BACKGROUND PAPER

GOVERNANCE and THE LAW




Illicit Financial Flows and
Governance: The Importance
of Disaggregation

Peter Reuter
University of Maryland
Disclaimer

This background paper was prepared for the World Development Report 2017 Governance and the Law.
It is made available here to communicate the results of the Bank’s work to the development community
with the least possible delay. The manuscript of this paper therefore has not been prepared in
accordance with the procedures appropriate to formally-edited texts. The findings, interpretations, and
conclusions expressed in this paper do not necessarily reflect the views of The World Bank, its Board of
Executive Directors, or the governments they represent.

The World Bank does not guarantee the accuracy of the data included in this work. The boundaries,
colors, denominations, and other information shown on any map in this work do not imply any judgment
on the part of The World Bank concerning the legal status of any territory or the endorsement or
acceptance of such boundaries.
                        Background paper for World Development Report 2017
         Illicit Financial Flows and Governance: The Importance of Disaggregation*
                                                 Peter Reuter

                       School of Public Policy and Department of Criminology
                                       University of Maryland
Abstract
The international movement of money illegally or illicitly generated in developing countries has
become a major issue in the development agenda. Reducing illicit financial flows (IFFs) is now a
component of Goal 16 of the 2015 Sustainable Development Goals, as well as a staple of
declarations from the G7 and G20. The purpose of this Working Paper is to fit the IFF issue and
the reform agenda it has generated into the governance framework.
The new prominence of IFFs is largely the result of the publication of estimates that these flows
are extremely large (exceeding $1 trillion annually), dwarfing the inflow of Official Development
Assistance (circa $100 billion annually). IFFs have five major sources (bribes, tax evasion,
criminal enterprise earnings, corporate profit shifting and currency regulation evasion) and a large
number of channels for movement of the moneys (e.g., bulk cash smuggling, shell corporations,
informal value transfer systems and trade based money laundering). For advocacy purposes it has
been useful to aggregate all these and to produce black box estimates. For policy and research
purposes it is more appropriate to disaggregate. For example, the proposal to require country-by-
country reporting of corporate profits strikes at multinational corporate profit shifting and has no
effect, for example, on the decision to send the earnings of human smugglers to secrecy
jurisdictions. The same specific targeting holds for analysis of the consequences of IFF themselves.
Whereas the movement of corrupt incomes overseas is likely to reduce the elite’s commitment to
strengthening the rule of law, corporate profit shifting will have minimal effect on the development
of laws and institutions.
The reform agenda for IFFs has four major components: country-by-country reporting of profits;
listing of beneficial ownership of assets; automatic exchange of tax information and anti-money
laundering (AML) provisions. The paper examines the governance problems confronted by this
agenda, drawing primarily on the experience of the 25 year old AML regime. Though supported
by all governments and with a potentially effective enforcement mechanism (blocking non-
complying countries from access to the international banking system), AML rules have been
consistently poorly implemented. This reflects the poor alignment of incentives and institutions,
both private and public. Even governments that have been the principal architects of the Financial
Action Task Force (FATF) system, have failed to implement many of the FATF
Recommendations, while many major international banks have paid large fines for systematically
flouting the rule. The IFF reform agenda needs to more carefully assess those governance problems
in order to be effective.




*DavinO’Regan provided valuable research assistance. Lidia Ceriani provide helpful comments on earlier drafts.
Michael added insights to the final draft
I. Introduction


After decades of billion dollar scandals around long-serving dictators removing vast fortunes from
their impoverished nations, the broader phenomenon of which this is part has acquired a label:
Illicit Financial Flows (IFFs). The term encompasses the international transfer of moneys
generated by bribery, tax evasion and illegal markets. However, the I is for illicit, not illegal; thus
it includes a variety of legally ambiguous transfers such as corporate revenues shifted to lower tax
jurisdictions through, for example, transfer pricing abuse. The law may not be violated if the
developing country does not have laws and regulations governing transfer pricing but the
transaction is normatively unacceptable.
IFFs have been the object of much attention from high level bodies such as the G20 and G7. The
increased attention from these bodies toward three specific innovations (automated exchange of
tax information, registration of beneficial ownership and country by country reporting of corporate
revenues) has been partly inspired by the prominence of IFFs in the last decade. The StAR
Initiative of the World Bank and UNODC is another manifestation of these efforts, as is the
heightened interest in Anti Money Laundering (AML) efforts. IFFs show up in Goal 16 of the
2015 Sustainable Development Goals (SDGs).1 The release of the Panama Papers in April 2016
gave even greater popular prominence to this issue.2
IFFs are of policy interest partly because they drain capital and tax revenues from developing
countries; multinational corporation transfer of profits from high tax developing jurisdictions to
low tax jurisdictions is particularly egregious in respect to reducing tax revenues. Moving assets
out of the country illegally enables tax evaders, corrupt officials and successful criminals to better
protect them from seizure by the government. These transfers also give the same malefactors
access to kinds of consumption and investment not available in their home countries; villas in the
south of France and fast cars to drive on good roads. Thus the transfers increase the rewards of
evading taxes, taking bribes and smuggling humans or drugs.
However, IFFs also threaten even more fundamental aspects of development. The rule of law is
partly dependent on the support of economic and political elites. If the elites are able to shelter
much of their wealth outside the country, they have a lesser stake in the development of property
rights, itself a central element of the rule of law. Broader public trust can also be eroded by the
revelation of elites moving money out of the country.
The purpose of this Note is to develop a better understanding of the drivers of the phenomenon
itself in terms of governance and also of the governance challenge in trying to reduce IFFs. It deals
with the distribution of power as a factor in both aspects. The discussion does not examine the
drivers or consequences of the activities that generate most IFFs, such as bribery and tax evasion.
Thus a statement that IFFs from criminal earnings does not affect development institutions is not
a statement about the adverse effects of a large criminal sector, which may indeed have very

1
  Goal 16.4 “By 2030, significantly reduce illicit financial and arms flows, strengthen the recovery and return of
stolen assets and combat all forms of organized crime”
2
  The Panama Papers refers to a cache of 11.5 million documents released on-line by the International Consortium
for Investigative Journalism; the documents are from the files of a Panamanian law firm (Mossack Fonseca) that
assisted in the creation of shell corporations in many jurisdictions. https://panamapapers.icij.org/
serious economic and development consequences. The paper deals only briefly with the contested
issues of definition and measurement.
Analytic Contribution
The genius of the advocacy movement around the IFF issue is the emphasis on aggregates. The
large estimates (now on the order of $1 trillion per annum) and the breadth of the phenomenon
help give it prominence. However analytic progress will be made through disaggregation both of
sources and of channels, the methods used to move funds internationally.
Consider a standard breakdown of sources: bribes, tax evasion, criminal earnings, corporate profit
shifting and currency regulation evasion (Jansky, 2015). This paper argues that the combination
of factors that determine what share of each source is moved internationally varies. Each source
has different development consequences, both direct and indirect. The sources vary also in terms
of channels used to move the funds internationally. Thus they may respond differently to the
various control mechanisms, such as beneficial ownership registration, automatic exchange of
information and AML. The implication is that for both research and policy purposes it is necessary
to disaggregate IFFs rather than to develop an agenda that treats them as a single phenomenon.
The remainder of this Introduction deals briefly with conceptual issues and the existing literature.
Section II identifies the many parties generating IFFs. That is followed by sections on the
importance of disaggregating IFFs for purposes (a) of understanding their drivers, (b) of
understanding their consequences for the development process and (c) for assessing the prospects
of various control proposals. The last of these sections integrates the distinctions between power
and capacity; functions and form; rules and the rule of law. That is followed by an examination of
the limitations of the global AML regime, as a useful analogy for purposes of considering why
current ambitious proposals might fail. The final section concludes briefly about the path forward
for research.
Literature Review: IFFs
The literature contains varying definitions of IFFs. Epstein (2005, p.7) defines IFFs as “capital
taken abroad in a hidden form, perhaps because it is illegal, or perhaps because it goes against
social norms, or perhaps because it might be vulnerable to economic or political threat”. So what
exactly makes these financial flows illicit? There are two leading interpretations: normative and
legal. The normative interpretation suggests that financial flows become illicit “not only because
they hinder development, but also because they are deemed illegitimate from the perspective of an
existing consensus about the social (developmental) good” (Blankenburg and Khan 2012, p. 32).
Conceptually this may be attractive but it defies operationalization. The legal interpretation, upon
which the empirical literature on IFFs is predominantly built, suggests that IFFs refer to money
that is “earned, transferred or used” in contravention to existing law (Kar and Freitas, 2012). That
is the definition used by most commentators on the issue.
The publication of Raymond Baker’s Capitalism’s Achilles Heel in 2005 is the starting point for
IFFs. Since then the original research on the topic has almost exclusively been on estimating its
scale. Global Financial Integrity, a non-profit organization founded by Baker after publication of
his book, has produced a stream of reports on the extent of IFFs both globally and for individual
countries3. There are almost no competing estimates4, though the GFI methodology has been
frequently critiqued, most specifically in Nitsch (2016). The research on the drivers, consequences
and policy aspects of IFFs has been minimal. There is more research on those aspects of related
phenomena, particularly capital flight and its macroeconomic consequences but we found no effort
to assess the relevance of that research for IFFs. Indeed, there is controversy as to the relationship
between capital flight and IFFs (e.g. Ajayi and Ndikumana, 2014).
What is new is the label and a concerted effort to specifically target the IFFs, separately from
efforts to go after the activities that generate the funds: corruption, tax evasion etc. As noted
already, stemming IFFs has also been the inspiration for efforts to develop a universal list of
beneficial owners of all assets, country-by-country reporting for corporate accounts and automatic
exchange of tax information across countries, as well as to strengthen AML efforts (e.g. Cobham,
2014). 5 No observer would claim that these efforts had yet reduced IFFs but they represent
potentially important developments.
Little attention has been given to IFF composition or the factors affecting variation across countries
and over time. Baker’s original estimate attributed only about 5% to corruption, with most of the
flows coming from “commercial activities”; criminal earnings accounted for about one third.
Consistent with that are the conclusions of The High Level Panel on Illicit Financial Flows from
Africa (2015), which asserted that only 5% of African IFFs came from corruption and that most
came from corporate activities; unfortunately it offered no empirical support for those contentions.
Otherwise the estimates have been of the black box variety that give no insight into the specific
sources. The GFI estimates are similarly blind with respect to the channels used, except that the
share involving trade mispricing was estimated to be over 80% in the 2015 publication (Kar and
Spanjers, 2015), compared to less than 44% when estimates were made in 2011 (Kar and Freitas,
2011), perhaps as a consequence of a change in methodology.6


II. The Generators
We begin by noting that a variety of actors generate IFFs. For expositional purposes we identify
four distinct groups, though some individuals may participate in more than one of these activities.
        (1) Multinational corporations. Possibly acting within the law, perhaps within law that was
procured through corrupt acts or overt lobbying (e.g. Global Witness, 2006), MNCs may move
funds from developing countries, to low tax jurisdictions through Transfer Pricing Abuse (TPA)
or other methods (Leite, 2012). The TPA may be legal simply because many developing nations
have, at least until recently, not had rules governing transfer pricing (Leite, 2012).


3
  See for example Kar and Freitas (2011) and Kar (2012) .
4
  Ndikumana has produced a number of estimates of a similarly defined measure for Africa; see e.g., Boyce and
Ndikumana (2012) and Ajayi and Ndikumana (2014)
5
  See for example the 2015 G7 communique: “We recognize the importance of beneficial ownership transparency
for combatting tax evasion, corruption and other activities generating illicit flows of finance and commit to
providing updates on the implementation of our national action plans. We reiterate our commitment to work with
developing countries on the international tax agenda and will continue to assist them in building their tax
administration capacities.”
https://sustainabledevelopment.un.org/content/documents/7320LEADERS%20STATEMENT_FINAL_CLEAN.pdf
6
    Nitsch (2016) shows that shifts in the implementation details by GFI over the years.
       (2) Corrupt officials. They may earn their bribes through any of a variety of activities such
as charging for rights that should be freely available, improper contract awards, extortion for not
imposing proper criminal sanctions.7 It is assumed that only very senior corrupt officials steal
enough to make it worth moving the resulting funds overseas.
       (3) Tax evaders. The activities that generate the funds are legal; the illegality is solely in
the concealment for tax purposes. The evaders can be individuals or privately held companies.
         (4) Criminal entrepreneurs. Drug traffickers, human smugglers and other illegal market
enterprises can generate large revenues. There is a reasonable question as to whether these generate
so much outflows from developing countries as illicit inflows, since the bulk of their earnings are
generated in high price rich markets; Thoumi and Anzola (2012) provide an analysis pointing to
this for the Colombian cocaine trade. 8
The first group, MNCs, have distinct incentives and methods that need to be treated separately.
The following discussion primarily covers individuals in the categories (2) to (4).
Note that we have omitted the list of actors those who facilitate the flows at the destination,
including: governments that support secrecy laws; regulatory/supervisory agencies that fail to
enforce compliance with AML laws and rules; businessmen who collaborate in the mispricing of
trade transactions, to name just a few. In effect, we assume that there is a perfectly elastic supply
of opportunities for making illegal international transactions, once a source country participant has
decided to undertake one. Section V will discuss how this supply might be constrained by future
actions.
The analysis here takes the incentives for the sources as given. For example, tax evasion will be
higher if tax rates rise, cet. par. We ignore that effect of tax rate variation on evasion: our interest
is only in whether a rise in tax rates affects the probability that a dollar held by a corrupt official,
tax evader or criminal entrepreneur is held domestically or sent overseas. For the moment the
analysis ignores possible feedback; preventing the export of earnings to countries where the fruits
might be more enjoyable may reduce the incentives for tax evasion, bribe taking or criminal
enterprise.
Evasion of currency controls for purchase of legal services or goods overseas is also an element of
IFFs, as this constitutes an illegal use of legal income. Consider the Bangladeshi doctor who wants
her daughter to attend private school in England ; she will have to evade controls which cap
purchase of foreign exchange in one year to $1,000.9 This will be a component of estimated IFFs,
even if the income was generated legally, properly taxed and is used for an entirely worthy purpose.
Currency controls in effect generate yet a fifth source of IFFs; the wholly legal who is evading a
specific rule with a possibly questionable purpose/rationale. Even if those controls are perceived
as being in the best interest of the country, which is inconsistent with professional opinion (Engel,
2012), those who disagree may believe the controls lack legitimacy and thus act against them. The

7
  For a classification of corrupt acts see EU Criminal Law Convention on Corruption
http://www.coe.int/en/web/conventions/full-list/-/conventions/treaty/173
8
  A new OECD study shows variation among West African illegal markets in the share that is outflow or inflow
(OECD, 2016). Soudijn and Reuter (2016) shows large inflows to Colombia from the smuggling of cocaine to the
Netherlands.
9
 See https://www.bb.org.bd/openpdf.php There are other rules and regulations governing foreign exchange for
health/medical reasons.
process by which the laws were enacted may be fair but the outcome may be perceived as unfair
by losers.


III. Factors influencing the generation of IFFs:
The focus is on identifying factors that influence the decision to send money out of the country for
each of the sources. We suggest that there will be some variation, factors that are more important
for one source than for others.
Incentives: Individuals
There are multiple incentives for moving illegally generated money internationally rather than
holding it domestically.
 1. Portfolio diversification. Whether or not there is any taint of illegality to the generation of the
funds or to the transfer abroad, the owner may wish to reduce the riskiness of his portfolio by
acquiring assets in other countries whose economic fortunes are only weakly correlated with those
of his home country. The tax evader, bribe taker etc. is assumed to seek to maximize returns on
investment.10 Similarly, the failure of the state to provide a stable investment environment will
encourage exit of capital, legal or illegal. A stable investment environment is one that generates
predictable exchange rates, interest rates, returns on investment and tax rates, all necessary for
investor confidence.
These factors are part of general macroeconomic management, not specific policies directed at
IFFs. However, they suggest that IFFs will increase, along with other outward capital flows, if a
developing country’s macroeconomic condition deteriorates or if investment rules change in ways
that reduce the attractiveness of the home country relative to others.
 2. Hiding assets from government. Funds held overseas are presumably more difficult for the
domestic government to track and confiscate. There are more legal hoops for the government to
jump through in both phases; establishing ownership of assets and then persuading a foreign court
to hand them back. (Manton and Daniel, 2012; Chene, 2006).
Again there are multiple factors, not specific to IFFs, that may affect the extent to which corrupt
officials etc. will seek to hide assets overseas. One is the transparency of the domestic financial
system. For example, if it is easy to create domestic nominee bank accounts or shell corporations
for real estate holdings, then foreign assets may not be more attractive in that sense. Moreover,
foreign assets can be harder to liquidate or access in an emergency compared to assets concealed
at home. Thus opacity of the domestic system of financial regulation may reduce incentives for
sending funds out of the country.
More competent law enforcement will also increase the incentive to move money elsewhere. A
country in which law enforcement is weak poses little threat of seizure of illegally gained assets.
        3. Political stability. The prior two paragraphs assume that the government has legitimacy
and follows legal procedures. In many developing countries there have been successive predatory
governments, kleptocratic regimes whose primary goal is the enrichment of senior government
officials. Under those circumstances corrupt officials specifically may seek to avoid confiscation

10
  This is a simplifying and conventional assumption but profit should be broadly interpreted. Perhaps an important
source of utility for the miscreant is seeking safety from people as capricious and rent-seeking as him/herself
of assets by a successor government. Political instability in the predatory state may mean that even
when my party is in power I worry that the next predator might seize my illegally generated assets.
11
   Another possibility is that corrupt elites in a stable country, with a long-term view of the future,
could reach agreement that there will be no seizures of corrupt assets with turnover of government;
the agreement might only be tacit. 12 This factor seems less relevant to criminal entrepreneurs and
tax evaders.
         4. Currency controls. Restrictions on repatriation of corporate profits will of course
increase the incentives for corporate profit shifting. Clearly the stricter the currency controls for
individuals, the greater the incentive to violate them for purposes that might otherwise be regarded
as legitimate. Will currency controls affect the behavior of the other generators of IFFs (i.e. those
whose income sources are illegal)? Such restrictions make it more difficult to move assets overseas
legitimately and encourage illegal methods instead but an assumption of the literature is that illegal
income is not moved legally anyway, since that would provide information for the government, so
the effect might be slight.13
Incentives: Multinational Corporations
A number of scandals in Europe, such as those involving Starbucks and Google in the UK14 have
revealed the power of large multinational corporations to manipulate the tax systems of even the
most powerful states (Murphy, 2012). It would be surprising if these and other multinational
corporations were not able to manipulate the much weaker systems of developing countries to
similarly reduce their tax burdens. Mispricing of internal transactions, whether they be the prices
of intermediate goods (raw copper or cut flowers) or payments for intangibles such as good will
or patents, are standard means for accomplishing this. What factors are likely to increase the extent
of such manipulation?
Eden (2012), reviewing the slender empirical literature on the topic, identifies three distinct groups
of factors: tax rates, foreign exchange restrictions and political risk. Higher tax rates, more
restrictions on the convertibility of the local currency or on the amount that can be repatriated in a
year, will lower confidence in the capacity of the domestic government. Political instability will
also have an effect.15 All will encourage multinational corporations to move income to other
countries through manipulation of internal transactions.


IV. The Consequences of IFFs
Institutional consequences


11
   This generates an hypothesis. As a kleptocrat ages, IFFs from this source should rise as probability of new regime
increases. Robert Mugabe’s continued dominance in Zimbabwe into his 90s suggests the dangers of relying on age
for these purposes..
12
   Kenya may offer an example of such an agreement, as there have been three transitions without seizure (Jomo
Kenyatta, Daniel Arap Moi, Mwai Kibaki, Uhuru Kenyatta).
13
   If criminal income is laundered domestically, so as to become a de facto legal asset in the source country, then it
may be sent overseas legally. It would still conceptually constitute an IFF but it would not be included in the GFI
estimates.
14
   Murphy (2012).
15
   Whether shifts from one kleptocrat to another has an effect depends on the extent to which the MNC’s contracts
are tied to the current regime.
The mechanisms connecting the component flows of IFFs to elite and popular support for the rule
of law and property rights differ. Thus each source may have specific institutional consequences.
Bribes
Bribes (shorthand for all forms of corrupt earnings) sent overseas reduce elite support for the
development of the state because less of the elite’s wealth is dependent on the domestic property
rights regime. This assumes that bribery is sufficiently common that corrupt officials and their
associates make up a significant component of the economic/political elite. In many countries the
state is the source of much of the wealth, e.g. through the allocation of monopoly franchises for
telecoms or financial services.16 Government service itself is often a path to wealth in such
societies.
Popular support will be affected only to the extent that the foreign placement of bribes becomes
known.17 In some instances the revelation of large scale corrupt wealth occurs through the
discovery of foreign holdings as the result of the initiative of the country in which the stolen assets
are held. (US DOJ, 2014; Chrisafis, 2012). This raises an empirical question: does the removal of
the wealth create anger additional to that arising from the revelation of the corruption itself? There
is no systematic evidence on this.18
There is a possible secondary effect because it may be harder to monitor foreign funds as compared
to domestic funds used for financing of political campaigns. Election results may be bought
through the use of foreign and unregulated funds.19 There are also spill-overs to other countries. A
kleptocrat may use foreign funds to help support friendly rulers in other countries, to ensure that
if he needs to flee there is a friendly refuge in the region.20
Tax Evasion
For direct effects, the analysis of tax evasion (whatever the tax: income, sales, customs, VAT) is
probably identical to that for bribes. In very corrupt states there is often a commingling of the

16
  Gandhi and Walton (2012). T find that half of all billionaires operate in “rentier” sectors of the economy, often
through state promoted monopolies/oligopolies. The Economist (2014) builds off of this to look at trends in other
countries. Reuters (2014) focuses on television and telecom monopolies in Mexico.
17
   Michael Levi (personal communication) argues that popular support is eroded by suspicion of such transfers;
optimistically one might hope that reality encourages suspicion.
18
   The Panama Papers provided details of foreign accounts by officials in many developing countries. We examined
major newspapers in Nigeria(Vanguard; Premium Times), Kenya(East African; the Nation), and South Africa (Mail
& Guardianbut did not find many op/eds or columns aboutthem. There was little reference to the Panama papers and
commentary was generally defeatist in tone: consider this from a Nigerian paper “what does it matter that further
evidence of how he laundered his loot in assets transferred to shell companies hidden in a tax haven abroad should
have newly surfaced, thanks to the Panama Papers leak?” In other words, the revelations don’t mean much given the
level of looting and corruption that is already taking place: http://allafrica.com/stories/201604130104.html
19
   Consider the following account of events before the 2015 Nigerian election. ’Available statistics indicate that
Nigeria has become the largest importer of U.S. dollars,’ the regulator said in Friday's notice …. Corruption in the
build-up to Nigeria's 2015 election is partly responsible for the increase, Governor Lamido Sanusi said at the central
bank's Monetary Policy Committee meeting on Tuesday, adding that it is "absolutely wrong" for bureaux de changes
to buy hundreds of millions of dollars without accountability.” http://www.reuters.com/article/nigeria-
moneylaundering-idUSL5N0HN1EL20130927
20
   “[F]unds held abroad can be used to curry favour in other countries which might later provide a safe haven if the
kleptocrat has to exit.” (OECD, 2014; p.31)
economic elite, who have enough income to motivate tax evasion, with the political elite who
might have large incomes from bribes. It is possible that the evasion of taxes is one of the principal
protected activities for which officials obtain bribes. 21
What is the effect of revelation that evaded taxes have been taken out of the country? The Panama
Papers and the Swiss Leaks (2015 disclosure of secret accounts held by HSBC in Switzerland)
have provided just such revelations, on a very large scale for some countries such as Argentina
and Mexico.22 No systematic study has examined the effects on popular support for the state; they
may do little more than confirm widespread cynicism about elites in those countries where elite
corruption is already well known but have an important effect in others where elite ethics are less
suspect.
Criminal Incomes
We assume that though drug dealers and human smugglers make payments to corrupt police and
border officials, they themselves are not in the same political and economic elite.23 Neither before
entering drug dealing nor afterwards do they exert influence on the political system beyond
obtaining protection of their criminal activities. Nor are the public likely to see their export of
earnings as representing a failure of that system distinct from their capacity to operate criminally.
Thus there are no direct or indirect effects on institutional development. That is not to say that
higher criminal incomes are not detrimental; it is just that their removal from the country has no
additional consequences.
Currency Rule violations
In nations with currency controls, the purchase of foreign goods or services beyond what the
government permits is included in the standard definition of IFFs. To the extent that the elite is
able to obtain good education for their children overseas they will invest less effort in improving
the schooling in their home country. That is specific to this particular example of a currency
violation. It is not apparent that these payments will substantially affect their attitudes toward the
strengthening of the rule of law or the property rights regime.
The popular response to revelations of these payments may be an undermining of trust in the
political system. In this example, the rich are able to obtain advantages for their children
unavailable to others. The violation of the currency control itself is the source of anger; neither
what is purchased nor the source of the income is seen as problematic. The response might be
different and harsher if the violation involved purchase of luxury goods.

21
   Note though that many developing countries make little effort to tax high wealth individuals, so that bribes may
hardly be necessary. For example, the International Center for Tax and Development reports on a study of Uganda
which “examined lawyers from top commercial law firms and found that out of a group of 60, only 12 had paid
personal income tax in 2012, and only 13 did the following year. The team also investigated individuals who paid
large import duties. They found that although 12 people paid more than $180,000 in customs duties in 2014, none of
them had paid personal income tax.” http://www.ictd.ac/blogs/entry/why-african-govs-should-tax-the-rich
22
   See for example, https://www.icij.org/project/swiss-leaks/new-countries-seek-hsbc-data-and-undeclared-cash
23
   In some countries that assumption is violated. Consider Tajikistan, which generates a substantial share of GDP
through heroin trafficking from Afghanistan to Russia and Eastern Europe (Paoli, Greenfield and Reuter, 2010). It is
widely alleged that the heroin trade is operated by high level officials and that this constitutes a substantial share of
corrupt officials’ income. In Myanmar during the 1970s and 1980s, when the heroin trafficker Khun Sa controlled
that trade, many senior officials may also have had direct involvement. Impressionistically, direct control of the drug
trade by officials seems to be true of only a few countries.
Corporate Profit Shifting Abuse
Assume that a multinational corporation violates norms of transfer pricing as laid down by the
OECD but that there are not yet domestic laws implementing the OECD guidelines (as is true in
many developing nations). Thus there is no domestic illegality. 24
MNC’s have an interest in the strength of the rule of law in countries in which they operate. Is that
affected by the extent to which they pay taxes in those countries? They have many other
motivations for caring about the rule of law. It affects for example how appropriately they are
regulated and how fairly and efficiently disputes with workers and contractors are settled and
whether well connected local rivals are favored. Thus reducing recorded domestic profit would
seem to have only a modest effect on the commitment of the corporation to the development of
the State. A counter-argument is that often a small number of MNCs account for a large share of
total MNC contribution to GDP and in particular exports. In those situations the MNCs may prefer
a weak and biddable state to one that is competent, predictable and rule enforcing.25
Popular response to revelation of profit shifting might be important. It may create a well-justified
suspicion that the State has been incompetent and/or corrupt in its dealing with the corporation.
This is particularly likely to be true in nations in which a small number of MNCs account for a
large share of exports and/or exploitation of natural resources.
Figure 1 attempts to capture the hypothesized effects, identifying the principal underlying
mechanisms. The entries should be regarded as speculative; they lack either a strong empirical or
theoretical base.




24
   It may be a violation in the home country of the MNC but a cursory examination of the Canadian law suggested
that it did not require application of the OECD arms length principle in another country.
25
   Many historical examples can be offered. United Fruit thoroughly corrupted governments in Guatemala and
elsewhere in Latin America over decades (Kurtz-phelan, 2008), just as Firestone did in Liberia (Miller and Jones,
2014).
Figure 1. Mechanisms and institutional outcomes of IFFs
Investment and Tax Base Effects
It is not clear that there is as much variation in the direct effects of the different sources. The
relevant parameters are the savings and consumption patterns of each type of generator.
Unfortunately there is no systematic information about the consumption patterns of corrupt
officials, tax evaders or criminal entrepreneurs. Relying on little more than stereotypes, and taking
into account their risk-taking personalities, we hypothesize that criminal entrepreneurs, with
shorter life expectancy and at risk of spending a portion of their career in prison (at least in some
countries) are likely to have lower savings rates than the others. Criminal entrepreneurs may spend
more on transitory pleasures rather than acquire valuable domestic capital goods (lavish housing
and furnishings) that the corrupt political elite often flaunt. What occasionally shows up is how
criminal entrepreneurs and corrupt officials spend and invest money overseas. 26 Domestic
corruption cases may throw up interesting information in local newspapers but we have been
unable to find any effort at systematic compilation of anecdotes to provide even a data base of
stories. Bribes, though perhaps legally subject to tax, are not in fact likely to have taxes paid, even
if kept in the country
Again, corporate profit shifting is distinct; the income shifted as a result is not subject to domestic
taxes and thus is fully subtracted from the national tax base. Whether this affects investment is
much harder to determine; foreign investors may favor countries whose effective tax rates are
lower because of the ability to move a larger share of profits to other jurisdictions.
Figure 2 attempts to summarize the factors that influence the decision to move specific sources out
of the country of origin and the mechanisms linking them. These should be treated as crude
hypotheses intended to generate discussion. They are in no sense empirically tested links.




26
  Unfortunately the information about expenditures is limited to those assets that are left at the time the
kleptocrat is pursued; expenditure patterns are unknown. Moreover, with the exception of Sani Abacha, the assets
discovered from the most prominent kleptocrats, such as Marcos and Mobutu, have been very small relative to the
wealth that they are supposed to have stolen from their countries. For example, it is claimed that Mobutu stole $4
billion from Zaire (http://www.washingtonpost.com/wp-srv/inatl/africa/zaire/swiss.htm but the total assets
identified for recovery amount only to $37 million.
Figure 2. Factors determining the direct effects of specific IFF sources
V. Channels
A considerable variety of methods are available for an individual wishing to move money out of a
country illegally. The many Typologies Reports of the Financial Action Task Force provide
illustrations of many of these methods, from the unsophisticated (bulk cash smuggling) to the very
elaborate (e.g. Trade Base Money Laundering).27 While any of them could be used for any of the
sources, Reuter and Truman (2004) suggest that each source seems to favor specific channels over
others. Reuter and Truman’s empirical analysis is based on the descriptions provided in the
Typologies reports available at the time, and should be taken as merely indicative for a variety of
reasons. Table 1 reproduces the Reuter and Truman findings, using Typologies reports up to
2004.28




27
   The Typologies Reports can be found at http://www.fatf-
gafi.org/publications/methodsandtrends/?hf=10&b=0&s=desc(fatf_releasedate)
28
   Amongst the limitations of the Typologies Reports as a data base are: (a) the highly schematic presentation of
examples, which does not permit assessment of the appropriateness of classification decisions by the report authors
(b) the dominance of developed countries as a source of data and (c) the non-random sampling of cases for inclusion
in the reports.
Consider the distribution of channels for the two sources with the most observations, drug
smuggling and fraud. Drug smuggling makes substantially more use of simple methods such as
cash smuggling, money orders, cashiers’ checks and structured deposits (15% vs. 8%), whereas
fraud shows more reliance on shell companies (10% vs. 4%). Given the non-representative nature
of the sample and the small number of cases, not too much stress should be given to the specific
numbers here. 29
FATF (2011) similarly noted that corrupt funds are likely to flow through different channels than
those derived from drug trafficking or other criminal markets. Drug revenues are initially
generated in cash transactions; the central task for the drug distributors is to manage this large and
regular flow of currency (e.g. Woolner, 1994 and van Duyne, 2002). The laundering problem for
corrupt high level officials is distinct; payments are likely to be in electronic form, though cash is
certainly not unknown.30 High level officials, particularly heads of state, are also likely to have at
their disposal methods that are not accessible to others. Sani Abacha, the president of Nigeria,
periodically had the Central Bank of Nigeria wire funds into his personal accounts in Jersey
(CITE). The FATF documents on Trade Based Money Laundering (e.g. Asia and Pacific Group.,
2012) offer many examples of different schemes; in none of the examples is the predicate crime
corruption.31
This variation matters because the efforts to control IFFs, discussed below, are likely to have
differential effects across channels. Country-by-country reporting of corporate profits aims one
specific target; the prevention of abusive corporate tax shifting, It will do nothing to effect for
example, trade based money laundering or bulk cash smuggling. This is taken up further in the
next section.


VI. Controlling IFFs
IFFs have both a source country and a destination country; they often also involve transit countries.
Destination countries are primarily wealthy nations, since they offer secure facilities, a wide range
of investment opportunities and political stability. Transit countries are often the small secrecy
jurisdictions such as the British Virgin Islands or Jersey that are heavily dependent on handling
these transactions; note though that recent efforts to measure financial secrecy have put
Switzerland, Hong Kong and the United States at the top.32 The flows can best be stemmed by
internationally co-ordinated actions involving both source and destination countries, as reflected
in the resolutions of the G20 on the issue of IFFs.



29
   Other studies also show the dominance of cash in the drug economy; see for example van Duyne and Levi (2005)
and EMCDDA & Europol (2016).
30
   FATF (2011) devotes a section to cash transactions but these seem small scale in comparison to the totals for the
kleptocrats, typically hundreds of thousands of dollars compared to the tens or hundreds of millions stolen. It notes
that “the grand corruption cases would not be expected to have significant amounts of cash. A cash payment to a
PEP would break the chain of bank records, of course, but it would require the PEP to run the gauntlet of AML/CFT
controls designed to combat placement of illegally-derived cash into the system….” (p.24)
31
   Another oddity of this study is that some of the examples involve complex trade document falsification not for
laundering purposes but as the means of carrying out a fraud.
32
   See Tax Justice Network Financial Secrecy Index http://www.financialsecrecyindex.com/introduction/fsi-2015-
results
There are five main interventions explicitly aimed to reduce IFFs.33
        1. Anti-money laundering (AML) laws and programs. These attempt to (a) prevent
offenders from turning illegally generated moneys into legal funds that can be used for any
investment or consumption purpose (b) use the effort to launder moneys to apprehend and punish
offenders, including those professionals who help the primary offenders move, conceal or
transform the proceeds of crime. Almost all nations have in place laws and institutions that have
received some degree of approval from the Financial Action Task Force (FATF), a G7-created
entity that sets the rules for such matters.
        2. Stolen asset and recovery procedures. A range of laws and programs aim to facilitate the
return of assets stolen from national coffers by corrupt officials.
       3. Automatic exchange of information between countries. Under these agreements, each
country’s banks are required to provide the other country with information about accounts that are
held by that country’s residents.
4. The development of new rules regarding country-by-country reporting of corporate profits,
intended to prevent corporate profit-shifting abuses.; see OECD (2015).
5. The development of beneficial ownership registries, which would ensure that ownership of
financial assets, and a broad array of real assets, could not be hidden.
Domestic IFF control efforts in developing countries can have an influence but this is likely to be
limited (Levi, 2012). It appears that most revelations of IFFs, in particular related to stolen assets,
are the results of investigations initiated by the ultimate destination countries rather than by the
developing countries themselves or the small transit countries, notably the secrecy jurisdictions.
This is hardly surprising. The assets are often, perhaps mostly, stolen by members or associates of
the ruling elite; the government is thus unlikely to initiate a recovery request unless there has been
a falling out between the corrupt official and the rest of the government 34 or following a media
expose that may prompt a lukewarm or even corrupted request for assistance in name only.
Impressionistically, Switzerland, the United States and the United Kingdom, despite their high
scores on the financial secrecy index, account for a large share of all the asset recovery cases
involving corrupt officials. This may be a consequence of the fact that so many of the stolen assets
end up at least partially in those three countries.35 In the StAR data base, these three are identified
as accounting for just over half of all the cases.36 There is no more complete data base that would
permit assessment of whether Switzerland, the UK and US should be viewed as doing more than
their proportionate share of asset recovery efforts; the relevant metric might be the share of global
financial assets or the share of a broader class of convertible assets.

33
   See for example the 2015 Financing for Development commitment on IFFs: to “substantially reduce and
eventually eliminate IFFs” by i) combatting tax evasion and corruption through strengthened national regulation and
increased international cooperation; ii) reducing opportunities for tax avoidance; and iii) enhancing disclosure
practices and transparency in both source and destination countries, e.g. through transparency in financial
transactions between governments and companies to relevant tax authorities.”
34
   Even under those circumstances, the government takes a risk that the now disgraced official can be an effective
informant against others who may have stolen assets.
35
   The high degree of NGO activism in the UK and the USA may also play a role.
36
   As of May 23 2016, in the field” Jurisdiction of asset location”: there were 47 jurisdictions and 236 cases. Top 5
jurisdictions or location by count of cases: USA 65, Switzerland 34, UK 26, Jersey 9, France 7.
As already noted, the controls are likely to be quite channel-specific. The withdrawal of the 500
Euro note by the European Central Bank, scheduled for 2018, will make money laundering more
difficult for drug smugglers, who have to transfer large quantities of cash across international
borders. 37 However, it is unlikely to have a consequence for MNC’s evading domestic taxes
through transfer pricing abuses.


VII. The Governance Dimension38
A goal of this paper is to situate the problem of IFFs, and the efforts to curb them, in the framework
of governance challenges. Every aspect of IFFs has a connection with the problems of poor
governance.
We offer three examples representing the variety of connections.


Think of function, not only form
As Section 1 of the 2017 World Development Report explores, importing forms is not enough to
change the facts. Kenya is often cited as a major destination for the proceeds from piracy in the
Indian Ocean and a key transit point for terrorist funds to neighbouring Somalia (Yikona, 2013).
With the goal of fighting illicit financial activities, in particular money laundering and terrorism
financing, Kenya established the Financial Reporting Centre (FRC) in 2012. While creating this
institution enabled the country to get off the Financial Action Task Force’s (FATF) list of countries
that might be sanctioned for non-compliance, little seems to have changed. There is substantial
evidence of high level corruption in the government in recent years (Persson, Rothstein and
Teorell, 2013), yet no cases have been brought against senior officials or intermediaries as a result
of violations of the money laundering regulations. To make the change meaningful, the
administration would have had to pursue many of its own senior members or those of the previous
administration with whom there are complex ties.


Think power, not only capacity
When assessing the issue of IFFs, the lack of capacity of developing countries in managing
complex laws and regulations is often stressed as the cause. This is for instance the case of
multinational corporations and abusive transfer pricing; some developing countries do not have
any transfer pricing rules, thus ensuring that the multinational corporation can transfer as much of
its profits elsewhere as it wishes. However low capacity is not necessarily the inevitable result of
low per capita income. Multinational corporations are often the initiating actors in these matters,
providing corrupt incentives for the government to not develop these rules (e.g Global Witness,
2006). At the source of low capacity, there is often a power issue: it is in the interest of someone
to keep the capacity low because it allows them to extract some rents (Leite, 2012). Lack of
political will has indeed been recognized as more difficult to overcome than any legal, institutional

37
   Soudijn and Reuter (2016) studied the records of cash smuggling organizations servicing Colombian cocaine
traffickers in the Netherlands. All converted the cash receipts into 500 Euro notes at considerable expense (3%)
before packaging for shipment to South America.
38
   This section was co-authored by Lidia Ceriani.
or operational issues in this context (Stephenson et al. 2011)


Think of the role of law, not only the rule of law
Promoting compliance with international laws and regulation can be manipulated by incumbents
for gaining legitimacy, while continuing to do business as usual. One important requirement of the
FATFi Anti Money Laundering (AML) regime is to facilitate requests for information and co-
operation from other nations pursuing money laundering cases involving the country’s nationals
(Recommendations 35-40). Most nations have in fact adopted statutes and regulations that conform
to the FATF rules. However, there are many developing nations in which the authorities have a
long record of dragging out the process of co-operation so long that in fact the laws are effectively
nullified (Chêne, 2006). For example, take the case of mutual legal assistance (MLA), which is an
essential tool in cases of corruption with international aspects, including foreign bribery and
money laundering. As pointed out by Terracol and Dell (2015), countries might be reluctant to
execute an MLA request when there is no credible commitment from the authorities to use the
information and evidence provided for convicting the corrupt individual. They may also fear in
some cases that the purpose of the request is simply corrupt rent extraction.
The activities of the head of the Nigerian Economic and Financial Crimes Commission (EFCC) in
the early part of the last decade illustrates positively this need to consider the incentives of
international partners (Ribadu and Center for Global Development, 2010). The EFCC director was
able to get US/UK support after some early successes and through direct personal efforts to
develop a rapport with counterparts: “The EFCC had to prove it was serious about financial crime
busting and cooperation. Its positive response to foreign requests for help boosted confidence. So
did its early successes with fraud cases.. and the compensation of foreign victims…. In 2003, with
a few successes under the EFCC’s belt, I visited diplomatic envoys from the EU, the UK, and the
United States in Abuja to get their support…International cooperation facilitated the recovery of
stolen funds.” A number of prominent cases were brought by Scotland Yard (with funding from
DFID) against Nigerian officials who had moved the stolen assets to the UK. These efforts
allegedly had indirect benefits as well. “Successful cooperation with foreign law enforcement
agencies also resulted in the voluntary repatriation to Nigeria of billions of dollars stashed abroad.
Those with dirty assets found it much more difficult to hide their loot outside of Nigeria and were
no longer willing to risk having them frozen and forfeited.” 39


VII. Prospects for Effective Reform
IFFs have come to the fore as a goad to policy reform. They are not a discovery of dispassionate
research, an accumulation of evidence across multiple studies. The reform agenda is ambitious but
is much more prominent in the international dialogue about aid and financial policy than might
have been expected ten years ago when IFF studies were first published. It is fair though to ask
how likely it is that if the proposals were adopted that they would in fact have much consequence
for IFFs. Governance challenges are probably the most critical for assessing the plausibility of the
reform agenda.

39
  It is intriguing that, with all the other jurisdictions in which money can be hidden, the crackdown in the UK was
apparently enough to bring back stolen money to Nigeria.
The proposed reforms are promising in principle. They represent a set of reforms that as a package
offer just the kind of transparency that has been lacking for so long in the financial systems of
many individual nations and, more importantly, of the global financial system as a whole. But
theoretical promise characterizes a vast array of policy initiatives. Governance is often the weak
point of such broad reform schemes.
Learning from Anti Money Laundering Efforts40
To see the problems likely to be encountered, it is worth examining the experience of the global
anti money laundering regime. That regime is now a 25-year old system, with a set of
Recommendations (Standards) that have been fairly since those promulgated in 2002 after the
World Trade Center attack (FATF, 2012). It is a well resourced and high profile effort to
accomplish many of the same goals that are needed to reduce IFFs, since moving money out of a
country usually requires laundering.
AML has a well articulated theoretical rationale. It is very difficult to enjoy the fruits of crime
(broadly interpreted) without making use of the formal financial system, whether it be banks,
stockbroking firms, or insurers. The AML system is a set of rules that require these institutions to
assist law enforcement in detecting possible criminal sources of funds. A bank, the most commonly
used institution, must learn the true identity of the new customer (KYC) and establish that the
proposed account or transaction fits with the claimed legal purpose of the account. Politically
Exposed Persons (PEPs) – both domestic and foreign - are to be subjected to Enhanced Due
Diligence, since they are at higher risk of having corruptly obtained funds. A bank may refuse to
accept a new customer, inform the national Financial Intelligence Unit of a suspicious transaction
and/or close the account of a customer of whom it is now suspicious. This ought to make the
laundering of proceeds of crime substantially more difficult, as reflected for example in the higher
price and lower quality of available services.
The Financial Action Task Force an OECD affiliate of 34 mostly rich nations, has promulgated a
set of 40 Recommendations, that are to be adopted by individual nations. They cover all kinds of
financial institutions and Designated Non Financial Businesses and Professions (DNFBP). For
example, one Recommendation (23(a)) requires that lawyers and some other DNFBP report
suspicious transactions, as is required of banks.41 Recommendation 24 is that the beneficial
ownership of corporations be made public.42




40
   This section is adapted from Reuter (2014)
41
   “Lawyers, notaries, other independent legal professionals and accountants should be required to report suspicious
transactions when, on behalf of or for a client, they engage in a financial transaction in relation to the activities
described in paragraph (d) of Recommendation 22. Countries are strongly encouraged to extend the reporting
requirement to the rest of the professional activities of accountants, including auditing.”
42
  “Transparency and beneficial ownership of legal persons *Countries should take measures to prevent the misuse
of legal persons for money laundering or terrorist financing. Countries should ensure that there is adequate, accurate
and timely information on the beneficial ownership and control of legal persons that can be obtained or accessed in a
timely fashion by competent authorities. In particular, countries that have legal persons that are able to issue bearer
shares or bearer share warrants, or which allow nominee shareholders or nominee directors, should take effective
measures to ensure that they are not misused for money laundering or terrorist financing….”
Almost all countries43 have submitted themselves to Mutual Evaluation Reviews (MERs), either
through FATF or through what are oddly titled “FATF Style Regional Bodies” (FSRBs). 44 The 4th
round of MERs started in 2014 and will only be completed in 2020; it makes some effort at
assessing effectiveness, though not very stringently (Halliday, Levi and Reuter, 2014). For most
countries we still have to rely on the 3rd round of MERs, conducted between 2004 and 2012, which
aimed to assess the extent to which the country had implemented the 40 Recommendations in law
and had created institutions to make the laws work.
What is striking is that so few countries, even among those that were the progenitors of the system,
such as the United States, France and the United Kingdom, had anything close to a perfect
scorecard. For example, the U.S. is in gross violation of the beneficial ownership obligation, as
states (notably Delaware and Nevada, but they are only the most egregious) readily and massively
offer shell corporation services (De Willebois et al. 2011).45 Nor is this, by any account, a minor
violation; for AML purposes ascertaining beneficial ownership is widely accepted as a key
condition. A recent assessment of donor country compliance with FATF Recommendations by the
OECD (2014) found that not a single country among the donors was fully Compliant with all the
Recommendations on implementation of beneficial ownership registration and 37% were
classified as Non Compliant.46 Overall, the Report found modest compliance on many of the Key
and Core Recommendations. For developing countries the record is likely to be substantially
worse.47
This weak record does not reflect a lack of enforcement tools. FATF has the authority to put a
country on a “black list”, officially the list of “High risk and non-cooperative jurisdictions”. For
those in the highest category (in 2014 just two: Iran and the Democratic People’s Republic of
Korea) FATF asks “ its members and other jurisdictions to apply counter-measures to protect the
international financial system from the on-going and substantial money laundering and terrorist
financing (ML/FT) risks emanating from the jurisdictions”48, in other words to apply financial
sanctions. Another FATF list, with four countries on it in 2014, is subject to higher scrutiny but
not sanctions; this imposes high costs on financial institutions and perhaps on their customers if
they are not given easy access to correspondent banking privileges elsewhere.



43
   A number of jurisdictions other than nations have also been evaluated, such as Hong Kong and the Vatican.
44
   Individual MERs can be found at http://www.fatf-gafi.org/countries/
45
   Following release of the Panama Papers, the administration proposed a number of changes that would require
banks to establish beneficial ownership. All but one are proposals. The only change that will be definitely be
implemented attracted immediate and compelling criticism “The internationally accepted definition of beneficial
owner—the actual human being who truly gains from the company’s equity—is altered in the law. Under the new
rules, anyone who owns less than 25% of the company need not be reported, and the appointed president of a shell
company can be listed as the beneficial owner.” http://qz.com/678956/a-design-flaw-in-obamas-new-panama-
papers-rule-could-help-shell-companies-dodge-cops-and-taxes/
46
   A country could be Compliant, Largely Compliant, Partially Compliant or Non-Compliant for any specific
Recommendation.
47
   A study for the West African regional FSRB found ““With regard to the outco me of the ME, the overall result
shows very low performance in terms of compliance with the FATF 40+9 Recommendations, which was about
twenty percent (20%). The performance is even less illuminating with regard to compliance with the sixteen (16)
core and key Recommendations considered as the building blocks of an effective AML/CFT system, which was less
than two percent (1.7%).” Inter-Governmental Action Group against Money Laundering in West Africa, 2014
48
   http://www.fatf-gafi.org/topics/high-riskandnon-cooperativejurisdictions/documents/public-statement-june-
2014.html
There is evidence that nations care about the risk that of being placed on the list. In 2000 when the
first “black list” was developed, it had 15 jurisdictions on it. Some countries took drastic actions
to avoid being on that initial list For example, Austria, with a long tradition of anonymous accounts
that facilitated tax evasion, prohibited them in 2001 in order to avoid being blacklisted. By 2006
only Burma was still listed.
The FATF regime has changed the routines of the financial sector, broadly defined, in important
ways (e.g. Favarel-Garigues, Godefroy and Lascoumes, 2011). All major banks have large staffs
of AML compliance officers.49 The documentation requirement to become or, in the case of long-
standing customers, to remain a customer of a bank has become, in many countries, quite
demanding.50 There is evidence that the major international banks are more cautious in offering
correspondent banking services to banks in countries that are regarded as high risk for terrorist
finance or money laundering (Economist, 2014; Commonwealth Secretariat, 2016). There is a
concern that it may have adversely affected development finance in those countries (CGD, 2015).
Other financial institutions have also adopted much more elaborate KYC rules than they did in the
past and are taking ongoing monitoring of clients more seriously than in the past (though the
evidence here is largely anecdotal)51.
Yet for all this, there is little evidence that opportunities for money laundering have been restricted.
Indeed, one astonishing feature is that many major international banks flagrantly flout the regime.
Put aside the sanctions violations of banks such as HSBC and BNP Paribas52; the sanctions against
Iran, Sudan and Cuba were part of US foreign policy and at least some senior bank executives in
foreign capitals were explicit that they did not feel they should be tools of the foreign policy of
another country.53 But HSBC, Wachovia (now part of Wells Fargo) and JP Morgan Chase, all
leading international banks, have all been caught laundering money that almost certainly came
from Latin American drug dealers54. In 2011, ten years after the revelations of how easily Sani
Abacha had deposited stolen assets in the British banking system, the Financial Services Authority
(now the Financial Conduct Authority) found that about three quarters of banks in the UK, most
of them private banks serving wealthy customers, did not implement rules to identify Politically
Exposed Persons, such as the sons of senior Nigerian officials, whose wealth might come from
misappropriation of publicly owned assets (Financial Services Authority, 2011). Under FATF
rules, such persons should be subject to enhanced due diligence.



49
  “While J.P. Morgan doesn't disclose its total number of compliance employees, it is expected to have 8,000 people
focused on anti-money-laundering efforts alone, according to the bank.” Ensign(2014).
50
   I am on the board of an academic society which lost its UK bank account because it tried to pay the travel
expenses of an Iranian speaker at its annual conference. Months later, the society is still searching for a bank willing
to handle its affairs.
51
   A new consulting report for the UK suggests that there is less of this wholesale dropping of clients than is usually
alleged and that many factors other than AML contribute to those events that do occur (Artingstall, Dove, Howell
and Levi, 2016)
52
   On HSBC see http://uk.reuters.com/article/2012/12/11/us-hsbc-probe-idUSBRE8BA05M20121211; on BNP
Paribas see http://www.bloomberg.com/news/2014-07-09/bnp-paribas-pleads-guilty-in-u-s-to-violating-
sanctions.html
53
   In most instances there were similar UN sanctions which reflected the same sentiment and for which the UK and
France provided support.
54
   For example on JP Morgan Chase see http://www.publicintegrity.org/2013/04/30/12577/jpmorgan-chase-s-record-
highlights-doubts-about-big-banks-devotion-fighting-flow
What is one to make of such massive indifference to a regulatory regime which has in principle
almost universal approval? Using AML to limit the ability of terrorists to finance their operations,
to catch and punish major drug dealers or corrupt officials in poor countries is hardly controversial.
Yet corporations that are often thought to be of the highest respectability, and to value such a
reputation, chose to routinely violate the laws. Changing the “culture” of banks is a common theme
both in the academic literature and in public statements about improving bank compliance with
AML, though there is less agreement on what levers are more or less effective in doing so. 55
Nor do the violations by banks represent the sole way in which the AML system is being undercut.
A multitude of methods are available to conceal the criminal origins of money in order to invest
in the legitimate economy and thus enjoy the fruits of crime undisturbed. Some involve formal
institutions and reflect strategies to defeat the bank AML controls (e.g. shell corporations, nominee
accounts), while others involve informal institutions that bypass the banking system (e.g. informal
value transfer systems).56
The AML system created by FATF thus represents an instance of a carefully designed and
universally accepted legal system to control an important method of protecting the fruits of crimes
such as bribe taking and tax evasion. No one can point to any evidence that the system has had its
intended effects of reducing the underlying crimes. The AML system certainly provides
prosecutors with additional charges that enable them to bring to justice some miscreants (and their
professional ‘enablers’) who might otherwise be able to avoid prosecution. It also generates data
bases that help investigations. However these are hardly transformative capabilities. Not even
FATF dares argue that those with criminal earnings have much more difficulty in laundering them
into usable legal forms.
Much of the failure of the system can be understood through relatively simple analysis of
incentives and norms. Banks seek clients who are willing to pay well for individualized services.
The belief that other banks might be willing to take on such clients without Enhanced Due
Diligence, is a strong motive for ignoring those rules, as revealed in the FSA 2011 study and the
notorious acceptance of moneys from President Abacha of Nigeria during the 1990s. The fines,
until recently, were modest enough and the probability of detection small enough that many banks
took on accounts and transactions that were at a minimum high risk and often blatantly involved
predicate offenses.
The failure of the national legal and regulatory systems is perhaps more complex to explain. One
factor may be the relative novelty of the responsibility given to the bank regulatory officials, to
assess the individual bank performance with respect to the bank’s policing activity (Favarel-
Garrigues,Godefroy, and Lascoumes, 2011). Money laundering violations do not directly threaten
the operations of the bank itself but rather failure in its duty to aid society in catching criminals.
Another factor is that legislation is weakened through the usual process of regulatory capture.
AML is both expensive and market contracting for the individual bank. Financial industries have
been effective in finding both theoretical and empirical arguments for making individual country
laws unnecessarily weak (e.g. Wallison, 2015). The lack of an empirical literature showing that



55
   See for example the statement of the outside expert monitoring HSBC’s compliance with AML.
https://www.zacks.com/stock/news/212190/hsbc-monitor-says-money-laundering-controls-inadequate
56
   A sense of the variety of money laundering schemes can be found in a description of 100 cases collected by the
Egmont Group in 2000: http://www.u4.no/recommended-reading/fius-in-action-100-cases-from-the-egmont-group/
AML does in fact reduce any of the claimed harms facilitates the task of the financial industry in
undermining proposed controls.
The AML experience is not promising and as an important component of the IFF control agenda,
it has a claim to particular attention in trying to project the consequences of that agenda. However,
there are other similar areas in which there is more evidence of progress. For example, Automatic
Exchange of Tax Information (AEOI) has now become routinized among a number of wealthy
nations and apparently has cut down on certain kinds of tax evasion (OECD, 2012). Secrecy
jurisdictions such as the Cayman Islands and the British Virgin Islands have made important
changes to reduce opacity of their financial systems.57
Which are the more relevant comparisons for assessing the likely consequences of the proposed
reforms? The package of reforms contemplated by the IFF warriors are complex. They offer
multiple opportunities for any of the players (clients, financial institutions, corporations,
regulators, national governments) to develop strategies to thwart them. The much criticized
beneficial ownership rule just imposed by the Obama administration (see n. 37 above) is indicative
of the ways in which good intentions can be undercut by interested parties while superficially
meeting the goals of the reformers.


Concluding Comments
This paper has been relentlessly speculative. That reflects the state of understanding of the IFF
phenomenon. The concept is poorly defined; the inclusion of aggressive tax avoidance schemes is
an instance of a controversy that has large implications for both measurement and policy. 58 The
available estimates, coming from one organization using a single methodology, seem inflated to
many observers. The relationship between the underlying concept of IFFs and the estimates is
obscure. Nothing is known about the relative importance of component sources or of the channels
that are used to move the funds overseas, which will surely vary over time and across countries.
Discussions of the likely effect of different control measures is just an exchange of impressions
rather than the result of any systematic analysis.
This profound ignorance is not the consequence of confusing research results. It reflects the lack
of any sustained research agenda. This paper has echoed many criticisms of the GFI estimates but
GFI is the only organization that has consistently studied the phenomenon. It is hard to identify
major institutional contributions apart from GFI’s and case studies of advocacy organizations such
as the Tax Justice Network or Global Witness. Only a handful of researchers outside of GFI have
written more than a couple of papers on IFFs (e.g. Alex Cobham and Leonce Ndikumana).
Whatever the criticisms of the existing estimates, there is no doubt that IFFs are substantial enough
to merit close attention. With reduction of IFFs enshrined in the Sustainable Development Goal
16, general ignorance about the phenomenon has become an embarrassment. For reform efforts, a



57
  OECD Global Forum on Transparency and Exchange of Information for Tax Purposes (2015).
58
  “There is disagreement, however, on whether tax avoidance should be considered an illicit flow. For example, the
report of the High Level Panel on Illicit Financial Flows from Africa includes “tax abuse” which comprises “evad -
ing or aggressively avoiding tax” in its broad inter pretation of illicit flows. Other sources and institutions exclude
tax avoidance from the definition of IFFs on the grounds that legal exploitation of tax loopholes is not illicit.” (UN.
2016)
good understanding of the governance issues and the obstacles to aligning the interests and
capacities of the many participants is crucial.
References


Ajayi, S. Ibi, and Léonce Ndikumana, eds. Capital flight from Africa: Causes, effects, and policy
issues. OUP Oxford, 2014.


Artingstall, D; Dove, N, Howell, J. and M. Levi (2016) Drivers & Impacts of Derisking: A study
of representative views and data in the UK, by John Howell & Co. Ltd. for the Financial Conduct
Authority London

Asia and Pacific Group on Money Laundering (2012) Typology Report on Trade Based Money
Laundering                  Sydney,                Australia            http://www.fatf-
gafi.org/media/fatf/documents/reports/Trade_Based_ML_APGReport.pdf


Blankenburg, S. and M. Khan (2012) “Governance and Illicit Flows,” in P. Reuter (Ed.) Draining
development? Controlling flows of illicit funds from developing countries, The World Bank.


Boyce, James K., and Léonce Ndikumana. "Capital Flight from Sub-Saharan African Countries:
Updated Estimates, 1970-2010." PERI Working Papers (2012).


Chêne, M. (2006) "Mutual legal assistance treaties and money laundering." Transparency
International, U 4.


Chrisafis, A. (2012, February 6). France impounds African autocrats’ “ill-gotten gains.” The
Guardian. Retrieved from https://www.theguardian.com/world/2012/feb/06/france-africa-
autocrats-corruption-inquiry


Cobham, A. (2014) Benefits and Costs of the IFF Targets for the post-2015 Development Agenda
Copenhagen                                                                         Consensus
http://www.copenhagenconsensus.com/sites/default/files/iff_assessment_-_cobham_0.pdf


Commonwealth Secretariat (2016) Disconnecting from Global Finance The Impact of AML/CFT
Regulations          in             Commonwealth                Developing          Countries
http://thecommonwealth.org/sites/default/files/inline/DisconnectingfromGlobalFinance2016.pdf


Corruption Watch, UK (2013) Deception in High Places: The Angola-Russian Debt Deal London
The Economist. (2014, March 15). Planet Plutocrat. The Economist. Retrieved from
http://www.economist.com/news/international/21599041-countries-where-politically-connected-
businessmen-are-most-likely-prosper-planet
Engel, C. M. (2012). Capital controls: what have we learned? (BIS Working Paper 68f). Bank for
International Settlements. Retrieved from http://www.bis.org/publ/bppdf/bispap68f.pdf


Ensign, R. L. (2014Lenders Bolster Risk and Compliance Staff. Wall Street Journal. May 5


Epstein, G. A., ed. (2005), Capital Flight and Capital Controls in Developing Countries.
Cheltenham, U.K.: Edward Elgar


European Monitoring Center on Drugs and Drug Addiction and Europol (2016) EU Drug Markets
Report:                                   Strategic                             Overview
file:///Users/peter/Downloads/strategic_overview_eu_drug_markets_report_2016%20(1).pdf


FATF (2011) Laundering the Proceeds of Corruption Paris, France


FATF (2012), International Standards on Combating Money Laundering and the Financing of
Terrorism & Proliferation, updated June 2016, FATF, Paris, France, www.fatf-
gafi.org/recommendations.html


Favarel-Garrigues, G., Godefroy, T. and Lascoumes, P., 2011. Reluctant partners? Banks in the
fight against money laundering and terrorism financing in France. Security Dialogue, 42(2),
pp.179-196.


The Financial Services Authority (2011) Banks’ management of high money-laundering risk
situations. How banks deal with high-risk customers (including politically exposed persons),
correspondent banking relationships and wire transfers. London

Gandhi, A., & Walton, M. (2012). Where Do India’s Billionaires Get Their Wealth?’. Economic
and Political Weekly, 47(40), 10–14.

Global Witness Heavy Mittal A State within a State: The inequitable Mineral Development
Agreement between the Government of Liberia and Mittal Steel Holdings NV
file:///Users/peter/Downloads/mittal_steel_en_oct_2006_low_res.pdf
Halliday, T., Levi, M. and Reuter. (2014) Assessing the Assessors: How Well do the International
Monetary Fund and the Financial Action Task Force Evaluate National Efforts to Control Money
Laundering?                     American                      Bar                    Foundation
http://www.lexglobal.org/files/Report_Global%20Surveillance%20of%20Dirty%20Money%201
.30.2014.pdf


Herkenrath, Marc (2014) “Illicit Financial Flows and their Developmental Impacts: An
Overview,” International Development Policy/Revue Internationale de Politique de Development,
Articles and Debates, 5.3, https://poldev.revues.org/1863


Inter-Governmental Action Group against Money Laundering in West Africa (GIABA) (2014).
Improving Compliance with International AML/CFT Standards: Report on the Strategic Review
of the First Round of Mutual Evaluations of GIABA Member States (2007 - 2012). Dakar, Senegal:
(GIABA


Kar, Dev. "Mexico: Illicit Financial Flows, Macroeconomic Imbalances, and the Underground
Economy." Macroeconomic Imbalances, and the Underground Economy (January 2, 2012)
(2012).’


Kar, D, and S Freitas. Illicit financial flows from developing countries over the decade ending
2009. Vol. 100. Washington, DC: Global Financial Integrity, 2011.


Kar, D. and Freitas, S., 2012. Illicit Financial Flows from China and the Role of Trade
Misinvoicing. Global Financial Integrity, Washington, DC.


Kar, D, and Spanjers (2015). Illicit financial flows from developing countries 2004-2013
Washington, DC: Global Financial Integrity


Jansky (2013) “Illicit Financial Flows and the 2013 Commitment to Development Index,” CGD
Policy Paper 034, Center for Global Development. http://www.cgdev.org/publication/illicit-
financialflows-and-2013-commitment-development-index .


Kurtz-phelan, D. (2008, March 2). Bananas: How the United Fruit Company Shaped the World -
Peter Chapman - Book Review. The New York Times. Retrieved from
http://www.nytimes.com/2008/03/02/books/review/Kurtz-Phelan-t.html
Leite, C. (2012) “The Role of Transfer Pricing in Illicit Financial Flows” in P. Reuter (Ed.)
Draining development? Controlling flows of illicit funds from developing countries, Washington,
DC, The World Bank.


Levi, M (2012) “How Well do Anti-Money Laundering Controls Work in Developing Countries?”
in Reuter(Ed.) Draining development? Controlling flows of illicit funds from developing countries,
The World Bank.


Levi, M. and van Duyne, P.C., 2005. Drugs and Money: Managing the Drug Trade and Crime
Money in Europe. Routledge.


Manton, J and T. Daniel (2012) “The Kleptocrat’s Portfolio Decision” in Reuter (Ed.) Draining
development? Controlling flows of illicit funds from developing countries, The World Bank.


Miller, T. C., & Jones, J. (2014, November 18). Firestone and the Warlord. ProPublica. Retrieved
from https://www.propublica.org/article/firestone-and-the-warlord-intro


Murphy, R. (2012, November 13). Amazon, Google and Starbucks are struggling to defend their
tax          avoidance.         The           Guardian.          Retrieved            from
https://www.theguardian.com/commentisfree/2012/nov/13/amazon-google-starbucks-tax-
avoidance


Nitsch, V. (2015) “Trillion Dollar Estimate: Illicit Financial Flows from Developing Countries”
Darmstadt Discussion Papers in Economics #227


OECD. (2012). Automatic Exchange of Information: What Is It, How It Works, Benefits, What
Remains to Be Done. Paris: Organization for Economic Cooperation and Development. Retrieved
from http://www.oecd.org/ctp/exchange-of-tax-information/automatic-exchange-of-information-
report.pdf


OECD (2014) Measuring OECD Responses to Illicit Financial Flows from Developing Countries
Paris, OECD
OECD (2015) Action 13: Guidance on the Implementation of Transfer Pricing Documentation and
Country-by-Country Reporting
https://www.oecd.org/ctp/beps-action-13-guidance-implementation-tp-documentation-cbc-
reporting.pdf


OECD/AfDB (November 2016) Illicit Financial Flows: The economy of illicit trade in West
Africa, Paris: OECD Publishing.


OECD Global Forum on Transparency and Exchange of Information for Tax Purposes. (2015).
Tax Transparency 2015. Paris: Organization for Economic Cooperation and Development.
Retrieved from http://www.oecd.org/tax/transparency/global-forum-annual-report-2015.pdf


Paoli, Greenfield and Reuter (2009) The World Heroin Market: Can Supply be Cut? New York,
Oxford University Press


Persson, Anna, Bo Rothstein, and Jan Teorell. "Why anticorruption reforms fail —systemic
corruption as a collective action problem." Governance 26.3 (2013): 449-471.


Reuter (2014) “Benefits and Costs of the IFF Targets for the Post-2015 Development
Agenda”         Commentary          for          the           Copenhagen           Consensus
http://www.copenhagenconsensus.com/sites/default/files/iff_perspective_-_reuter.pdf


Reuters. (2014, July 14). Mexico’s president signs telecoms reform rules into law. Reuters.
Retrieved from http://www.reuters.com/article/us-mexico-reforms-idUSKBN0FJ2DU20140714


Ribadu, Nuhu, and Center for Global Development. (2010) Show Me the Money: Leveraging Anti-
Money Laundering Tools to Fight Corruption in Nigeria: An Insider Story. Washington, D.C.:
Center for Global Development,


Soudijn, M. and P. Reuter (2016) “Cash and Carry: The High Cost of Cash Smuggling in the Drug
Trade” Crime, Law and Social Change


Stephenson, K. M., Gray, L., Power, R., Brun, J.-P., Dunker, G., and Panjer, M. (2011). Barriers
to Asset Recovery: An Analysis of the Key Barriers and Recommendations for Action. Washington
DC: World Bank Publications.
Terracol and Dell (2015), “Mutual legal assistance and corruption,” Transparency International,
U4, 17.


Thoumi, Francisco E., and Marcela Anzola (2012) "Illicit capital flows and money laundering in
Colombia.". in P. Reuter (Ed.) Draining development? Controlling flows of illicit funds from
developing countries, Washington, DC The World Bank.

United Nations (2016) Inaugural 2016 Report of the Inter-agency Task Force on Financing for
Development – Addis Ababa Action Agenda: Monitoring commitments and actions
http://www.un.org/esa/ffd/wp-content/uploads/2016/03/2016-IATF-Chapter2A.pdf

US Department of Justice. (2014, August 7). Press Release: U.S. Forfeits Over $480 Million Stolen
by Former Nigerian Dictator in Largest Forfeiture Ever Obtained Through a Kleptocracy Action.
United States Department of Justice. Retrieved from https://www.justice.gov/opa/pr/us-forfeits-
over-480-million-stolen-former-nigerian-dictator-largest-forfeiture-ever-obtained

van der Does de Willebois, E.; Halter, E.M., Harrison, R.A., Won Park & JC Sharman (2011) The
Puppet Masters: How the Corrupt Use Legal Structures to Hide Stolen Assets and What to Do
About It Washington, DC, World Bank http://star.worldbank.org/star/publication/puppet-masters

van Duyne, P., Crime-entrepreneurs and financial management. In: P.C. van Duyne, K. von Lampe
en N. Passas (eds.), Upperworld and Underworld in Cross-border Crime, Nijmegen, Wolf Legal
Publishers, 2002

Wallison, P. J. (2015). Is the Dodd-Frank Act Responsible for the Economy’s Slow Recovery from
the Financial Crisis and the Ensuing Recession? Presented at the Ninth Hillsdale College Free
Market     Forum,        Hillsdale    College.   Retrieved      from    https://www.aei.org/wp-
content/uploads/2015/11/Wallison-Hillsdale-Paper.pdf


Woolner, A. (1994) Washed in Gold: The Story Behind the Biggest Money Laundering
Investigation in US History New York, Simon and Schuster


Yikona S. (2013) Pirate Trails: Tracking the Illicit Financial Flows from Pirate Activities Off the
Horn of Africa. Washington, D.C., World Bank Publications