%2multibyte Version: 5.50.0.2960 CodePage: 1252
\documentclass[12pt]{article}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\usepackage{amssymb}
\usepackage{amsmath}
\usepackage{harvard}
\setcounter{MaxMatrixCols}{10}
%TCIDATA{OutputFilter=LATEX.DLL}
%TCIDATA{Version=5.50.0.2960}
%TCIDATA{Codepage=1252}
%TCIDATA{}
%TCIDATA{BibliographyScheme=Manual}
%TCIDATA{Created=Tuesday, October 02, 2012 10:19:32}
%TCIDATA{LastRevised=Wednesday, December 18, 2013 17:46:58}
%TCIDATA{}
%TCIDATA{}
%TCIDATA{Language=American English}
%TCIDATA{CSTFile=LaTeX article (bright).cst}
\newtheorem{theorem}{Theorem}
\newtheorem{acknowledgement}[theorem]{Acknowledgement}
\newtheorem{algorithm}[theorem]{Algorithm}
\newtheorem{axiom}[theorem]{Axiom}
\newtheorem{case}[theorem]{Case}
\newtheorem{claim}[theorem]{Claim}
\newtheorem{conclusion}[theorem]{Conclusion}
\newtheorem{condition}[theorem]{Condition}
\newtheorem{conjecture}[theorem]{Conjecture}
\newtheorem{corollary}[theorem]{Corollary}
\newtheorem{criterion}[theorem]{Criterion}
\newtheorem{definition}[theorem]{Definition}
\newtheorem{example}[theorem]{Example}
\newtheorem{exercise}[theorem]{Exercise}
\newtheorem{lemma}[theorem]{Lemma}
\newtheorem{notation}[theorem]{Notation}
\newtheorem{problem}[theorem]{Problem}
\newtheorem{proposition}[theorem]{Proposition}
\newtheorem{remark}[theorem]{Remark}
\newtheorem{solution}[theorem]{Solution}
\newtheorem{summary}[theorem]{Summary}
\newenvironment{proof}[1][Proof]{\noindent\textbf{#1.} }{\ \rule{0.5em}{0.5em}}
\let\latexcite=\cite
\input{tcilatex}
\begin{document}
\title{{\LARGE From tax evasion to tax planning\thanks{%
We acknowledge useful comments to participants of the Canadian Economics
Association Conference, Montreal, 2013. The usual disclaimer applies.}}}
\author{Arnaud Bourgain\thanks{%
Corresponding author: arnaud.bourgain@uni.lu}, Patrice Pieretti, Skerdilajda
Zanaj \\
%EndAName
CREA, University of Luxembourg}
\date{July 9, 2013}
\maketitle
\begin{abstract}
This paper analyzes within a simple model how a removal of bank secrecy can
impact tax revenues and banks'profitability,\ assuming that offshore centers
are able to offer sophisticated but legal, tax planning. Two alternative
regimes are considered. A first, in which there is strict bank secrecy and a
second, where there is international information exchange for tax purposes.
In particular, we show that sharing tax information with onshore countries
can be a dominant strategy for an OFC\ if there is enough scope for
providing tax planning. Moreover, a partial reduction of tax liabilities can
already prompt OFCs to voluntarily exchange relevant tax information. We
also discuss the conditions under which the possible removal of bank secrecy
may reduce or increase the onshore country's tax revenue.
KEY words: offshore centers, tax planning, tax evasion
JEL classifications: F21, H26, H87
\end{abstract}
\section{Introduction}
Since the mid-1990s, there have been numerous attacks against bank secrecy
and opaque financial structures that have been accused of extensive tax
evasion (OECD 1998, FSF 2000). In the wake of the financial crisis of 2008,
an anti-evasion action was re-launched by the G-20 that urges approximately
100 OECD and non-OECD countries (including most Offshore Financial Centers -
OFCs) to sign "Tax Information Exchange Agreements" (TIEAs).\footnote{%
More than 700 agreements were signed on August 10, 2011 (OECD 2011) between
OFCs and OECD countries. These agreements require jurisdictions to exchange
information on request without restrictions due to bank secrecy or domestic
tax interest requirements.
\par
\bigskip } Actions were also taken unilaterally by US authorities after they
accused Swiss banks (UBS and Credit Suisse Group) of aiding tax evasion
schemes. With the Foreign Account Tax Compliance Act (FATCA) of March 2010
and an initial agreement signed between the US and Switzerland in February
2013, events are speeding up\footnote{%
The FATCA agreement requires that foreign institutions report information
about financial accounts held by US taxpayers directly to the US tax
authorities.}. Since April 2013, five European governments have promoted
FATCA models as "the new international standard" and have envisaged their
implementation as the basis for a multilateral exchange of information.
These actions have prompted some sizeable OFCs (e.g., Luxembourg, BVI,
Bermudas) to announce their willingness to introduce an automatic
information exchange. Will these actions be the end of OFCs' business? Will
this approach substantially mitigate tax leakages experienced by countries
pressing for more tax transparency? Which economic forces encourage OFCs to
accept the abrogation of bank secrecy on non-resident deposits?
\emph{These questions are particulary relevant since in the current period
of strong pressure against tax evasion in OFCs, we observe an impressive
developement of sophisticated tax planning strategies. Holding and other
special entities are more and more used by rich individual investors fot tax
mitigation. }
\emph{Quotation Tax justicve network here?}
\emph{This trend is in line with the mesaure of the absence of significant
repatriation of funds despite the recent OFCs crackdown (Johannesen and
Zucman 2014)}
The aim of our paper is to address an issue that has attracted little
attention in the current debate. If onshore countries' foremost aim is to
increase their tax revenue, it is legitimate to question whether the removal
of bank secrecy can achieve this aim given that there exist alternative,
albeit sophisticated, ways to mitigate the tax burden. To this end, we
develop a simple model to analyze how removing bank secrecy may affect tax
revenues and banks' profitability, assuming that offshore centers are able
to offer sophisticated but not illegal tax planning. The focus is on a small
OFC that is able to attract investors located in the rest of the world. Two
regimes are considered. In the first scenario, we assume that the OFC enjoys
strict bank secrecy and is thus attractive to tax evaders. However, the
onshore economy is able to pressure the tax haven by shaming and blaming it.
This pressure damages its reputation in the hope of limiting tax evasion. In
the second case, we consider a scenario in which the OFC agrees to share
information with the onshore economy for tax purposes. Sophisticated tax
planning appears as an alternative to tax evasion. By accepting the
information exchange, the reputation of OFCs remains sound, and the onshore
countries are deprived of one weapon in the fight against tax leakages
caused by legal offshoring practices. In the following, we adopt a positive
approach rather than a normative one by assuming that the competing
jurisdictions are self-interested. The OFC is motivated by maximizing its
banks' profits, whereas the onshore government is concerned with maximizing
its net tax revenue. \footnote{%
This is a suitable way to characterize the objective of onshore governments
given the current economic situation, as noted by Nicod\`{e}me (2009).}
The main results obtained in this paper can be summarized as follows. We
first show that sharing tax information with onshore countries can be a
dominant strategy for an OFC if there is enough scope for providing tax
planning to its clientele without infringing tax laws. It follows that the
willingness of an OFC to provide tax information does not necessarily lead
to its closure. Furthermore, we show that partial tax saving resulting from
offshore investments may prompt OFCs to voluntarily exchange relevant tax
information. The required share of tax mitigation leading to this
cooperative behavior depends on exogenous parameters, such as the level of
international financial integration. Finally, we highlight a surprising
result. We show that the removal of bank secrecy may, under some conditions,
reduce the onshore country's tax revenue.
Our paper is related to the recent literature on tax avoidance through tax
havens, although these contributions are mainly focused on tax planning
strategies implemented by MNEs. In this context, Slemrod and Wilson (2009)
demonstrate the parasitic effect of tax havens on other countries' welfare.
In their setting, tax havens waste resources by providing tax evasion
services to firms, and tax administrations incur expenditures to limit tax
evasion. Other authors (Hong and Smart, 2010; Desai, 2006), however,
highlight the beneficial effects of tax havens. For example, by reducing
their tax burdens through the use of offshore entities, MNEs may enhance
their activity in non-tax havens. Johannesen (2010) analyzes the effect of
tax havens on low and high tax jurisdictions within a framework of imperfect
competition. In particular, he shows that an equilibrium may arise in which
the tax rate of the low tax country is increased, whereas the tax bases of
non-tax havens decrease. Elsayyad and Konrad (2012) analyze how fighting tax
havens, particularly by imposing tax information exchange agreements,
modifies the competition pattern between tax havens. If initiatives are
adopted in a sequential way, they show that the resulting exit of some tax
havens increases the market concentration among the remaining tax havens,
which become more profitable and more resistant to complying; thus, actions
taken against tax havens may be welfare reducing for the OECD. Our paper
also examines the possible adverse effects caused by the reaction of tax
havens to initiatives against them. However, contrary to Elsayyad and Konrad
(2012), our paper focuses on changes within a tax haven that are more or
less able to substitute (legal) tax planning activities for traditional tax
evasion services rather than on changes in competition between tax havens.
The offshore business does not necessarily disappear with official
compliance with information exchange agreements, but its nature may change.
Therefore, in contrast to Johannesen (2010) and Elsayyad and Konrad (2012),
we do not focus on competition between different tax havens.
The work of Bacchetta and Espinoza (2000) bears some resemblance to our
paper. These authors derive general conditions under which two possibly
asymmetric countries may add a bilateral information-exchange clause in a
tax treaty. Analogously, our model attempts to identify conditions under
which an OFC is (unilaterally) willing to transmit tax information to a
high-tax country. Their model is based on bilateral exchange and
necessitates repeated games to make information exchange sustainable. In our
paper, however, the willingness of the OFC to cooperate can be reached in a
one-shot game because giving up bank secrecy can, under specific conditions,
make legal tax planning more profitable than pure tax evasion.
The paper is organized as follows. \emph{Section 2 presents the shift away
from tax evasion with bank secrecy to sophisticated tax planning.}\ In
Section 3, we model how an OFC competes with the onshore world by providing
tax benefits. Within this context, we successively consider strict bank
secrecy and information sharing. Section 4 discusses the conditions under
which the OFC may voluntarily agree to give up bank secrecy. Then, we
analyze how the possible decision of the OFC to comply with international
tax exchange rules can affect the onshore tax revenue. Section 5 concludes.
\section{\protect\bigskip Towards tax planning strategies without bank
secrecy}
In a recent report (2009), the OECD expresses concern about High Net Worth
Individuals (HNWIs) posing particular challenges to tax administrations
because of the complexity of their affairs, the amounts of tax revenue that
are at stake, and, especially, the opportunity to undertake aggressive tax
planning. Ten years ago, noting that the attractiveness of Switzerland for
tax evasion was declining, Geiger and H\"{u}rzeler (2003) explained that
Swiss banks were adapting to this development by improving, among other
abilities, expertise in international tax and estate planning instruments.
The Economist notes \emph{\ }that Liechtenstein "is one of the more
proactive" to rethink its strategy. "It signs as many tax treaties as
possible [...] and offering a tempting array of vehicles, including trusts".%
\footnote{%
"Leaky devils. Tax havens start to reassess their business models", The
Economist, April 13, 2013.}
Interestingly, bank secrecy seems to be inessential for tax mitigation. The
financial press abounds with statements by bankers, tax lawyers, and tax
advisers praising the various sophisticated structures that legally lower
tax bills in taxpayers' home countries without depending on bank secrecy.
According to Kenneth Rubinstein (2010), a New Yorker lawyer, "secrecy has no
place in proper tax planning"; "rather than falling for promises of secrecy
from unscrupulous marketers, investors should seek guidance from qualified
tax counsel and ensure that their international assets are structured in a
tax-compliant manner". Individual investors are also inclined to use\ more
and more sophisticated methods that were initially created for the tax
planning of Multi-National Enterprises (MNEs). \emph{\ }A legal means of
international tax planning for individual investors can be achieved by using
special entities in international financial centers. In this case, we are
confronted with tax planning rather than tax evasion, although the dividing
line is not always clear (Slemrod and Yitzhaki, 2001). According to the OECD
(2009, p.26), "Wealthy investors are often highly mobile and may be
attracted to countries perceived as offering a favorable taxation
environment. This may include such factors as no or low capital gains tax on
the disposition of privately held assets and the presence of a good
treaty-network. Rather than changing their tax residence, wealthy investors
may hold investments through no or nominally taxed offshore entities with a
view to mitigate tax on foreign source income or gains".
Among the various international tax planning strategies, the holding
corporation is an interesting vehicle used by individuals and MNEs (Gravelle
2009, Mc Cann 2006, Mintz and Weichenrieder 2010). For example, Schmidt and
Lady (2007) explain in detail how US HNWIs use holding companies and the
corresponding tax rules. In particular, OFCs have designed holding companies
to exploit Double Tax Treaties (DTTs) \footnote{%
DTTs are agreements between two states (since the 1920s) designed to protect
investors against the risk of double taxation. DTT networks have increased
in parallel with the development of foreign direct investments.}\emph{\ }and
the EU Directive "Parent-Subsidiaries"(1990)\footnote{%
Modified in 2003.} that have encouraged extensive treaty shopping (Avi-Yonah
and Panayi 2010). Interestingly, there has recently been a proliferation of
DTTs. Rawling (2007) argues that the recent initiatives of the OECD, the EU,
and the IMF for information exchange agreements have encouraged bilateral
DTTs as part of or separate from TIEAs, and many OFCs have concluded DTTs
that they did not previously have. Indeed, more than 3000 bilateral tax
treaties connecting approximately 180 countries are in force today (Rixen
2010, Rixen 2011). The OECD treaty model represents the general consensus on
international taxation, but the rules have become more sophisticated and
complex over time (Rixen 2010), creating loopholes that are exploited by
low-tax jurisdictions. The EU Directive "Parent-Subsidiaries" intends to
eliminate tax obstacles to profit distribution between groups of companies
in the EU by abolishing the withholding of taxes on dividends between
subsidiaries and parents.\footnote{%
This double taxation represented discrimination compared with a situation in
which the two entities were located in the same country.} To benefit from
these exemptions, the parent and subsidiaries must be fully taxable, and a
minimum of permanent shareholding is required. Special entities called
"conduit companies" that exploit DTTs and the Parent-Subsidiaries Directive
have seen exponential growth during the last decade, reflecting the
development of tax planning. The number of Luxembourg financial companies,
called SOPARFI \footnote{%
SOci\'{e}t\'{e} de PArticipation FInanci\`{e}re}(see Appendix), increased
from 2800 in 2000 to 55 000 in 2011. Dutch financial holding companies were
evaluated at 42 072 in 2007, and similar or comparable legal vehicles exist
in Cyprus, Malta, Switzerland, and Caribbean OFCs (often called
International Business Companies).
\section{\protect\bigskip Model setting}
Consider an OFC located in country $F$ competing with a financial center
located in country $H$. We assume that banks located in a jurisdiction
compete in unison with banks located in another country.\footnote{%
The aim is to neglect competition among banks within a joint location to
focus exclusively on the interaction between banking centers. Consequently,
only one representative bank for each jurisdiction will be considered.} The
OFC attracts investors living in country $H$, whereas the reverse does not
occur. Investors who reside in the OFC are supposed to keep their capital in
the OFC. This assumption emphasizes that OFCs generally have very low
populations and offer intermediation services predominantly to large foreign
(onshore) economies. Investors of the onshore country $H$ are heterogeneous
and uniformly distributed, with unit density on the interval $[0,1]$
according to their attachment to home, indexed by $x$. The closer an
individual is to the origin, the more she is attached to her home country.
Each individual is endowed with one unit of capital she can invest in the
home banking system or in the OFC.
An individual of type \emph{\ }$x\in \left[ 0,1\right] $\emph{\ }who
offshores her money incurs a cost equal to a moving cost\emph{\ }$k>0$\emph{%
\ }times\emph{\ }$x$. The coefficient $k$\emph{\ }can be viewed as a measure
of the degree of international financial integration.\emph{\ }The government
of country\emph{\ }$H$\emph{\ }taxes capital according to the home country
principle, whereas the OFC does not impose a tax on capital. To analyze how
the OFC and the onshore economy compete for investors, we consider two
alternative regimes: strict bank secrecy and information sharing. Under
strict bank secrecy, the OFC does not share any type of information about
its clientele with country $H$, whereas under the second scenario,
information sharing takes place. Importantly, in the second regime, legal
tax planning can emerge as an alternative to tax evasion, as detailed in
Section $\left( \ref{Info}\right) $.\emph{\ }In each regime, investors can
invest their money either at home or in the OFC. If they opt for the second
choice, two asset types are available: one in which sophisticated tax
mitigation strategies are not needed and one that involves tax planning
techniques. Note that we only focus on the tax motive when investors are
supposed to offshore their money. Consequently, the ordinary asset will
dominate the sophisticated asset in the case of pure bank secrecy, whereas
tax planning will dominate the ordinary asset when there is automatic
information sharing for tax purposes.
The onshore region and the OFC interact at different levels. First,
investors from the onshore economy are enticed to avoid taxes by offshoring
their money. The ensuing tax loss induces the onshore economy to react by
setting an appropriate tax rate and deciding on actions to damage the OFC's
reputation. Second, there is competition between the banks of both
countries. In this context, we assume that the interest rates offered to
investors result endogenously from a non-cooperative game between both
banking systems. The consequence is that these rates reflect the relative
tax attractiveness of the OFC, which, in turn, depends on the possible
existence of bank secrecy and the ease with which tax planning opportunities
can be provided. \footnote{%
That interest rates reflect international taxation conditions is consistent
with English and Shahin (1994), who find that, following the passage in the
late 1980s of two laws that effectively removed banking secrecy for cases of
insider trading and money laundering, Swiss banks raised deposit rates by 53
and 105 basis points, respectively. According to Besson (2004, p.64), Swiss
banks can afford to charge higher-than-average fees by virtue of their
high-end image, their reputation for financial strength, and, finally,
banking secrecy. In addition to fees, banking secrecy affects deposits rates.%
} The onshore government is assumed to maximize its tax revenues while
anticipating (i) how investors react to differences in taxes and to
reputational costs when investing in the OFC and (ii) how banks compete to
attract investors.
\subsection{Strict bank secrecy}
Under this regime, the OFC provides strict bank secrecy to its investors
while the onshore center does not. Offshore tax dodgers are thus supposed to
be perfectly sheltered from their home tax administration. This perfect
opacity allows tax evasion by investing in non sophisticated assets like
riskless deposits.\footnote{%
Investments in risky financial assets could be considered. Because the focus
of the paper is exclusively on the tax aspect of offshore investments we
shall avoid this complication.} Because strict bank secrecy makes tax
evasion difficult to detect, authorities of onshore countries try\ to make
OFCs less attractive for tax evasion by actions\footnote{%
Recently, the fight against tax evasion has become a major priority in
developed countries and pressure (like blacklisting or blame and shame
campaigns) against tax havens has increased. These actions are intended to
entice the OFC to exchange tax information.} intended to disparage their
reputation. Consequently, the tax evaders will endure a premium that
increases with the loss of reputation (Picard and Pieretti, 2011) of the
non-cooperative tax haven in which they invest.\footnote{%
Indeed, Sharman (2001 p.12) observes that, "investors tend to avoid or leave
jurisdictions with bad reputations not only out of concern that their money
will be misappropriated, but also because firms risk harming their own
reputations, as reflected in their share prices."
\par
{}}
Two options are then available to the $H$ country's resident. Either the
investor puts her money in the home bank where she incurs a tax or she
evades taxes by investing in the OFC. One unit of wealth invested at home by
an individual of type $x\in\left[ 0,1\right] $ yields
\begin{equation*}
V_{H}=r_{H}-t,
\end{equation*}
where $r_{H}$ is the rate of return and $t$ the tax rate. If the same
individual invests in the OFC, she avoids the home tax but has to incur in
addition of a reputation cost, a moving cost which reflects her attachment
to home and the ease with which money can be transferred abroad.
It follows that one unit of wealth invested by the individual of type $x$ in
the offshore financial center yields a return $r_{F}$ diminished by the
mobility cost $kx$ and a reputation harm $\alpha$. The corresponding
indirect utility is given by%
\begin{equation*}
V_{F}=r_{F}-kx-\alpha
\end{equation*}
Given the utility of the different options, the individual of type $x\in%
\left[ 0,1\right] $ chooses to invest in the country which offers the
highest net return. It follows that the individuals of type $x\in\left[ 0,%
\overline{x}\right) $ where
\begin{equation*}
\overline{x}=\frac{r_{F}-r_{H}+t-\alpha}{k}
\end{equation*}
opt for tax evasion and those of type $x\in\left[ \overline{x},1\right] $
decide to invest at home. As a result, the investment supply to the home
banking place equals to $D_{F}=\overline{x}$ and the supply to the tax haven
equals $D_{H}=1-\overline{x}$.
\subparagraph{International banking competition}
The banking systems of countries $H$\ and $F$\ raise funds from investors
and offer respectively the interest rates $r_{H}$ and $r_{F}$. The collected
funds by the banks are invested into risk-free assets that yield a given
(world) rate of return $r$. The banks' profit functions in the countries $H$%
\ and $F$\ are given as follows%
\begin{equation*}
\Pi_{H}=(r-r_{H})D_{H}\text{ \ and \ }\Pi_{F}=(r-r_{F})D_{F}.
\end{equation*}
Each banking center selects its return rate supposing that the rate of its
rival is given. The equilibrium rates are
\begin{equation*}
r_{H}=r-\frac{2k-t+\alpha}{3}\text{ \ \ and \ }r_{F}=r-\frac{k+t-\alpha}{3}.
\end{equation*}
The corresponding deposits supplied at home and abroad are respectively
\begin{equation*}
D_{H}=\frac{2k-t+\alpha}{3k}\text{ \ and\emph{\ \ }}D_{F}=\frac{k+t-\alpha }{%
3k}.
\end{equation*}
\subparagraph{Onshore government decision}
We suppose that the investor's reputation harm is a policy variable that
depends on the pressure that the onshore jurisdiction is able to exert place
on the OFC, for example, by blacklisting or campaigning about the risks of
tax evasion. We further consider that the cost of exerting pressure is given
by a quadratic function $C(\alpha)=\alpha^{2}/2$. The convexity of the cost
function may reflect the increasing difficulty to exert pressure which can
in particular rely on the existence of institutional limitations. For
example, a successful crackdown on bank secrecy requires collective action
of onshore jurisdictions which can be hard to sustain or can lead to
unacceptable infringement on the sovereignty of other states. Finally, we
assume that policy makers of the home country maximize their net tax income $%
T_{H}^{E}(t,\alpha)=tD_{H}-C(\alpha)$\ with respect to the tax rate $t$\ and
the pressure variable $\alpha$. This way to characterize the onshore
government's objective is consistent with the current period of global
crisis forcing countries to fix their fiscal imbalances (Nicod\`{e}me,
2009). Solving the maximization problem yields the equilibrium values
\begin{equation}
\alpha^{\ast}=\frac{2k}{6k-1}\text{ } \label{alpha}
\end{equation}
and
\begin{equation}
\text{\ }t^{\ast}=k+\frac{\alpha^{\ast}}{2}=\frac{6k^{2}}{6k-1}
\label{tstar}
\end{equation}
with $t^{\ast}<1\Leftrightarrow1/51/3,$ whereas a higher $k$ induces lower taxation if $%
k<1/3.$ In the first case, the direct effect dominates, while the indirect
effect dominates in the second case. The two forces equalize at $k=1/3.$
\bigskip
Using $\left( \ref{alpha}\right) $ and (\ref{tstar}), the equilibrium
deposit supplies become
\begin{equation*}
D_{H}^{\ast}=\frac{2k}{6k-1}\text{ and \ }D_{F}^{\ast}=\frac{4k-1}{6k-1}.
\end{equation*}
The supplies are positively signed if $k>1/4$. It follows that if $k\leq1/4$
there is no tax evasion ($D_{F}=0$). The reason is that capital mobility in
this interval encourages the onshore government to pressure the
uncooperative tax haven to such an extent that investors apprehend tax
evasion because of very high reputation costs. The equilibrium interest
rates are%
\begin{equation*}
r_{H}^{\ast}=r-\frac{2k^{2}}{6k-1}\text{ and }r_{F}^{\ast}=r-k\frac {4k-1}{%
6k-1}.
\end{equation*}
Because the OFC offers tax shelter, one could expect that the interest rate
is always higher in the onshore banking center. This is however not always
the case. Indeed,\ we have $r_{H}^{\ast}r_{F}^{\ast},$\ if $k>1/2$.\ When financial integration is high
, i.e. $k<1/2,$\ international pressure $\alpha$\ is high to counteract tax
evasion and thus investors become more reluctant to offshore their money.
This prompts the onshore bank to offer a lower interest rate than its
offshore rival.
The corresponding equilibrium profits are given by
\begin{equation*}
\Pi_{H}^{\ast}=\frac{4k^{3}}{\left( 6k-1\right) ^{2}}\text{\ \ and \ }%
\Pi_{F}^{\ast}=\frac{k\left( 4k-1\right) ^{2}}{\left( 6k-1\right) ^{2}}
\end{equation*}
The net tax income $B_{H}=T_{H}-\alpha^{2}/2$ of the onshore country is
\begin{equation*}
B_{H}^{\ast}=\frac{2k^{2}}{6k-1}
\end{equation*}
It is straightforward to show that $B_{H}^{\ast}=\frac{1}{3}\ t^{\ast}$.
This implies that at equilibrium, the net tax revenue of the onshore country
$B_{H}^{\ast}$\ increases with the mobility cost if $k>1/3$\ but decreases
with $k$\ if $\frac{1}{4}0$.
\subparagraph{The onshore government' decision}
The onshore government chooses the tax rate $t$ that maximizes
\begin{equation*}
T_{H}=t\left[ \widehat{D}_{H}+D_{F}^{l}+\left( 1-b\right) D_{F}^{s}\right]
\end{equation*}
This yields the equilibrium tax rate%
\begin{equation}
\bar{t}=\frac{\theta}{b}-\frac{\gamma}{2} \label{tbar}
\end{equation}
It follows that the equilibrium tax rate decreases with $b$\ and $\gamma $.
The intuition is straightforward. If tax planning becomes more attractive ($%
b $\ increases) and/or the rate of return of sophisticated assets improves
following an increase in $\gamma $, the onshore administration will react by
reducing the tax burden to reduce capital outflows. \ It also appears that
the tax rate increases with $\theta $. The reason is that a tougher anti tax
haven policy (increased $\theta $) which increases the risk associated with
tax planning raises the onshore jurisdiction's power to tax. Note also that
the condition $1>\bar{t}>0$\ is satisfied if $\ \theta \in \left( b\frac{%
\gamma }{2},b\frac{\gamma }{2}+b\right) $. The equilibrium return rates
offered to the investors in their in the OFC and in their home country are
\begin{equation*}
\text{ }\bar{r}_{F}^{s}=R-\left( \frac{\theta}{2}+\frac{k}{3}+\frac{1}{4}%
b\gamma\right) ,\;\bar{r}_{F}^{l}=r-\frac{k}{3}\;\text{and}\ \ \bar{r}_{H}=r-%
\frac{2k}{3}
\end{equation*}
We assume that $r$\ is high enough so that the margins of the different
types of investments are always positive. Surprisingly the interest rate
offered by the offshore center decreases with the risk parameter $\theta.$
As highlighted above, an increase in $\theta$ raises the tax rate $t$\ which
makes tax planning more attractive. This finally explains why the offshore
financial center is able to lower its interest rate. The equilibrium
investments choices are%
\begin{equation*}
\overline{D}_{F}^{s}=\frac{1}{3},\;\overline{D}_{F}^{l}=0\;\text{and\emph{\ }%
}\emph{\ \ }\overline{D}_{H}=\frac{2}{3}.
\end{equation*}
The banks' equilibrium profits are
\begin{equation*}
\overline{\Pi}_{F}=\left( R-\bar{r}_{F}^{s}\right) \overline{D}%
_{F}^{s}=\allowbreak\frac{4k+6\theta+3b\gamma}{36}\text{ and\ \ }\overline{%
\Pi }_{H}=\frac{4k}{9}.
\end{equation*}
The equilibrium tax income of the onshore economy equals%
\begin{equation*}
\overline{T}_{H}=\frac{\left( 2\theta-b\gamma\right) \left( 3-b\right) }{6b}
\end{equation*}
Notice the tax revenue in the onshore country augments with $\theta$. It
follows that making tax planning more risky improves the onshore country's
tax revenue.$\allowbreak$
\section{Removing bank secrecy}
In the following we analyze the incentives of an OFC to voluntarily\textbf{\
}disclose tax information. Then we focus on how the removal of bank secrecy
impacts the tax revenue in the onshore country.
\subsection{Incentives to provide tax information}
In this section, we consider the conditions under which the offshore center
has an incentive to abandon bank secrecy and to comply with international
information sharing rules. We assume that the OFC chooses the regime that
entails the highest bank profitability. For this reason, we focus on the
difference $\overline{\Pi }_{F}-\Pi _{F}^{\ast }$\ which equals
\begin{equation*}
\Psi(b,k,\theta)=\allowbreak\frac{1}{12}\gamma b+\left( \frac{1}{9}k+\frac {1%
}{6}\theta-\Pi_{F}^{\ast}\right)
\end{equation*}
It follows that $\Psi(b,k,\theta)>0$ if $b>\widehat{b}$ where $\widehat{b}=%
\frac{12}{\gamma}\Pi_{F}^{\ast}-\frac{2}{3\gamma}\left( 2k+3\theta\right) .$
We can state the following proposition.
\begin{proposition}
If the share of tax benefit $b$ is high enough ($b>\widehat{b}$), the OFC
has an incentive to accept information sharing for tax purpose only.
\end{proposition}
A direct implication of this proposition is that the willingness of an OFC
to provide tax information does not necessarily lead to its closure. This is
because the OFC may be able to set up legal structures designed to provide
tax planning as an alternative to illegal tax evasion. However, the
threshold beyond which the OFC agrees to comply with tax information
regulations is not immutable. In particular, it changes with the risk
parameter $\theta $, which is particularly influenced by anti-avoidance tax
rules designed by onshore governments and k, the level of international
financial integration.
First, we see that the threshold value $\widehat{b}$ decreases with $\theta $%
. In other words, it is more likely that the OFC will cooperate by
exchanging tax information when tax planning becomes riskier. This is a
surprising result because one would expect that riskier tax planning would
decrease the willingness of the OFC to give up bank secrecy. This issue
becomes clearer, however, if we bear in mind that an increase in $\theta \ $%
makes investors more reluctant to offshore their money, which, in turn,
gives the onshore government more power to tax and the onshore banks more
liberty to lower interest rates. As a result, the OFC also has an incentive
to lower the return it offers to tax planners if the elasticity of the
investment supply it faces is sufficiently low, as is the case in our model%
\footnote{%
Remember that in our model the equilibrium demand for tax planning is
independent of the offered return.}. Finally, tax planning becomes more
profitable for offshore banks, and the OFC is thus more inclined to opt for
information sharing.
\bigskip
\textbf{Corollary:} \textit{The riskier the tax planning, the likelier the
offshore center will agree to share tax information.}
\bigskip
How does increasing capital mobility affect the OFC's decision to abandon
bank secrecy? Because increasing financial openness (lower $k$) decreases
banks' profits in each scenario, it follows that the total effect of $k$ on
the difference $\overline{\Pi }_{F}-\Pi _{F}^{\ast }$\ and, hence, on the
threshold value\ $\widehat{b}$\ is ambiguous. Indeed, when $k>\overline{k}$\
we have $\frac{\partial \widehat{b}}{\partial k}>0,$\ and $\frac{\partial
\widehat{b}}{\partial k}<0$ otherwise. This means that decreasing mobility
costs (higher mobility) make the OFC more willing to abandon bank secrecy
when mobility costs are high ( $k>\overline{k}$).\textbf{\ }However, when
mobility costs are low ($k<\overline{k}$), we have the opposite effect. In
the first case ( $k>\overline{k}$), more intense international competition
forces banks to increase their interest rates, which damages the
profitability of the offshore banks relatively more when they provide strict
bank secrecy. However, when capital mobility is high ($k<\overline{k}$), the
onshore jurisdiction exerts considerable effort to damage the OFC's
reputation. As we saw above, this raises the onshore tax rate significantly.
Accordingly, the OFC can maintain its attractiveness without significantly
increasing its interest rate. Consequently, higher capital mobility damages
its profitability comparatively less in the case of bank secrecy. The OFC is
therefore less willing to lift its bank secrecy.
\subsection{Does information sharing always improve onshore tax revenue ?}
Before analysing how a possible removal of bank secrecy affects the tax
revenue of the onshore country, it is interesting to highlight some
surprising results. Our model shows that the regime of information sharing
compared with strict bank secrecy may be consistent with higher offshore
investments ($\overline{D}_{F}^{s}$\ $>$\ $D_{F}^{\ast }$\ ) if capital
mobility is high enough ($k<1/3$), and with a lower tax rate for appropriate
parameter values.\footnote{%
Indeed, direct comparison (\ref{tbar}) and (\ref{tstar}) shows that the
difference $t^{\ast }-\bar{t}=\allowbreak 6\frac{k^{2}}{6k-1}-\bar{t}$ has
two complex roots if $\ \frac{2\theta -b\gamma }{b}<\frac{4}{3}$.
\par
$\bigskip $%
\par
{}} Hence, switching to information sharing can entail ambiguous effects on
the onshore country's budget. Let us now analyze in more detail the
highlighted effects. Toward this end, we define the following$\,\ \Phi
(b,k,\theta )=\overline{T}_{H}-B_{H}^{\ast }$\ , which equals%
\begin{equation*}
\Phi (b,k,\theta )=\frac{\left( 2\theta -b\gamma \right) \left( 3-b\right) }{%
6b}-B_{H}^{\ast }=\allowbreak \frac{1}{6b}\left( \gamma b^{2}-\left(
6B_{H}^{\ast }+2\theta +3\gamma \right) b+6\theta \right) .
\end{equation*}
Solving the equality $\Phi (b,k,\theta )=0$\ for $b$, we show in Appendix 2
that there is only one real root denoted by $\overline{b}$ that satisfies
the condition $\overline{b}\in \left[ 0,1\right] $. It follows that $\Phi
(b,k,\theta )>0$\ for $b\in \left[ 0,\overline{b}\right] $. In other words,
if the tax benefit measured by $b$\ exceeds the threshold $\overline{b}$, we
obtain $\Phi (b,k,\theta )<0$.
This leads us to the following proposition
\begin{proposition}
Information sharing increases onshore tax revenues if and only if the tax
benefit resulting from tax planning is not too high ($b<\overline{b}$).
\end{proposition}
The above analysis shows that the effectiveness of tax planning for reducing
tax liabilities is crucial in gauging the success of the regime of
information exchange. In particular, if tax havens are sufficiently skilled
at setting up legal structures for tax optimization purposes, pushing for
the removal of bank secrecy with the aim of maximizing tax income may be
inconsistent. This aspect has been neglected in the current scientific and
political debate. However, the threshold value $\overline{b}$ can be
increased if the onshore world is able to augment the risk parameter $\theta
$,\ because $\frac{\partial \overline{b}}{\partial \theta }>0$. This can
occur if onshore governments are able to improve\ international tax
legislation to deter the use of increasingly sophisticated forms of tax
avoidance.
It is also interesting to observe how financial integration modifies the
threshold $\overline{b}$\ beyond which the regime of information sharing is
not desirable for a tax-maximizing country. To this end, we calculate the
derivative of $\overline{b}$\ with respect to $k$. This yields $\frac{%
\partial \overline{b}}{\partial k}>0$\ if $\frac{1}{4}\frac{1}{3}$. In other
words, when mobility costs are "low" ($\frac{1}{4}\frac{1}{3}$), we observe the opposite effect. How can we explain this
result ? Because $\overline{T}_{H}$\ does not depend on $k$, changes in
capital mobility only impact tax revenue in the bank secrecy regime. In
Section 2.1., we saw that a change in mobility costs has an ambiguous effect
on the tax rate and thus on the tax revenue of the onshore county. When
capital mobility is high ($\frac{1}{4}\frac{1}{3}$). Consequently,
increased capital mobility makes the information exchange regime less
attractive from the point of view of tax revenue in the first case and more
attractive in the second case. That the onshore country may be worse off in
terms of tax revenue is initially surprising. However, it is important to
keep in mind that by agreeing to exchange information, the OFC is no longer
exposed to international pressures aimed at damaging its reputation.
Accordingly, the onshore country lacks an important weapon in the fight
against tax leakages caused by legal offshoring practices. As noted by
Rawlings (2007, p.58), "Through complying with these initiatives OFC states
have reinscribed their reputation and political soundness in the eyes of
investors and have become jurisdictions characterized by `good governance'
meeting the highest international standards [....] These multilateral
initiatives have had the reverse effect of what they originally intended:
through allowing OFCs to demonstrate their good governance to the world they
maintain their client base and sustain an ongoing fiscal competition between
states for tax revenues".
\section{Conclusion}
We observe that although there is a crackdown on bank secrecy that favors
tax evasion, tax planning using sophisticated legal structures continues to
prosper. Most of these techniques, which were initially designed for
multinational corporations, are generally based on a network of
international tax treaties. Thus, they allow wealth management, which
utilizes large groups of tax consultants, to mitigate tax liabilities
without bank secrecy.
Our paper models competition between OFCs and onshore banking centers by
considering, successively, strict bank secrecy with pure tax evasion and tax
information exchange with tax planning. Two interesting results are
highlighted. First, an OFC can voluntarily abandon strict bank secrecy
without closing down its activity. This decision hinges on the ability of
the OFC to offer tax benefits by legal but sophisticated means. The model
defines a parameter to capture this ability that must exceed a given level.
In particular, this threshold depends on the legal uncertainty resulting
from efforts deployed by onshore countries and/or international bodies to
refine and implement anti-avoidance regulations. Surprisingly, the model
shows that under conditions of high capital mobility, these attempts can
make OFCs less prone to abandon strict bank secrecy. Second, it may occur
that tax revenue earned by onshore countries shrinks with the removal of
strict bank secrecy if OFCs are able to sufficiently and legally reduce the
tax liabilities of their customers. This threshold can be increased, but not
necessarily eliminated, by strengthening international regulations. This
surprising result indicates a possible inconsistency in the effort of
onshore countries pushing tax havens to exchange relevant tax information
for exclusively fiscal goals.
It follows that solely targeting the abolition of strict bank secrecy may
not be enough. Combating bank secrecy policy is a complex task that requires
a clear understanding of modern OFCs and their capacity to adapt their tax
minimization strategies to ever-changing regulatory environments. However,
this ability to implement increasingly sophisticated tax planning devices is
not shared equally. Thus, there is scope for vertical differentiation among
the various OFCs with the eventual possibility that some of them will have
to close. Our model does not account for this extension, which should be
addressed in a future work. Furthermore, the objectives that guide onshore
countries in their fight against tax havens should be clarified. Are tax
equity considerations the only intentions pursued? Or, as the facts suggest,
do onshore countries primarily want to increase their tax revenue?
Addressing this second objective is likely to impact important strategic
variables such as taxes and, in turn, to induce the types of results
highlighted in our paper.
\section*{Appendix 1: Example of international tax planning}
Here, we show how tax planning is made possible without the existence of
bank secrecy. To this end, we illustrate (see Figure below) how tax
structuring can be achieved within the Luxembourg financial holding company
SOPARFI (Soci\'{e}t\'{e} de participation financi\`{e}re). Similar or
comparable legal vehicles exist in Cyprus, Switzerland, the Netherlands, the
UK, Malta, the Netherlands Antilles, and Barbados. The SOPARFI is basically
a taxable Luxembourg company and is thus eligible to benefit from DTTs and
the EU Parent-Subsidiaries directive. The main activity of a SOPARFI is to
acquire and hold shares of other companies.
\FRAME{fhF}{5.4302in}{3.1393in}{0pt}{}{}{Figure}{\special{language
"Scientific Word";type "GRAPHIC";maintain-aspect-ratio TRUE;display
"USEDEF";valid_file "T";width 5.4302in;height 3.1393in;depth
0pt;original-width 5.3731in;original-height 3.0943in;cropleft "0";croptop
"1";cropright "1";cropbottom "0";tempfilename
'MY0GY300.wmf';tempfile-properties "PR";}}
Let us envisage a simplified example of tax planning for an individual
investor. An individual investor founds a SOPARFI in Luxembourg that
acquires corporate shares owned by the founding investor in an onshore
country (France, for example). The corporate income of the French company is
taxed at a normal rate (34.4\%, including social contributions). If the
investor perceives dividends directly in France, he will be subject to a
withholding tax on dividends (31.5\%, including social contributions). In
our case, the dividend is perceived by the Luxembourg SOPARFI, which
benefits from an extensive network of double-taxation treaties and from the
EU Parent-Subsidiary directive. Consequently, the dividends must be paid in
Luxembourg, where the normal withholding tax rate equals 15\%.
Further tax mitigation options are available in Luxembourg. For example, the
dividends can be invested in a "Special Investment Fund" (SIF) dedicated to
professionals and other well-informed investors. The SIF is entitled to
full-tax exemption, except that it must pay a small subscription fee and an
annual charge of 0.01\% on all its assets. Another possibility for a SOPARFI
to avoid withholding taxation is to transfer the dividend income to a
financial company in another low-tax jurisdiction. In this case, the
Luxembourg tax authorities require that the foreign tax rate be at least
11\%. Furthermore, the SOPARFI can avoid the withholding tax by indebting
itself to another financial company owned by the founder of the SOPARFI.
This avoidance scheme is facilitated by the fact that a high debt-equity
ratio (85/15) is generally acceptable to the Luxembourg tax authorities for
a shareholding activity. It follows that high debt contracting is allowed,
and the resulting interest paid is tax deductible without being subject to a
withholding tax. Finally, the financial profits of the SOPARFI can be
invested in real estate or used for various purchases in another country or
in the origin country.
\section*{Appendix 2}
Solving $\Phi(b,k,\theta)=0$ with respect to $b$ yields the real roots of $\
\Phi(b,k,\theta)=\allowbreak\frac{1}{6b}\left( \gamma b^{2}-\left(
6B_{H}^{\ast}+2\theta+3\gamma\right) b+6\theta\right) =0$ are\
\begin{align*}
\overline{b} & =\frac{1}{2\gamma}\left( 6B+2\theta+3\gamma-\sqrt {%
24B\theta+36B\gamma+36B^{2}+\left( 2\theta-3\gamma\right) ^{2}}\right) \ \ \
\ \ \\
\ \ \overline{\overline{b}} & =\frac{1}{2\gamma}\left( 6B+2\theta +3\gamma+%
\sqrt{24B\theta+36B\gamma+36B^{2}+\left( 2\theta-3\gamma\right) ^{2}}\right)
\end{align*}
$\allowbreak$
It is easy to check that $\overline{\overline{b}}>1$ since $B_{H}^{\ast}$
and $\gamma$ are strictly positive. It also appears that $\overline{b}>0$ .
In addition, we have $\overline{b}\leq1$ if $B_{H}^{\ast}\geq\allowbreak%
\left( \frac{2\theta-\gamma}{3}\right) $, which is always the case since we
assume that $\theta\in\left( b\frac{\gamma}{2},b\frac{\gamma}{2}+b\right) $.
It follows that $\overline{b}$ is the only admissible root of $\ \Phi
(b,k,\theta)=0$ satisfying the condition $\overline{b}\in\left[ 0,1\right] $%
. As a consequence, $\ \Phi(b,k,\theta)>0$ if $b<\overline{b}$ and $%
\Phi(b,k,\theta)<0$ otherwise.
\section*{References}
\begin{enumerate}
\item Avi-Yonah R., Panayi Ch. HJI (2010), Rethinking treaty-shopping
lessons for the European Union, University of Michigan Law School, Working
Paper n%
%TCIMACRO{\U{b0} }%
%BeginExpansion
${{}^\circ}$
%EndExpansion
10-002.
\item Bacchetta Ph., Espinoza M.P. (2000),\ Exchange-of-Information Clauses
in International Tax Treaties, International Tax and Public Finance, 7,
275--293.
\item Besson S. (2004), Le Secret Bancaire, Presses Polytechniques et
Universitaires Romandes, Lausanne.
\item Desai M.A., Foley C.F., Hines J.R. (2006), The demand for tax haven
operations. Journal of Public Economics, 90, 513-531.
\item Duff D.G. (2009), Tax avoidance in the 21st century, in Evans C. and
Krever R. (eds) Australian tax reform in retrospect and prospect, Thomson
Reuters Australia Limited, Sydney, 477-501.
\item Duff D.G. (2010) "Responses to tax treaty shopping:\ A comparative
evaluation", UBC Faculty of Law, Vancouver.
\item English M., Shahin W. (1994), Investigating the Interest Rate Impact
of Changing Secret Bank Deposit Laws: Switzerland, Journal of Banking and
Finance 18, 461-475.
\item Elsayyad M. Konrad K.A. (2012), Fighting multiple tax havens, Journal
of International Economics, Vol. 86 (2), 295-305.
\item Evers M., de Graaf A. (2009), Limiting Benefit Shopping: Use and Abuse
of EC Law, EC Tax Review, Vol. 18 (6), 279-298.
\item Financial Stability Forum, FSF. (2000), Report of the Working Group on
Offshore Financial Centres, Report, 5 April.
\item Geiger and H\"{u}rzeler (2003), The transformation of the Swiss
private banking market, Journal of Financial Transformation, Vol. 9, 93-103.
\item Gravelle J.G. (2009), Tax Havens: International Tax Avoidance and
Evasion, National Tax Journal, Vol. LXII, n%
%TCIMACRO{\U{b0} }%
%BeginExpansion
${{}^\circ}$
%EndExpansion
4, p. 727-753.
\item Hampton M.P. Christensen J. (2002), Offshore Pariahs? Small Island
Economies, Tax Havens, and the Re-configuration of Global Finance, World
Development, 30 (9), 1657--1673.
\item Johannesen N. (2010), Imperfect Tax Competition for Profits,
Asymmetric Equilibria and Beneficial Tax. Havens, Journal of International
Economics, 81, 253-264
\item Johannesen N., Zucman G. (2014), The End of Bank Secrecy? An
Evaluation of the G20 Tax Haven Crackdown, American Economic Journal,
forthcoming.
\item Kessler W., Eicke R. (2007), Germany: Treaty shop until you drop, Tax
Notes International, Vol 46 (4), p.377.
\item Hong Q., Smart M. (2010),\ In praise of Tax Havens: International Tax
Planning and Foreign Direct. Investment, European Economic Review, 54(1),
82-95.
\item Kudrle R. T. (2008), The OECD's Harmful tax competition initiative and
the tax havens: From bombshell to damp squib, Global Economy Journal, Vol. 8
(1), 1-23.
\item McLaren J., Passant J., (2010),Tax havens: Do they have a future
providing banking and financial services?, Canberra Law Review 1.\
\item McCann H. (2006), Offshore finance, Cambridge University Press.
\item Mintz J.M., Weichenrieder A. (2010), Indirect Side of Direct
Investment. Multinational company finance and taxation, CESifo Book Series,
MIT Press.
\item Nicod\`{e}me G. (2009), On recent development in fighting harmful tax
practices,\ National Tax Journal, Vol LXII (4), 775-771.
\item OECD (1998), Harmful Tax Competition: An Emerging Global Issue.
\item OECD (2004), Statement on transparency and exchange of information for
tax purposes.
\item OECD (2008), Study into the Role of Tax Intermediaries.
\item OECD (2010), Promoting transparency and exchange of information for
tax purposes, 3 September.
\item OECD (2011), Global Forum on Transparency and Exchange of Information
for Tax Purposes: Information Brief, 14 September.
\item Picard P., Pieretti P. (2011), Bank secrecy, illicit money and
offshore financial centers, Journal of Public Economics, 95, 942-955.
\item Rawlings G. (2005) Mobile people, mobile capital and tax neutrality:
Sustaining a market for offshore finance centres, Accounting Forum 29,
289-310.
\item Rawlings G. (2007), Taxes and Transnational Treaties: Responsive
Regulation and the Reassertion of Offshore Sovereignty, Law and Policy, 29
(1), 51--66.
\item Rixen Th. (2010), Bilateralism or multilateralism? The political
economy of avoiding international double taxation, European Journal of
International Relations, 16 (4), p. 589-614.
\item Rixen Th. (2011), From double tax avoidance to tax competition:
Explaining the institutional trajectory of international tax governance,
Review of International Political Economy, 18 (2), 197-227.
\item Rubinstein K. (2010), Secrecy Has No Place in Proper Tax Planning,
Family Wealth Report, New York, 8 March.
\item Schmidt P.M. and Lady T.C. (2007), Individuals' Use of Offshore
Holding Companies, The Tax Adviser, August, 462-465.
\item Slemrod J., Wilson J. (2009), Tax competition with parasitic tax
havens, Journal of Public Economics 93 (11--12), 1261--1270.\
\item Slemrod J., Yitzhaki S. (2002) \textquotedblleft Tax Avoidance,
Evasion and Administration\textquotedblright\ in A. J. Auerbach and M.
Feldstein (eds.) Handbook of Public Economics, Vol. 3, North-Holland:
Amsterdam, 1423-1470.
\item Tax Justice Network (2009), Tax information exchange arrangements, Tax
Justice Briefing, May.
\end{enumerate}
\end{document}