Working Paper 404
The Financial Secrecy Index: Shedding
New Light on the Geography of Secrecy
Both academic research and public policy debate around tax havens and offshore nance typically
suffer from a lack of denitional consistency. Unsurprisingly then, there is little agreement about
which jurisdictions ought to be considered as tax havens—or which policy measures would result
in their not being so considered. In this article we explore and make operational an alternative
concept, that of a secrecy jurisdiction and present the ndings of the resulting Financial Secrecy
Index (FSI). The FSI ranks countries and jurisdictions according to their contribution to opacity
in global nancial ows, revealing a quite different geography of nancial secrecy from the image
of small island tax havens that may still dominate popular perceptions and some of the literature
on offshore nance. Some major (secrecy-supplying) economies now come into focus. Instead of
a binary division between tax havens and others, the results show a secrecy spectrum, on which all
jurisdictions can be situated, and that adjustment lfor the scale of business is necessary in order to
compare impact propensity. This approach has the potential to support more precise and granular
research ndings and policy recommendations.
JEL Codes: F36, F65
Alex Cobham, Petr Janský, and Markus Meinzer
The Financial Secrecy Index: Shedding New Light on the
Geography of Secrecy
Tax Justice Network
Institute of Economic Studies, Faculty of Social Sciences, Charles
University in Prague
Tax Justice Network
A version of this paper is published in Economic Geography (July 2015).
John Christensen, Moran Harari, Andres Knobel, Richard Murphy, Nick
Shaxson and Sol Picciotto are important contributors to the theoretical and
practical development of the FSI, and we are grateful for their support. We
are also grateful for the valuable comments of Dariusz Wojcik, the editors
of Economic Geography and anonymous reviewers both there and at the
Center for Global Development. Alex Cobham was a research fellow at the
Center for Global Development while the paper was being prepared.
CGD is grateful for contributions from the Omidyar Network, Open
Society Foundations, and the Joffe Charitable Trust in support of this
Alex Cobham, Petr Janský, and Markus Meinzer. 2015. "The Financial Secrecy Index: Shedding New
Light on the Geography of Secrecy." CGD Working Paper 404. Washington, DC: Center for Global
Center for Global Development
2055 L Street NW
Washington, DC 20036
(f ) 202.416.4050
The Center for Global Development is an independent, nonprot policy
research organization dedicated to reducing global poverty and inequality
and to making globalization work for the poor. Use and dissemination of
this Working Paper is encouraged; however, reproduced copies may not be
used for commercial purposes. Further usage is permitted under the terms
of the Creative Commons License.
The views expressed in CGD Working Papers are those of the authors and
should not be attributed to the board of directors or funders of the Center
for Global Development.
Defining Tax Havens: Approaches and Implications ................................................................ 3
Secrecy Scores, from Policy Measures .......................................................................................... 9
Global Scale: The Provision of Financial Services .................................................................... 13
The FSI: A New Geography of Financial Secrecy .................................................................... 16
Conclusions ..................................................................................................................................... 20
Many citizens of developing (and developed) countries now have easy access to tax havens and the result is
that these countries are losing to tax havens almost three times what they get from developed countries in aid.
—Jeffrey Owens, Director, Organzation for Economic Cooperation and Development
(OECD) for Tax Policy and Administration, in Owens (2009)
We will set down new measures to crack down on those tax havens that siphon money from developing
countries, money that could otherwise be spent on bed nets, vaccinations, economic development and jobs.
—Gordon Brown, UK Prime Minister setting out the G20 agenda, in (Brown 2009)
Where are tax havens? In its special issue on tax havens, the Economist in 2013 (Valencia
2013) acknowledges the many ways in which the term, as much as its sibling offshore financial
center, is blurred—to the point of conceding that Delaware, a U.S. state, can be a tax haven.
In this article we argue that the term tax haven is an ill-defined misnomer, which has
supported the creation of a misleading dichotomous economic geography of tax pariahs.
Much as Sidaway and Pryke (2000) find the term emerging economy to be both interest driven
and lacking a convincing definition, the imprecision of the term tax haven has led to various
problems. In policy making, it has not only allowed questionable pressure on a group of
typically small, politically isolated jurisdictions, but it has also underpinned the failure, to
date, to find a comprehensive global response to the financial secrecy that thwarts the
effective taxation of income and profit, and facilitates money laundering, abuses of market
regulations, and the financing of terrorism.
In academic literature, the lack of clear and agreed definitions on tax havens and offshore
finance has contributed to important and systematic weaknesses in the resulting analyses,
whether in international political economy, economic geography, or international economics.
The most obvious problem to stem from this failure of definition is the difficulty posed for
the robustness of results, when the category of tax haven is not so much disputed as taken
for granted without explicit definition. Without clear and verifiable criteria on how lists of
tax havens have been derived, studies such as Hines and Rice (1994) or Johannesen and
Zucman (2014) expose themselves to the risk of creating invalid results by falling prey to
selection bias in the construction of their data.
The term offshore financial center (or later simply offshore) has widely assumed the successor role
to the term tax haven, at least in economic geography (Johns 1983; Cobb 1998; Roberts
1994; Hampton 1996; Hudson 1998a, 2000; Maurer 2008; Warf 2002). However, it has
arguably failed to break free from some of the constraints imposed by both the imprecise
and binary nature of the terminology (e.g., the apparently contradictory results in the studies
of Kudrle (2013) and Haberly and Wójcik (2014) on the relevance of time zone differences
as causal factors for determining the use of offshore witness to the wanting robustness of
empirical research findings relying on offshore as an independent variable). A solid base of
comparable research findings is unlikely to emerge without greater consistency of definition.
This article’s objective is to introduce the concept of a secrecy jurisdiction to economic
geography. We argue that more robust research findings and greater definitional consistency
are likely to emerge only when the focus of attention is shifted away from tax aspects or
offshoreness onto (specific, measureable components of) the financial secrecy that is offered by
jurisdictions. Largely underexplored and overlooked by academics in general, and economic
geographers in particular, the issue of financial secrecy merits greater attention since it is an
inherent part of most, if not all, of the economic activity undertaken offshore. For this
purpose, we propose a new framework of analysis whose backbone is a secrecy jurisdiction.
A secrecy jurisdiction’s central characteristics relate to the legislative provision of financial
secrecy to those who are physically resident elsewhere. We propose criteria that reflect both
the specific choices made by jurisdictions and the potential importance of those choices for
other jurisdictions. The resulting Financial Secrecy Index (FSI) thereby captures both the
intensity of jurisdictions’ commitment to financial secrecy, and their external scale, giving a
ranking of tax haven importance according to what Held et al. (1999) term impact propensity.
Once explicit, detailed, and verifiable criteria are applied, the results cast doubt over the
common, dichotomous distinction between countries and tax havens or offshore financial
centers. Rather, all reviewed countries offer various components of financial secrecy,
suggesting a secrecy spectrum upon which all jurisdictions can be situated. The mapping of
financial secrecy is not, therefore, an exercise in separating sheep from goats. Like offshore
for Wójcik (2012b), it is a matter of degree.
The geography of financial secrecy revealed by the FSI confirms some of the conventional
wisdom. For example, Switzerland, Luxembourg, Hong Kong, Cayman Islands, and
Singapore rank as the top five jurisdictions responsible for global financial secrecy and
associated harm. More surprisingly perhaps, the United States ranks sixth and Germany
eighth, and if the entire British sphere of influence was ascribed to London, the United
Kingdom would rank far above all other jurisdictions as the single greatest provider of
financial secrecy worldwide.
In terms of policy making, these results point to the fundamental importance of G8 nations
leading by example, if they wish to make serious progress on areas such as offshore tax
evasion, money laundering, and other forms of high-level corruption. The FSI suggests that
the traditional, subjective lists of tax havens have given undue weight to relatively secretive
but globally less important players—while the range of financial secrecy components are
found to extend across most major economies.
This article contributes to the body of literature linking geographic approaches to various
policy fields (e.g., Swords 2013; Dixon 2014, Kitchin et al. 2013; Loopmans 2008), which has
been growing since Martin (2001) decried the missing agenda of policy-relevant economic
geography research. The FSI establishes a critical, geographic, and policy-relevant
perspective on the issues of offshore finance and tax havens. In addition, it contributes to
the emerging strand of literature around the geography of secrecy or transparency by
providing further indicators of transparency (Wójcik 2012b).
The article proceeds in four sections. The first section addresses the issue of defining tax
havens, surveying the various approaches taken over time, and ultimately reaches a
preference for the term secrecy jurisdiction. In the second section, we develop a set of metrics
for this definition, on the basis of internationally comparable data. The third section outlines
the approach taken to generate a measure of the relative scale of each jurisdiction in the
global trade in financial services. In the fourth section, we combine the measures of secrecy
and scale to propose a ranking, the FSI, and demonstrate how the implied geography of
financial secrecy differs from that of a number of the main blacklists that are, or have been,
in use. A brief conclusion reflects on policy and theoretical implications and offers
suggestions for future research.
Defining Tax Havens: Approaches and Implications
The term “tax haven” has been loosely defined to include any country having a low or zero
rate of tax on all or certain categories of income, and offering a certain level of banking or
commercial secrecy. Applied literally, however, this definition would sweep in many
industrialized countries not generally considered tax havens, including the United States.
(Gordon 1981, 14)
Therefore, the broadest definition of a tax haven would include any country whose tax laws
interact with those of another so as to make it possible to produce a reduction of tax liability
in that other country. By such a definition virtually any country might be a “haven” in
relation to another. (Picciotto 1992, 132)
For rigorous analysis of the impact of jurisdictions offering financial secrecy, a specific and
objectively quantifiable definition is needed.1 The most common term—tax haven—is
probably also the most problematic. In 1981, the Gordon Report to the U.S. Treasury finds
that there was no single, clear objective test that permits the identification of a country as a
tax haven—instead offering a range of potential definitions, which could potentially include
any jurisdiction (Gordon 1981). It is interesting to note from the quote above that Gordon
effectively rules out any definition that might include the United States as a tax haven. While
originally understood to imply a jurisdiction with lower tax rates than elsewhere, the term
came to be used to cover jurisdictions with a great range of functions, many largely unrelated
to taxation. Gordon stresses opacity: "By definition, all of the jurisdictions with which we are
concerned afford some level of secrecy or confidentiality to persons transacting business,
particularly with banks" (1981, 15).
More recent literature has sought, more or less unsatisfactorily, to identify more specific
definitions by drawing out subcategories. Eden and Kudrle (2005), for example, identify one
group of havens based on type of taxation, following Palan (2002), and one based on
activity, following Avi-Yonah (2000) and Kudrle and Eden (2003). Palan, Murphy, and
Chavagneux (2010) create an ideal typology of tax havens refined by the niche strategies each
tax haven may engage in. Notwithstanding the intersecting nature and complication of these
1 This section draws on Cobham (2012).
various definitions, analysis under the heading tax haven tends to focus, understandably, on
tax aspects. This view is most commonly associated with the Organization for Economic
Cooperation and Development (OECD). While an earlier report (OECD 1987) focuses on
reputation, there is somewhat more precision in an OECD (1998) report. Specifically, the
1998 report emphasizes no, or only nominal, taxes as the starting point for the identification
of a tax haven, but it also emphasizes the lack of an effective exchange of information, lack
of transparency, and the absence of substantial activities.
The overarching rationale for the existence of tax havens that emerges from this approach is
the provision of relief to businesses or individuals from the rates of tax that apply elsewhere.
To achieve this, either the economic activity (in substance) has to be moved to a new
location from the original jurisdiction, or alternatively taxing rights have to be transferred by
other means (manipulation of the form).
This dichotomous approach, separating jurisdictions into nonhavens and varying categories of
tax havens, remains fraught with difficulty for research purposes. Two high-profile economic
articles, two decades apart, illustrate the issue. Hines and Rice (1994) and Johannesen and
Zucman (2014) assess the impact of tax havens on U.S. corporate tax and the true net
foreign asset positions of rich countries, respectively. Hines and Rice (1994: 40) note the
absence of a clear definition, and that “this vague characterization makes the process of
classifying tax haven countries somewhat arbitrary,” before combining IRS and other lists,
along with some ad hoc decisions around scale of finance. Johannesen and Zucman (2014)
apply a list drawn from work undertaken by the OECD over the course of many years,
which the authors have adjusted in vague terms.2 Both articles, however, draw clear
conclusions about the scale of impact of tax havens.
In early work by economic geographers on the subject, the term tax haven has been described
as a narrow, outdated and possibly stigmatizing label, which the authors mostly discarded in
favor of the (then) new, more neutral and broader term offshore financial center (e.g., Roberts
1994; Cobb 1998; Hampton 1996; Warf 2002; Hudson 1998b). Another implicit rationale for
the shift toward using offshore financial center instead of tax haven was the greater
relevance in the global economy resonating with the former term. This trend of ascribing a
growing role to what is understood as offshore finance is encapsulated well by Maurer’s
(2008: 160) famous quote: “Far from a marginal or exotic backwater of the global economy,
offshore in many ways is the global economy.”
The uncertainty stemming from a dichotomous approach as to what should be rightfully
labeled onshore or offshore has, however, been inherited from the tax haven terminology.
By some, offshore is used to indicate virtually all cross-border economic phenomena, such as
in the literature on offshoring (Clark and Monk 2013; Grossman and Rossi-Hansberg 2008).
Others have used the terminology to include some subset of cross-border economic activity
2 Similarly, Zucman (2014) provides no definition but offers a broad discussion of some typical activities
and a list in the appendix.
by focusing on certain characteristics (such as low regulation, low taxation, or secrecy)
(Wainwright 2013) or by comparing those characteristics with onshore (Roberts 1994).
Hampton (1996) differentiates between tax havens and three types of offshore financial
centers. While the former are defined by “no, or at best, low, direct and indirect tax rates
compared with the other jurisdictions,” the latter are seen as centers that host “financial
activities that are separated from major regulating units (states) by geography and/or by
legislation” (Hampton 1996 4–5, 10). However, as the author acknowledges, the
operationalization of both terms, as well as differentiating between them, remains very
The lines of the offshore/onshore dichotomy blur further in Hudson's (1998a) work. He
defines offshore as meaning “beyond the regulatory reach of the onshore authority,” and
frames the setting up of International Banking Facilities (IBFs) in New York as an attempt
to create “offshores onshore,” adding valuable complexity but further eroding the
conceptual clarity of a dichotomous divide Hudson's (1998a, 6). Wójcik (2012a, 7) explicitly
acknowledges that being an offshore jurisdiction or not “cannot be answered with a simple
yes or no. Just like world cityness, it is a matter of degree.” In a similar vein, Coe, Lai, and
Wójcik (2014, 765) discuss the problems of drawing a clear-cut division between offshore
and onshore by pointing to midshore finance centers, which are a chimera of both, or the
counterintuitive finding that “some onshore jurisdictions (e.g. Delaware, Miami) could be
more lax than offshore ones.”
The latest approach for empirical analyses around offshore is exemplified by (Wójcik 2012a,
(7), who defines offshore jurisdictions as “jurisdictions that specialize in attracting the
registration of [investment vehicles] with foreign sponsors.” Emphasis is placed on the term
investment vehicle, which appears to exclude a priori important banking centers, such as
Switzerland or Germany, by focusing heavily on the place of registration of certain legal
entities such as shell companies, trusts, special purpose vehicles, and mutual funds. The
operationalization of an offshore jurisdiction employed by Wójcik (2012a) relies on a
consensual approach originally pioneered by Palan, Murphy, and Chavagneux (2009), later
relabeled expert agreement (Haberly and Wójcik 2014).
This expert agreement approach relies on a metalist of tax havens, fed by a review and
scoring of the numbers of hits by 11 lists of tax havens and offshore financial centers
compiled over the course of over 30 years by different international organizations and
researchers (Haberly and Wójcik 2014). The authors use varying levels of expert agreement
around tax haven listings and offshore financial centers to empirically test the offshoreness
of foreign direct investment (FDI), acknowledging the possible futility in insisting on a
conceptual division between tax havens and offshore financial centers. Instead, the authors
maintain (Haberly and Wójcik 2014, 5) that “What defines offshore finance, however, is less
the jurisdiction within which transactions are booked or conducted, than their conduct in a
networked transnational legal space produced by the lack of a clear legal basis for
multinational activity.” As this suggests, understanding tax havens and offshore finance
requires an analysis of extraterritorial impact. The important challenge thus appears to be
how to move from a realization that offshore is a pervasive aspect of the world economy,
rather than a group of troublesome (small) jurisdictions, to a definition that can be made
operational for research and policy purposes.
Beyond economic geography, offshore financial center (or OFC) is preferred, for example,
by the International Monetary Fund (IMF), the mandate of which is more closely aligned to
issues of international financial regulatory oversight and stability than to issues of tax. Palan
(1998, 64) explores some of the difficulties of consistent definition in this case, noting that in
the financial literature “offshore is used […] to describe unregulated international finance
[…] Rather confusingly, however, the International Monetary Fund and the Bank for
International Settlements consider only tax havens as Offshore Financial Centres, though
the City of London, which does not qualify as a tax haven, is considered the hub of global
An important IMF Working Paper by Zoromé (2007, 7) discusses the definitional issues in
some detail, proposing a specific, measurable definition: “an OFC is a country or jurisdiction
that provides financial services to nonresidents on a scale that is incommensurate with the
size and the financing of its domestic economy.” He goes on to identify such OFCs by
examining the ratio of net financial service exports to gross domestic product (GDP) from
IMF balance of payments data and by looking at jurisdictions with especially high values (an
approach that we discuss further and build on later in this article).
The key difference between the IMF’s preexisting list and Zoromé’s (2007) findings is the
addition of the United Kingdom, which neatly illustrates the value of using objective criteria:
a level playing field (including politically uncomfortable findings) may be more likely to
emerge. Where Hudson (1998b), for example, is explicit about London as the long-standing
home to offshore business—most obviously, the Eurodollar market—neither the London
nor the United Kingdom appears on any of the common lists. Despite the advantage
achieved by using quantitative criteria, Zoromé’s (2007) approach retains the preference for
a binary list of locations of concern.
The third main term used—and increasingly so since it was defined and promoted by
Murphy (2008)—is secrecy jurisdiction. It is not entirely clear when the term was used for the
first time, but, according to Peet and Dickson (1979), it featured in a report by the U.S.
House of Representatives (1970). The focus remains on specific actions taken, but by
employing the word jurisdiction, the legal realm is emphasized. This follows the logic of
Palan (2002), who discusses the commercialization of sovereignty: the decision by certain
jurisdictions to obtain economic advantage by allowing selected political decisions (over, for
example, the taxation of nonresidents) to be dictated by those likely to benefit from the
decision (for example, financial, legal, and accounting practitioners). 3
3 The idea that political decision making can be distorted, so that becoming a secrecy jurisdiction may
damage democratic representation, is explored further under the name the finance curse (Shaxson and Christensen
The emphasis on secrecy is necessary, Murphy (2008) argues, because it is this that allows
nonresidents to take advantage of favorable features in the jurisdiction’s legal framework
with the confidence that they will not run afoul of the legal system in the places where they
reside. There are thus two key characteristics that define a secrecy jurisdiction:
• “The secrecy jurisdiction creates regulation that they know is primarily of benefit
and use to those not resident in their geographical domain”
• “The creation of a deliberate, and legally backed, veil of secrecy that ensures that
those from outside the jurisdiction making use of its regulation cannot be identified
to be doing so.” (Murphy 2008, 6)
By focusing on what makes them attractive, the secrecy jurisdiction concept therefore relies,
above all, on an assessment of the comparative advantage of the jurisdictions in question.
The route the secrecy jurisdictions have chosen, in order to attract (the declaration of)
foreign economic or financial activity is the provision of relatively favorable terms to
nonresident users. In effect, this indicates a reliance on regulatory arbitrage (potentially, but
not necessarily, including tax regulation).
To be successful over time, such behavior should be hidden as far as possible from the views
of regulators in those other jurisdictions, elsewhere, who may take countermeasures to frustrate
the arbitrage. A major role of secrecy therefore is to facilitate changes in the form, but not
the substance, of economic activity so that for regulatory purposes, it appears to take place
elsewhere. In the extreme, structures are established such that activity appears for regulatory
purposes to take place nowhere (Murphy 2008). For example, the recent U.S. Senate
hearings into Apple discovered that the information technology giant had managed to create
corporate entities in Ireland, which for tax purposes had no jurisdiction—most significantly,
Apple Operations International, which reported net income of $30 billion from 2009 to
2012 and filed no corporate tax return anywhere (U.S. Senate Permanent Subcommittee on
The ideal approach for the identification of secrecy jurisdictions might therefore contain two
separate components: one reflecting each jurisdiction’s (objectively measurable) performance
against key indicators of secrecy—that is, how far they have gone in terms of Murphy’s
(2008) second criterion above—and one reflecting each jurisdiction’s importance in the
global provision of financial services to nonresidents (i.e., their quantifiable scale)—that is,
their success according to Murphy’s first criterion. Equally, these components can be
considered in the terms of Held et al. (1999) as measures of intensity and extensity,
combined to show impact propensity. In addition they combine emphasis on internal policy
decisions, and—in line with van Hulten (2012)—extraterritorial reach. In the following two
sections, we lay out the basis for our attempt to assess each component.
This approach has two main theoretical and conceptual advantages over the other two
terminologies. First, by focusing on secrecy and transparency, the empirical determination of
a jurisdiction’s intensity of providing secrecy becomes inherently easier than for tax or other
regulatory aspects. Since properly enforced transparency should be easily observable in many
cases, the comparative evaluation of a jurisdiction’s policies becomes more feasible. The
resulting secrecy spectrum on which a jurisdiction’s policies can be positioned results in
overcoming the dichotomy trap, a second major advantage over the other terminologies.
There is a potential, conceptual drawback to this approach. Popular views rely heavily on tax:
for example, the Cayman Islands are a tax haven because of the absence of any taxes on
individual income and corporate profits, and regardless of any other characteristics such as
transparency.4 Arguably this viewpoint confirms the weakness of the term tax haven, for even
in this example, the concern would not be with the Cayman Islands’ competing, through low
tax rates, to attract real activity. Rather, the concern is that the Cayman Islands may attract
profits or incomes that are, in fact, derived from economic activity taking place elsewhere: so
that the central feature of the behavior is not to offer lower tax for the same activity but to
separate the recording and accounting of the tax base from the jurisdiction where it actually
arises. What makes the low or zero tax rates attractive for this type of process, as opposed to
the relocation of real activity, is the potential to hide relevant details from the jurisdiction
where the tax base arises but from which it has now been separated.
Similarly, consider recent ‘Luxleaks’ (ICIJ 2014) revelations about near-zero taxation
agreements for certain financial activity of multinationals in Luxembourg. Irrespective of
their lawfulness, they resulted in major tax losses in other jurisdictions; but while this had
been known in some circles for some time, it is only the current wave of public transparency
that has resulted in political pressure to make such activity impossible. As such, the
unacceptable feature (for other EU countries) of the process was not the low tax rates, but
rather the ability to hide the large shifts of tax base. (Whether the key to acceptability was
hiding this from tax authorities, or from citizens, is an interesting research question.)
Equally, revelations about Irish tax treatment of major multinationals (e.g. Pinsent Masons,
2014)) has caused intense pressure for adjustment of the approach. While again the low or
zero tax rate provided the ultimate benefit for business, it was the lack of transparency that
made the arrangements politically sustainable. In both the Luxembourg and Ireland cases,
the true tax rate was itself hidden so that any external assessment based on the statutory rate
or on effective rate constructed from public data would not have reflected the full tax haven-
ness of these states.
An alternative approach to the secrecy jurisdiction focus could be to consider a jurisdiction’s
tax haven-ness as depending on the degree to which it is able to attract the tax base of
economic activity that takes place elsewhere. This would align with an ongoing policy
process: at the behest of the G8 and G20 groups of countries, the OECD is currently in the
middle of a two-year process, the Base Erosion and Profit Shifting (BEPS) initiative, which
has the explicit aim of reforming international corporate tax rules to achieve better alignment
4 We are grateful to an anonymous reviewer for highlighting this view.
between the location of corporate profits and the underlying, real economic activity (OECD
2013). BEPS Action Point 11 (out of 15) requires creation of a baseline estimate, hitherto
lacking, on the extent of misalignment.
Current work using survey data on U.S .multinationals (IMF 2014) and global balance sheet
data (Cobham and Loretz 2014) identifies a set of jurisdictions that systematically obtain a
disproportionately high share of the corporate tax base in relation to their hosting of (real)
economic activity: for example, Luxembourg, Ireland, and the Netherlands are identified in
both studies. While the samples in these studies are dominated (in both home and host
economies) by rich countries, it is conceivable that future work will overcome these
constraints in order to produce a more balanced, global picture of the jurisdictions that lead
in this measurable aspect of tax haven-ness. Even then, of course, being a hub for corporate
profit shifting is just one aspect of haven-ness (probably the most researched so far as in
Karkinsky and Riedel (2012) or Janský and Prats (forthcoming)); other measures would be
needed to capture, for example, jurisdictions’ role in the evasion of personal income and
wealth taxation (see, e.g., Zucman 2014).
As the secrecy jurisdiction has not yet been used by economic geographers except for a
cursory mention from Wójcik (2012a), it is an objective of this article to establish the
concept. For the remainder of this article we define secrecy jurisdiction in line with Meinzer
(2012a, 1) as a jurisdiction that “provides facilities that enable people or entities to escape or
undermine the laws, rules and regulations of other jurisdictions elsewhere, using secrecy as a
Secrecy Scores, from Policy Measures
A situation of financial transparency may be characterized (1) by relevant information being
placed on public record for all stakeholders to access; (2) by access on certain private
financial data only by authorized authorities (such as tax administrations, police, etc.); or (3)
by collecting, analyzing and sharing relevant information effectively with foreign
counterparts. These are the areas in which we address the creation, by policy, of secrecy.
We have constructed 15 explicit, detailed, and verifiable indicators that measure the secrecy
provided to nonresidents in the laws and regulations of jurisdictions. As a proxy for secrecy
provided to nonresident investors, these key financial secrecy indicators (KFSI) change over
time subject to refinement and data availability. Taken together, these indicators result in one
compound secrecy score allocated to each jurisdiction. The scores are normalized to a range
of zero (perfect transparency) to 100 (complete secrecy) and in practice vary between 32.4
(Sweden) and 88 (Samoa). For the FSI 2013, 82 jurisdictions are included, and the data set
used for this article includes an additional five jurisdictions, bringing the total to 87.5
5 The relevant data on five additional countries were generated for the Center for Global Development, to
be used as part of the Commitment to Development Index, which ranks rich countries on the development
impact of their policies and incorporates the FSI (Janský forthcoming).
The data set underlying the 15 KFSIs is available online for review, and linked to underlying
sources (FSI 2013a). The main and preferred data sources were official and public reports by
the OECD; the associated Global Forum on Transparency and Exchange of Information for
Tax Purposes (hereafter Global Forum; Meinzer 2012b); the Financial Action Task Force
(FATF); IMF; and the U.S. State Department’s annual International Narcotics Control
Strategy Report (e.g., U.S. Department of State (2013), which in one volume contains
country reviews, including specific and comparative anti–money laundering data.
In addition, specialist tax databases and websites such as by the International Bureau of
Fiscal Documentation, PriceWaterhouseCoopers (Worldwide Tax Summaries), Lowtax.net,
and others have been consulted.6 Furthermore, surveys have been sent to the ministries of
finance and the financial intelligence units of all 87 reviewed jurisdictions, which included
targeted questions about the jurisdiction’s tax and regulatory system. The questionnaires sent
to the ministries of finance and to the financial intelligence units can be viewed online: see
FSI (2013b) and FSI (2013c), respectively. All jurisdictions had the opportunity to provide
up-to-date information by answering the questionnaires.
Out of a maximum of 202 variables available in the database for each jurisdiction, up to 49
are used to compute the secrecy score. Each of the 15 indicators is weighed equally. For
some indicators, data availability and comparability is a problem. For instance, a publication
by the OECD (2013) with specific comparative information on tax administrations used for
two of the 15 indicators contains information for a total of 52 countries, out of which only
34 are included in the FSI 2013. For these two indicators, this leaves 48 countries of the FSI
2013 without a primary data source. If a jurisdiction did not respond to the questionnaires,
and if (in some cases) follow-up enquiries with local researchers did not yield additional
insights, this absence of data is reflected in the database by marking the relevant field as
unknown. However, when constructing the indicators, the jurisdictions without data have
been assessed under these circumstances as if their policies with respect to the particular
indicator under assessment provide secrecy. Absence of data was awarded a secrecy score.
The guiding principle for data collection was to always look for and assess the lowest
standard (or denominator) of transparency available in each jurisdiction. For example, if a
jurisdiction offered three different types of companies, two of which required financial
statements to be published online, but the third is not required to disclose this information,
then we have answered the particular question about the online availability of accounts with
The 15 KFSIs can be grouped around four broad dimensions of secrecy: (1) knowledge of
beneficial ownership (three KFSIs); (2) corporate transparency (three KFSIs); (3) efficiency
of tax and financial regulation (four KFSIs); and (4) international standards and cooperation
(five KFSIs). A brief discussion of the four groups follows below; a more complete
6 These databases were accessible at the following addresses: http://www.ibfd.org/IBFD-Tax-
description of each indicator is provided in the Methodology report, available online (Tax
Justice Network 2013a).
For the first group of indicators, the notion of beneficial ownership of assets and legal
entities and structures has its roots in the anti–money laundering discourse that began in the
1990s (Blum et al. 1998; Cuellar 2003; Levi 2002; Pieth and Aiolfi 2003; Carrington and
Shams 2008; Unger 2007; UN Office on Drugs and Crime 2007).
The FATF (2012, 110) defines beneficial owners as the “natural person(s) who ultimately
owns or controls a customer and/or the natural person on whose behalf a transaction is
being conducted. It also includes those persons who exercise ultimate effective control over
a legal person or arrangement.” This view is shared only partly by the international tax
community. In a report published at the request of the Financial Stability Forum, OECD
(2001) explicitly uses the notion of a beneficial owner being a natural person. Contrary to
this, the influential model tax convention of the OECD (2008) suggests that a beneficial
owner can be a legal entity.
The OECD’s annual tax cooperation reports ((OECD 2006, 2007, 2008, 2009a, 2010))
2006–10 also illustrate the confusion here. While OECD (2006, 148) clearly defines the term
legal owner, it refers to the term beneficial owner only in circular logic: “Legal ownership refers to
the registered owner of the share, which may be an individual, but also a nominee, a trust or
a company, etc. Beneficial ownership reporting requirements refers to a range of reporting
requirements that require further information when the legal owner is not also the beneficial
For the purposes of the FSI, we apply the concept of beneficial ownership broadly, as
defined by the FATF, to bank accounts (KFSI 1), trusts and foundations (KFSI 2), and
corporate entities with limited liability (KFSI 3).
The second dimension of financial secrecy relates to companies. Given the pervasiveness of
companies in offshore finance as the basic vehicle to commit crimes and engage in abusive
behavior, and considering their privileges granted by society, for instance, in terms of limited
liability, it can be argued that corporations ought to be subject to a higher standard of
transparency than merely submitting information to some registry. In order to prevent
market failures and distortions through information asymmetries, the public at large,
regulators, investors, and consumers should be able to easily find out about the activities of
any corporate vehicle along various dimensions. KFSI 4 assesses if beneficial, or at the very
least, legal ownership is accessible over the Internet for less than 10US$/€. KFSI 5 reviews
whether the financial statements of each type of company with limited liability is accessible
online again for less than 10US$/€. KFSI 6, in turn, asks if countries require companies to
submit and publish certain financial data on a country-by-country basis.
Third, we are concerned with the efficiency of tax and financial regulation. While at first
glance, efficient tax or financial regulation is not related directly to financial secrecy, one way
of preserving secrecy in financial matters is to encourage a culture of noncompliance by,
among others, not monitoring domestic economic actors by failure to collect basic
information (KFSI 7). Similarly, dispensing with basic tools for efficient tax administration
(such as the reliance on taxpayer identification numbers for matching information from
various sources) can help to encourage noncompliance (KFSI 8). Furthermore, if countries
create strong incentives for other countries to enter into bilateral tax treaties, this opens new
doors for tax avoidance and increases secrecy through complexity in international taxation
(McGauran 2013; Weyzig 2012; Rixen 2008; Picciotto 1992). On the other hand, countries
can also create strong incentives for other nations to lower their tax rates and thereby
encourage investors from all over the world to seek low or zero tax rates, which, in turn,
invite undeclared, secretive investments for tax evasion or avoidance purposes (KFSI 9).
Finally, compliance with international standards and the level of international cooperation is
assessed. Over the last decades, international efforts at enhancing cooperation in financial
matters have increased either by hard international law or through best practice standards and
associated evaluations of their implementation (soft law; Abbott and Snidal 2000). Most
relevant for assessing financial secrecy are the evolving anti–money laundering regimes
(KFSI 11), various tax information exchange initiatives (KFSI 12 and 13), as well as generic
international judicial cooperation (KFSI 15), as an important law enforcement tool mostly
for high-profile crimes beyond simple tax evasion. Furthermore, a series of thematic
international conventions contain commitments related to financial transparency (KFSI 14).
A possible drawback to the secrecy jurisdiction approach is the following. The conceptual
basis allows objective, verifiable criteria to be used in place of the expert list approach that
has been necessary to make any progress with the term tax haven. However, the choice of
criteria is necessarily subjective, as in any index. While the criteria reflect a range of
international standards and related mechanisms, any given observer could reasonably make
a case for focusing only on some aspects—on, say, the extent of company beneficial
ownership information and its international exchange, while setting aside banking secrecy
and much else.
While the eventual choice of FSI criteria has developed over time through wide engagement
with country and thematic experts, the basis for this particular choice is similar to that for
expert lists of tax havens. One difference, of course, is that the process itself and the criteria
are entirely transparent and verifiable, allowing any observers to corroborate the degree of
secrecy of any particular jurisdiction or, instead, to fashion their particular choice of criteria
into an alternative secrecy score.
In what follows, we present the FSI as published and consider how the resulting geography
of secrecy differs from other analyses. At the same time, we recognize that narrower,
broader, or differently weighted combinations of secrecy components would yield
(sometimes substantial) variations. Equally, the FSI could be seen as a complementary
instrument to the analysis of tax rates, for example. However, for the reasons discussed
above, robust measures of haven-ness based on public tax rate data alone are likely to remain
elusive, even if definitional issues can be resolved.
Global Scale: The Provision of Financial Services
We are interested in which countries affect financial secrecy globally, rather than in countries
with high secrecy scores, but without significant impact. Therefore the second component of
the FSI is the global scale weight (GSW) attributed to each jurisdiction, and this is based on
the assessment of the size of each jurisdiction’s share of the global market for financial
services provided to nonresident clients. We explain how this assessment is made, before
considering potential criticisms of the approach. Our methodology for the calculation of the
GSW builds on the work of Zoromé (2007). Zoromé relies on the relative intensity of the
provision of financial services to nonresidents by taking a measure of financial services
exports and scaling by jurisdictional GDP.
Here we are concerned not so much with intensity (domestically), but impact (globally), so
we measure the market share of each jurisdiction (that is, each jurisdiction’s provision of
financial services to nonresidents, as a ratio to the total global provision of services to
nonresidents across all jurisdictions, rather than as a ratio to the jurisdiction’s own GDP). As
Cobham (2012) shows, taking global contribution rather than relative intensity in the
provision of financial services to nonresidents leads to quite a different picture: with 2007
data, the former criterion points to Cayman Islands, Luxembourg, Switzerland, the United
Kingdom, and the United States, while the latter points, instead, to Bermuda, Cayman
Islands, Guernsey, Jersey, and Luxembourg.
The global scale weights are based on publicly available data about the trade in international
financial services of each jurisdiction. The preferred data source is the IMF’s Balance of
Payments Statistics (BOPS), which provides data on international trade in financial services,
and this extends to 53 of our 87 jurisdictions. We employ data from BOPS based on two
different manuals, BPM5 (IMF 1993) and BPM6 (IMF 2013a). When available—mostly
years 2005 to 2011—we use data on the basis of BPM 6. Otherwise—mostly for years prior
to 2005—we use an earlier edition, BPM 5. We do not find substantial empirical differences
between the two. For 2011, the recent year with most available data, the BOPS cover 116
jurisdictions for exports.7
For the rest of the sample, we extrapolate from IMF data on stocks of internationally held
financial assets to derive trade or flow estimates (again following Zoromé, 2007). Data on
stocks of portfolio assets and liabilities are taken from two IMF sources: the Coordinated
Portfolio Investment Survey (CPIS) (IMF 2013b) and the International Investment Position
(IIP) (IMF 2013c) statistics, of which the latter is part of the BOPS. CPIS data for 2011
cover 76 jurisdictions for total portfolio assets, and 215 jurisdictions for total portfolio
liabilities, which are derived from reported assets. IIP data for 2011 cover 112 jurisdictions,
7 The 2013 index is based on data available in mid-2013. More timely updates of this important data set
would, in general, allow more recent data to be used.
and is filtered (again following Zoromé, 2007) to exclude FDI, reserve assets, and all assets
belonging to general government and monetary authorities.
There is an argument for preferring liability data to asset data, since it ought to reflect—for
example—that French clients holding assets in German banks create a German services
export and a German liability. However, there are two reasons to use assets. First, it is assets
that are directly reported by jurisdictions. These data are therefore more likely to capture the
full range of assets, rather than liability data, which are inferred by inverting the stated asset
claims of other jurisdictions, and hence are likely to be incomplete. Second, a jurisdiction’s
overseas assets, beyond a certain point dictated by their domestic economic structure (a
different point for the United States compared to that for the island of Jersey, for example),
will be managed on behalf of nonresidents and hence also indirectly reflect the export of
financial services. As would be expected given the nature of financial markets, there is a
strong correlation between assets and liabilities where data for both are present.
We use liabilities data to extrapolate values of assets where neither assets nor financial
services exports are reported. The adjusted data on stocks of assets are then used to estimate
current flows of financial services. We aim to improve on the IMF extrapolation by using a
panel of data (2001–11) rather than a single year on which to base the extrapolation, which
appears to allow marginally more accurate estimation of flows from stock data. The implied
coefficients (all significant at the 1 percent level) are very similar regardless of the
specification chosen, including fixed-effects panel regressions. We ultimately select a pooled
ordinary least squares (OLS) regression to allow the constant to be constrained to zero
(allowing a nonzero constant only trivially affects the goodness of fit, which is between 0.83
and 0.85 under each specification we consider).
We also use liabilities data to assess the reasonableness of reported assets, which leads us to
identify a discrepancy specific to the Cayman Islands. Here the recorded value for
liabilities—that is, that based on the recording of other jurisdictions—far exceeds the
declared value for assets. To see this, we consider the difference in recorded values of
liabilities minus assets, as a ratio to jurisdictions’ GDP. This allows us to scale the size of the
difference according to jurisdiction so that, for example, Jersey is not necessarily more likely
to stand out than the United States. We use GDP from the World Bank’s World
Development Indicators (World Bank 2013) or, when not available, from the CIA’s World
Factbook (CIA 2013). Also, where necessary we use the values of GDP from the closest year
The ten highest recorded values of liabilities minus assets as a ratio to jurisdictions’ GDP all
relate to one jurisdiction: the Cayman Islands. For only one other jurisdiction is there a ratio
greater than 10 in any year (for the Netherlands Antilles that no longer exists). For all 11 of
the Cayman observations from 2001 to 2011, the ratio exceeds 250, with the highest values
(in excess of 500 times GDP) all recorded in the most recent years.
This feature of Cayman-declared data is confirmed by IMF researchers Lane and Milesi-
Ferretti (2010) and by Zucman (2014), who noted that it results from the Cayman Islands—
unlike all other major reporters—reporting only on its banks’ portfolio holdings and
excluding those of its large hedge fund industry.
We therefore impute a value for Cayman Island assets. We proceed with the assumption
that the liabilities data—as recorded by all other reporting jurisdictions—is the most accurate
reflection of the Caymans’ activity and therefore extrapolate an alternative asset measure.
To do this, we perform a simple OLS regression of our asset value on CPIS reported
liabilities, with no constant, using the pooled data for all jurisdictions except the Cayman
Islands, from 2001 to 2011. Taking the coefficient (2.05) as the average ratio of assets to
liabilities in our data set, we multiply the 2001–11 values for Cayman Island liabilities by this
to obtain a value for Cayman Island assets, which we believe reflects more closely the actual
scale of Cayman Island activity in offshore financial services. Given the IMF analysis (Lane
and Milesi-Ferretti 2010), this is likely if anything to be an underestimate.
In total, we are able to create flow data (true or extrapolated) for a total of 217 jurisdictions,
which we believe cover the majority of the global provision of financial services to
nonresidents (and a vast majority of the total of 245 jurisdictions considered in our analysis).
Finally, we can use the total level of financial service exports for the 217 jurisdictions and
take the exports of each of the FSI jurisdictions with available data as a share of this global
total. This creates a global scale weight reflecting the relative importance of each jurisdiction.
The Global Scale Weight is defined as
The total global scale weight for the 80 FSI jurisdictions with data is 96.85; rising to 97.27
when we include the additional five countries assessed separately.
It is important to note that this weighting alone does not imply harboring or supporting
inappropriate behavior by the jurisdictions in question. Arguably, those near the top should
be congratulated on their success in the field of international trade in financial services
(although in light of recent examples, such as Iceland, Ireland, and Cyprus, they may, of
course, also want to consider the extent of their reliance on this risky sector). Rather, the
GSW is an indicator of the potential for a jurisdiction to contribute to the global problem of
financial secrecy, if secrecy is chosen in the range of policy areas discussed above.
We believe that this methodology represents the most robust possible use of the available
data, given its limitations, as a means to evaluate the relative contribution of different
jurisdictions to the global total of financial services provided to nonresidents. The fact that
researchers must follow such a convoluted path to reach this point is evidence of the failure
of policy makers to ensure that global financial institutions and national regulators have
access to the necessary data to track and understand international finance.
The FSI: A New Geography of Financial Secrecy
The FSI reveals a new geography of financial secrecy, with two main features. First, the FSI
reveals the dominant role of a number of major economies—in contrast with the emphasis
on small island states that tax haven lists prepared by multilateral organizations have long
exhibited. Second, the FSI shows a contrasting view of corruption to that of the most high-
profile alternatives such as Transparency International’s Corruption Perceptions Index (CPI)
(Transparency International 2012).
The final step in creation of the FSI is to combine the ranking by scale of activity with the
secrecy scores, in order to generate a single number by which jurisdictions can be ranked,
reflecting the potential global harm done by each. As with the choice of secrecy indicators
and their relative weighting in the secrecy score, and with the focus on financial services
exports to determine relative scale, the method of combination cannot be objective.
Underlying the choice made is a desire for neither secrecy nor scale to dominate the final
In practice, there is significantly more variation in the scale weighting than the secrecy score,
so we transform the two to generate a series with variations of a similar order. The simplest
transformations that achieve this are to take the cube of the secrecy score and the cube root
of the scale weight so that for each country
The full index for 2013 is available online (Tax Justice Network 2013b). Table 1 compares
the top 10 jurisdictions on the FSI, with those ranked separately by the secrecy score and by
GSW. Clear differences in the geography of secrecy or of corruption are apparent: GSWs
point to the largest financial centers, secrecy scores point to the smallest, traditionally
noncooperative jurisdictions, while the FSI itself combines the last two to provide a picture
of scale-weighted secrecy. Some major economies now come into focus, reflecting their
importance in the global provision of financial services. The most secretive jurisdictions are
of so little importance that they do not make the top 10 of the FSI overall; but most of the
biggest players by scale are also sufficiently secretive to feature in the FSI top 10. Only the
United Kingdome is sufficiently transparent to move far down the FSI (with a secrecy score
just below 40, it ranks twenty-first in the FSI despite being responsible for 18.5 percent of
global financial services exports).
Researchers using the index should, of course, consider the particular aims of their own
work before deciding on the appropriate measure to use. Research focusing on the relative
risk of illicit financial flows in transactions with different jurisdictions, for example, may
require pure secrecy scores. In contrast, understanding global changes in secrecy may require
a weighting, such as that in the index, in order not to be unduly swayed by the experience of
a few small, highly secretive jurisdictions. The combined FSI also allows for comparison of
the extraterritorial importance of jurisdictions’ financial secrecy.
Table 1 shows two related indices: the CPI(Transparency International 2012), which
combines 13 different sources based on expert opinion surveys to rank countries according
to the perception of corruption and has been criticized for presenting only the perceptions
of an international, largely corporate elite (Christensen 2007; Cobham 2013); and the Basel
Anti–Money Laundering Index (BAMLI) (Basel Institute on Governance 2013), which is
more obviously similar to the FSI and rates countries according to money laundering and
terrorist financing risk, on the basis of components including international organizations’
ratings. We use the detailed BAMLI Expert Edition Data, as of July 15, 2013. Note that the
BAMLI includes components based on scores from the CPI (10 percent) and the FSI (25
In the BAMLI meanwhile, and above all in the CPI, some of the lowest-income countries
perform worst. Simple regressions of each index or component on per capita income
confirm this pattern: there is a significant positive correlation for the CPI, with income
explaining 57 percent of variation in corruption, and the BAMLI (R² of 37 percent). Secrecy
scores also tend to be worse for lower-income countries, but income only explains 20 percent
of the variation in secrecy; for the overall FSI, the pattern disappears, with explanatory
power of income falling to just 8 percent.8
8 Regressions not reported; available on request.
Top Ten Jurisdictions by FSI, FSI Components, and Other Indices
Ranking by FSI Secrecy
GSW BAMLI CPI
1 Switzerland Samoa United States Somalia Afghanistan
2 Luxembourg Vanuatu
Kingdom Afghanistan Korea, DR
Islands St. Lucia Switzerland Cambodia Sudan
Islands Tajikistan Myanmar
6 United States Liberia Germany Iraq Turkmenista
Islands Singapore Guinea-
8 Germany Barbados
9 Jersey Belize Ireland Eritrea Venezuela,
10 Japan San
Marino France Myanmar Burundi
69.0 83.4 59.3 n/a n/a
GSW 58.9% 0.07% 80.4% 0.023% 0.014%
Note: FSI and BAMLI results for 2013, CPI results for 2012. Secrecy scores have not been calculated for any of
the top 10 countries by BAMLI or by CPI.
In Figure 1, we compare the FSI results with 14 current and historic lists of tax havens by
average secrecy score of included jurisdictions and total global scale weight. The information
on 11 lists come directly from Murphy (2009), Irish (1982), Hines Jr. and Rice (1994),
Financial Stability Forum (2000), IMF (2000), OECD (2000), FATF (2000, 2002),
Hampton and Christensen (2005), Lowtax.Net (2008), Zoromé (2007), and Levin (2007).
We also include three more recent lists: that of the U.S. Government Accountability Office
(2008); OECD (2009), and ActionAid UK (2013), as used by the Enough Food For
Everyone IF campaign, which saw more than 100 nongovernmental organizations campaign
beginning in 2012 for the United Kingdom and other governments to deliver policy changes
at the 2013 G8 summit. Six small jurisdictions that appear separately on one or more lists are
dropped because we either do not analyze them (Anjouan, Campione d'Italia, Ingushetia,
and Turkish Republic of Northern Cyprus), or include them elsewhere (Alderney and Sark).
In addition, we include the top 10 jurisdictions by scale, by secrecy, and on the FSI overall.
With only one exception, the listed jurisdictions account in total for a smaller share of the
GSW than the 10 biggest jurisdictions in the FSI—while their average secrecy is generally,
but not always, somewhat higher than the average secrecy score for either the whole FSI or
the top 10. The lists, almost without exception, have focused attention on smaller, somewhat
more secretive jurisdictions—to the exclusion of only somewhat more transparent, much
While this assessment is far from definitive, two main conclusions are suggested. One is that
measures of de facto and de jure compliance with specific anticorruption measures—
whether in the BAMLI or FSI secrecy score components—seem much less strongly
correlated with per capita income levels than is the CPI. The other is that by including a
measure of the scale of jurisdictions’ potential contribution to the global problem of
secretive flows, rather than seeing each jurisdiction in isolation, the FSI highlights the major
financial players—instead, perhaps, of jurisdictions with poor performance but minimal
impact on others. In this way the FSI presents a new view of the geography of financial
secrecy: one that highlights the influence that jurisdictions exert extraterritorially through
Figure 1. Tax haven lists and the FSI (by secrecy and scale).
Top 10 FSI
Top 10 FSI
55 60 65 70 75 80 85
Global scale weight
Average secrecy score
The FSI reflects an effort to assess financial secrecy on the basis of verifiable, empirical data.
As such, it shows a spectrum of secrecy rather than a binary distinction between tax havens
and others. The resulting global mapping reflects the pervasiveness of secrecy and the
leading role of some major economies including those of the United States and the United
Kingdom. This article’s theoretical contribution lies in two strands of literature. Martin's
(2001) landmark discussion of the “missing agenda” of policy-relevant economic geography
research has created a body of literature that theorizes around institutional change (Varró
2014; Isserman and Markusen 2013; Woods and Gardner 2011, among many others) or
relates geographic approaches with specific policy fields such as industrial agglomeration
(Swords 2013), finance (Dixon 2014), social media (Kitchin et al. 2013) or urban planning
(Loopmans 2008). The FSI contributes to both strands of policy-relevant economic
geography by providing an economic geographic perspective in the policy field of
international taxation and “tax competition.” At the same time, the FSI argues that a shift is
required from a narrow tax focus onto broader financial secrecy and transparency matters in
order to facilitate effective policy change. Because increased financial transparency has the
potential for educating and mobilizing the electorate about the harm caused through
financial secrecy, there is greater likelihood for democratic societies to overcome the
resistance of powerful vested interests in favor of maintaining the status quo (Meinzer
In an earlier work on an ill-defined but popular term, Sidaway and Pryke (2000: 187)
consider the case of emerging markets. Among their findings is that the use of the term to
reflect the strange and exotic other “belies deeper continuities with colonial geographical
imaginations”; in other words, the use of the term, and its uncertain definition, reflects, to
some extent, a power dynamic and a set of interests.
The parallel here is that the use of the term tax havens by policy makers is almost uniquely
associated with expressions of dismay and belligerence (cracking down, or shutting havens),
or of denial and otherness (the common refrain, we are not a tax haven).9 While many of the
jurisdictions in question are revealed in the FSI to be highly secretive, and sometimes to play
a potentially major role in global secrecy, the difference in our approach is that major
economies are ranked by the same standard—rather than being able to rely on political
power to ensure they remain outside any lists compiled.
The largely futile attempts to tackle tax havens over the last decades bear witness to the
inadequacy of the chosen terminology and methods. Johannesen and Zucman (2014, 65)
show that the recent crackdown only modestly affected offshore funds, and at best “caused a
relocation of deposits [to] the benefit of the least compliant havens.” We argue that the
misguided division into tax havens and others lies at the heart of this failure to provide a
more comprehensive (and effective) response. In contrast, the policy agenda developed at
9 For a collection of recent statements of this form from jurisdictions, see
the G20 and more recently at the 2013 G8 summit mirrors the shift undertaken by the FSI
in focusing on financial secrecy instead of direct tax aspects, and hence in starting with major
economies rather than small financial centers.
It is not inconceivable that a rigorous, widely held definition of tax havens could emerge;
and over time, advances in data could allow such a definition to become robustly measurable
in a way that supports more nuanced findings and more detailed research and policy analysis.
At present, however, only the FSI or some variation on this approach appears to offer that
The shift of emphasis away from tax, which is embodied by the FSI, leads to a second,
emerging strand of economic geography literature on the geography of transparency (Wójcik
2012b). As Wójcik (2012b) finds for country-by-country reporting by multinational
companies, the FSI seeks on a broader basis to “help keep alive a public deliberation on the
architecture of the international tax system.” The FSI’s criteria-based approach, and the
resulting spectrum of secrecy, offers the potential to inform more sustainable and effective
policies for changes. In a similar way, it could also contribute to more robust research
findings than those that rely on tax haven lists. The detailed secrecy scores can also allow
researchers to explore whether particular types of secrecy play a particular role in
determining, for example, the benefits, or otherwise, of particular economic and financial
flows (e.g., is economic growth more or less likely to result from FDI made through
jurisdictions that allow secrecy about company ownership?).
Further extensions could include the development of country-specific rankings, recognizing
that different secrecy jurisdictions will be more relevant for some countries than for others.
The construction of such a ranking would rely on the same scoring of secrecy but would
substitute for GSWs with weights to reflect the importance of bilateral partner jurisdictions
for the country in question—so we might call this a bilateral FSI. Such an analysis carried out
for the Czech Republic, using the 2011 FSI, revealed a top five of Austria, United States,
Belgium, the Netherlands, and Panama.
This approach can identify country-specific vulnerabilities, revealing further detail about the
geography of financial secrecy. As Cobham (2014) illustrates for a range of African
countries, it is also possible to use other bilateral economic data in order to rank
vulnerabilities in other areas (e.g., to compare the risk a country faces in its direct and
portfolio investment). This kind of analysis could be particularly useful for countries with
limited resources to tackle illicit financial flows, by highlighting for policy makers the most
relevant secrecy jurisdictions for a given country and type of economic activity.
Abbott, K. W., and Snidal, D. 2000. Hard and soft law in international governance.
International Organization 54:421–56.
ActionAid UK. 2013. FTSE 100 tax haven tracker: The new data. Available online:
Avi-Yonah, R. S. 2000. Globalization, tax competition, and the fiscal crisis of the welfare
state. Harvard Law Review 113:1573–1676.
Basel Institute on Governance. (2013). Basel AML Anti–Money Laundering Index.
Blum, J. A., Levi, M., Naylor, R. T., and Williams, P. 1998. Financial havens, banking secrecy and
money-laundering. New York: UN Office for Drug Control and Crime Prevention.
Available online: http://amnet.org.il/joomla/attachments/UN-
Brown, G. 2009. Declaration on strengthening the financial system, G-20 London Summit. Available
Carrington, I., and Shams, H. 2008. Elements of an effective AML/CFT framework: legal,
regulatory, and best institutional practices to prevent threats to financial stability and
integrity. In Seminar on current development in monetary and financial law,October 23–27 October.
Vol. 5, ed. International Monetary Fund, 377-413. Washington, DC: IMF.
Christensen, J. 2007. Mirror, mirror on the wall, who’s the most corrupt of all. Tax Justice
Network. A paper prepared for the World Social forum 2007 in Nairobi.
CIA. 2013. World Factbook. Retrieved from https://www.cia.gov/library/publications/the-
Clark, G. L., and Monk, A. H. 2013. The scope of financial institutions: in-sourcing,
outsourcing and off-shoring. Journal of Economic Geography 13:279–98.
Cobb, S. C. 1998. Global finance and the growth of offshore financial centers: The manx
experience. Geoforum 29:7–21.
Cobham, A. 2012. Tax havens and illicit flows. In Draining development? Controlling flows of illicit
funds from developing countries, ed. P.Reuter, 337–72. Washington, DC: World Bank.
Available online: https://openknowledge.worldbank.org/handle/10986/2242.
______. 2013. Corrupting perceptions. Foreign Policy, July 22.
______. 2014. The impacts of illicit financial flows on peace and security in Africa. Tana
High Level Forum on Peace and Security in Africa Discussion Paper. Available online:
Cobham, A., and Loretz, S. L. 2014. International distribution of the corporate tax base:
Implications of different apportionment factors under unitary taxation. International
Centre for Tax and Development Working paper 24. Available online:
Coe, N. M., Lai, K. P., and Wójcik, D. 2014. Integrating finance into global production
networks. Regional Studies 48:761–77.
Cuellar, M.-F. 2003. The tenuous relationship between the fight against money laundering
and the disruption of criminal finance. Journal of Criminal Law and Criminology 93:311–466.
Dixon, A. D. 2014. The new geography of capitalism: firms, finance, and society. Oxford: Oxford
Eden, L., and Kudrle, R. T. 2005. Tax havens: Renegade states in the international tax
regime?*. Law & Policy 27:100–27.
FATF. 2000. Report on non-cooperative countries and territories. Paris: Financial Action Task Force.
Available online: http://www.fatf-
______. 2002. Report on non-cooperative countries and territories. Paris: Financial Action Task
Force. Available online: http://www.fatf-
______. 2012. International standards on combating money laundering and the financing of terrorism &
proliferation—The FATF recommendations. Available online: http://www.fatf-
FSI. 2013a. Mapping financial secrecy. Available online:
______. 2013b. The questionnaire sent to the ministries of finance. Available online:
______. 2013c. The questionnair sent to the financial intelligence unit. Available online:
Financial Stability Forum. 2000. Report of the working group on offshore centres. Basel: Financial
Stability Forum. Available online:
Gordon, R. A. 1981. Tax havens and their use by United States taxpayers: An overview. Washington,
DC: Books for Business.
Grossman, G. M., and Rossi-Hansberg, E. 2008. Trading tasks: A simple theory of
offshoring. American Economic Review 98:1978–97.
Haberly, D., and Wójcik, D. 2014. Tax havens and the production of offshore FDI: an
empirical analysis. Journal of Economic Geography, 15:75–101
Hampton, M., and Christensen, J. 2005. Tax havens of the world. In Tax us if you can, ed.
(pp.). Tax Justice Network, 36–37. Available online:
Hampton, M. P. 1996. The offshore interface: Tax havens in the global economy. Bassingstoke, UK:
Held, D., McGrew, A., Goldblatt, D., and Perraton, J. 1999. Global Transformations: Politics,
Economics, and Culture (1 edition.). Stanford, Calif.: Stanford University Press.
Hines Jr., J. R., and Rice, E. M. 1994. Fiscal paradise: Foreign tax havens and American
business. Quarterly Journal of Economics 109:149–82.
Hudson, Alan C. 1998a. Offshore Onshore: Re-Shaping the Financial Regulatory Landscape,
in: Martin, Ron (Ed.): Money and the Space Economy, London, 139-145.
Hudson, Alan C. 1998b. Placing Trust, Trusting Place: on the Social Construction of
Offshore Financial Centres, in: Political Geography 17: 8, 915-937.
______. 2000. Offshoreness, Globalization and sovereignty: A postmodern geo-political
economy? Transactions of the Institute of British Geographers 25:269–83.
ICIJ. 2014. Luxembourg leaks: Global companies’ secrets exposed. International Consortium of
Investigative Journalists. http://www.icij.org/project/luxembourg-leaks
IMF. 1993. Fifth Edition of the IMF’s Balance of Payments and International Investment
Position Manual (BPM6).
______. 2000. Offshore financial centers. IMF Background Paper. Washington, DC:
International IMF. Available online:
______. 2013a). Sixth Edition of the IMF’s Balance of Payments and International
Investment Position Manual (BPM6).
______. 2013b). Coordinated Portfolio Investment Survey. http://cpis.imf.org/
______. 2013c). International Investment Position.
______. 2014. Spillovers in international corporate taxation. IMF Policy paper. Available
Irish, C. R. 1982. Tax havens. Vanderbilt Journal of Transnational Law 15:449–510.
Isserman, N., and Markusen, A. 2013. Shaping the future through narrative: The third sector,
arts and culture. International Regional Science Review 36:115–36.
Janský, P. Forthcoming. Updating the rich countries’ commitment to development index:
How they help poorer ones through curbing illicit financial flows. In Social Indicators
Janský, P., and Prats, A. Forthcoming. International profit shifting out of developing
countries and the role of tax havens. Development Policy Review.
Johannesen, N., and Zucman, G. 2014. The end of bank secrecy? An evaluation of the G20
tax haven crackdown. American Economic Journal: Economic Policy 6:65–91.
Johns, R. A. 1983. Tax havens and offshore finance: A study of transnational economic development.
New York: St. Martin’s Press.
Karkinsky, T., and Riedel, N. 2012. Corporate taxation and the choice of patent location
within multinational firms. Journal of International Economics 88:176–85.
Kitchin, R., Linehan, D., O’Callaghan, C., and Lawton, P. 2013. Public geographies through
social media. Dialogues in Human Geography 3:56–72.
Kudrle, R. T. 2013. The future of offshore finance as a global policy problem. Deconstructing
Offshore Finance, Oxford, 2-3 September 2013.
Kudrle, R. T., and Eden, L. 2003. Campaign against tax havens: Will it last? Will it work?
Stanford Journal of Law, Business and Finance 9:37-69.
Lane, P. R., and Milesi-Ferretti, G. M. 2010. Cross-border investment in small international
financial centers. IMF working paper 38. Available online:
Levi, M. 2002. Money laundering and its regulation. Annals of the American Academy of Political
and Social Science 582:181–94.
Levin, C. 2007. A bill to restrict the use of offshore tax havens and abusive tax shelters to
inappropriately avoid federal taxation, and for other purposes law proposal. US Senate,
1st Session, 110th Congress, S.681. Washington, DC: US Senate. Available online:
Loopmans, M. 2008. Relevance, gentrification and the development of a new hegemony on
urban policies in Antwerp, Belgium. Urban Studies 45:2499–2519.
Lowtax.Net. 2008. Lowtax.Net, web site promoting secrecy jurisdictions.
Martin, R. 2001. Geography and public policy: The case of the missing agenda. Progress in
Human Geography 25:189–210.
Maurer, B. 2008. Re-regulating offshore finance? Geography Compass, 2:155–75.
McGauran, K. 2013. Should the Netherlands sign tax treaties with developing countries?
Available online: http://www.somo.nl/publications-
Meinzer, M. 2012a. Where to draw the line? Identifying secrecy for applied research. Tax Justice
Network. Available online: http://www.taxjustice.net/cms/upload/pdf/FSI_2012_Cut-
______. 2012b. The creeping futility of the global forum’s peer reviews. Tax Justice
Network. Available online: http://www.taxjustice.net/cms/upload/GlobalForum2012-
______. Forthcoming. Towards an international yardstick for identifying tax havens and
facilitating reform. In Global tax governance—normative and institutional issues, ed. Thomas
Rixen and Peter Dietsch. London: European Consortium for Political Research.
Murphy, R. 2008. Finding the screcy world: Rethinking the language of “offshore.” London: Tax
Research LLP. http://www.taxresearch.org.uk/Documents/Finding.pdf
______. 2009. Where are the world’s secrecy jurisdictions?. Tax Justice Network. Available online:
OECD. 1987. Issues in International taxation. International tax avoidance and evasion. Paris: OECD.
OECD. (1998). Harmful tax competition: an emerging global issue. Paris: OECD.
______. 2000. Towards global tax co-operation: Report to the 2000 Ministerial Council meeting and
recommendations by the Committee on Fiscal Affairs: Progress in identifying and eliminating harmful
tax practices. Paris: OECD. Available online:
______. 2001. Behind the corporate veil: Using corporate entities for illicit purposes. Paris: OECD.
Available online: http://www.oecd.org/daf/ca/43703185.pdf.
______. 2006. Tax Co-operation 2007: Towards a Level Playing Field. Paris: OECD.
______. 2007. Tax Co-operation 2007: Towards a Level Playing Field. Paris: OECD.
______. 2008. Tax Co-operation 2008: Towards a Level Playing Field. Paris: OECD.
______. 2008. OECD Model tax convention on income and on capital—An overview of available
products. Paris: OECD.
______. 2009a. Tax Co-operation 2009: Towards a Level Playing Field. Paris: OECD.
______. 2009b. A progress report on the jurisdictions surveyed by the OECD Global Forum in
implementing the internationally agreed tax standard. Progress made as at 2nd April 2009. Paris:
OECD. Available online: http://www.oecd.org/ctp/42497950.pdf.
______. 2010. Tax Co-operation 2010: Towards a Level Playing Field. Paris: OECD.
______. 2013. Action plan on base erosion and profit shifting. Paris: OECD. Available online:
Owens, J. 2009. Tax and development: Why tax is important for development. Tax Justice
Focus 4:1–3. Available online: http://www.taxjustice.net/cms/upload/pdf/TJF_4-
Palan, R. 1998. Trying to have your cake and eating it: how and why the state system has
created offshore. International Studies Quarterly 42:625–43.
______. 2002. Tax havens and the commercialization of state sovereignty. International
Palan, R., Murphy, R., and Chavagneux, C. 2009. Tax havens: How globalization really works
Ithaca, NY: Cornell University Press.
Peet, C. D. J., and Dickson, T. P. 1979. The United States. International Business Lawyer 7:280-
Picciotto, S. 1992. International business taxation, A study in the internationalization of business
regulation. London: Weidenfeld and Nicholson.
Pieth, M., and Aiolfi, G. 2003. Anti-money laundering: levelling the playing field. Basel: Basel
Institute on Governance.
Pinsent Masons. 2014. European Commission: agreements between Apple and Irish tax
authorities may have been illegal state aid. Out-Law.com. http://www.out-
Rixen, T. 2008. The political economy of international tax governance. Basingstoke, UK: Palgrave
Roberts, S. 1994. Fictitious capital, fictitious spaces: The geography of offshore financial
flows. In Money, power and space, ed. S. Corbridge; N. J. Thrift, and R. Martin, 91–115.
Shaxson, N., and Christensen, J. 2013. The finance curse: How oversized financial sectors attack
democracy and corrupt economics. Margate, UK: Commonwealth Publishing.
Sidaway, J. D., and Pryke, M. 2000. The strange geographies of’emerging markets’.
Transactions of the Institute of British Geographers, 187–201.
Swords, J. 2013. Michael Porter’s cluster theory as a local and regional development tool:
The rise and fall of cluster policy in the UK. Local Economy 28:369–83.
Tax Justice Network. 2013a. Financial Secrecy Index 2013 methodology. Available online:
______. 2013b. Financial Secrecy Index. Available online:
Transparency International. (2012). Corruption Perceptions Index 2012. Retrieved from
Unger, B. 2007. Money laundering, tax evasion, capital flight, tax havens, the rule based and
the risk based approach: Keep it simple. Paper presented at the Seminar on Money
Laundering, Tax Evasion and Financial Regulation, Transnational Institute, June 12–13,
Amsterdam. Available online:
UN Office on Drugs and Crime. 2007. An overview of the UN conventions and the
international standards concerning anti-money laundering legislation. Available online:
U.S. Department of State. 2013. 2013 International Narcotics Control Strategy Report.
U.S. Government Accountability Office. 2008. International taxation: Large U.S.
corporations and federal contractors with subsidiaries in jurisdictions listed as tax havens
or financial privacy jurisdictions. Available online:
U.S. House of Representatives. 1970. The United States House of Representatives report on the Bank
Secrecy Act No. 91-975. United States House of Representatives.
U.S. Senate Permanent Subcommittee on Investigations. 2013. Offshore profit shifting and
the U.S. Tax Code—Part 2 (Apple Inc.). Available online:
Valencia, M. 2013. Storm survivors. Special reports of The Economist on offshore finance. The
Economist, February 16. Available online: http://www.economist.com/news/special-
Van Hulten, A. 201. Remapping the fiscal state after the global financial crisis. Economic
Geography, 88(3), 231–253.
Varró, K. 2014. Making (more) sense of political-economic geographies of continuity and
change: Dialoguing across ontological divides. Progress in Human Geography doi:
Wainwright, T. 2013. Emerging onshore-offshore industries: The case of asset-backed
securities markets in Europe. Deconstructing Offshore Finance, September 2–3,
Oxford. Available online:
Warf, B. 2002. Tailored for Panama: Offshore banking at the crossroads of the Americas.
Geografiska Annaler: Series B, Human Geography 84:33–47.
Weyzig, F. 2012. Tax treaty shopping: Structural determinants of foreign direct investment
routed through the Netherlands. International Tax and Public Finance 20:910–37.
Wójcik, D. 2012a. Where governance fails: Advanced business services and the offshore
world. Progress in Human Geography 37:330–47.
______. 2012b. Accounting for globalization: Evaluating the effectiveness of country-by-
country reporting. Employment, Work and Finance working paper no. 08. Available
Woods, M., and Gardner, G. 2011. Applied policy research and critical human geography
Some reflections on swimming in murky waters. Dialogues in Human Geography 1:198–214.
World Bank. (2013). World Development Indicators. Retrieved from
Zoromé, A. 2007. Concept of offshore financial centers: in search of an operational
definition IMF working paper no. 87. Available online:
Zucman, G. 2014. Taxing across borders: Tracking personal wealth and corporate profits.
Journal of Economic Perspectives 28:121–48.