
analysis. ETR’s are most useful in capturing and comparing the tax burden of firms and
industries, and for ascertaining the level of tax avoidance amongst a sample of firms. All
ETR’s involve some measure of a firm’s tax liability divided by some measure of their pretax income or cash-flow. There have been various measures of both the tax liability and the
income or cash-flow used in the tax avoidance literature, with the variation being dependent
on the research question being asked.
When deciding on the most suitable measure of ETR, Porcano (1986) acknowledged that
there was controversy about whether the average or marginal effective rate should be used.
Fullerton and Henderson (1983) had found low correlation between the two measures, and
Fullerton (1984) who examined the causes of the differences between the measures,
concluded that average effective tax rates are more appropriate for measuring cash flows, and
therefore, distributional burdens, while marginal effective tax rates were more useful for
analysing capital investment incentives.
In order that an ETR be useful for research, Weiss (1979) considered three characteristics of
the calculation that should be included. Firstly, the data used to calculate an ETR for a given
year should only reflect the liability generated with respect to that year’s income, and should
not reflect carryovers of credits or net operating losses from previous years. Secondly, the
ETR should concentrate on the liability generated by domestic income as it is only domestic
Page 58
Corporate tax avoidance
Submission 75
income that is affected by the U.S. tax system, and thirdly, the denominator in the calculation
should correspond as closely as possible to the concept of “economic income”.
Hanlon and Heitzman (2010) draw attention to conforming tax avoidance, as this also creates
problems for measuring tax avoidance. Conforming tax avoidance is the use of methods that
artificially reduce both tax income and book income.
While there are incentives for
managers to maximise book income, there are also instances where managers have actually
paid tax on overstated income (Erickson, Hanlon and Maydew, 2004).
There are also
incentives for the opposite situation where accounting income is artificially reduced to
provide tax benefits. In certain circumstances, firms may not be constrained by having to
maximise profits, such as private firms, and during periods following the installation of a new
CEO. None of the ETR measures will capture these activities as both book income and the
tax expense will be reduced.
Despite these limitations, ETR’s have become the most widely used proxies for tax avoidance
in academic literature. They capture a broad range of tax avoidance activities and confirm
potential levels of tax avoidance.
3.1.1
Zimmerman (1983)
Stickney and McGee (1982) were probably the first to examine the ETR’s of individual firms
rather than aggregated industry data, using a large sample of firms. They relied on data from
the financial statements alone as opposed to tax data from the IRS or a combination of both.
Their study, using multivariate analysis, was to “determine whether combinations of factors
systematically explain differences in effective tax rates” (p.127).
They examined the
Page 59
Corporate tax avoidance
Submission 75
determinants or characteristics of the variation of ETR’s for 1,236 US firms in 1978 & 1980.
The determinants or characteristics tested were:

the degree of capital intensity;

the extent of natural resource involvement;

the proportion of debt in the capital structure;

the size of firm; and

the extent of foreign operations.
Conflicting results regarding the association between low ETR’s and firm size from previous
studies such as Stickney and McGee (1982), Siegfried (1972; 1974) and Salamon and
Siegfried (1977), led Zimmerman (1983) to undertake a detailed investigation into the issue
of ETR’s and firms size. Based on research by Alchian and Kessel (1962) and Jensen and
Meckling (1978), Zimmerman (1983) hypothesised that larger firms are subject to greater
government scrutiny and wealth transfers, than smaller firms. Therefore, in order to reduce
these political costs, large firms are more likely to choose income reducing accounting
procedures than small firms (Dhaliwal, Salamon and Smith [1982]; Watts [1977]; Watts and
Zimmerman [1979]). Zimmerman (1983) predicted that the largest firms would therefore
have the highest ETR’s.
In determining the association between firm size and tax-related political costs, proxied by
variations in firms’ ETR’s, Zimmerman (1983) used both financial statement data and IRS
data in separate comparative tests, calculating ETR’s as:
ETR = Income taxes / Operating cashflows
Page 60
Corporate tax avoidance
Submission 75
Income taxes included state, federal and foreign reported income tax expense less the change
in deferred tax, representing the firm’s total worldwide income tax liability for the year.
Operating cash-flows were used in the denominator, as they exclude the accrual accounting
procedures in pre-tax book income or taxable income that had been found to vary with firm
size (Hagerman and Zmijewski 1979). Operating cash-flows is calculated as the difference
between sales and the cost of goods sold. Adjustments to other cash-flows such as selling
and administrative expenses were ignored in order to maintain comparability between
financial statements and the IRS data. Due to different capital structures and capital-tolabour ratios, this created systematic differences in ETR’s across industries. Subsequent
studies using Zimmerman’s (1983) measure took the operating cash-flow amount directly
from Compustat (Omer et al. 1991). This proxy can be estimated using Australian data.
3.1.2
U.S. Congress, Joint Committee on Taxation (1984)
During the second half of the 1970’s and early 1980’s, there was a change in focus of tax
research from an economics to an accounting emphasis. There was also a concurrent, wideranging debate taking place regarding the effectiveness and efficiency of the U.S. tax system
(Auerbach [1982], Cox [1980], Gabinet and Coffey [1977], Gravelle [1982], McClure [1975],
Severins [1976], Stiglizt [1976]).
The debate culminated in the U.S. Congress Joint
Committee on Taxation undertaking a review of the entire corporate tax system, with their
report being published in 1984. Amongst other matters, the Committee found that the
relative contribution of corporate taxes to government receipts had steadily declined over the
previous 32 years (Joint Committee on Taxation [1984], cited in Porcano [1986]). The
Committee was interested in quantifying the tax burden of U.S. corporations and industries,
including the distributional burdens and anomalies, and calculated the ETR’s as follows:
Page 61
Corporate tax avoidance
Submission 75
ETRit = (TXFEDt + TXFOt) / (IBt + MIIt + XIDOt - ESUBt)
As the Committee was only looking at the Federal tax burden, the measure of the tax liability
was the current Federal tax expense plus the current foreign tax expense. It includes the
foreign tax expense as this is included in taxable income for federal tax purposes. It excludes
the current portion of state and local tax expenses and current tax expense (or savings)
attributable to extraordinary items and discontinued operations.
Deferred taxes were
excluded as they often roll over from one year to the next during a period of growth or
inflation and may not be paid until many years in the future, if at all.
In addition,
corporations may overstate the accrued income tax liability and thus the provision for taxes in
order to provide a "cushion" for potential increases in tax liability resulting from IRS
examinations (Omer et al. 1991). The measure of income was defined as total world-wide
financial income before taxes, minus gains and losses from unconsolidated subsidiaries,
extraordinary items and discontinued operations.
This proxy can be estimated using
Australian data.
3.1.3
Citizens for Tax Justice (1985)
During this period, the Citizens for Tax Justice (CTJ) were also analysing corporate tax
behaviour. They were interested in understanding the variation in the tax burden of different
firms and industries and assessing the equity of the U.S. tax system. To do this they used a
simple ETR, calculated as follows:
ETRit = ΣTXPDt / ΣIBt
The CJT (1985) study, according to Callihan (1994), "calculated average ETR's of 250 large
corporations for the years 1981-1984 by taking taxes currently paid divided by pre-tax
Page 62
Corporate tax avoidance
Submission 75
profit." (p.27) This appears to be strikingly similar calculation as Dyreng et al.’s (2010)
long-run Cash ETR.
The CTJ found that the overall average ETR for this sample of
companies was only 15%, compared to the STR of 46% in effect during the sample period.
The low rates were mainly attributed to the use of accelerated depreciation and investment
tax credits. The proxy can be estimated using Australian data.
This study received harsh criticism from the academic field. Egger (1985) argued that by
focussing only on current tax payments, the CTJ ignored the effect of timing differences, thus
omitting a large portion of the actual tax liability. There can be substantial differences
between ETR's calculated including only current taxes paid and those calculated to include at
least some portion of deferred taxes. (Marovelli 1986) Gupta and Newberry (1997) criticised
the CTJ methodology as being mostly ad hoc, the sample was not representative of the whole
corporate sector, there was no control sample of smaller firms, and no statistical tests of
significance were undertaken.
Egger (1985) also added that there are sound economic
reasons for tax provisions such as accelerated depreciation.
In a study replicating and comparing the methodologies used by the US Congress, Joint
Committee on Taxation (1984), Citizens for Tax Justice (1981, 1983, 1984), and Tax Notes
(1986), Spooner (1986) observed that the types of problems encountered are sample
selection, point estimates versus trends, classification of companies within an industry, other
taxes, the matching of tax and the income on which it is imposed and the difficulty of
computing separate U.S. and foreign rates. Spooner (1986) found that calculating ETR's for
a single firm for a single year can be problematic as they can swing dramatically from year to
year caused by changes in investment patterns, profitability and tax laws, and by the effect of
net operating losses carry-forwards or tax refunds. The study concluded that, “[i]f these types
Page 63
Corporate tax avoidance
Submission 75
of studies are published every year and effective tax rates are averaged over a longer period,
thus making trends more apparent, these studies should become increasingly valuable.”
(p.299)
3.1.4
Shevlin (1987)
Shevlin (1987) examined the factors that influence a firm's decision to fund R&D through a
limited partnership or in-house. As there are two motivations, the tax motivation and the offbalance sheet motivation, Shevlin (1987) was interested in which provided the main or
primary motivation and if there was a point where there was a cross-over in the relative
importance of the two motivations. An ETR was used to examine the tax motivation and was
calculated as follows.
ETRit = (TXTt - ΔTXDBt) / (PIt - [DTXDBt / STRt])
The tax payable is defined as the tax expense less the change in the deferred tax account. This
ignores Zimmerman's (1983) inclusion of the tax deferred investment tax credits, calculated
as total current tax expense plus deferred tax expense minus the change in the deferred tax
liability (Omer et al. 1991). The denominator is pre-tax income minus the change in the
deferred tax account grossed up by the statutory marginal tax rate. Pre-tax income is defined
as reported income plus income (loss) from minority interests. Grossing up the change in
deferred tax liability by the STR eliminates income derived from timing differences. This
calculation results in a smaller denominator compared to Zimmerman (1983) and more
closely reflects the taxable income of the firm (Omer et al. 1991). This proxy can be
estimated in Australia. However, the metrics used in the U.S. do not line up totally with the
Australian disclosures.
Page 64
Corporate tax avoidance
Submission 75
Shevlin (1987) found that the "results indicate that the tax variable adds incremental
explanatory power to the logit model with firm size and thus offer support to the tax
motivation hypothesis … to use an R&D [Limited Partnership]." (p.502). Shevlin (1987)
conceded that, as this study is examining the incentives to engage in limited partnerships for
R&D investment, it is the marginal tax rate instead of the effective tax rate that should be
used, but that is difficult to estimate. Shevlin (1987) also noted that from the results of this
study "implicit" taxes16 need to be included in the definition of tax planning.
At the same time that Gupta and Newberrry (1992) were investigating the effectiveness of
TRA86, Shevlin and Porter (1992) re-examined the evidence presented by the Citizens for
Tax Justice (1988) regarding the success of the reforms. The success was claimed by the
increase in the ETR’s of a group of the largest firms. The results were consistent with the
CTJ (1988) findings that the ETR's of large firms increased after TRA86.
3.1.5
Manzon and Plesko (1984)
In order to examine the degree to which the Economic Recovery Act, 1981 (ERTA) and the
TRA86 had altered the level and distribution of ETR’s, Manzon and Smith (1994) calculated
two separate ETR’s. One was the U.S. Federal ETR that proxied the tax liability with the
sum of current federal income tax, and used the sum of pre-tax income for the period as the
accounting income. This proxy cannot be estimated in Australia. It was calculated as:
Federal_ETRit = TXFEDt / PIt
16
eg. interest from local government bonds is tax-free but corporations also accept a lower rate of return from
these investments.
Page 65
Corporate tax avoidance
Submission 75
The second measure was the world-wide ETR that used the total tax less the deferred portion
for the period as the tax liability, and used the same definition of the accounting income as
the first measure. This proxy can be estimated in Australia but may contain inconstancies
with the deferred tax disclosure. It is calculated as:
Worldwide_ETRit = (TXTt - TXDt) / PIt
Manzon and Smith (1994) used a sample of all firms reported on the Compustat database
with sufficient information to calculate the ETR measures. The measurement periods were
1978-1980 (pre-ERTA), 1982-1985 (post-ERTA), and 1988-1990 (post-TRA86), and firms
with negative total reported income over an entire measurement period were excluded. The
results showed that ERTA led to a significant reduction in effective tax rates, especially for
larger firms with significant capital investments, but did not alter the variation between firms
in different industries. TRA86 reversed this decline and significantly increased ETR's along
with a reduction in the variation in ETR's among firms in different industries and size
groupings. Smaller firms had higher ETR's in both pre- and post-ERTA periods conflicting
with Zimmerman's (1983) findings regarding the political costs to larger firms causing higher
taxes, and therefore, higher ETR’s.
3.1.6
Dyreng, Hanlon and Maydew (2008)
There was little activity in pursuing new ETR’s until Dyreng, Hanlon and Maydew (2008)
started examining the level of tax avoidance over longer periods than the single year period
used in most research. The use of single year estimates increased the number of observations
for statistical purposes but led to high levels of volatility in the ETR’s, especially for ETR’s
Page 66
Corporate tax avoidance
Submission 75
based on cash taxes paid. This analysis had not been undertaken since the Citizens for Tax
Justice examined the two hundred largest U.S. companies in the early 1980’s.
Dyreng, Hanlon and Maydew (2008) were attempting to understand what abilities firms have
to avoid income taxes over long periods of time and to evaluate whether single-year ETR's
are predictive of tax avoidance activities over a longer period. They calculated single year
estimates using the following formula;
ETRit = TXPDt / (PIt - SPIt)
The average cash ETR’s were estimated for both five-year and ten-year periods using the
following formula.
ETRit = Σ(TXPDt) / Σ(PIt - SPIt)
The use of Cash Taxes Paid identifies tax avoidance associated with permanent differences.
Permanent differences can cause estimates of ETR’s based on the Current Tax Expense being
overstated. The cash ETR is also not affected by changes to the valuation allowance or tax
cushion.
Using the average of Cash Taxes Paid over a longer period overcomes the
difficulties with interpretation and inferences, caused by tax refunds and adjustments for
previous periods, as well as problems associated with the estimation of Deferred Taxes. The
use of the average Pre-tax Income over a longer period overcomes the problem of negative
denominators (loss firm-years) as this can obscure inferences about a firm's tax avoidance
activities. Dyreng et al. also removed the effects of Special Items on the Pre-Tax Profit as
they wanted to examine the tax effect on the underlying operations of the firms. These
proxies can be estimated using Australian data.
Page 67
Corporate tax avoidance
Submission 75
Dyreng, Hanlon and Maydew (2008) analysed the ETR’s of a large sample of U.S. listed
companies for a ten year period from 1995 to 2004. They found that a quarter of the sample
firms were able to maintain an ETR below 20% during the period. They also found that
annual cash ETR's are not good predictors of long-run cash ETR's and therefore, are not
accurate predictors for long-term tax avoidance. Low annual cash ETR's are more persistent
than high cash ETR's, and low, long-run cash ETR's appear to be spread across most
industries but with some clustering. Dyreng et al. (2008) included the proviso that implicit
taxes are not included in the analysis, and that under certain conditions, avoidance of explicit
taxes comes at the price of bearing implicit taxes, such as reductions in pre-tax rates of return.
Frank, Lynch and Rego criticised this measure as only reflecting tax avoidance rather than
tax avoidance.
However, that will depend on the definitions used for these concepts.
Regardless of the criticisms, this measure has become one of the main proxies used to
estimate tax avoidance in a variety of environments (Lisowsky, 2010; Cook, Moser & Omer,
2014; Lennox, C., P. Lisowsky, and J. Pittman, 2013; Chen, Chen, Cheng and Shevlin, 2010).
3.1.7
Chen, Chen, Cheng and Shevlin (2010)
Chen et al. (2010), noted that cash ETR’s reflect both permanent and temporary book-tax
differences, but that the temporary effects will reverse when estimated as an average or total
over a longer period of time. Therefore, there is no need to make adjustments for timing
differences. The single year estimate captures both permanent and temporary differences,
and the long-term rate captures mainly the permanent differences. They included the Special
Items in the Pre-Tax Profit as they were interested in the tax avoidance of the firm as a
whole. The formula the used is:
Page 68
Corporate tax avoidance
Submission 75
ETRit = TXPDt / PIt
This is the most simple of the cash ETR’s, being cash taxes paid as a proportion of the total
pre-tax profits. As with most recent tax research in accounting, Chen et al. also used the
GAAP ETR to confirm their results, and to estimate the effects of some accruals on the firm’s
ETR’s and tax liabilities. This proxy can be estimated using Australian data.
3.1.8
Conclusions
Global ETR’s proxy for a wide range of tax avoidance methods but cannot measure specific
methods of tax avoidance. Therefore, ETR’s have been primarily used in research studies
that define tax avoidance in the broad or general sense. In Australia, only the GAAP ETR,
and other ETR’s that use global income and tax paid metrics are able to be calculated due to
insufficient disclosures compared to the U.S. or those proposed by the OECD. This also
creates problems of interpreting global ETR’s. It must be noted that there are legitimate
business activities that reduce the global ETR, for example, REIT’s, R&D expenditures,
accelerated depreciation and capital allowances.
Neither GAAP ETR's, nor cash ETR's can guarantee that the tax expense, or tax paid, is
matched to the correct book income in the denominator. For example, tax paid on reassessments for tax audits in previous years will appear in the numerator but the equivalent
adjustment to the income will not be included in the denominator. Long-run ETR's can
overcome many of the mismatching problems, especially when using cash taxes paid.
Appropriate disclosures and a long-run perspective, allows for the calculation of more
specific ETR’s, such as domestic and overseas based, and more specifically on a country-by-
Page 69
Corporate tax avoidance
Submission 75
country basis. Additionally, specific ETR’s can narrow down the range of activities and
methods being employed by companies.
Page 70
Corporate tax avoidance
Submission 75
3.2
Book-Tax Gaps or Differences (BTG’s)
Book-tax gaps are broad measures similar to ETR’s, and are based on similar metrics. There
are also similar limitations on their use in Australia due to restricted disclosures. However,
they are useful to determine the magnitude of an individual company’s tax burden and effect
of potential tax avoidance activities.
One of the first references to BTG’s was by Plesko, Boynton and Boone (1996), when
commenting on the U.S. General Accounting Office’s estimated gap between tax receipts by
the IRS and the amount that could theoretically be collected based on aggregated income
data. Plesko (1999) went further and compared some of the most commonly used ETR’s to
the actual tax returns of firms in order to examine the level accuracy in the measurements of
the tax burden. Plesko (1999) concluded that the “use of financial statement information will
introduce substantial error into the measurement of taxable income.” (p.172) From analysing
the different measures of the tax liability, he concluded that the current tax expense
performed the most accurately. He also added that it is the marginal, not the average tax rate
that is important when examining a firm’s response to changes in tax policy.
3.2.1
Desai (2003)
Following from the insights in Manzon and Plesko (2002), Desai (2003) examined the
changes to the relationship between book and tax income during the 1990’s.
He also
calculated the BTG’s using the following formula:
BTGit = - (PIt - [TXDIt / STRt]) - ([TXTt - TXDIt] / STRt)
Page 71
Corporate tax avoidance
Submission 75
In order to alleviate some of the criticisms of the Manzon and Plesko (2002) approach, Desai
(2003) deducted the deferred tax expense, grossed up by the statutory tax rate, from both the
pre-tax income and from the total tax expense, in order to align current income with the
current tax expense. The use of the deferred tax liability is problematic in Australia, as it is
not used in a consistent manner. Not all companies report this amount and it is difficult to
reconcile with the companies’ reported tax expenses.
Therefore, this BTG cannot be
estimated with the same level of assurance as in the U.S.
Desai’s (2003) findings were more emphatic that Manzon and Plesko (2002). He found that
the identifiable factors that had previously been used to account for existence of the BTG
were now insufficient to account for the magnitude of the gap. He estimated that, in 1998,
less than half the gap (approximately $154.4 billion) cannot be accounted for by the
historically relevant measures, such as capital allowance, employee stock options and
accelerated depreciation.
3.2.2
Conclusion
The book-tax gaps use very similar metrics to the ETR’s but present the information in a
different format. They allow for the firm-level examination of the tax burden and the
marginal effect of tax changes. Global BTG’s proxy for a wide range of tax avoidance
methods but are likewise limited when attempting to identify and measure specific methods
of tax avoidance, and are used to examine tax avoidance generally. Furthermore, given the
Australian financial disclosure regime, only global BTG’s can be calculated which capture all
tax minimisation activities, including those that are actively encouraged by the tax system.
Therefore, global BTG’s are not useful when examining or attempting to identify specific tax
avoidance behaviours. Global BTG’s capture all tax minimisation activities and this can
Page 72
Corporate tax avoidance
Submission 75
create results that are difficult to interpret with respect to tax avoidance. As with ETR’s,
there are legitimate business activities that reduce the global BTG, for example, REIT’s,
R&D expenditures, accelerated depreciation and capital allowances. BTG’s are also a prerequisite for the regression-based techniques that are used to isolate particular tax
minimisation activities and behaviours.
3.3
Regression-based Measures
Regression-based measures attempt to isolate the most aggressive components of a
company’s tax strategies, from benign effects. They do this by using all know determinants
of BTG’s and finding out how much of the gap is caused by these. The remaining gap is
usually attributed to aggressive tax avoidance activities, with the level of aggressiveness
determined by the extent, number and types of determinants included in the model. In theory,
these types of measures should be able to isolate a particular activity by including other tax
avoidance measures.
However, this analysis is limited by the inaccuracy of the other
measures and can be seen as only adding noise to an already noisy measure.
3.3.1
Desai and Dharmapala (2006)
In 2006, Desai and Dharmapala examined whether a link existed between the increase in the
BTG and the growth of high powered incentives for managers. As the main source of the
legitimate differences between book and tax income is through accruals, Desai and
Dharmapala (2006) regressed total accruals on the BTG from Manzon and Plesko (2002) as
shown in the following formula:
Residual BTG = μi + εit from the regression,
Page 73
Corporate tax avoidance
Submission 75
BTGit = β0 + β1TACCit + μi + εit
The resultant error terms are regarded as the residual level of the BTG. This residual BTG is
their measure of tax avoidance and they assert that it captures the most aggressive tax
minimisation such as tax sheltering activities. They found that higher powered incentives
were associated with lower levels of tax sheltering, based on well-incentivised managers
being less likely to extract rents, or divert funds, at the expense of shareholders, especially in
well-governed firms. This model can be estimated using Australian data. However it suffers
from the same limitation due to the inclusion of BTG’s and a lack of data in Australia for the
deferred tax expense and the deferred tax liability.
The use of regression analysis introduces the additional problem of endogeneity. In this case,
the incentive compensation may be affected by the level of tax sheltering already taking place
in the firm. In their review of tax research in accounting, Hanlon and Heitzman (2010)
identify the lack of a good structural model for both BTG’s and ETR’s make models such as
Desai and Dharlmapala (2006) open to additional error. The problem with identifying the
“known” determinants of tax sheltering activity is that they might just be a by-product of a
separate, non-tax decision. Desai and Dharmapala (2006) acknowledge the short-comings of
their measure and to address the causality issues, would need a natural experiment to validate
their results. However, they interpret the residual book-tax gap as a more precise measure of
tax sheltering activity than previous BTG’s and ETR’s.
3.3.2
Frank, Lynch and Rego (2009)
Using a similar approach to Desai and Dharmaplala (2006), Frank, Lyinch and Rego (2009)
developed a measure based on the measures used to estimate the level of earnings
Page 74
Corporate tax avoidance
Submission 75
manipulation through accruals. They were examining the association between aggressive tax
reporting and aggressive financial reporting, using their measure to evaluate the effectiveness
of other current measures of tax avoidance in identifying tax shelter activity. The measure is
designated as the discretionary book-tax gap (or DTAX) and supposedly captures only those
tax aggressive activities that are at the discretion of management, such as accruals. Frank et
al.’s model uses the residual from regressing total permanent differences on nondiscretionary
items that are known to cause permanent differences (e.g., intangible assets) and other
statutory adjustments (e.g., state taxes) but are likely unrelated to tax reporting
aggressiveness. This is the tax avoidance equivalent of the performance-matched, modified
Jones (1991) model for estimating the level of earning management in the financial
statements of a firm. They estimate the permanent differences between book and tax income
with the following formula:
BTGit = {BIit – [(CFTEit + CFORit) / STRit]} – (DTEit / STRit)
This formula uses only current tax expense, both domestic and foreign, and deducts the
grossed up, deferred tax expense from pre-tax income. The determinants that are included
are intangible assets including goodwill due to the different tax and accounting treatments of
these assets, unconsolidated interests and minority interests that also have different
treatments, state-level income taxes that create non–discretionary differences by reducing
taxable income, changes in the net operating losses carried forward because of the
relationship to the valuation allowance, and the level of permanent differences from the
previous year, to control for non–discretionary permanent differences that persist through
time, such as municipal bond interest and tax credits. This model can also be estimated using
Page 75
Corporate tax avoidance
Submission 75
Australian data but has the same limitations due to the unreliable nature of the deferred tax
liability under Australian disclosure rules.
Using a logit regression, Frank, Lynch and Rego (2009) validated their measure against
known instances of tax sheltering activity and found that it was superior in identifying firms
engaging in these activities to the other measures they tested. These other measures included
Desai and Dharmapala’s (2006) residual BTG, Manzon and Plesko’s BTG, and the GAAP
ETR. Cash ETR’s and long run ETR’s were not included in the comparison. However,
DTAX has suffered from the same criticisms as Desai and Dharmapala’s (2006) residual
BTG, regarding as lack of a good structural model of BTG’s introducing additional errors to
an already “noisy” measure, the difficulty in identifying all known determinants and that it
does not capture conforming tax avoidance or deferral strategies (Hanlon and Heitzman,
2010)
The regression-based measures have not been taken up very extensively by tax researchers in
accounting because they are viewed as fairly “noisy” measures (Hanlon and Heitzman, 2010).
However, they have been included in the battery of measures that has become the normal
method of evaluating the level of tax avoidance. Researchers estimate a selection of different
measures and compare the results, sometimes to get the result that best fits their theoretical
hypotheses, but mostly to identify the different methods that have been used that might give
rise to anomalous results. Differences between certain measures allows for inferences as to
the methods being employed. However, as the different measures capture different aspects of
tax avoidance, as identified in the definitions and methods sections, some of the more
relevant results may be lost when using multiple measures.
Page 76
Corporate tax avoidance
Submission 75
3.4
Characteristics-based Measures
These measures are based on the characteristics of firms that are known to engage in some
tax specific avoidance activities. They are used to compliment the ETR and BTG measures
as they can estimate the probability that companies are engaging in some of the more
aggressive tax avoidance measures, such as transfer pricing and thin capitalisation. Some of
these measures, such as the predicted level of UTB’s and the probability of a company being
involved in tax shelter activities, were developed in the U.S. and cannot be estimated in
Australia due to lower disclosure standards regarding some of the information required to
calculate them. With greater disclosure and transparency, not only would Australian analysts
and researchers be able to estimate the U.S. measures, but other characteristics-based
measures could also be developed.
3.4.1
Transfer Pricing Index (TPRICE)
The transfer pricing index (TPRICE) was developed by Richardson, Taylor and Lanis (2013)
based on the characteristics of firms that were known to have engaged in abusive transfer
pricing, relative to a control group of other firms. The TPRICE is similar to algorithms used
by tax authorities in Australia, and elsewhere, to more effectively target their audit and
review work (Richardson et al., 2013). It uses a sum-score approach based on six specific
items that can be accessed from publicly available data, and can be used to rank firms
according to their propensity, or ability, to use transfer pricing mechanisms to transfer profits
to low taxing jurisdictions. The formula to calculate this measure is:
TPRICE
=
1.672
+
0.264*Size
+
0.079*Profit
+
0.095*Leverage
+
0.025*Intangibles + 0.071*Multinational + 0.006*Tax Haven
Page 77
Corporate tax avoidance
Submission 75
Where:
Size = Natural log of Total Assets;
Profit = Natural log of Net Profit
Leverage = long-term debt divided by total assets;
Intangibles = R&D expenditure divided by total assets;
Multinational = the number of foreign subsidiaries divided by the total number of
subsidiaries;
Tax Haven = a dummy variable of 1 if the firm has at least one subsidiary company
incorporated in an OECD (2006) listed tax haven, otherwise 0;
For each firm, a TPRICE score is calculated with the highest score representing the greatest
potential for aggressive transfer pricing. It does not indicate if a firm is actually abusing
transfer pricing mechanisms, but it does indicate the propensity, or opportunity, for firms to
engage in aggressive transfer pricing. This measure compliments broad measures of tax
avoidance such as global ETR’s, helps to explain the mechanisms that might cause low
ETR’s or high BTG’s, and vice versa. They also identify companies that require a more
extensive examination of their financial statements, especially those items that are only
disclosed through the notes to the statements, or disclosures to the stock market.
3.4.2
Maximum Allowable Debt (MAD) Ratio
The maximum allowable debt (MAD) ratio is not a direct proxy for tax avoidance but an
estimate of the thin capitalization within a company based on characteristics debt in a
Page 78
Corporate tax avoidance
Submission 75
company’s capital structure. Thin capitalisation refers to a corporate investment decision by
the firm to finance business operations primarily with debt capital rather than equity capital in
its capital structure. The procedure used to calculate the MAD ratio is outlined in Taylor and
Richardson (2013). Specifically, it calculates a firm's thin capitalization position using the
safe harbor test. This test involves calculation of a safe harbor debt amount (SHDA) using
the method outlined in ITAA 97. The SHDA is 75% of the average asset value of Australian
operations, net of non-interest bearing liabilities and investments in associates. Proxy
measures for each of the variables used to calculate the SHDA are employed as follows:
SHDA = (Average Total Assets – non-IBL) x 0.75
Where:
non-IBL = non-interest bearing liabilities.
The proxy measure of maximum allowable debt for tax purposes is then calculated as
follows:
Maximum Allowable Debt (MAD) Ratio = Average Debt / SHDA
Where:
Average Debt = total interest bearing liabilities (IBL)
Firms with a MAD ratio in excess of one are potentially non-compliant with the thin
capitalization rules contained in ITAA97 (ATO, 2005).
While firms are considered
compliant with the thin capitalization rules when they have a MAD ratio of less than one, a
firm may have a MAD ratio close to, or above the upper limit of one despite being entirely
Page 79
Corporate tax avoidance
Submission 75
compliant with the thin capitalisation rules, due to changes in business conditions, or the way
financial statement elements. The MAD ratio is the basis of a measure of thin capitalization
(TCAP) and, therefore, the ability of a company to engage in profit stripping to a low tax
jurisdiction. This measure does not identify illegal activities by companies but is used to
complement the more broad measures such as global ETR’s and BTG’s, and help explain
activities that produce results indicating a high level of tax avoidance.
3.5
Conclusions
In their review of tax research in accounting, Hanlon and Heitzman (2010) come to the
conclusion that not all measures are appropriate for all research questions. They also mention
the difficulty in constructing a widely accepted definition of tax avoidance. This lack of
definition creates problem with measuring tax avoidance and with interpreting the results in
general and with respect to specific tax avoidance activities.
One of the main problems with computing estimates of taxable income from financial
statements is due to a lack of disclosure about the sources of taxable income and the actual
taxes paid, or to be paid, from the current year's earnings. Attenuation of this problem can be
achieved by more detailed disclosures as outlined in the next section.
Page 80
Corporate tax avoidance
Submission 75
4. Recommendations and Conclusions
A tax system requires the balancing of two competing objectives. One is the distributional
role and the other is the efficiency of the system. As with most systems, a tax system that
relies on voluntary compliance is far more efficient than one that is enforced. A major factor
that determines the level of voluntary compliance is the level of trust in the tax authorities
and the taxation system (Kirchler et al., 2008). Trust in a system is increased with higher
levels of transparency.
In Scandinavian countries, tax returns are public records, and
voluntary tax compliance is very high despite the relatively high tax rates. Trust in a tax
system is also dependent on how equitably the system shares the tax burden (Shaviro, 2014),
and the procedural fairness with which the system is run (van Djike and Verboon, 2009). It is
these philosophies that have guided the Australian Tax Office’s engagement with corporate
taxpayers. The payoff for tax authorities is the unilateral decision by Rio Tinto Ltd to close
down most of its tax haven subsidiaries and be as open and transparent as possible about its
business dealings, including its tax.
The idea that closer public scrutiny lowers corporate tax rates has been evidenced by recent
research. A campaign in the U.K. that identified companies with subsidiaries in tax havens is
estimated to have raised the average GAAP ETR in the U.K. by three percentage points. In
the U.S., companies that have been identified as being highly tax aggressive have on average,
reported higher GAAP ETR’s in the periods following the reports. In 2010, Rio Tinto
voluntarily adopted a much higher degree of transparency in reporting on its operations. The
precise reasons for this change have not been divulged but it is presumed to involve gaining
greater levels of support from other stakeholders in the regions in which it operates. It
realises the importance of its reputation when negotiating new leases and operations. BHP
Page 81
Corporate tax avoidance
Submission 75
Billiton Ltd has announced that it will follow the Rio Tinto Ltd lead, and it is hoped that this
will put pressure on other companies to follow suit. Such disclosures have the potential to
create a competitive advantage for companies that are transparent and make full and
meaningful disclosures. Therefore additional tax disclosures, ceteris paribus, may create a
deterrent to tax avoidance if nothing else. More importantly however, the analysis above of
the definitions, methods and measures of tax avoidance indicates that there is some
disconnect between the measures with respect to the methods and definitions of tax
avoidance. It is our contention that as a result it has therefore been very difficult for any
interested party to conclusively assess the tax behaviour of any Australian company and very
easy for Australian companies that are suspected of avoiding their fair share of tax to defend
their position. To improve the balance between those assessing the tax behaviour of
Australian companies and the companies themselves we recommend three very specific and
practical disclosure policies, which can be implemented within the context of the financial
reporting regime in Australia. The greater and the more specific are the disclosure of profits
of and tax paid by Australian companies, with foreign operations in particular, in Australia
and overseas and preferably on a country by country basis. To some extent U.S. companies
are currently required to provide that information combined with a reconciliation between
effective and statutory tax rates in their financial reports. Our final recommendation, which a
requirement in the U.S., is for Australian companies to disclose unreported tax benefits.
From an academic point of view, there is never enough disclosure by companies. However,
in the case of tax reporting, for both the tax authorities and the financial statements, there is
little information available to determine the overall level of corporate tax avoidance, or to
identify likely causes of the cross-sectional variations in ETR’s and BTG’s. Many of the
Page 82
Corporate tax avoidance
Submission 75
proxies used for tax research in the U.S. use data that are not disclosed in Australia. Australia
does not even require profits and taxes to be separated into foreign and domestic sourced.
The higher level of disclosures in the U.S. is one of the main reasons that tax research in the
U.S. is greatly advanced on the state of tax research in Australia and the rest of the world.
However, even in the U.S. there are some non-disclosures that have the potential to lay bare
some of the most aggressive types of tax minimisation strategies, such as transfer pricing and
thin capitalisation. The disclosure of profits and tax by geographic segments has the potential
to achieve greater scrutiny of offshore intra-company transfers, and deter the abuse of the
system. Even the simple disclosure of the portion of profits and taxes that are sourced from
overseas would provide an enhanced understanding of the operation of corporate tax system
in Australia.
Currently, companies only have to report revenues by geographical source, and then, only if
they are of material importance to the financial statements.
The current geographical
breakdown is based on internal control and reporting mechanisms that already exist, in order
to minimise reporting and compliance costs. This has led to ad hoc use of definitions and
different classifications of geographic segments, making this financial statement information
hard to compare or analyse. However, while companies may not aggregate geographical
segment data, or store it in a format suitable for reporting, they do need to retain and have
ready access to this information. To ensure their compliance with tax laws, along with other
laws and regulations, companies require this financial information on a country-by-country
basis, for every country in which they operate. Therefore, this disclosure should not put a
large compliance burden on companies. Rio Tinto Ltd has continued to post persistent profits
despite the costs incurred in making these disclosures.
Page 83
Corporate tax avoidance
Submission 75
From a tax avoidance and transparency perspective, geographic segment reporting should at
least include total revenues, profits generated, and taxes paid in each country that the
corporation operates in. These additional disclosures will allow the computation and analysis
of the operations of companies in each geographical segment in which they operate. If
companies value their reputations, they will not be inclined to report 50% of their profits
being generated in tax havens, where they pay no tax, while 100% of their revenues are
sourced elsewhere. Corporations spend fortunes establishing their brands over long periods
of time, and are very sensitive to bad publicity. Tax avoidance, which used to be viewed as a
right and a virtue, is now having a moral label attached to it following comments by Richard
Goyder of Westfarmers, and Gerry Harvey from Harvey Norman (SMH, 2/10/2014).
There are legitimate business activities that companies can use that affect the global measures
of tax avoidance, such REIT’s, R&D expenditures, accelerated depreciation and capital
allowances. With the additional proposed geographical reporting disclosures combined with
a reconciliation of effective with statutory tax rates these legitimate activities can be
separately analysed and studies can focus on the most aggressive forms of tax avoidance,
such as transfer pricing and thin capitalisation.
Companies assume many positions with respect to their tax affairs, some based on a sound
foundation of tax law and facts, while other tax positions are riskier. Whether a tax position
taken by a company is legal or not, can depend on the future outcome of judicial deliberations
or negotiations with tax authorities.
Companies are not required to make disclosures
regarding these positions in their financial statements and they are known as unreported tax
benefits (UTB’s). In 2008, the U.S. Financial Accounting Standards Board introduced a
requirement that all firms report the likelihood of being able to sustain these positions
Page 84
Corporate tax avoidance
Submission 75
through a tax audit, including judicial review. This disclosure has given U.S. investors,
analysts and tax researchers a powerful tool to examine both the aggressiveness and
magnitude of tax positions that have been adopted by companies. New measures of tax
avoidance have already been derived from these disclosures, including a model that can
predict the probability that a company’s tax position will be denied. A similar disclosure
should be required in Australia, as corporations taking aggressive tax positions can have a
detrimental effect on shareholder value. The disclosure of UTB’s provides an estimate of the
level of risk encapsulated in a company’s tax strategies. In addition to results from ETR’s
and BTG’s, the disclosure of UTB’s provides a complimentary and powerful tool to
determine the likelihood that a company with low ETR’s or high BTG’s is engaging in
aggressive tax planning. Therefore, UTB’s can overcome some of the limitations associated
with interpreting ETR’s and BTG’s, especially those based on global metrics, and will enable
the calculation for characteristics-based measures, such as the predicted UTB measure used
in the U.S.
In response to a September 2014 report by the Tax Justice Network, that identified and
quantified cases of suspected tax avoidance in Australia, there was a flurry of denials by
companies named in the report, justifying their tax payments and strategies. One factor that
can legitimately reduce corporate tax liabilities is the ability of companies to deduct up to
150% of the value of complying research and development (R&D) expenditures. Some
companies are suggesting that this can explain a five to eight percentage point drop in their
ETR’s. It is impossible to verify such claims as companies are not required to disclose their
R&D expenditures, unless it is material to their financial position or performance. However,
an item with a tax effect of more than two percentage points should probably be regarded as a
Page 85
Corporate tax avoidance
Submission 75
material item. More clarity and disclosure around items with special tax treatment, such as
R&D expenditures and capital allowances, should be a reciprocal part of these generous
concessions. In order to claim these concessions, companies already have to keep detailed
records, so the recordkeeping and disclosure costs should not impose onerous costs on
businesses. The inclusion of these financial disclosures allows analysts and other researchers
to account for the effect of legitimate tax minimisation sschemes, providing a more accurate
analysis and valuation of a company’s operating and investing strategies. Disclosure of these
activities will lead to better measures, such as adjusted ETR’s and BTG’s, and additional
complimentary, characteristics-based measures, such as predicted UTB’s and the probability
of shelter activity, that were developed in the U.S.
The methods used by corporations to avoid tax are endless in variety and creativity.
However, as with bribery and corruption, the opening up of these schemes and arrangements
to the light of public scrutiny is the most effective method to bring them to an end. The mood
of the general population of taxpayers toward corporate tax avoidance, which was once
tolerated, seems to be changing.
Ethical and sustainable business is the future of
corporations, and this does not entail secrecy surrounding tax disclosures and activities.
Page 86
Corporate tax avoidance
Submission 75
5. References
Alchian, A.A. and Kessel, R.A., 1962, ‘Competition, monopoly and the pursuit of money’,
Aspects of Labor Economics, Vol.14, pp. 157-83.
Allingham, M.G. and Sandmo, A., 1972, ‘Income Tax Evasion: A Theoretical Analysis’,
Journal of Public Economics, Vol.1, pp. 323-338.
Avi-Yonah, R., 2006, ‘Corporate Social Responsibility and Strategic Tax Behavior’, Law and
Economics Working Papers, University of Michigan Law School, pp. 1-28.
Avi-Yonah, R., 2009, ‘Taxation, Corporate Social Responsibility, and the Business
Enterprise’, Comparative Research in Law and Political Economy, Vol. 5, No. 3, pp. 117.
Balakrishnan, K., Blouin, J. and Guay, W., 2012, ‘Does Tax Aggressiveness Reduce
Corporate Transparency?’, University of Pennsylvania Working Paper.
Bankman, J., 1999, ‘The Tax Shelter Problem’, National Tax Journal, Vol. LVII, No. 4, pp.
925-936.
Blumenthal, M., Christian, C., & Slemrod, J., 2001, ‘Do normative appeals affect tax
compliance? Evidence from a controlled experiment in Minnesota.’, National Tax
Journal, Vol. 54, pp. 125-38.
Borek, C.T, Frattarelli, A. and Hart, O., 2014, Harvard University Law School, Working
Paper.
Page 87
Corporate tax avoidance
Submission 75
Braithwaite, J., 2005, Markets in Vice, Markets in Virtue, New York: Oxford University
Press.
Chen, S., Chen, X., Cheng, Q. and Shevlin, T., 2010, ‘Are family firms more tax aggressive
than non-family firms?’, Journal of Financial Economics, Vol. 95, pp. 41–61.
Chen, K-P. and Chu, C.Y.C., 2005, Internal control versus external manipulation: a model of
corporate income tax evasion’, The RAND Journal of Economics, Vol. 36, No. 1, PP.
151-164.
Chi, S.S., Pincus, M. and Teoh, S.H., 2014, ‘Mispricing of Book-Tax Differences and the
Trading Behavior of Short Sellers and Insiders’, The Accounting Review, Vol. 89, No.
2, pp. 511-543.
Christensen, J. and Murphy, R., 2004, ‘The Social Irresponsibility of Corporate Tax
Avoidance: Taking CSR to the bottom line’, Development, Vol. 47, No. 3, pp. 37-44.
Chyz, 2013, ‘Personally tax aggressive executives and corporate tax sheltering’, Journal of
Accounting and Economics, Vol. 56, pp. 311–328.
Citizens for Tax Justice, 2014, ‘Inverting Corporations Should Be Required to Pay Taxes
Owed on Profits Held Offshore’, www.ctj.org, 11 Sep 2014.
Clausing, K., 2014, ‘Corporate Inversions’, Tax Policy Centre, Urban Institute and Brookings
Institution, 20 Aug 2014.
Crocker, K.J. and Slemrod, J., 2005, ‘Corporate tax evasion with agency costs’, Journal of
Public Economics, Vol. 89, pp. 1593–1610.
Page 88
Corporate tax avoidance
Submission 75
Desai, M. A., 2003, ‘The Divergence between Book Income and Tax Income’, in Poterba,
J.M., Ed., Tax Policy and the Economy, Volume 17, MIT Press, Cambridge MA.
Desai, M.A. and Dharmapala, D., 2009, ‘Corporate Tax Avoidance and Firm Value’, The
Review of Economics and Statistics, Vol. 91, No. 3, pp. 537–546.
Donohoe, M.P. and Knechel, W.R., 2014, ‘Does Corporate Tax Aggressiveness Influence
Audit Pricing?’, Contemporary Accounting Research, Vol. 31, No. 1, pp. 284-308.
Dunbar, A., Higgins, D.M., Phillips, J.D. and Plesko, G.A., 2010, ‘What do measures of tax
aggressiveness measure?’, Conference Paper, 103rd Annual Conference on Taxation,
National Tax Association.
Dyreng, S. D., Hanlon, M. and Maydew, E.L., 2008, ‘Long-Run Corporate Tax Avoidance’,
The Accounting Review, Vol. 83, No. 1, pp. 66-82.
Dyreng, S. D., and Lindsey, B.P., 2009, ‘Using Financial Accounting Data to Examine the
Effect of Foreign Operations Located in Tax Havens and Other Countries on U.S.
Multinational Firms' Tax Rates’, Journal of Accounting Research, Vol. 47, No. 5, pp.
1283-1316.
Dyreng, S. D., Lindsey, B.P. and Thornock, J.R., 2013, ‘Exploring the role Delaware plays as
a domestic tax haven’, Journal of Financial Economics, Vol. 108, No. 3, pp. 751-772.
Economic Policy Institute, 2014, ‘Myths and Facts About Corporate Taxes, Part 3: Are
American Companies’ Profits Trapped Overseas?’, http://www.epi.org/blog/myths-andfacts-about-corporate-taxes-part-3, accessed 5 Feb 2015.
Page 89
Corporate tax avoidance
Submission 75
Edwards, A., Kravet, T.D. and Wilson, R.J., 2014, ‘Trapped cash and the profitability of
foreign acquisitions’, Rotman School of Management Working Paper, (1983292).
Erickson, M., Hanlon, M. and Maydew, E.L., 2004, ‘How much will firms pay for earnings
that do not exist? Evidence of taxes paid on allegedly fraudulent earnings’, The
Accounting Review, Vol. 79, No. 2, pp. 387-408.
Financial Accountability & Corporate Transparency Coalition, 2014, Comments on The
Department of Treasury’s Notice 2014 - 52 ,accessed at http://thefactcoalition.org/wpcontent/uploads/2014/12/Treasury-Comment-from-FACT-FINAL.pdf, on 5 Feb 2015.
Frank, M.M., Lynch, L.J. and Rego, S.O., 2009, ‘Tax Reporting Aggressiveness and Its
Relation to Aggressive Financial Reporting’, The Accounting Review, Vol. 84, No. 2,
pp. 467–496.
Freedman, J., Loomer, G. and Vella, J., 2009, ‘Corporate Tax Risk and Tax Avoidance: New
Approaches‘, British Tax Review, Vol. 1, pp. 74-116.
Frey, B.S., 1997, ‘A Constitution for Knaves Crowds Out Civic Virtues’, The Economic
Journal, Vol. 107, No. 443, pp. 1043–1053.
Frischmann, Shevlin, T. and Wilson, R., 2008, ‘Economic consequences of increasing the
conformity in accounting for uncertain tax benefits’, Journal of Accounting and
Economics, Vol. 46, pp. 261–278.
Fullerton, D., 1984, ‘Which effective tax rate?’, National Tax Journal, Vol. 37, pp. 23-41.
Page 90
Corporate tax avoidance
Submission 75
Fullerton, D. and Henderson, Y.K., 1983, ‘Long Run Effects of the Accelerated Cost
Recovery System’, National Bureau of Economic Research Working Paper No. 828,
Cambridge, Mass.
Gallemore, J., Maydew, E.L. and Thornock, J.R., 2014, ‘The Reputational Costs of Tax
Avoidance’, Contemporary Accounting Research, Vol. 31, No. 4, pp. 1103-1133.
Gergen, M.P., 2002, ‘The Logic of Deterrence: Corporate Tax Shelters’, 55 Tax L. Rev. 255
2001-2002
Graham, J.R. and Tucker, A.L ., 2006, ‘Tax shelters and corporate debt policy’, Journal of
Financial Economics, Vol. 81, pp. 563–594.
Greenberg, J., 1984, ‘Avoiding Tax Avoidance: A (Repeated) Game-Theoretic Approach’,
Journal of Economic Theory, Vol. 32, pp. 1-13.
Hanlon, M. and Heitzman, S., 2010, ‘A review of tax research’, Journal of Accounting and
Economics, Vol. 50, pp. 127-178.
Hanlon, M. and Slemrod, J., 2009,’ What does tax aggressiveness signal? Evidence from
stock price reactions to news about tax shelter involvement’, Journal of Public
Economics, Vol. 93, Iss. 1–2, pp. 126–141.
Inger, K.K., 2014, ‘Relative Valuation of Alternative Methods of Tax Avoidance’, The
Journal of the American Taxation Association, Vol. 36, No. 1, pp. 27-55.
Page 91
Corporate tax avoidance
Submission 75
Jennings, R., Weaver, C.D. and Mayhew, W.J., 2012, ‘The Extent of Implicit Taxes at the
Corporate Level and the Effect of TRA86’, Contemporary Accounting Research, Vol.
29, No. 4, pp. 1021-1059.
Jensen, M.C. and Meckling, W.H., 1976, ‘Theory of the Firm: Managerial Behavior, Agency
Costs and Ownership Structure’, Journal of Financial Economics, Vol. 3, pp. 305-360.
Kirchler, E., Hoelzl, E. and Wahl, I., 2008, ‘Enforced versus voluntary tax compliance: The
“slippery slope” framework’, Journal of Economic Psychology, Vol. 29, No. 2, pp. 210225.
Landolf, U., 2006, ‘Tax and corporate responsibility’, International Tax Review, Vol. 29, pp.
6-9.
Landsberger, M. and Meilijson, I., 1982, ‘Incentive generating state dependent penalty
system: the case of income tax evasion’, Journal of Public Economics, Vol. 19, pp.
333-352.
Lanis, R. and Richardson, G., 2013, ‘Corporate social responsibility and tax aggressiveness: a
test of legitimacy theory’, Accounting, Auditing & Accountability Journal, Vol. 26, No.
1, pp. 75-100.
Lennox, C., Lisowsky, P. and Pittman, J., 2013, ‘Tax Aggressiveness and Accounting
Fraud‘, Journal of Accounting Research, Vol.51, No. 4, pp. 739–778,
Lisowsky, P., 2010, ‘Seeking Shelter: Empirically Modeling Tax Shelters Using Financial
Statement Information’, The Accounting Review, Vol. 85, No. 5, pp. 1693-1720.
Page 92
Corporate tax avoidance
Submission 75
Lisowsky, P., Robinson, L. and Schmidt, A., 2013, ‘Do Publicly Disclosed Tax Reserves Tell
Us About Privately Disclosed Tax Shelter Activity?’, Journal of Accounting Research,
Vol. 51, No. 3, pp. 583-629.
Lopez, T.J., Regier, P.R. and Lee, T., 1998, ‘Identifying Tax-Induced Earnings Management
Around TRA 86 as a Function of Prior Tax-Aggressive Behavior’, The Journal of the
American Taxation Association, Vol. 20, No. 2, pp. 37-56.
McGuire, S.T., Omer, T.C. and Wilde, J.H., 2014, ‘Investment Opportunity Sets, Operating
Uncertainty, and Capital Market Pressure: Determinants of Investments in Tax Shelter
Activities?’, The Journal of the American Taxation Association, Vol. 36, No. 1, pp. 126.
Murphy, T., 2000, ‘Part IVA: From here to … ’, Journal of Australian Taxation, May/June
2000, pp.198-207.
OECD, 2008, Study into the Role of Tax Intermediaries, ISBN-978-92-64-04179-0
OECD, 2015, ‘Follow-Up Work on BEPS Action 6: Prevent Treaty Abuse’, Comments
Received on Public Discussion Draft, 12 Jan 2015.
Manzon, G.B. and Plesko, G.A., 2002, ‘The Relation Between Financial and Tax Reporting
Measures of Income’, Tax Law Review, Vol. 55, pp. 175-214.
Plesko, G.A., 2003, ‘An evaluation of alternative measures of corporate tax rates’, Journal of
Accounting and Economics, Vol. 35, No. 2, pp. 201-226.
Page 93
Corporate tax avoidance
Submission 75
Porcano, T., 1986, ‘Corporate tax rates: Progressive, proportional, or regressive’, Journal of
the American Taxation Association, Vol. 7, No. 2, pp. 17-31.
Rego, S.O. and Wilson, R., 2012, ‘Equity Risk Incentives and Corporate Tax
Aggressiveness’, Journal of Accounting Research, Vol. 50, No. 3, pp. 775-809.
Richardson, G., Taylor, G. and Lanis, R., 2013, ‘The impact of board of director oversight
characteristics on corporate tax aggressiveness: An empirical analysis’, Journal of
Accounting and Public Policy, Vol. 32, pp. 68–88
Saavedra, D., 2013, ‘Analysis of Unsuccessful Tax Avoiders’, Working Paper, Massachusetts
Institute of Technology, Dec 2013.
Salamon, L.M. and Siegfried, J.J., 1977, ‘Economic power and political influence: The
impact of industry structure on public policy’, American Political Science Review, Vol.
71, No. 3, pp. 1026-1043.
Scholz, J. T., and Lubell, M. (1998), ‘Trust and taxpaying: Testing the heuristic approach to
collective action’, American Journal of Political Science, Vol. 42, No. 2, pp. 398-417.
Shackelford, D.A., and Shevlin, T., 2001, ‘Empirical tax research in accounting’, Journal of
Accounting and Economics, Vol. 31, No. 1, pp. 321-387.
Shackelford, D.A., Slemrod, J. and Sallee, J.M., 2007, ‘A unifying model of how the tax
system and generally accepted accounting principles affect corporate behavior’, (No.
w12873). National Bureau of Economic Research.
Page 94
Corporate tax avoidance
Submission 75
Shaviro, D.N., 2001, ‘Evaluating the Social Costs of Corporate Tax Shelters’, 55 Tax L. Rev.
445 (2001).
Shevlin, T., 1987, ‘Taxes and off-balance-sheet financing: research and development limited
partnerships’, The Accounting Review, pp. 480-509.
Siegfried, J.J., 1974, ‘Effective average US corporation income tax rates’, National Tax
Journal, pp. 245-259.
Sikka, P., 2010, ‘Smoke and mirrors: Corporate social responsibility and tax avoidance’,
Accounting Forum, Vol. 34, No. 3–4, pp. 153–168.
Slemrod, J., 2007, ‘Cheating Ourselves: The Economics of Tax Evasion’, Journal of
Economic Perspectives, Vol. 21, No. 1, pp. 25–48.
Stickney, C.P., and McGee, V.E., 1982, ‘Effective corporate tax rates the effect of size,
capital intensity, leverage, and other factors’, Journal of Accounting and Public Policy,
Vol. 1, No. 2, pp. 125-152.
Taylor, G. and Richardson, G., 2013, ‘The determinants of thinly capitalized tax avoidance
structures: Evidence from Australian firms’, Journal of International Accounting,
Auditing and Taxation, Vol. 22, pp. 12– 25.
Taylor, G., & Tower, G., 2009, ‘Impact of Adoption of IFRS on the Thinly Capitalised
Position of Australian Companies’, eJournal of Tax Research, Vol. 7, No. 1.
Page 95
Corporate tax avoidance
Submission 75
Taylor, G., Tower, G. and Van Der Zahn, M., 2011, ‘The influence of international taxation
structures on corporate financial disclosure patterns’, Accounting Forum, Vol. 35, No.
1, pp. 32–46.
U.S. Congress, Joint Committee on Taxation, 1999, ‘Description and Analysis of PresentLaw Tax Rules and Recent Proposals Relating to Corporate Tax Shelters’, Appendix II
to JCX-82-99.
U.S Senate, Permanent Subcommittee on Investigations, 2013, ‘Offshore Profit Shifting and
the U.S. Tax Code Part 2 (Apple Inc.)’, Exhibits for Hearing, Washington, DC.
U.S. Treasury, Office of Tax Policy, 2000, ‘The Deferral of Income Earned Through U.S.
Controlled Foreign Corporations: A Policy Study’, Department of Treasury, Dec 2000.
van Dijke, M. and Verboon, P., 2010, ‘Trust in authorities as a boundary condition to
procedural fairness effects on tax compliance’, Journal of Economic Psychology, Vol.
31, No. 1, pp. 80-91.
Weisbach, D.A., 2002, ‘Ten Truths About Tax Shelters’, 55 Tax L. Rev. 215
Weiss, R.D., 1979, ‘Effective Corporation Income Tax Rates’, National Tax Journal, Vol. 32,
pp. 380-390.
Williams, D.F., 2007, ‘Tax and Corporate Social Responsibility’, Discussion Paper, KPMG’s
Tax Business School.
Wilson, R., 2009, ‘An Examination of Corporate Tax Shelter Participants’, The Accounting
Review, Vol. 84, No.3, pp. 969-999.
Page 96
Corporate tax avoidance
Submission 75
Zimmerman, J.L., 1983, ‘Taxes and firm size’, Journal of Accounting and Economics, Vol. 5,
pp. 119-149.
Page 97