GEORGETOWN UNIVERSITY

APR 2 4 2007

LAW LIBRARY
INSTITUTE FOR INTERNATIONAL ECONOMICS

INTL
m
6768
. R48
20©4
 t NT1-
£11,2
HV

CHASING

MONEY

The Fight Against

Money laundering
INSTITUTE FOR INTERNATIONAL ECONOMICS

iioru

C H A S I N G ” \\

MONEY
The Fight Against

Mdney Laundering

Peter Reuter and Edwin M. Truman

Washington, DC
November 2004
Peter Reuter is a professor in the School of INSTITUTE FOR INTERNATIONAL
Public Affairs and in the Department of ECONOMICS
Criminology at the University of Maryland. 1750 Massachusetts Avenue, NW
Since July 1999, he has been the editor of the Washington, DC 20036-1903
Journal of Policy Analysis and Management. (202) 328-9000 FAX: (202) 659-3225
He was a senior economist in tire Wash¬ www.iie.com
ington office of the RAND Corporation
C. Fred Bergsten, Director
(1981-93). He founded and directed RAND's Valerie Norville, Director of Publications
Drug Policy Research Center (1989-93). His
early research focused on the organization of and Web Development
illegal markets and resulted in the publica¬ Edward Tureen, Director of Marketing
tion of Disorganized Crime: The Economics
of the Visible Hand (MIT Press, 1983), which Typesetting by Circle Graphics
won the Leslie Wilkins award for the most Printing by Automated Graphic Systems, Inc.
outstanding book of the year in criminology
and criminal justice. He has served as a con¬ Copyright © 2004 by the Institute for
sultant to numerous government agencies International Economics. All rights
(including the US General Accounting reserved. No part of this book may be
Office, the White House Office of National reproduced or utilized in any form or by
Drug Control Policy, the National Institute any means, electronic or mechanical,
of Justice, and the Substance Abuse and including photocopying, recording, or by
Mental Health Services Administration) and information storage or retrieval system,
to foreign organizations including the without permission from the Institute.
United Nations Drug Control Program and
the British Department of Health. For reprints /permission to photocopy please
contact the APS customer service depart¬
Edwin M. Truman, senior fellow, was assis¬ ment at Copyright Clearance Center, Inc.,
tant secretary of the US Treasury for inter¬ 222 Rosewood Drive, Danvers, MA 01923;
national affairs (1998-2000). He was staff or email requests to: [email protected]
director of the Division of International Fi¬
nance of the Board of Governors of the Fed¬ Printed in the United States of America
eral Reserve System (1987-98) and director
of the division from 1977 to 1987. From 1983 06 05 04 5 4 3 2 1
to 1998, he was one of three economists on
the staff of the Federal Open Market Com¬ Library of Congress Cataloging-in-
mittee. He has been a member of numerous Publication Data
international groups working on interna¬
tional economic and financial issues, includ¬ Reuter, Peter
ing the Financial Stability Forum's Working Chasing dirty money : progress on
Group on Highly Leveraged Institutions anti-money laundering / Peter Reuter,
(1999-2000), the G-22 Working Party on Edwin M. Truman,
Transparency and Accountability (1998),
the G-10-sponsored Working Party on Includes bibliographical references and
Financial Stability in Emerging Market index.cm.
p.
Economies (1996-97), the G-10 Working ISBN 0-88132-370-5
Group on the Resolution of Sovereign 1. Money laundering. 2. Money
Liquidity Crises (1995-96), and the G-7 laundering — Prevention. 3. Money
Working Group on Exchange Market Inter¬ laundering — United States — Prevention.
vention (1982-83). He has published on I. Truman, Edwin M. II. Title.
international monetary economics, interna¬
HV6768.T78 2004
tional debt problems, economic develop¬
ment, and European economic integration. 363.25’968 — dc22 2004054940

The views expressed in this publication are those of the authors. This publication is part
of the overall program of the Institute, as endorsed by its Board of Directors, but does
not necessarily reflect the views of individual members of the Board or the Advisory
Committee.
 To
Caroline Isber and Tracy Philbrick Truman
Contents

Preface xi

Acknowledgments xv

1 Chasing Dirty Money 1

Money Laundering and Its Control 3
Conclusions and Recommendations 6

2 How Much Money Is Laundered? 9

Macroeconomic Estimates 11
Microeconomic Estimates 19
Conclusions 23

3 Money Laundering: Methods and Markets 25

Laundering Mechanisms 27
Which Methods Are Used for Which Crimes? 32
Who Provides the Laundering Services? 34
Markets for Laundering Services 35
Classification of Offenses 40
Conclusions 43

4 The Anti-Money Laundering Regime 45

Prevention and Enforcement 46
Current US Anti-Money Laundering Regime 49
US National Money-Laundering Strategies 70

vii
 The Global Anti-Money Laundering Regime 77
Costs of the US Anti-Money Laundering Regime 93

5 Combating Predicate Crimes Involved in Money Laundering 105

Enforcement and Predicate Crimes 105
Enforcement in the United Kingdom and Other Nations 114
Outcome Measures and Analytic Frameworks 119
Conclusions 128

6 Protecting Financial System Integrity 129

Financial System Integrity as an AML Goal 130
Evaluating Progress 132
Conclusions 136

7 Combating Global “Public Bads” 139

Terrorism 139
Corruption and Kleptocracy 148
Failed States 156

8 Improving the Global AML Regime 171

Scope of the AML Regime 172
Means to What Ends? 175
Implementation Challenges for the United States 177
Global Implementation Challenges 182
International Cooperation 183
Domestic Law Enforcement 186
Research Agenda 188
Final Comments 191

References 193

Glossary and Acronyms 201

Index 209

Tables
Table 2.1 Taxonomy of underground economy activities 12
Table 2.2 Estimates of the underground economies of
21 OECD countries, 1989-2001 14
Table 2.3 Estimated earnings from criminal activity in the
United States, 1965-2000 20
Table 2.4 Estimated criminal proceeds in the United States, 1990 22
Table 3.1 Frequency of predicate offenses and methods 33
Table 3.2 Examples of money-laundering costs 37

viii
Table 3.3 Taxonomy of predicate crimes for money laundering 41
Table 4.1 Evolution of the AML regime in the United States,
Europe, and globally 50
Table 4.2 Prevention pillar of the US anti-money laundering
regime 54
Table 4.3 Distribution of US National Money Laundering
Strategy action items across AML goals, 1999-2003 74
Table 4.4 Distribution of US National Money Laundering
Strategy action items across AML prevention and
enforcement elements, 1999-2003 75
Table 5.1 Defendants charged with money laundering,
1994-2001 109
Table 5.2 Adjudications and convictions in cases with money
laundering as the most serious offense, 1994-2001 110
Table 5.3 Other offenses for which convicted federal money
launderers were sentenced, 1995, 2000 111
Table 5.4 Suspicious activity reports by country, 1994-2002 116
Table 7.1 Failed and failing states in 2002 158
Table 7.2 Inclusion of countries in INCSR ratings and FATF
reviews 161
Table 7.3 Failed and failing states and INCSR
ratings 162
Table 7.4 Money laundering and corruption 164

Figures
Figure 2.1 Estimated criminal income as percent of US GDP,
1965-2000 21
Figure 4.1 Pillars of the anti-money laundering regime 47
Figure 4.2 Share of cross-border assets of offshore financial centers,
1992-2003 89
Figure 4.3 Trends in US federal prevention and enforcement real
outlays, fiscal 1985-2004 96
Figure 5.1 Suspicious activity reports filed, 1996-2003 107
Figure 5.2 Total fines and restitution for money laundering in the
United States, 1990-2001 1 14

Boxes
Box 1.1 Anti-money laundering and combating terrorism
financing 2
Box 3.1 26
Laundering methods of a drug trafficker
Box 3.2 Embezzlement and (self-) money laundering 27
Box 3.3
"Underground" banking that finances human smuggling 28
Box 3.4 29
Pilfering by a media baron
Box 3.5 30
Other money-laundering methods
Box 4.1 Accountants, auditors, and anti-money laundering
regimes 62

IX
 73
Box 4.2
Electronic finance and money laundering
Box 4.3
The FATF Non-Cooperative Countries and Territories
Initiative
86
Box 5.1
Detecting money laundering through failures to file
108
suspicious activity reports
Box 7.1 160
"Failed" or "failing" states

x
Preface

The structure and functioning of the international financial system in an

increasingly globalized world has received considerable attention from the
Institute for International Economics throughout its existence. Morris
Goldstein has written extensively on the topic, including The Case for an
International Banking Standard in 1997, which was shortly adopted by the
world's authorities. Wendy Dobson and Gary C. Hufbauer analyzed World
Capital Markets: Challenge to the G-10 in 2001. Kimberly A. Elliott previewed
some of the issues addressed in this study in her Corruption and the Global
Economy in 1997.
This study by Peter Reuter and Ted Truman addresses an additional and
extremely important element of the international financial system: money
laundering. Money laundering is a quintessential global issue that com¬
bines many of our previous themes. But combating it requires a coopera¬
tive, global, and special anti-money laundering (AML) regime, and this
study provides the first comprehensive effort to assess the effectiveness of
the existing regime. The authors first describe the activity of money laun¬
dering. They then provide an overview of the global AML regime as it has
evolved over the past 15 years. Finally, they assess the regime's effective¬
ness in addressing three underlying goals: reducing crime, protecting the
integrity of the core financial system, and controlling three types of global
"public bads" — terrorism, corruption, and failed states. They find evidence
the regime has helped increase the integrity of the banking system in the
United States and many major financial centers. Elowever, there is an
absence of systematic evidence that it has made money laundering more
than marginally more difficult for those who need to clean dirty money or
finance terrorism.

XI
 The authors conclude with recommendations for improvements in the
US and global AML regime during its next phase, which will probably be
one of consolidation following its rapid expansion to date. They stress the
need for increased international cooperation on tax evasion, not covered
by the current US AML regime, to solidify support for the international
AML regime. They emphasize the role for greater financial as well as tech¬
nical assistance for poorer countries to tighten the global fight against
money laundering and terrorism financing. They recommend that the
United States volunteer for an IMF/ World Bank assessment of its own
financial system, including regulations affecting money laundering and
terrorism, because of the central global role of the United States. They also
recommend the revival of the National Money Laundering Strategy, which
the United States previously submitted to the Congress and published reg¬
ularly, but with changes from those produced from 1999 to 2003. They con¬
clude with a systematic research agenda to identify the costs of the AML
regime and improve its efficiency going forward.
Peter Reuter is a professor in the School of Public Policy and the
Department of Criminology at the University of Maryland. He is a widely
cited expert on drug policy and coauthor of Drug War Heresies: Learning
from Other Vices, Times, and Places (Cambridge University Press, 2001).
Edwin M. Truman has been a senior fellow at the Institute for International
Economics since 2001, following a distinguished career of more than 25
years in the US government, principally the Federal Reserve and the
Treasury. This is his second major Institute study, following Inflation
Targeting in the World Economy, which was published in 2003.
The Institute for International Economics is a private, nonprofit institu¬
tion for the study and discussion of international economic policy. Its
purpose is to analyze important issues in that area and to develop and
communicate practical new approaches for dealing with them. The
Institute is completely nonpartisan.
The Institute is funded largely by philanthropic foundations. Major
institutional grants are now being received from the William M. Keck, Jr.
Foundation and the Starr Foundation. A number of other foundations and
private corporations contribute to the highly diversified financial resources
of the Institute. About 18 percent of the Institute's resources in our latest
fiscal year were provided by contributors outside the United States, includ¬
ing about 8 percent from Japan. Major support for this project was pro¬
vided by Carla A. Hills and Roderick M. Hills.
The Institute's Board of Directors bears overall responsibilities for the
Institute and gives general guidance and approval to its research program,
including the identification of topics that are likely to become important
over the medium run (one to three years), and which should be addressed
by the Institute. The director, working closely with the staff and outside
Advisory Committee, is responsible for the development of particular pro¬
jects and makes the final decision to publish an individual study.

xii
 The Institute hopes that its studies and other activities will contribute to
building a stronger foundation for international economic policy around
the world. We invite readers of these publications to let us know how they
think we can best accomplish this objective.

C. Fred Bergsten
Director
November 2004
 INSTITUTE FOR INTERNATIONAL ECONOMICS
1750 Massachusetts Avenue, NW, Washington, DC 20036-1903
(202) 328-9000 Fax: (202) 659-3225

C. Fred Bergsten, Director

BOARD OF DIRECTORS ADVISORY COMMITTEE

"Peter G. Peterson, Chairman Richard N. Cooper, Chairman

"Anthony M. Solomon, Chairman,
Executive Committee
Isher Judge Ahluwalia
Robert Baldwin
Leszek Balcerowicz
Conrad Black Barry P. Bosworth
Susan M. Collins
Bill Bradley
Chen Yuan Wendy Dobson
Juergen B. Donges
"George David Barry Eichengreen
"Jessica Einhorn
Jeffrey A. Frankel
"Stanley Fischer Daniel Gros
Jacob Frenkel
Stephan Haggard
Maurice R. Greenberg David D. Hale
"Carla A. Hills Dale E. Hathaway
Nobuyuki Idei Takatoshi Ito
Karen Katen
W. M. Keck □ John Jackson
Peter B. Kenen
Lee Kuan Yew
Anne O. Krueger
Donald F. McHenry Paul R. Krugman
Mario Monti
Minoru Murofushi Roger M. Kubarych
Jessica T. Mathews
Hutham Olayan Rachel McCulloch
Paul O'Neill
Thierry de Montbrial
"James W. Owens Sylvia Ostry
Frank H. Pearl
Karl Otto Pohl Tommaso Padoa-Schioppa
Dani Rodrik
"Joseph E. Robert, Jr. Kenneth Rogoff
David Rockefeller
David M. Rubenstein Jeffrey D. Sachs
Nicholas Stern
Renato Ruggiero
Edward W. Scott, Jr. Joseph E. Stiglitz
William White
George Soros Alan Wm. Wolff
Lawrence H. Summers
Peter D. Sutherland
Jean Claude Trichet
Laura D' Andrea Tyson
Paul A. Volcker
"Dennis Weatherstone
Edward E. Whitacre, Jr.
Marina v.N. Whitman
Ernesto Zedillo

Ex officio
"C. Fred Bergsten
Nancy Birdsall
Richard N. Cooper

Honorary Directors
Alan Greenspan
Reginald H. Jones
Frank E. Loy
George P . Shultz " Member of the Executive Committee
Acknowledgments

We worked on this study for three years, which was much longer than we
anticipated at the start. Once we got into it, the topic expanded. In the
process of our work, we were fortunate in the substantial assistance,
encouragement, and advice we received from a wide range of people who
have been active participants in the anti-money laundering regime, mostly
in the United States and the United Kingdom, but in other countries as
well. We spoke with more than 75 bank officials, financial regulators, other
government officials, international officials, consultants, prosecutors, and
investigators. Many of them had worn several hats, and some participated
in our two study group meetings. They were all very generous with their
time and advice, but there are too many of them to name separately. None
should feel responsible for the views or conclusions expressed in this
study, but we acknowledge their substantive contributions.
We do want to thank explicitly five people who read and commented on
our final manuscript: Michael Levi, Robert E. Litan, Daniel K. Tarullo,
David R. Truman, and William F. Wechsler. Their guidance was much
appreciated, but they, too, should not be held responsible for any of our
sins of omission or commission. Three major contributors to this study
were Frank Gaenssmantel, Fabrizio Iacobellis, and Jeri Smith-Ready. With¬
out their diligent research assistance and dedication, this study would have
taken many more months, if not years. We also thank Valerie Norville and
her team of Marla Banov, Madona Devasahayam, and Kathryn Sweetman,
who skillfully guided the final birthing process.
Our final thanks go to our spouses for whose support and toleration we
are continually grateful and to whom we have dedicated the product of our
collaboration.

xv
 1
Chasing Dirty Money

Money laundering is the conversion of criminal incomes into assets that

cannot be traced back to the underlying crime. Over the past three decades,
the number and scope of laws and regulations aimed at combating money
laundering have expanded dramatically.
The anti-money laundering (AML) effort by the United States began
with the passage in 1970 of the Bank Secrecy Act, which was largely domes¬
tic in nature and covered only depository institutions. Since then, anti¬
money laundering has become a highly structured international regime
that regulates a wide variety of institutions. Not all of them — casinos are
an example — are normally viewed as part of the financial system.
The emergence of international terrorism as a major policy concern in
recent years has led to a further ratcheting up to cover yet more institutions
and activities. Box 1.1 examines the connections between the AML regime
and efforts to combat terrorist financing.
Growth of the global anti-money laundering regime has generated rela¬
tively little public controversy. The banking sector initially resisted
increased government interference in its relationships with clients, but the
sector has since learned how to accommodate AML requirements in ways
that impose relatively modest costs and inconveniences on both banks and
their customers. Fears about the effects on the international competitiveness
of US banks have also faded as other nations have imposed similar regimes.
Privacy considerations have rarely been a major issue, despite the fact that
the structure represents a considerable investment of public authority and
public and private resources in the collection of information.
Notwithstanding the increased authority and investment represented by
the anti-money laundering regime, few assessments have been carried out
1
 Box 1.1 Anti-money laundering and combating
terrorism financing
The attack on the United States on September 1 1 , 2001 , led to stepped-up efforts to
move the war against terrorism and its financing to the forefront of national and inter¬
national anti-money laundering regimes. However, long before the tragic events of 9/1 1 ,
international initiatives to control money laundering incorporated efforts to combat ter¬
rorist financing. For example, a number of countries already had explicitly included the
financing of terrorism as a predicate or underlying offense in their anti-money launder¬
ing regimes, and the Financial Action Task Force (FATF) and the Egmont Group of
Financial Intelligence Units (FlUs) reviewed a number of cases involving terrorist financ¬
ing. The United States reported to the FATF that it designated 30 foreign organizations
as terrorist organizations in 1997, and seized $1.4 million in cash and property in con¬
nection with an antiterrorism case in 1998. In 2000, the United States and the FATF
highlighted the potential connection between the financing of terrorism and hawala and
other informal value transfer systems.
The tools developed nationally and internationally as part of the anti-money laun¬
dering regime can also be used in dealing with the financing of terrorism. First, regime
tools can be used as investigative devices to learn something not only about the ori¬
gins of funds but also their destinations. Customer due diligence, for example, can help
determine not only who customers are but also what they do, where their money comes
from, what they are doing with it, and where it is transferred to. Second, the regime
can be used as a prosecutorial device, as in a 1998 US confiscation case involving a
scheme to finance terrorism in the Middle East, or in the more recent US case involv¬
ing a Chicago-based charitable organization, Benevolence International Foundation
(even though the money laundering charge in that case was dropped as part of a plea
bargain). Third, combating the financing of terrorism involves close international coop¬
eration in the exchange of information, blocking funds, and closing down channels
used to transfer funds.

(box continues next page)

of either its achievements or consequences. This study's aim is to begin the

task of evaluating the effectiveness of the global anti-money laundering
regime. It describes the phenomenon of money laundering itself, to the
extent that the available fragments of information allow, as well as the sta¬
tus of the current AML regime. This is followed by an analysis of its effec¬
tiveness in achieving three goals: reducing crime, protecting the integrity
of the core financial system, and controlling three types of global "public
bads" — terrorism, corruption, and failed states. The study concludes with
recommendations on how the AML system and analysis of its effectiveness
could both be improved.
The process of preparing this study revealed that there is a dearth of
quantitative data about money laundering and efforts to control it. Nor has
there been much analysis of what few data exist. The available information
consists of case descriptions, raw accounts of law enforcement events (such
as convictions on money-laundering charges or numbers of reports of sus¬
picious activities), and anecdotes from investigators, prosecutors, or, on
rare occasions, the criminals themselves. The academic literature falls

2 CHASING DIRTY MONEY

 Box 1.1 (continued)
There are differences as well, of course, between the general anti-money laundering
regime and the specific variant applied to terrorist financing. First, terrorist financing gen¬
erally (although not exclusively) involves financial flows that originate in legitimate activ¬
ities to support illegitimate activities, rather than the reverse process in which funds from
illicit activities are made to appear licit. Although most other money laundering originates
with illegitimate activities, even here one traditional technique is the exploitation of legit¬
imate activities, especially those handling large amounts of cash such as casinos or gro¬
cery stores. This points to the importance of financial institutions not only knowing their
customers but also knowing what those customers are doing, where they get their
money, and where it is being sent.
Second, terrorist financing typically involves smaller amounts of money than does tra¬
ditional money laundering, often far less than $1 00,000. Combating such relatively small-
scale laundering can be far more difficult — sometimes like looking for a needle in the
haystack.
Third, the stakes are higher in combating terrorist financing in that the amounts
involved may be small, but the potential benefits to society from prevention and confis¬
cation are huge. Thus, the objective is not to contain or reduce but to eliminate the activ¬
ity because the benefit-cost ratio of doing so is high.
Finally, while the goal in most other money-laundering activities can be linked to some
degree or other to the profit motive, in terrorist financing the profit motive (other than cost
minimization) is largely replaced by noneconomic motives, particularly political ones. This
may further hamper detection.
The basic question is the extent to which the authorities can proactively use the
anti-money laundering regime to attack and eliminate terrorist financing and terrorism
itself. The simple answer is that the regime can make a major contribution to combating
terrorism, but some of the differences sketched out above imply the need for, at the very
least, more intense application of existing anti-money laundering instruments as well as
the use of supplementary mechanisms.

into three broad categories: (1) practical law review articles primarily
directed toward identifying the necessary components of an effective
AML regime and explaining the complex statutes in force to control money
laundering; (2) criminological and historical analyses, many of which are
highly judgmental and value-driven; and (3) crude economic analyses of
the extent of money laundering.

Money Laundering and Its Control

Money laundering is conventionally divided into three phases: placement

of funds derived from an illegal activity, layering of those funds by passing
them through many institutions and jurisdictions to disguise their origin,
and integration of the funds into an economy where they appear to be legit¬
imate. Although the anti-money laundering regime has many objectives —
including the aforementioned goals of reducing crime, preserving financial
system integrity and controlling terrorism, corruption, and failed states —

CHASING DIRTY MONEY 3

those objectives are for the most part compatible and do not present oper¬
ational conflicts.
No credible estimates are available as to the volume of money launder¬
ing, or its distribution across countries and activities (chapter 2). Certainly
the aggregate annual figure is in the hundreds of billions of dollars, but
whether that figure is a small number of hundreds or more than a trillion
is unknown. The vagueness of such estimates is a result both of disagree¬
ments over how to conceptualize money laundering and of weaknesses in
the techniques used to quantify it. As a consequence, estimated changes in
the volume of money laundered cannot be used as a measure to judge the
effectiveness of the global anti-money laundering regime.
Moreover, aggregate figures conceal as much as they reveal. The ad¬
verse social consequences of a million dollars laundered to finance a ter¬
rorist act, on the one hand, and a million-dollar embezzlement, on the other,
are so different that adding together the two figures would not produce a
useful statistic for policy purposes. What is needed — but not available —
is reliable figures for the major types of offenses that generate the total
amount.
Money can be laundered in many different ways that can involve a vari¬
ety of businesses and professions (chapter 3). Major drug traffickers face a
unique money-laundering problem — namely, the need to clean large quan¬
tities of currency (much of it in small bills) on a frequent basis. Most other
criminal offenses generate funds that can be more easily concealed.
Surprisingly little evidence exists that much money laundering involves
professionals who provide services to multiple clients. Many cases involve
laundering by the offenders themselves (in embezzlement cases, for exam¬
ple) or relationships between an offender and someone who carries out a
few transactions solely for that person.

The underlying or "predicate" crimes that make it necessary to launder

proceeds can be divided into five categories: drug trafficking, other "blue-
collar" crimes, white-collar crimes, bribery and corruption, and terrorism.
These crimes differ in terms of their reliance on cash, the quantities of
money involved, the severity of their negative social impact, and whom
they affect. As a result, policy decisions may have different consequences
for each category. At least for some activities and offenders — most notably
major drug traffickers — good-quality money-laundering services appear
to be hard to find. They are certainly expensive, with regular reports of
laundering costs as high as 4 to 8 percent of the gross amounts.
As discussed in chapter 4, the AML regime consists of a prevention pillar
(customer due diligence, reporting, regulation and supervision, and sanc¬
tions) and an enforcement pillar (a list of predicate crimes, investigation,
prosecution and punishment, and confiscation). Globally, the prevention
pillar has developed more rapidly, while in many nations the enforcement
pillar is weak. International financial institutions — primarily the Inter-

4 CHASING DIRTY MONEY

national Monetary Fund (IMF) and the World Bank — now play a major
role in assessing primarily the implementation of the prevention pillar
throughout the world.
Chapter 4 describes in considerable detail the AML regime in the United
States and its evolution. It summarizes the prevention pillar's coverage of
various financial and nonfinancial entities, and then contrasts prevailing
coverage with that of the mid-1980s. An examination of the five national
money-laundering strategies presented to the US Congress between 1999
and 2003 reinforces a number of key points about the structure and evolu¬
tion of the US regime, particularly the two-pillar framework and the ele¬
ments of each pillar. The chapter also reviews efforts over the past
15 years to establish a global AML regime and compares and contrasts the
US AML regime with regimes in other countries. The chapter concludes
with consideration of the gross financial cost of the US AML regime to the
government, private-sector institutions, and the general public. On the basis
of several assumptions and a few rough guesses, the conclusion is that the
cost is substantial but not overwhelming — on the order of $7 billion in
2003, or about $25 per capita.
Chapters 5 through 7 assess the effectiveness of the global AML regime
and its progress with respect to the three goals of reducing crime, protect¬
ing the integrity of the core financial system, and controlling terrorism, cor¬
ruption, and failed states. Applying a single framework to assess an AML
regime with respect to each of these goals is not the best way to carry out
such an evaluation; instead, those measures deemed most appropriate to
judge the effectiveness with respect to each goal are used on a case-by-case
basis. Under current circumstances, only indirect measures of effectiveness
can be applied.
Chapter 5 argues that enforcement activities under the US AML regime
have not been intense. While the number of suspicious activity reports filed
has risen rapidly in recent years, as has the value of assets confiscated, total
seizures and forfeitures amount to an extremely small sum (approximately
$700 million annually in the United States) when compared with the crude
estimates of the total amounts laundered. Moreover, there has not been an
increase in the number of federal convictions for money laundering. A very
speculative estimate of the risk of conviction faced by money launderers is
about 5 percent annually. Data from other industrialized nations indicate
even lower levels of enforcement.

It is natural for economists to think of the AML regime as an effort to con¬

trol an illegal market, in this case the market for money-laundering ser¬
vices. However, using that framework to understand better the functioning
and effectiveness of the AML regime results in surprising findings. The
available evidence suggests that most money laundering is not carried out
as a separate activity by professionals, but rather is often part of the under¬
lying offense or involves ad hoc assistance. This implies that price signals

CHASING DIRTY MONEY 5

 may be very weak and that market analysis may not provide useful in¬
sights. On the other hand, it may well be that the market framework needs
to be more thoroughly analyzed, a worthwhile task for future study. Both
theoretical and empirical work is needed to determine whether it is in fact
useful to think of money-laundering controls in terms of the demand for,
and supply of, illegal services with an implicit or explicit price.
For this study, assessing the effectiveness of the AML regime in reduc¬
ing crime meant relying on indirect indicators such as suspicious activity
reports, prosecutions and convictions, forfeitures and seizures, and prices
paid for money-laundering services. The indicators provide some support
for the proposition that the AML regime has contributed to the overall
effectiveness of law enforcement by providing an additional tool.
With respect to protecting the integrity of the core financial system
(chapter 6), the AML regime established in major jurisdictions over the past
15 years has changed how banks and other financial institutions do busi¬

ness. Today's AML regime has induced banks to take their obligation to
avoid direct contact with criminal money seriously. Banks generally have
implemented reporting systems and developed monitoring techniques that
make them much less attractive for the placement phase of money laun¬
dering. However, while the global system that has emerged presents tan¬
gible obstacles to using banks and mainstream financial institutions for the
placement of funds, the effectiveness of the AML regime with respect to
the layering phase of money laundering is much more difficult to assess.
The AML regime today appears to be reasonably effective in protecting
the integrity of the core financial system in major financial centers. How¬
ever, it was not possible for this study to apply systematically the preferred
assessment instrument, which is close examination and cross-classification
of money-laundering cases, to determine the size of the financial institu¬
tions involved and the nature of their involvement.

With respect to each of the global "public bads"— terrorism, corruption,

and failed states — chapter 7 concludes that each is individually complex
enough that an AML regime can only contribute modestly to combat it. On
terrorism, for example, the standard of zero tolerance, while defensible, is
essentially impossible to achieve. By the indicator of amounts frozen or
seized, the global AML regime has had limited success in combating ter¬
rorism since the end of 2001. Moreover, international cooperation in this
area has been uneven.

Conclusions and Recommendations

The anti-money laundering regime and its associated laws and regulations
represent a means to multiple ends. Money laundering is not itself the tar¬
get; the regime primarily aims to reduce the activities that generate the
money to be laundered (e.g., drug dealing, corruption, terrorism). Preserving

6 CHASING DIRTY MONEY

the integrity of the core financial system is a different type of AML goal, as
the aim is not so much to reduce money laundering as to move it to other
channels.

A central policy question is whether the anti-money laundering regime

needs to expand further, given the regime's rapid growth across countries
and financial and nonfinancial businesses and professions in recent years.
To date, very modest evidence that a particular channel (for example, real
estate brokers or life insurance agents) has been used to launder money has
provided a justification for bringing that channel into the net of AML reg¬
ulation. Little systematic evidence has been advanced that these extensions
of the AML regime, with the costs they impose on legitimate businesses
and their customers, will do more than marginally inconvenience those
who need to launder the proceeds of their crimes.
In the years ahead, it is likely that the pace of expansion of the AML
regime will slow and that the focus will shift to improving global imple¬
mentation of the current regime. As part of this consolidation process, in¬
creased cooperation will be important in a number of areas. Cooperation
between the public and private sectors is critical, since the current flow of
information is primarily from private to public, without significant feed¬
back. Another important area for cooperation involves technical and finan¬
cial assistance to poorer jurisdictions, in which an effective AML regime is
essentially a luxury good.
The international community will also have to continue to grapple with
the substantial differences in objectives, regulatory structures, and philoso¬
phies that impede effective coordination. Ratification and implementation
by the major nations of the new UN Convention Against Corruption will
be an important signal of willingness to cooperate on these matters.
For its part, the United States faces numerous challenges going forward,
such as satisfying the revised Forty Recommendations issued in 2003 by
the Financial Action Task Force, which was established by the G-7 summit
in Paris in 1989 to examine measures to combat money laundering. Among
the recommendations that have prompted debate in the United States are
those to expand coverage of the AML regime to lawyers, accountants, and
auditors, and to deal with special purpose vehicles, legal structures that are
sometimes used to disguise beneficial owners of assets.
The United States also needs to demonstrate its commitment to a strong
global AML regime by voluntarily submitting to a full IMF/World Bank
assessment of its financial sector, including regulations affecting money
laundering and terrorism financing. For the same reason, the United States
should expand the list of crimes committed abroad (including tax evasion)
that can lead to money-laundering prosecutions domestically.
Another recommendation is that the US executive branch resume prepar¬
ing a National Money Laundering Strategy on a regular basis, but in a
different manner than the five strategies produced from 1999 to 2003.
While the strategy does not need to be redone annually, it should provide

CHASING DIRTY MONEY 7

 systematic reports on progress in implementing the objectives identified in
preceding strategies, along with more analytical assessments of how well
the system is working.
The US government should also find ways to encourage better use of the
database of suspicious activity reports, which at present appears to be an
evidentiary supplement rather than a source of new cases. Banking regu¬
lators need to create a database of cases involving financial institutions to
examine the extent of the money-laundering threat to the core financial sys¬
tem and to assess progress in containing that threat.
Finally, the global AML regime needs to be strengthened through devel¬
opment of a systematic research program using economic tools, starting
with more sophisticated assessment of the costs of the AML regime. Other
important research-related activities include creating a database of exist¬
ing cases that provides a detailed description of the prices, methods, and
predicate crimes involved. This would represent a first step toward ana¬
lyzing the existence and mechanics of the market for money-laundering
services. The market-model framework for money laundering needs to be
better developed; of particular importance is whether the model can incor¬
porate more opportunistic modes of converting the proceeds of crime into
forms that cannot be traced.
Scholars are inclined to emphasize the importance of research, but in the
case of money laundering and finding ways to combat it, the need for
greater research is particularly acute. The fact is that, to date, an elaborate
system of laws and regulations that affects the lives of millions of people
and imposes several billion dollars in costs annually on the American pub¬
lic has been based to a substantial degree on untested assumptions that do
not look particularly plausible. While the failure to evaluate systematically
the AML regime has not as yet impeded its expansion either in the United
States or elsewhere, at some stage it should and most likely will. The sys¬
tem needs careful examination before any further expansion is actively
contemplated.

8 CHASING DIRTY MONEY

How Much Money Is Laundered?

Conceptually, money laundering is straightforward: the effort to conceal

the origins of illegally obtained funds that have been converted for legiti¬
mate purposes. As defined by the US General Accounting Office (GAO
1996, 1): "Money laundering is the act of converting money gained from
illegal activity, such as drug smuggling, into money that appears legitimate
and in which the source cannot be traced to the illegal activity." Terrorism
financing requires inverting this definition, as it more typically involves
money from legal pursuits that is converted into forms that facilitate acts
of violence for political purposes.
Estimating how much money is actually laundered in the United States,
any other country, or globally is extremely difficult. A sustained effort
between 1996 and 2000 by the Financial Action Task Force (FATF) to produce
such estimates failed.1 In fact, no direct estimates exist of how much money
passes through the financial system, whether broadly or narrowly defined,
for the purposes of converting illegal gains into a nontraceable form. Finan¬
cial firms lack both incentive and tools to estimate the extent of laundering in
their accounts, so it is unlikely that any such figure will ever be produced,
though changes in technology might help financial institutions in this respect.
What is known is the amount of laundered money the US government
identifies through its investigations. According to the 2002 National Money
Laundering Strategy — an annual report from 1999 to 2003 by the US Trea¬
sury to Congress on anti-money laundering efforts — seizures of money

1. One of the authors of this study (Peter Reuter) was involved in the latter stages of this
effort, which did not result in any official publication.

9
laundering-related assets in fiscal 2001 amounted to $386 million, while
the corresponding figure for forfeited assets was $241 million. Considering
the billions of laundered dollars believed to be out there, a few hundred
million dollars annually is a negligible share of the true total.
A number of estimates have been published of the potential demand for
money laundering, i.e., the quantity of funds generated through either
crime or legal activities that are concealed from authorities in order to evade
taxes as well as what Vito Tanzi (1980) in his classic article calls "various
governmental restrictions on the activities of economic agents." Estimating
this demand can be done through either a macroeconomic or microeco¬
nomic approach. The various macroeconomic methods might be expected
to produce a rough upper bound of how much money is laundered, simply
because it is impossible to make fine distinctions in these studies. The micro-
economic method, which involves summing up the income generated by
major crime categories, provides a loose lower bound on the demand for
money-laundering services.
A review of the two methods comes to a simple conclusion: neither yields
estimates of the volume of laundered money that can be considered as any¬
thing more than an indicative order of magnitude. Such figures are useful
to confirm that the phenomenon of money laundering is of sufficient scale
to warrant public policy attention, but their quality is not good enough to
provide guidance for policy. Moreover, the macroeconomic estimates are
methodologically flawed, while the available microeconomic estimates
lack credible empirical foundations.
These negative conclusions regarding the accuracy of these estimates are
based on a fairly detailed and technical review. Confronting this reality is
critical because ignorance about the volume of money laundering is an
important conditioning reality for policymakers as well as for any assess¬
ment of the AML regime's effectiveness. Without a measure of how much
money is laundered, the most obvious and direct outcome measure for an
anti-money laundering regime — reductions in the targeted activity — is not
available. More indirect measures must be developed. It seems unlikely to
us that direct measurement of the total volume can be improved in the fore¬
seeable future. Even measures of predicate crimes, the principal targets of
the control regime, are crude.
Such a negative assessment, however, does not imply an endorsement
of policymaking-by-anecdote in this or any other area. To the contrary,
a recurring theme in this study is that better use could and should be
made of available data on particular aspects of money laundering, and
greater thought should be given to collecting and assembling relevant sta¬
tistics that aid policymakers. Even these efforts, however, are unlikely to
produce a global aggregate of the amount of money laundered that can be
used to monitor progress in control efforts. As discussed in chapters 5 and
8, more disaggregated measures are needed to assess the seriousness of the
money-laundering problem.

10 CHASING DIRTY MONEY

Macroeconomic Estimates

The macroeconomic approach to measuring the extent of money laun¬

dering is based on a broad definition that assumes that any revenue on
which no tax is paid — be it from a legal or illegal activity — will need to
be laundered in some way.2 In this view, the demand for money launder¬
ing is related to the monetary component of the so-called underground
economy.
The study of what has been called the underground, shadow, or hidden
economy first emerged in the late 1970s (Gutmann 1977, Feige 1979). It has
since evolved into a considerable body of literature that proposes a host of
different techniques for estimating the size of the phenomenon. More re¬
cently, one particular estimation method, the currency-demand approach,
has been applied frequently enough to allow for comparison of many dif¬
ferent countries over one or two decades.
The first step of this method, developed by Tanzi (1980), estimates a
currency-demand function for the concerned country. The dependent
variable is the ratio of currency holdings to a broad measure of the money
supply such as M2, and on the right-hand side of the equation are series
for the following:

■ tax rates (higher taxes are expected to induce use of currency to evade
them),

■ share of wages and salaries in personal income (these types of income

are more often than others paid in cash),

h interest rates (as the opportrmity cost of holding currency), and

h real per capita income (as a proxy for financial infrastructure, the
degree of urbanization and mobility; higher income decreases the use
of cash).

The second step, which is based on the estimated equation and actual
values of M2, can determine for each year the level that currency holdings
would have been with taxes at either the lowest level of the sample period
or at zero. The difference between this estimated figure and the actual
observed cash holdings is then attributed either to the entire underground
economy when a zero tax rate is used, or to its increase since the year with
taxes at their lowest level. Assuming that income velocity of currency cir¬
culation is the same in the underground and legal economies, the third

2. Theoretically, it would seem reasonable to assume that no taxes are paid on income from
illegal activities, but there are rare exceptions, such as when an offender chooses to launder
criminal earnings through a legitimate business and pay some taxes.

HOW MUCH MONEY IS LAUNDERED? 11

Table 2.1 Taxonomy of underground economy activities
Activities Monetary transactions Nonmonetary transactions
Illegal Trade in stolen goods, drug manu¬ Barter (drugs, stolen goods, smug¬
activities facturing and dealing, prostitution, gling, etc.); production of drugs for
gambling, smuggling, fraud own use; theft for own use
Tax Tax Tax Tax
evasion avoidance evasion avoidance
Legal Unreported income Barter of legal All do-it-yourself
activities from self-employment; Employeefringe
counts, dis¬ services and bor help
work and neigh¬
wages, salaries, and benefits
assets from unreported
work related to legal goods
goods and services

Sources: Mirus and Smith (1997) and Schneider and Enste (2000).

step computes the corresponding estimates of the total volume of the

underground economy or its increase.3
What is the currency-demand approach measuring, and how well does
it match the broad working definition of money laundering used in this
study? Clearly, the estimates are measures of a cash-based underground
economy dependent on particular levels of taxes and extent of tax evasion,
which is in line with the definition. Tanzi (1980) himself defines the under¬
ground economy as the "gross national product that, because of unre¬
porting and/or underreporting, is not measured by official statistics."
The impact of tax evasion on national accounts is unclear, and depends on
both the particular taxes evaded (e.g., personal income taxes may be more
easily evaded than property taxes) and on the methods used for estimat¬
ing each of the components of the national accounts. As a result, Tanzi's
definition is probably too general to match the specifics of his own method,
but precise enough to provide a reasonable upperbound estimate of total
money laundered.
Friedrich Schneider and Dominik Enste (2000) also apply the currency-
demand approach. They present a "reasonable consensus definition" of the
underground economy based on an earlier schema tization by Rolf Mirus and
Roger Smith (1997). Their taxonomy (table 2.1) better describes what the
currency-demand approach estimates, and also illustrates conceptually a
broad measure of money laundered. Schneider and Enste distinguish
between monetary and nonmonetary transactions. The latter, by definition,
cannot be measured by the currency-demand approach and cannot in the
first instance give rise to any money to be laundered. Schneider and Enste also

3. A more restricted variation on this method is the currency-ratio approach described in

Feige (1997). It uses the ratio of currency to demand deposits and makes the restricting
assumption that this ratio is affected only by changes in the underground economy.

12 CHASING DIRTY MONEY

distinguish between legal and illegal activities. With the reasonable assump¬
tion that no taxes are paid on illegal activities, it is the sum of the legal and
illegal categories of monetary transactions in the underground economy that
give rise to money laundering in its broadest definition.4
Schneider and Enste provide roughly comparable estimates of the
underground economy for many countries based on the currency-demand
approach. Using their 2000 study as well as a later paper by Schneider
(2002), table 2.2 shows figures for 21 member countries of the Orga¬
nization for Economic Cooperation and Development (OECD) for
selected years since 1989. The figures illustrate the extraordinary magni¬
tude of money-laundering estimates using this approach. On the basis of
these estimates, the combined underground economies of these 21 coun¬
tries in 1997 totaled more than $3 trillion annually, and for single nations
the underground economy represented an average of 16 to 17 percent of
GDP. For all the years listed, the amounts estimated are more than 7 per¬
cent of global GDP, and since 1994 the figure exceeds 10 percent for most
years.5 This is substantially above the 2 to 5 percent of global GDP cited in
1998 by Michel Camdessus, then managing director of the International
Monetary Fund, as a "consensus range" for the scale of money-laundering
transactions. Even this lower guesstimate was described by Camdessus
(1998) as "beyond imagination."
Differences across countries may be partly explained by differences in
taxpayer cultures and by tax rates themselves. For example, southern
European nations (Greece, Italy, Portugal, Spain) and Belgium, all of which
are thought of as having a tradition of low tax compliance, have the high¬
est proportion of GDP estimated as coming from the underground econ¬
omy, with figures above 20 percent for all years since 1994. The heavily
taxed Scandinavian countries follow them closely. The United States,
Austria, and Switzerland have much lower rates, less than 10 percent.
Not surprisingly, considering the size of US GDP, the United States leads
the field in absolute terms, with $650 billion to $800 billion in 1995 dollars,
followed by Japan and Germany. Italy and Spain are the only countries with
both underground economies accounting for more than 20 percent of GDP
and contributions to the worldwide underground economy in absolute
numbers of more than $100 billion.
The absolute and percentage estimates are shocking if taken as measures
of money laundering. However, they are frail even in their own terms as
measures of what evades government taxation and other restrictions, and
even more frail as the basis for estimating the volume of money laundered.

4. Transaction measures are broader than value-added measures, however, in that the for¬
mer encompass the gross expenditures at all stages in the production of output, while the lat¬
ter are net of inputs. Thus, gross transaction measures of the underground economy tend to
overstate its size relative to value-added measures of the total economy.

5. Based on world GDP in constant 1995 dollars. See World Bank (2003b).

HOW MUCH MONEY IS LAUNDERED? 13

 LO
" S
LO °
.© 0 CL it e
LO
CO CD CO CO co cd
CM cd co
CM CM CO
0
O 06
co CO Cvj t z: Q_
CO CO CO q D
- T— T— CO CO CO
c 0O
— CO CM d d CM Cvi 03 cvi d 0
TJ CO CO CO CM CO CO CM CO CO 03 O c
(0 O —
cd 0 C/3.
=0 “0 o
0 O CO
0 OQ *- 10
CM id CO co cd O
LO 03
CO
0 0
LO 03 06
O
cd h-
0 cvi T- CM CM 10
E
0 9-
c D C3 1 T 0 O
— - CM CM
0 LO Is-
LO «- _ 03 o
10 CM
0 LO
CM to C30 LO T—
03 CM cvi 03 CM LO d 03 P
Q) 0 7;
T— LO LO 0
CD i?
1— 1— 1— Co to 00
0 *♦— T— CO co 0
Q. O
CO q io 10 cd
T-.1 — LO 1 —co
c/) CO CM o
03 CO cd CO CM t— co cd 0
CO t q CO CO cvi
q q 00 T- CM O
c 0 CO CM K CO d LO 06 CM o CO
Cvi
OO
q
co ooS _g
0
— TJ 0 CO CM 0 co Cvi o
1 — h- CO CM 10 y— « D ^
.2=5 —
q T CM c c: o ^
CD id
03 03 T— 0
O CO T- jo
03 to T— CM
03 C St 1 — CM CM "" 1 — I— co
cd E
o
0 0 03 CM cd cd IO 1- Is- 1^
■ 0
c
CO CM O CO * in h-
0 0 CM q 0
o T“ d 06 06 m in i—
T- CM CM
CM 03 cvi cvi 03 id d £ TO
o CM T“ CM cvj
LO IO—
r— inT
0
i_
0
CVJ CO o
0 H- 0 1- IO 1— 0 03 JO
I 0- 0 CO id LO CM 0
CM r^. co E -Q
oo
CD 0
03 ^ 0 cd
Cvj q
cd CO
03
1— t- d cd
co C *r c
o
CO <
LO
cd 1 — CO CM N N q1 — q
CD •02 »-
CM CO d d 03 cd 03 0
c 0 Is- c\T
—
n -o
— 03 CO CM CO CM0 CO CM CM 03 T— = o o
“3 C
CM CO CO s 0 =3 o
</> 03
IO Is- V— N- o CM
d>
03
CD •£
O ^r
CO
03
00 co
T-
TD 0o .Q
-Q - 0
T— Q_ <
r^
03 CO T— cd cd cd
cd 03 0
T—CO
c O O
LO
0 03
cd tL CO r--
C
0
CO
3 Q) QQ- cd 00 CM xf T 00 Co 0 N- CM
h- q T- CM CVJ ~ q O
c
c 03 03 r- 03 cd cdT
O T— T- d LO id d 0 0 .c
CM 0 c _C
CL o
o CO T-id CD
ID M- id LO h- co O
0. 0 LO 03 CD cd cd
Q co LO 0 ? a>
to 00 CO CO CM cd CM co c
O LO CM 00 to CM ° C o 0
UJ cd 03 03 q CO
Y- y^ t-q q i— 0 d
~
•—2 0 CM d CM
03 N- N*
d 03 lo qcd
CO T- dCO
CM q *r *D t/)
C nro cvT •- 0 o
O O — CM 00 CO CM CM CM CM CM co o
CM CO 03 CM q T- o
co T- O 03
CM
04 T-
0 T—
03
<3- CQ *- LO id CO T—Is-
03 CO 00 co >
2>c a
CO CO CM CO cd cd 0 o a 0
0 0 cd cd Is- cd T^ o o c
V) 0) O 0 03
cd CM
q CM co Is-
1- Cvj c\j
CM CM q O 0 LU
LO
T“ C CL CM
a> T— T— d 10
T~
CM d
Q-05 z
® H 0)
1— Is- 1
0 c0~
03 LO _ IN Cc
E 1— CO co
0

o 0 *♦—
CL O LO LO
CO CO
LO
CO
q ^ co
03
CM 03
Is- k_
c © CO cd cd CO co 00 0
o •02 »-
LO LO
CO 00 CM co
CM
Y-»
co q
t- cd
q q
LO
r-- Is-
2 iZ 0
CO
~
o c 00 d T^— d CM CM CM
d 03 03 1^ 03 c
CD — "D CM CM 0 03 T- cvi O 03 -C
■o

O — CM
T— ■q r-
CM CO 03
Is—
q CO 2* y~
-♦— Si To~
oo TO
03
C_3
10 C C
c 03
O LO LO CM
y—
El
3 CD m- r^-
CL
03 — T~ co
CM Oo>-£1— Q
0
O T CO Is- CM CO 0 Jc
i_ CO 1 —
O O cd —
1O 0
O cd t-; S 03 2
CD Q) O
,_ C D- lo d
CM
q q 0
CD N N
03 q 0 q O
0 Q) 0 ■O c
~o

i— cd CM
d CM o
cvi 03 d "" d d 0 -Q --c
CD 1—CM 1^ O T— 0
~o

T- O c
T3 O in CO CM ID CD cn
O »- 03 0 —~
0. 0 id id CO co Q- o-T
C CM 10 CM
0 k_
3 CO CM CO
"D

cd Is- lo q CM O 0
CO 03 LO 03 CM CM cvi cvi °o 0 CO X CL
~ CM CO q q q 1— q
a) c 00 CO d CM
d Is- 1- T- 00 d q 03 03
*D

JO d 0
CO co 03 O O 0 ^ 0
1— T— 03 CM T— c
0 T3= O O
10
CO
03
CD 'S
O
CO
CO q 03 03
CT
J E
q o
(0 03 CO co CM 1— 1— c
~o

CO CD CL
a) O (J 1— CM 06 00 CM “CD
3 >%
M- id CD 0
ro O Q
1- c Q- O 00
CO CM 00 Tt O
03
CM q CM q
d cd 03 CM
E d o £
®
'3

cvi CvJ
T— O o
~o

</> O *■ 03 O(/) ^o J*
o
a 0 03 y—
UJ t- Y- T— E c f— life P 0
tL 0
0 0
co
04 0 C 0 0)
0 c c 0 0 c/3
T— §
04 0
C
0 0 c -C 0 c 0
c
0 0
Is-
0
o
k.
0 E0
0 c
0 0 i_ 0 > =3 .£ 0 v_ 03 0 -Q E 0 CO
0
1— L_ 13 0 0 T3 O O c 0 N 0
n
c>. C E 0 c
0 JO T “ N 0 0 O
Q_ E 0
k- 0 ^ c
to D 0 03 c 0 .E 0 L_
0 ■*—> 0 0 > 0 >,0 T~ 0 i_ 0 0 0 LU
O 3 =3 0 0 0 L_ 0 0 0 0 0 0 0 003 Cl £ 5 c O E O > 0) CO “D
O i— 2Z2:aa)(D T3 C/33 O)
<<cooQaaoe; LU
Z) < 0 p-
1—
l§ T—
14 T—
'
By definition, the entire right-hand, nonmonetary side of table 2.1 is not
included in these estimates of money laundering, as these items were ex¬
cluded by the estimation method used by Schneider and Enste (2000) and
do not fit conceptually.
Moreover, the currency-demand approach has been extensively and con¬
vincingly critiqued (Tanzi 1999). First, its strong focus on taxes to explain
the size of the underground economy ignores the criminal trades, which
are not a function of tax rates but of other factors (such as the demand for
illegal drugs); it measures only the lower left-hand side of table 2.1. For
example, removing legal restrictions on casino gaming in the 1990s in many
American states may have reduced the size of the illegal gambling sector.
Second, the currency-demand approach assumes that the underground
economy consists of only cash transactions. Studies in Norway have sug¬
gested that cash may account for only about 80 percent of such activities
(Isachsen and Strom 1980, 1985). Moreover, some money laundering, such
as that associated with financial fraud, is done primarily, if not exclusively,
in noncash monetary transactions such as electronic transfers. Third, the
currency-demand approach assumes that income velocity is the same in the
hidden and the official sectors. As the two are likely to have very different
structures, that assumption is almost certainly false.
Fourth, the estimates presented in table 2.2 are based on the strong
assumption of no underground economy in the year with lowest taxation.
When designing the currency-demand approach, Tanzi explained that pre¬
dicting a currency-to-M2 ratio based on the lowest level of taxation over
the sample period would indicate the increase in the underground econ¬
omy since the year of that lowest tax level. Theoretically, Tanzi maintained,
one would need to put taxes at zero to estimate the entire underground
economy. Schneider and Enste are less explicit in laying out the framework
for their econometric work, but they do mention that a weak point of the
approach is the assumption of no underground economy in the base year.
The fact that the already enormous estimates would be further inflated if
based on a currency-to-M2 ratio predicted with zero taxes raises serious
doubts about the accuracy of the estimates.
Finally, substantial amounts of currency are held outside certain coun¬
tries. For countries with currencies used internationally (the United States,
Germany [prior to 1999], the euro countries, the United Kingdom, Japan,
and Switzerland to some degree), this factor might change the results sub¬
stantially, because currency held abroad should be excluded to obtain a
net volume of the domestic underground economy. If currency abroad is
not part of the domestic underground economy, it might well be in the for¬
eign underground economy. This implies that this method is not reliable
for purposes of measuring the volume of money laundering in the coun¬
tries issuing international currencies, but in principle may be appropriate
for constructing a global aggregate if one could reallocate the currency
appropriately.

HOW MUCH MONEY IS LAUNDERED? 15

 Unfortunately, there are no promising ways around these problems. For
example, it is difficult to find a proxy for illegal activities to include in the
equation. Nor is there any promising approach for estimating income veloc¬
ity in the hidden sector.
The reliance on currency as the basis for measuring money laundering
resulting from illegal drugs imparts the right bias by excluding the nonmon¬
etary sector and taking into account the huge amounts of cash in the drug
trade. That advantage will be lost with other macroeconomic approaches.
There are contending macroeconomic approaches. One alternative me¬
thod is based on the discrepancies between different components of na¬
tional income accounts. For example, nonreporting or underreporting may
affect income statistics differently than other national accounts categories
such as consumption. Second, one can compare official data on the labor
force with data based on household surveys, and take the difference as an
indicator of underground economic activity. The third alternative option
is the so-called physical input approach. Based on total real physical inputs
(most typically electricity), it estimates total GDP, subtracts official GDP
from it, and attributes the difference to underground activity. Lastly, the
underground economy has been estimated with an econometric method

that allows for estimating an unobserved variable "between" certain

observable causes and effects.6
When used individually, all of the alternative approaches have obvious
and very substantial shortcomings. For example, simple comparisons of
different national accounts categories may not be meaningful because this
procedure conflates measurement error with unrecorded activity. Labor
surveys may be biased, physical inputs measure only "underground in¬
dustrial" activity, and the associated econometric modeling requires a set
of strong assumptions.
The results themselves also reduce the credibility of the various eco¬
nomic approaches. Derek Blades and David Roberts (2002) make telling
comments in reviewing various estimates. Their main critique is that
nonobserved activities are highly concentrated in certain sectors of the
economy, such as retail trade, taxis, trucking, and restaurants, while other
sectors such as power generation, heavy industry, or air transport are
intrinsically less vulnerable. Taking this into account, the high estimates of
the size of underground economies would imply that much larger shares
of the susceptible sectors are underground, which hardly seems credible.
Considering, then, that the different shortcomings tend to bias the re¬
sults, and sometimes in different directions, it is best to take the different
figures with several grains of salt. At best, the various estimates suggest
that there is substantial potential demand for money-laundering services,
but there is little basis for concluding whether it amounts to hundreds of
billions or trillions of dollars.

6. See Dixon (1999) for a brief review of various methods.

16 CHASING DIRTY MONEY

 Before moving to the discussion of microeconomic estimates of the
amount of money laundering, it is important to recognize the substantial
work that has been done on estimating underground economic activities in
the context of the compilation of national economic accounts. Most coun¬
tries today use the national accounting standards and definitions set out in
the System of National Accounts (1993), and referred to as SNA93. A com¬
patible but somewhat more specific set of rules is in force for countries in
the European Union (EU), namely the European System of Accounts
(Eurostat 1995), known as ESA95. To determine what needs to be included

in complete national accounts, both systems use the notion of a "produc¬

tion boundary." Everything that lies within the boundary is considered to
have been produced, with some value added to it, and is to be included in
the accounts, while everything beyond the boundary does not count as pro¬
duction or value added and is to be excluded. Presented in a very simpli¬
fied way, according to SNA93 and ESA95, all goods and services produced
with inputs of capital, labor, or any materials lie within the production
boundary, with the sole exception of housing services provided by owner-
occupiers. Goods that are not produced with such inputs, such as natural
resources, are considered outside the boundary.7 The point is that for both
SNA93 and ESA95 the concept of a production boundary is independent of
whether an activity is illegal or performed in violation of certain regula¬
tions. Thus, many national statistical agencies have made considerable
efforts to estimate nonobserved economic activity in order to include it in
their GDP estimates. In particular, they have attempted to define in
detailed fashion the different sectors of the underground economy, and
to develop particular methods to measure each sector. Consequently, the
findings are much more reliable than estimates based on general macro-
economic approaches such as those discussed above.
Were they easily available and comparable, the national estimates could
be useful for developing an overall measure of money laundering. How¬
ever, although there has been considerable study and assessment by na¬
tional and international agencies of all the particular methods of examining
underground economies,8 practices at national levels vary widely. The most
inclusive effort to collect and compare national practices and estimation
results to date is a study by the UN Economic Commission for Europe
(UNECE 2002). The study includes detailed reports by 29 countries about
their methods of estimating underground activity. Eight are developed
market economies and OECD members that are part of the principal focus

7. See Blades and Roberts (2002) for a simple representation of the production boundary. A
more detailed account can be found in Jackson (2000, 120-34).

8. The most complete overview is probably the Handbook for Measurement of the Non-Observed
Economy published by the OECD (2002) in cooperation with the International Monetary Fund,
International Labor Organization, and the Statistical Committee of the Commonwealth of
Independent States.

HOW MUCH MONEY IS LAUNDERED? 17

of attention in this study: Belgium, Canada, Finland, Germany, Ireland,
Italy, the United Kingdom, and the United States.
Unfortunately, most of these countries, while adjusting for a host of dif¬
ferent nonobserved activities, usually estimate these types of economic
activities in a way that does not allow identification of a total figure for the
underground economy (for example, as a percentage of GDP). The first
exception is Italy, where the estimate of the underground economy as a
proportion of total estimated GDP was 15 percent for 1998, i.e., a little more
than half of what currency-demand-based macroeconomic estimates sug¬
gested for 1997 and 1999. The second exception is the United States, but in
this case the efforts are limited to a survey of the literature discussed below
in the section on microestimates. Moreover, US figures primarily cover the
period around 1980 and are thus outdated for purposes of this study.
Apart from these two countries, only Belgium and Canada provided esti¬
mates for the UNECE study of the correction in national accounts for
nonobserved activities. In Belgium, these activities were reported to
account for 3 to 4 percent of total estimated GDP in 1997, and in Canada
less than 2.7 percent in 1992. The figures for these two countries (less so for
Italy) are substantially below the estimates in table 2.2, demonstrating yet
again the weakness of the currency-demand approach.
Studies on underground economies related to national accounts also
have their weaknesses when used to measure the scale of money launder¬
ing, specifically with regard to certain important types of illegal activities.
First, some activities like theft or extortion do not fall within the produc¬
tion boundary. Second, even for the countries that do report on illegal
activities, the UNECE (2002) study reports that the estimates "can only be
regarded as indicative" due to "difficulties in measurement." Third, not¬
withstanding such exceptions as the United Kingdom and the United
States, most OECD members do not list any estimates for illegal activities
at all, which again limits comparability.9
Two main lessons emerge from our review of attempts to estimate the
amount of money laundering in the context of national accounts. On the one
hand, insofar as data are available, they only reinforce the previous skepti¬
cism about the results from currency-demand estimation methods. On the
other, the data show that there are sophisticated methods available that can
be refined and may contribute to more meaningful estimates in the future
of the size of the underground economy and volume of money laundering.

9. One reason for the limited coverage in OECD countries may be the confusion in EU norms.
While the ESA95 (enacted in a Council Regulation in 1996) requires inclusion of illegal activi¬
ties, a Commission Decision of 1994 excludes them from what is required of states to ensure
complete national accounts (Art. 1 Commission Decision 94/168/EC, Euratom). Interestingly,
Calzaroni (2001) notes that the Italian approach to estimating underground activity explicitly
states that illegal activities are not included in line with EU requirements.

18 CHASING DIRTY MONEY

Microeconomic Estimates

The microeconomic approach to estimating the demand for money laun¬

dering is in a sense a complement to the macroeconomic approaches,
which pay limited attention to estimating total earnings from criminal
activities, aside from tax evasion. The microeconomic approach focuses on
different types of crimes and on estimating the incomes from each. These
estimates normally do not include the informal economy or activities that,
though legal, are not reported in order to evade taxes. However, in princi¬
ple it would be possible to graft those estimates onto such measures. The
problems associated with the microeconomic approach basically involve
the paucity and unreliability of the data. In other words, what little data
are available are not worth much.
There have been two systematic efforts to provide estimates of incomes
generated by a broad range of criminal activities in the United States. Carl Simon
and Arm Witte (1982) cobbled together figures for the late 1970s. Indicative
of the uncertainty of their results, which they acknowledge, is the basis for
their estimate for income from prostitution. They started with estimates of
the number of acts of prostitution and the number that a full-time prostitute
would commit in the course of a year. This calculation resulted in an esti¬
mate of full-time-equivalent prostitutes of 80,000 to 500,000. Add in consid¬
erable uncertainty about the annual earnings of a prostitute, and the result
is an estimate of income that spans approximately an order of magnitude.
Under the auspices of the President's Commission on Organized Crime,
Wharton Econometrics Forecasting Associates, a US research firm, also
developed estimates of incomes from many different criminal activities
(Godshaw, Koppel, and Pancoast 1987). In its final report in 1987, the com¬
mission stated that organized crime produced an annual net income of
approximately $47 billion, costing Americans $18.2 billion and more than
400,000 legitimate jobs. These criminal groups — including traditional mob¬
sters, motorcycle gangs, and other emerging ethnic crime groups — had
nearly 300,000 members, with an additional half million people employed
in part-time, crime-related jobs. Nine of the commission's 18 members, it
should be pointed out, said that the final report's account of the income
sources of organized crime was inadequate.
The estimates used for this study for the proceeds from 34 crimes in the
United States cover one or more years during the period from 1965 to 2000.
The 30 sources used include both public agencies and private organiza¬
tions. For most crimes, estimates were available for only a few years. There
were data for 10 or fewer crimes for 14 of the years, for 11 to 15 crimes for
19 of the years, and for up to 22 crimes for the remaining 4 years. Sixteen
crimes had 10 or fewer years of data, while only nine crimes had more than
25 years' worth. The results are presented in table 2.3 and figure 2.1. Simple
linear projections were used to impute the missing years to generate esti¬
mates for all crimes for all years; some of the projections cover large gaps.

HOW MUCH MONEY IS LAUNDERED? 19

Table 2.3 Estimated earnings from criminal activity in the
United States, 1965-2000 (billions of current dollars)
Tax evasion included Tax evasion excluded
Estimated criminal Percent Estimated criminal Percent
Year income of GDP income of GDP
1965 49 6.8 26
18 2.5
1970 74 7.1 45 2.5
1975 118 7.2 78 2.8
2.7
1980 196 7.0 4.0
1985 342 166
8.1 209
1990 471 3.6
1995 595 8.1
8.0 206 2.8
2000 779 7.9 2.3
224
Note: Non-tax evasion crimes included trafficking in illicit drugs, human trafficking, burglary,
larceny-theft, motor vehicle theft, robbery, fraud, arson, nonarson fraud, counterfeiting, illegal
gambling, loan sharking, and prostitution. Tax evasion crimes included federal income, federal
profits, and excise tax evasion.
Sources: Office of National Drug Control Policy (2000, 2001); Simon and Witte (1982); GAO
(1980); Federal Bureau of Investigation’s annual Uniform Crime Reports ; Internal Revenue
Service; International Organization on Migration; Abt, Smith, and Christiansen (1985); Kaplan
and Matteis (1 967); Carlson et al. (1984); and Key (1 979).

A glimpse at one of the more complete years is presented in table 2.4. It

reveals the general composition of the US criminal economy according to
these estimates. In 1990, the most lucrative crime (in aggregate) was drug
trafficking (cocaine, heroin, marijuana, and other), which resulted in
approximately $97 billion in revenues. Fraud generated $60 billion while
gambling generated less than $8 billion. The total for all crimes with avail¬
able data amounted to roughly $470 billion, approximately 8 percent of US
GDP in 1990. To obtain a smoother time series for comparison, data points
were interpolated for the 18 crimes with the most consistently available fig¬
ures (table 2.3). According to these data, criminal income in the United
States rose from just under 7 percent of GDP in 1965 to around 8 percent in
1985, with the largest increase occurring in the early 1980s. However, when
tax evasion estimates were removed from the totals, the criminal share of
US GDP ranged from 2.5 to 4 percent, peaking in the late 1980s before drop¬
ping to the levels of the late 1960s.
The fraud estimate, by far the largest single item in most years, is partic¬
ularly frail. It comes from a report by the Association of Certified Fraud
Examiners (2002), which sent survey forms to 10,000 of its members, fewer
than 10 percent of whom responded. Respondents provided specific infor¬
mation about cases of which they were aware. But they also were asked to
estimate the percentage of revenues that would be lost in 2002 as a result of
occupational fraud and abuse. As the median figure was 6 percent, using an
estimate for US GDP of $10.4 trillion in 2002 leads to an estimate of
$625 billion. No effort was made to adjust for nonresponse or to ask whether
respondents were in fact in a position to make such estimates. Nor did the

20 CHASING DIRTY MONEY

Figure 2.1 Estimated criminal income as percent of US GDP,
1965-2000

1965 1970 1975 1980 1985 1990 1995 2000

Source: See table 2.3.

study consider whether GDP was the correct base to use for these calcula¬
tions. If each examiner estimated the share of the flow through his or her
corporation, then the right base was much larger, namely the total volume
of transactions through corporations.
Revenues from crimes involving stolen goods (burglary, larceny, and
robbery) are probably overestimated, since they are based on the reported
value of the stolen property. A victim may inflate the worth of an item to
receive a higher insurance payment. Even if the claimed amount is accu¬
rate, a fence or pawnbroker is unlikely to pay a thief the retail value for pil¬
fered goods; indeed, the standard figure used in research studies is that
fences pay 20 to 30 percent of the market value of the good, depending on
how easily it can be resold (Muscato 2003).
Even the estimates of revenues from drug sales, by far the most systemat¬
ically developed, have very broad confidence intervals, though the govern¬
ment only publishes point estimates. One can get a sense of the uncertainty
of these revenue estimates by examining revisions in the related estimates
of the number of drug addicts that are published as part of the same stud¬
ies. When calculated in 2000, the estimated number of heroin addicts for
1992 was 630,000; in 2001 this 1992 figure was revised up to 945,000 (Office
of National Drug Control Policy 2000, 2001). Nor is this an isolated exam¬
ple. The 2000 estimate for frequent methamphetamine users in 1998 was

HOW MUCH MONEY IS LAUNDERED? 21

 Table 2.4 Estimated criminal proceeds in the
United States, 1990
Proceeds in billions
Crime of current dollars Percent of total

Tax evasion 262.2 55.7

Cocaine trafficking 61.3 13.0
12.6
Fraud (nonarson) 59.3
Heroin trafficking 17.6
3.7
Prostitution 14.7
Loan sharking 14.0 3.0
2.9
Marijuana trafficking 13.5 3.1
8.0 1.6
Motor vehicle theft 1.7
7.6
Illegal gambling 4.8 1.0
Other drug trafficking
Larceny/theft 3.8 0.8
Burglary 3.5
Robbery 0.5 0.7
0.2
Human trafficking 0.2 0.04
Counterfeiting 0.02
Fraud arson 0.1 0.008
0.04
Total 471.1

Sources: Office of National Drug Control Policy (2000, 2001); Simon

and Witte (1982); GAO (1980); Federal Bureau of Investigation’s
annual Uniform Crime Reports ; Internal Revenue Service;
International Organization on Migration; Abt, Smith, and Christiansen
(1985); Kaplan and Matteis (1967); and Carlson et al. (1984).

356,000; one year later the figure was revised to 659,000. Estimating the
prevalence of rare behavior, particularly one that leads to erratic lifestyles,
is difficult, resulting in a corresponding uncertainty. Taking into account the
range in estimates in the numbers of drug addicts, the $70 billion estimate
of revenues from drug sales in 2000 is probably best thought of as some¬
where between $35 billion and $105 billion, with no particular central ten¬
dency (Office of National Drug Control Policy 2000, 2001).
Many of the estimates of criminal earnings have no known provenance.
The demand for numbers generates a supply (Reuter 1986). If having an
approximation of total income from the smuggling of human beings (for
example, illegal immigrants) is useful for a congressional hearing or regu¬
latory proceeding, then an organization will be found to produce such a
number. In reality, however, there are no shortcuts to reliably generating
this figure, and quantity and price estimates are critical. As it stands, there
are only broad estimates of the numbers of illegal immigrants, even
broader estimates of how many have purchased services from smugglers
,
and yet broader estimates of how much they paid on average. While it is
possible to do surveys in immigrant communities and obtain reasonable
response rates even from illegal immigrants (DaVanzo et al. 1994), it is not
known whether questions about the immigrant experience will generate
meaningful data.

22 CHASING DIRTY MONEY

 Outside of the United States, estimates of criminal earnings are sparser
and often equally implausible. Blades and Roberts (2002) report a small
number of such estimates for OECD and transition economies. Their fig¬
ures, admittedly partial and essentially guesstimates, are usually in the
range of 0.5 to 1.0 percent of GDP.
Even taken at face value, these numbers are only weakly related to
money laundering. Much of this income is earned by people who use the
cash to directly purchase legal goods without making use of any financial
institution. Small-time thieves earning $25,000 annually are unlikely to
make use of a bank or any other means of storing or transferring value.10 It
is impossible to estimate or even guess what share of these revenues will
require laundering.
Moreover, it is important in the context of estimating amounts of money
to be laundered to distinguish gross revenues from criminal activities and
the profits, after covering costs, from those activities. If criminals use cash
to purchase legitimate goods, such as airline tickets for couriers, then that
money laundering is quite different from most conventional notions of the
activity. In fact, the legitimate nonlabor inputs to criminal activity consti¬
tute a very modest share of total costs; almost all expenses are those of pur¬
chasing criminal labor, but the distinction may be important for some
offenses such as cigarette smuggling. A central question, then, is how much
of the income needs to be laundered? Perhaps money is laundered only
when it is exchanged into a different currency or when it tries to crawl up
into the legitimate world — the "above-ground" economy. By definition,
laundering turns dirty money into clean money. If the revenue from a crim¬
inal transaction remains in the underground economy, it can stay dirty
without stigma.

Conclusions

Neither of the two broad approaches to estimating how much money is

laundered — examining incomes in the broadly defined underground econ¬
omy or incomes from criminal activities — provides numbers that meet
minimal standards for policy guidance. The findings can support only the
broadest statements about the extent of money-laundering activities. The
underground economy, and even the criminal economy, probably amount
to hundreds of billions of dollars each in the United States. However, this
statement provides no possible guidance for assessing the effectiveness of

10. In most jurisdictions, the use of money acquired by illegal means to cover "living ex¬
penses" or operating costs is technically considered (self-) money laundering and often can
be prosecuted under anti-money laundering statutes. However, this is not an aspect of the
general phenomenon of money laundering that by itself would rise to the level of a public pol¬
icy problem.

HOW MUCH MONEY IS LAUNDERED? 23

money-laundering controls by comparing the volume of money laundered
across time or nations. If an estimate rises by 10 percent from one year to
the next, it is as likely to be the result of changes in coverage or estimating
technique as any change in the actual size of the underground economy or
of criminal earnings.
Moreover, as will be discussed in more detail in chapter 3, there are very
different social consequences of money laundering associated with dif¬
ferent offenses. Such offenses vary in the amount of associated harm per
million dollars, in their underlying nature, and in their distribution across
different segments of society. Therefore, a national estimate of the total
volume of money laundered would not necessarily have much value in
guiding policy or assessing policy effectiveness. For example, if an annual
estimate declined due to a reduction in large-value corporate fraud, but
at the same time low-value terrorism financing increased, policymakers
would not likely take any satisfaction from the reduction in the global
aggregate. This underlines the basic point that controlling money laun¬
dering is a tool directed at many, varied objectives. What is important, at
least in terms of measurement issues, is the amount of money laundered
in connection with an activity that the anti-money laundering regime is
specifically aiming to combat.

24 CHASING DIRTY MONEY

 3
Money Laundering:
Methods and Markets

Money laundering is usually described as having three sequential elements —

placement, layering, and integration — as defined in a report by the Board of
Governors of the Federal Reserve System (2002, 7):

The first stage in the process is placement. The placement stage involves the phys¬
ical movement of currency or other funds derived from illegal activities to a place
or into a form that is less suspicious to law enforcement authorities and more con¬
venient to the criminal. The proceeds are introduced into traditional or nontradi-
tional financial institutions or into the retail economy. The second stage is layering.
The layering stage involves the separation of proceeds from their illegal source by
using multiple complex financial transactions (e.g., wire transfers, monetary
instruments) to obscure the audit trail and hide the proceeds. The third stage in the
money laundering process is integration. During the integration stage, illegal pro¬
ceeds are converted into apparently legitimate business earnings through normal
financial or commercial operations.

Not all money-laundering transactions involve all three distinct phases,

and some may indeed involve more (van Duyne 2003). Nonetheless, the
three-stage classification is a useful decomposition of what can sometimes
be a complex process.
In contrast to most other types of crime, money laundering is notable for
the diversity of its forms, participants, and settings. It can involve the most
respectable of banks unwittingly providing services to customers with
apparently impeccable credentials. For example, Richard Scrushy, chair¬
man and CEO of HealthSouth, a major health care corporation, was
indicted on 85 counts, including fraud and money laundering. His finan¬
cial executives pleaded guilty to using false earnings reports to mislead
banks into providing a $1.25 billion credit line. Scrushy himself is alleged

25
 Box 3.1 Laundering methods of a drug trafficker
“Rick” launched his own drug trafficking operation using the funds of the cartel he once
served. With the help of former associates, he used several methods to launder the pro¬
ceeds. Cash shipments arrived by boat or plane and were promptly placed by couriers
into a range of bank accounts (a process known as “smurfing”), an activity that corre¬
sponds to the placement phase of money laundering. An agent then moved the funds to
the personal accounts of overseas intermediaries, each of whom arranged to transfer the
funds back into the country into accounts at the national central bank, which granted
authorization.
At this point, Rick would call the intermediary to cancel the transfer. The funds were
then withdrawn in cash from the intermediary’s account and wired back in country to
other accounts, using the authorization from the national central bank to explain the
origin of the funds. Without knowing it, the central bank was giving legitimacy to drug
monies.
After this layering phase, Rick purchased real estate with the funds, using lawyers, bank
managers, and other professionals, which moves the process to the integration phase. He
offered unusually high commission rates (3 to 5 percent) to gain the cooperation of the pro¬
fessionals with whom he was doing business. The real estate purchases were usually
made in the names of other individuals or companies.
Eventually, several of the banks noticed that his account activities were rather odd and
notified the national financial intelligence unit. An investigation revealed that Rick’s
scheme had laundered tens of millions of dollars over several years.

Source: Egmont Group (2000).

to have used personal checks, cashiers' checks, and wire transfers to pur¬
chase nearly $10 million worth of high-value goods and real estate during
the layering phase of this laundering operation.
Money laundering can also involve small nonfinancial businesses know¬
ingly providing similar services to violent criminals, as in the case of truck¬
ers smuggling large bundles of currency out of the country for drug
traffickers.
Money laundering does not require international transactions; there are
instances of purely domestic laundering.1 Nonetheless, a large number of
cases do involve the movement of funds across national borders. Though
governments have unique police powers at the border, those same borders
can impede the flow of information. Thus the description and analysis in this
chapter place heavy emphasis on the international dimensions of money
laundering.

1. Just to cite one example, in the United States v. Clyde Hood et al, Central District of Illinois, an
indictment returned on August 18, 2000, charged the defendants with fraud for collecting
checks from investors, who were promised a 5,000 percent return. Funds were deposited in
checking accounts and used to incorporate and support participants' businesses, as well as to
purchase real estate, all within the Mattoon, Illinois, area.

26 CHASING DIRTY MONEY

 Box 3.2 Embezzlement and (self-) money laundering
Several officials of the Washington, DC Teachers Union (WTU), including president
Barbara A. Bullock, were implicated in a recent scandal involving the theft of $4.6
million.
The astonishingly simple scheme had several concurrent elements. One involved

Bullock’s chauffeur, Leroy Holmes, who in February 2003 pleaded guilty to laundering
more than $1 .2 million. Many of the more than 200 checks Holmes cashed were made out

to creditors such as Verizon or the DC Treasurer, with the original payee’s name crossed
out and replaced with Holmes’ name. He often left Independence Federal Savings Bank
with his pockets stuffed with as much as $20,000 worth of bills. The bank never filed either
the required currency transaction report or suspicious activity report and may face inves¬
tigation for colluding in the union’s money-laundering plan.
In addition, the WTU made several payments totaling $450,000 for the “consulting ser¬
vices” of a phony company called Expressions Unlimited. One of the company’s partners,
Michael Martin, claimed to be Bullock’s hairdresser but has since pleaded guilty to money¬
laundering conspiracy charges.
Union credit cards were used to buy expensive clothing, electronic equipment, art¬
work, and other costly items. As of February 2004, Bullock had been sentenced to nine
years in prison following a guilty plea, and four others had been indicted.

Source: Washington Post (various editions, 2003 and 2004).

Boxes 3.1 through 3.4 are examples of money laundering that illustrate
the variety of clients, providers, and methods involved. The chapter then
goes into more detail about the “market'' for money laundering — what is
known about the providers and prices they charge. The final section pre¬
sents a typology of offenses intended to provide a structure for policy
analysis in dealing with the heterogeneous set of offenses that engender
money laundering.

Laundering Mechanisms

A striking feature of money laundering is the number of different meth¬

ods used to carry it out. Some of the major mechanisms described below
are associated with only one of the three phases of money laundering,
while others are usable in any of the phases of placement, layering, and
integration.
Four methods of money
laundering — cash smuggling, casinos and other
gambling venues, insurance policies, and securities — are described below
in some detail. A number of others that may be of importance are listed in
box 3.5. The descriptions draw heavily on the FATF's annual typologies
reports, which list notable cases that illustrate the variety of laundering
techniques used.

MONEY LAUNDERING: METHODS AND MARKETS 27

 Box 3.3 “Underground” banking that finances human
smuggling
A South Asian man ran a small business with an annual turnover of around $150,000.
His banks were understandably surprised to see that between $1 .7 million and $3.5 mil¬
lion flowed annually through his private accounts for three years. Their suspicious trans¬
action reports triggered investigations that revealed that the suspect’s business was the
headquarters of an international “underground bank” with “branches” in several Central
Asian and European countries. Along with small amounts intended to support relatives
in the transferring parties’ home countries, this illegal banking system was used to trans¬
fer large sums for smuggling people into Europe. In May 2000, the suspect and one of
his branch managers were arrested. He had squirreled away around $140,000 in cash
in a safe and had purchased his home for $400,000 in cash shortly before his arrest.

Source: FATF (2002b).

Cash Smuggling

One of the oldest placement techniques, common smuggling of currency,

seems to be on the rise. Bulk shipments are driven across the border or
hidden in cargo, even though it is illegal to export more than $10,000 in
currency from the United States without filing a Report of International
Transportation of Currency or Other Monetary Instruments (CMIR). Cri¬
minals have even been known to purchase shipping businesses so that they
can store cash inside the goods. Individual couriers transport cash in
checked or carry-on baggage or on their persons. Smugglers can also sim¬
ply use the mail or a shipping company such as UPS or FedEx. US customs
officials spend most of their resources inspecting people and cargo coming
into the United States, so it is relatively easy to ship currency to another
country.2 Also, cash stockpiling (allowing cash to accumulate while wait¬
ing for a smuggling opportunity) is thought to have increased, particularly
in port or border regions. If cash smuggling has grown overall, it may be
partially attributed to the success of banks' antilaundering measures.

Casinos and Other Gambling Venues

Casinos. Chips are bought with cash, then after a period of time during
which gambling may or may not take place, the chips are traded in for a
check from the casino, perhaps in the name of a third party. When a casino

2. The authority to search in the United States does not distinguish between entry and exit.
However, historically there has been more interest in preventing the entry than the exit of
inappropriate goods and people. Nonetheless, the US Customs Service does occasionally use
its authority for exit inspections.

28 CHASING DIRTY MONEY

 Box 3.4 Pilfering by a media baron
Flamboyant Czech-born British businessman Robert Maxwell used the New York Daily

News as a money-laundering device, tunneling nearly $240 million through the tabloid’s
accounts during the nine months he owned the newspaper. In an audacious embezzle¬
ment endeavor, he siphoned pension funds from Maxwell Group Newspaper PLC in
London and deposited them in accounts controlled by the Daily News’s parent company
in the United States. Within days, wire transfers would move the money to hundreds of
other companies that only he could access. Maxwell engineered several bank loans to

the newspaper, large portions of which never showed up on the publication’s ledgers.
After his mysterious drowning death in November 1991, allegations surfaced that
Maxwell also laundered money from weapons sales to Iran.

Source: Robinson (1996).

has establishments in different countries, it may serve as an unwitting

international launderer if a customer requests that his or her credit be made
available in a casino establishment in another country. In addition, tokens
themselves may be used to purchase goods and services or drugs.

Horse racing. Winning tickets are bought at a slight premium, allowing

the winner to collect his or her money without tax liability and enabling the
launderer to collect a check from the track. Relevant taxes will be deducted
from this amount.

Lotteries. As at horse tracks, winning tickets are purchased from the win¬
ners as they arrive at the lottery office to collect their winnings. In a case
believed to be a common type of operation, a launderer placed many low-
risk bets at various bookmakers within his city, ending up with a long-term
7 percent loss rate — an unusual pattern and poor record for a professional
gambler. He had the checks for the winnings made out to 14 bank accounts
in the names of 10 different third parties, some of whom happened to be
armed robbers and their immediate families (FATF 2002b).

Insurance Policies

Single premium insurance policies, for which the premium is paid in an

upfront lump sum rather than in annual installments, have increased in
popularity. Launderers or their clients purchase them and then redeem
them at a discount, paying the required fees and penalties and receiving a
"sanitized" check from the insurance company. Insurance policies can also
be used as guarantees for loans from financial institutions. Many insurance
products are sold through intermediaries; consequently, insurance com¬
panies themselves sometimes have no direct contact with the beneficiary.

MONEY LAUNDERING: METHODS AND MARKETS 29

 Box 3.5 Other money-laundering methods
Structuring or “smurfing.” This involves breaking down cash deposits into amounts
below the reporting threshold of $10,000. Couriers (“smurfs”) are used to make the
deposits in several banks or to buy cashier’s checks in small denominations.
Informal value transfer systems. These which include hawalas, an Arabic word for a
particular international underground banking system. Handed cash in country A, a hawal-
adar can turn it into cash (or sometimes gold) in country B. The hawala includes the com¬
plete service from placement to integration. Similar services are provided under other
names in other parts of the world, such as fe chi’en in China.

Wire and electronic funds transfers. These refer to a method through which banks trans¬
fer control of money by sending notification to another institution by cable (in the past) or
electronically. Such transfers remain a primary tool at all stages of the laundering process,
but particularly in layering operations. Funds can be transferred through several different
banks in several jurisdictions in order to blur the trail to the source of the funds. Or transfers
can be made from a large number of bank accounts, into which deposits have been made
by “smurfing” to a principal collecting account, often located abroad in an offshore financial
center.

Legitimate business ownership. Dirty money can be added to the cash revenues of a
legitimate business enterprise, particularly those that are already cash intensive, such as
restaurants, bars, and video rental stores. The extra money is simply added to the till. The
cost for this laundering method is the tax paid on the income. With companies whose trans¬
actions are better documented, invoices can be manipulated to simulate legitimacy. A used
car dealership, for example, may offer a customer a discount for paying cash, then report
the original sale price on the invoice, thus “explaining” the existence of the extra illicit cash.
A slightly more sophisticated scheme may allow a criminal to profit twice in setting up
a publicly traded front company with a legitimate commercial purpose — first from the
laundered funds commingled with those generated by the business, and second by sell¬
ing shares in this company to unwitting investors.

“Shell” corporations. These exist on paper but transact either no business or mini¬
mal business. A related concept, used mostly in the United States, is the special purpose

(box continues next page)

In addition, relatively complex cases involving single premium contracts

have recently been discovered, involving slower procedures and less liq¬
uid transactions. These longer-term processes offer criminals a lower risk
of detection — in essence, time itself provides the layering by separating
chronologically the predicate crime from the eventual payoff. Evidence
also suggests forays by money launderers, or those seeking to launder
money, into the reinsurance industry, attractive because of its relative lack
of regulation. Such transactions allow for more layering.

Securities

The securities sector is characterized by frequent and numerous transac¬

tions, and several mechanisms can be used to make proceeds appear as

30 CHASING DIRTY MONEY

 Box 3.5 (continued)
vehicle. These are set up, usually offshore, complete with bank accounts in which money
can reside during the layering phase. The shell corporation has many potential uses. One

example is to buy real estate or other assets, then sell them for a nominal sum to one’s
own shell corporation, which can then pass the funds on to an innocent third party for the
original purchase price.

Real estate transactions. These can cloak illicit sources of funds or serve as legitimate
front businesses, particularly if they are cash intensive. Properties may be bought and
sold under false names or by shell corporations and can readily serve as collateral in fur¬
ther layering transactions.

Purchase of goods. This practice can be particularly attractive for laundering, especially
certain items. Gold is popular because it is a universally accepted store of value, provides
anonymity, is easily changed in form, and holds possibilities of double invoicing, false ship¬
ments, and other fraudulent practices. Fine art and other valuable items such as rare stamps
are attractive for laundering purposes because false certificates of sale can be produced, or
phony reproductions of masterpieces purchased. Moreover, the objects are easily moved
internationally or resold at market value to integrate the funds.

Credit card advance payments. A credit card holder may make a large payment with
dirty money to the issuing bank, resulting in a negative balance due. The bank then pays
out the balance with a check, which can be deposited into a personal account as appar¬
ently clean money. In recent years, increased bank scrutiny of these transactions has
discouraged this money-laundering technique.

Currency exchange bureaus. These are not as heavily regulated as banks, and de
facto, at least, may not be regulated at all, so they are sometimes used for laundering.
Substantial foreign exchange transactions are said to be shifting from banks to these
small enterprises. Two main laundering techniques are used. The first is to change large
amounts of criminal proceeds in local currency into low-bulk European currency for phys¬
ical smuggling out of the country, and the second is electronic funds transfer to offshore
centers. In one reported case, a currency bureau reportedly exchanged the equivalent
of more than $50 million through a foreign bank without registering these transactions in
its official records.

legitimate earnings from the financial markets. In addition, securities trans¬

actions often are international. The sector most commonly is used during
the layering and integration phases, since most law-abiding brokers do not
accept cash transactions. However, this obstacle is not an issue for crimi¬
nals operating within the financial sector itself, such as embezzlers, insider
traders, or perpetrators of securities frauds, because their (usually non¬
cash) funds are already present in the financial system. During the layer¬
ing phase, a launderer can simply purchase securities with illicit funds
transferred from one or more accounts, then use the proceeds from selling
these securities as legitimate money.
Unlike regular securities, bearer securities (common in some European
countries) do not have a registered owner, and when they change hands the
transaction involves physically handing over the security, thus leaving no
paper trail. The security's owner is simply the person who possesses it. Many
MONEY LAUNDERING: METHODS AND MARKETS 31
 but not all countries and jurisdictions have phased out the use of bearer
shares because of their potential role in money laundering and tax evasion.
Another laundering mechanism is the completion of simultaneous "put"
and "call" transactions (in essence, "side bets" on a stock's gain or loss) on
behalf of the same client, who pays with dirty money. The broker pays out
the winning transaction with clean money (minus a commission) and
destroys the losing transaction to avoid suspicion. Technically, the client
has only broken even with this deal, but profit is not the ultimate objective.
In its annual typologies reports on recent trends in money laundering, the
Financial Action Task Force (FATF) reports that some countries have seen
a significant shift in laundering activities from the traditional banking sec¬
tor to the nonbank financial sector, as well as to nonfinancial businesses
and professions.3 Even where the nonbank financial sector is subject to
anti-money laundering rules, organizations in these sectors are less willing
to abide by them, a reluctance that likely accounts for the relative paucity of
suspicious transaction reports originating from the nonfinancial sector.
Legal and accounting professionals in particular cite privacy concerns, but
FATF experts suggest that the lack of public pressure may also play a role
(FATF 2002b).

Which Methods Are Used for Which Crimes?

A reasonable conjecture is that different methods are used for laundering the
proceeds from different predicate crimes. The annual typologies reports of
the FATF and a report published in 2000 by the Egmont Group of Financial
Intelligence Units describe recent cases that illustrate methods of laundering
and investigation. Given that these are simply reported cases, they do not
necessarily reflect the relative importance of different techniques. With that
qualification, the FATF and Egmont Group reports can be used to develop
a matrix matching 11 predicate crimes with 20 money-laundering methods
(table 3.1). There were 223 cases available for classification, and each case
involved one or more offenses and methods of laundering, thus producing
a total of 580 entries.
Three offense categories accounted for over 70 percent of entries: drugs
(185), fraud (125), and other kinds of smuggling (92). The types of launder¬
ing methods were more evenly distributed — wire transfers were involved
in 131 cases (22 percent), but no other single method was involved in more
than 75 cases. For the three major offense categories, the observations were
broadly distributed across methods.
While these findings offer some insights into the laundering methods
used for different offenses, the results should not be overemphasized.

3. The report offers no systematic evidence to support this statement, and it is difficult to
identify a current database that would allow any agency to do so. But the conjecture is plau¬
sible, and an analysis of a fuller sample of actual cases would shed some light on its accuracy.

32 CHASING DIRTY MONEY

 CO
t- 03 CO
CO
i- CD Tt O
lo t- m lo CO CM CM 03 CO
CD C\J CM CO CM CO CO CM CO
03
CO
CO
c CO
a) 5 CO CD -r-

C\J C\J CO
O c LO

°
1

CO

CD C\J

i- CM
CO LO O
tJ" CM CO
CM O Tt
CO CM CM LO
O h-
CM
c LO
H
x .25 CO CM
0 </>
0
cm r-
CM
CO
J-0) 4->CL CM CM

.*9 3
m o
o

r-.
■D

co
<D U)
O O
c £
£
■C •—
o3

E
T3 U)
c a>
C
(0 1= 3 CM
LO CM CD CM
CO CD i
II
co
a)
cn S-
o
c CM
CM
03
CO
a
CO LO
0
5=
O
3
0 CM
CD o
CO
CQ
O LO
-U

o LO
00
CO
CD CD
C
T- CO
a. O) 5o CM O CO CO LO CO o ^ CO
w
3

o o £
0

>*
o
c.
CD
CO
i
i_ co wf■° 0 §|0
D 0
CO CO
8 3
Q- co0
0
o 0 ■O w ^
CD
cr !c 0 0 c o
> 0 „c
0 o 0 0 0
c to
0 JZ 0 c
)O CD
0
_c
0 *5 c o
o 'o "2 OJ c
CL C
X 03 toC0 0 O C
CD
o CL 0
=5 9
.2 O
o
0
co i- co
CD -* > 0 iz co o0
0 0 > 0 0 ^ E
0 CL
o
o 0® 3o .g0
CO “8^
c 0 s
E .N o 0 0^0
0 0 o $= co CO
0
nco ° ^ co03 £S
m<
■h 3 0
o co ro "g
_ 30g. 0 §>lll0 0
>F 0 e
C 0

C/D O
3
o LL
0 O*r 0-
CC _ l <0 0 o
oc oE» o Q o ^5 .ca>

33
r- £ 0 O
Neither the FATF nor the Egmont Group makes any claim to be offering a
representative sample of cases. However, the information does have some
value. For example, the data show that drug traffickers and other smug¬
glers use a wide variety of methods for laundering the proceeds of their
crimes. More weakly, they suggest that some methods are not much used,
such as alternative banking systems and trusts and securities.4

Who Provides the Laundering Services?

The only information available as to who launders money comes from

criminal and civil investigations, and the data represent the interaction of
enforcement tactics with the underlying reality. Enforcement may aim pri¬
marily at operations that are more professional (because they are higher-
value targets) or less professional (because they are easier to catch). Drug
dealers' money launderers may get more attention because the dealers
themselves are under more intense scrutiny. A substantial share of all re¬
ported US money-laundering cases involve drugs (chapter 5). Thus, the fol¬
lowing observations about available cases are merely indicative.
The most obvious nexus between the criminal and financial realms would
be persons inside the financial institutions themselves. Bank employees can
be coerced or bribed not to file suspicious activity reports (SARs) or cur¬
rency transaction reports (CTRs). Alternatively, the forms may be filled
out, with the government's copy conveniently filed in the trash while the
other copy remains in a drawer in case of an investigation.5
Lawyers are thought to be among the most common laundering agents
or at least facilitators, though they have been at the center of few cases in
the United States. A lawyer can use his or her own name to acquire bank
accounts, credit cards, loan agreements, or other money-laundering tools
on behalf of the client. Lawyers can also establish shell corporations, trusts,
or partnerships. In the event of an investigation, lawyer-client confiden¬
tiality privileges can be invoked. In one case cited by the FATF in its 1997-98
typologies report, a lawyer charged a flat fee to launder money by setting
up annuity packages for his clients to hide the laundering. He also arranged
for credit cards in false names to be issued to his clients, who could use the
cards to make ATM cash withdrawals. The card issuer knew only the iden¬
tity of the lawyer and had no knowledge of the clients' identities.
Other professionals involved in money laundering include accountants,
notaries, financial advisers, stockbrokers, insurance agents, and real estate

4. A Dutch study reports some details on a sample of cases involving money laundering (van
Duyne 2003). The sample was dominated by drug cases and most involved relatively simple
means of laundering.

5. Electronic filing, which would eliminate this option, is not currently required, at least not
in the United States.

34 CHASING DIRTY MONEY

agents. A British report on serious and organized crime noted that in 2002,

"purchasing property in the UK was the most popular method identified,

involving roughly one in three serious and organized crime groups where
the method was known" (National Criminal Intelligence Service 2003, 53).

Markets for Laundering Services

Since money laundering is a criminal service offered in return for payment,

making laundering services more expensive would reduce their volume
and thus the volume of predicate crime. Price might thus serve as a per¬
formance indicator. Unfortunately, law enforcement agencies do not sys¬
tematically record price information acquired in the process of developing
money-laundering cases, since that information is not necessary to obtain
a conviction.6
Moreover, price is an ambiguous concept in this context. Apart from the
fact that some laundering agents provide only partial services (for exam¬
ple, placement or layering), there are at least two possible interpretations
of price: first, the fraction received by the launderer, including what he or
she paid to other service providers, and second, the share of the original
total amount that does not return to the owner's control. The latter share
could include tax payments, as in the case of a retail proprietor who might
charge only 5 percent for allowing the commingling of illegal funds with
his or her store's receipts, but then might have to add another 5 percent for
the sales tax that would be generated by these fraudulent receipts.7
The policy-relevant price is the second of these, i.e., the difference be¬
tween the amount laundered and the amount eventually kept by the
offender. Pushing offenders to use laundering methods that involve smaller
payments to launderers but higher total costs (for example, because of taxes)
to the predicate offender is indeed preferable to raising the revenues
received by launderers as a group; after all, the difference may include pay¬
ments to the public sector. Such substitution might occur if the government
mounted more sting operations aimed at customers.
The difference is by no means only of theoretical interest. Take, for exam¬
ple, one case cited by the Egmont Group (2000) of high-priced laundering
where most of the price did not accrue to the launderer. A credit manager
at a car loan company was suspicious about one of his customers. "Ray"
had just bought a luxury sports car worth about $55,000, financing the car
through the credit company for $40,000, and paying the balance in cash.

6. The 2002 US National Money Laundering Strategy noted the importance of collecting such
data.

7. It is possible, of course, that this laundering will generate income tax payments. This
depends on the skill of the firm in generating false expenses. However, the sales tax is an
unavoidable consequence of inflating gross retail revenues.

MONEY LAUNDERING: METHODS AND MARKETS 35

Records showed that Ray had taken out several loans over the past few
years, all for the same amount of money and with a large portion as a cash
deposit. In many cases the loans had been repaid early with cash. The
national financial intelligence unit realized Ray was laundering for a long-
established criminal organization, putting cash from the sale of drugs into
the banking system. He would resell the newly bought cars, obtaining
checks to deposit into a single bank account, in all totaling over $300,000.
The losses made on the loan and the drop in the automobiles' resale values
were the cost of obtaining "clean" money.
Information about the price of money-laundering services is scattered and
anecdotal. In the money-laundering activity targeted by Operation Polar
Cap, a coordinated law enforcement sting operation during the late 1980s,
the drug trafficker would pay only 4.5 percent to the government sting laun-
derer initially, but was willing to go to 5 percent if the laundering were done
rapidly (Woolner 1994, 43). Later in the operation there were reports of
much higher margins. Experienced investigators refer to a general price
range of 7 to 15 percent for laundering for drug dealers, but some reports
are inconsistent with such estimates. One National Money Laundering
Strategy (US Treasury 2002, 12) reported a study that found commission
rates varying between 4 and 8 percent but rising as high as 12 percent.
Other criminals pay much less for money-laundering services. For ex¬
ample, John Mathewson, who operated a Cayman Islands bank that laun¬
dered money for a number of white-collar offenders (e.g.. Medicare
fraudsters, recording pirates) and US tax evaders, charged a flat fee of
$5,000 for an account, plus a $3,000 per annum management fee (Fields
and Whitfield 2001). Mathewson, who provided a complete set of ser¬
vices, also kept 1 percent of the float that the clients' money earned when
held overnight by other banks (US Senate 2001a, b).
The price paid for a particular money laundering service apparently is
partly a function of the predicate crime and the volume of funds that
needs to be laundered. Whereas legitimate financial transactions generate
lower per-unit costs the larger they are, the opposite is true for money
laundering — the risk of detection is a major cost, and that risk will rise
with the quantity being laundered. On the other hand, a broker involved
in Colombian black market peso operations stated in an interview that he
charged less for larger volumes of money. He once garnered between
$600,000 and $700,000 (5 to 6 percent) on a $12 million transaction that
took two months to process.8
Table 3.2 provides information on a few cases for which some data on
prices are available. The data are merely illustrative and so sparse that no
inferences about price trends can be drawn.

8. Public Broadcast Service (PBS) interview www.pbs.org/wgbh/pages/frontline/shows/

drugs/ interviews/david.html.

36 CHASING DIRTY MONEY

 "O 0 t
CO 0 £ca) £2
c
o o
0 0 0 3 o
it -C> 0
O)?
0
~o ^ =20CD
TO TO
■g~0 Esz
>-
0 o

■oa> 25 t-,cd c 0 O 0 _
o 0 s0 i

'0
"O
_Q 0
o > 0 030 _X 0
° "2 c 0 o
■oC2
0 o- £
TO °g
=3 —CD O -C
o c 0
O c0

'0
CO o
o E
(- w
co
0 § ” o o
co
CD “
O)
•H* C
C c
O <d Jf; O c
0
0 <:
0 0 .. CO
0 CD C 4= o
c 0 TO 0X 0 0
0 £ o
0
O E 0
CL
E O .t=
0 E
O
E ° ■77
0 0 0 0
o JS O Q_ 0
0
LO 0
0 I*
(T< Q.
0 CL 0
0

O 1
8^ COO -O
, ■
a) 0 0 0
o -0
oo
Iw
CL °

C i. <D
—D to
0) "O c
>
OCO 1—
in-, CO
cn- TO
cC
CD o
E cdE o CD CD W CO c
3^TO§
g?©
P ^
c T“
5}u >O E 3 c CO T3
dollars) current (in costs money-laundering

0 0
'
C\J
■o CO CO CO¬„ CO
c 0 c
Z> 5 8 «
c CO
'4= o 0 0 20 o 0 0

” 0co~ c=3
0
0) 0 C JxC
0
o
0 C
80 p
0o 4=o
C 0

If Ecu oo
o
0 . 0
o = CD 0 O
^ TO 2i
C 03
0
0
< 0
oc 03
CO
-■§255
C X r .. O
0 CO o C 2
•i 0 tf 0 0
0 t 0 0
0 0
03 0 E 0TO§
o S03_
i8
■*-» C ^ 0 CO 0 § 0c m 0
Q ^ C
4J U
(/) CD
E .S .c . 0 0 0 03
0 .tZ
0 E -g_ TO TO
It =3 0 2 E §■ 0c_j
a-o Q _ I
O 0 C TO
i_5 0
o® ^ E ZO
_0 0
CL
a
< <
^CC

o ^
LO
0 —
C? = 0 C
OO o
II
3 4) I i CO
o
o
CD
S c 0"
0
o
o -o CD
CO
CD
CO
oo
CO 0
CO t
CO
OO 00
it

it
'S■5 gTO cCO c ■0
of Examples

lido co~ =3 to 03 0
_Q
03 iz0
=3 CO 0 cr
_Q
O =3
d> 3= 0
° 8 T3

£ °
3.2

o 0

-X E
c p O

c ^
Table

0 C
o
13 c 0 §
0 03 CD
0 c
o 0 3 X

> o 0 =>

37
0^
 cD T3
0 0 O)
tr 13
E1/3 o0>-=
8.8 OTOCl) cTO
03 >
CL TO ~ o coE CL
r= CL TO 0 »- > 0 £E E.
oo
> 2 H
CD
o.-§
O TO TO§ O -D =LQ ^ o
c O 03 0 0
> c. C
E
O)
o tc
E?
C $
03 ^ C
CO
0 § § c
E _ 2 TO

'0
o S) 0 0 >, 0 O _ o 0 0O
o 03 C 0 CL E §OJ03 2 E0 N 0 O 28 * rf
c -D S 8 g> E i5 co -C
o ?T3 O)

'0
> E 0 CL
Og i:CO TO^ to Q O o _Q
LL

TO fOl £ CL o
$ 0
« —o>
c r c c C
o o i— O
O Q) C 0
0 0 c 0
(continued)

c c
15 £ o 0 E 0 E
0
0 E
ov_ E o E
CL CD
4-> E
o TO o 0 o 0 0
0
CL 0
CL o 0 0

— 0
0 CL
1 ^
"O
00 rtE^

'j
~C k.
a> u<i) *3 c

z
ID
dollars)

D 0 > O O != O
Ej) 8. t .
■o

CD
O
2?
£ 5 «o c
O

< TO £ 0 0 .2
tB O i_ O
CD
50
§ CO" 0^
O CL
LU <
E
current

&

~D tf) to^
0 Q. Q. 00 1
JD — 0
undering

cCO C
0 SZ C CL
0 O 0 0
0 0
0 CL )_
O
Q) ■i 2
jc _£Z sz Q.
oX 0
0 0 ~ 0 0 O
Jr => LU w 8 “cn O0 C
wj 13 0 1c
0
(in

C CL< 0 CL _C
costs

s & ■t 0
C0 0^
CD
C 0
iZ
0 . .
C O
o 0^ O0
money-la

0 0
0 0 0 *0 O) E c
b>22 ®
C 0 0
CO TO E E
"O

0 0
c 0 E
cn
>
E £TO -c
TO 0i- TO 0
Q_TO
Q_
§ E 8 I ® CL _i
TO -O
_0 -O
c
c
C £ o
o

o
II c
o sO) Io'5
o
ltT
o ~o
CO
co
o
IICL 0 cd
"O

2 0
< it
Examples

_c

s g it
of

'■a to c
0)
c
O)
£*>
o =3 to
O) 0 O) 0
to 3T:O ® =3
V_
Table 3.2

£ ° N
0
o to 22 c
T3 0
Oc 2
ETO -g
TO g
JO *1 1 1®

0 •g cr co .T 0 CO
LL CD
52
"t
pg.1 £
38
 A large number of money laundering cases appear to involve oppor¬
tunistic laundering rather than professional services. Where someone apart
from the offender provides the service, he may provide it only to that
offender, perhaps because they are related or connected through some
other activity. Drug dealers appear to be more likely to purchase formal
money-laundering services.
We began this study assuming that money-laundering services were
provided by professional money launderers. Some would be engaged in
other legitimate activities, but the assumption was that money laundering
was a service that they provided to a number of clients, and that they were
willing to provide it to those who could demonstrate financial capability
and who seemed not to be working for the government. Such launderers
exist, but in reported cases they are surprisingly rare.9 A great deal of
money seems to be self-laundered. For example, box 3.4 briefly describes
laundering by Robert Maxwell, a flamboyant press lord in the United
Kingdom. Other people may have aided him, but no one was an indepen¬
dent provider of laundering services. Terrorist financing cases also seem to
involve people who belong very much to the cause rather than being mere
commercial providers.
This is certainly not the first study to raise this question about self¬
laundering or money laundering integrated with the underlying crime. A
decade ago, Australia's National Crime Authority stated: "Most money¬
laundering activity is carried out by the primary offender, not by 'profes¬
sional' launderers, although the use of comp licit individuals is often crucial
to the success of the money laundering schemes" (Gilmore, 1999, 128, citing
National Crime Authority, 1991, vii).
The question of whether there are large numbers of stand-alone money
launderers is important for both policy and research purposes. The ratio¬
nale for the current system is based in part on the claim that its design
allows for apprehending and punishing actors who have provided a criti¬
cal service for those who commit certain kinds of crimes, and who pre¬
viously were beyond the reach of the law. For research purposes, the
assumption of a substantial number of stand-alone launderers makes the
market a useful heuristic device for analyzing the effects of laws and pro¬
grams. As will be discussed in chapter 5, however, that assumption appears
not to be well justified by the facts.
If money laundering is done mostly by predicate offenders or by non-
specialized confederates, then the current regime accomplishes much
less. A central point in a study by Mariano-Florentino Cuellar (2003) is

9. The case of the Beacon Hill Service Corporation (Morgenthau 2004) is a conspicuous excep¬
tion to this generalization. Beacon Hill was an unlicensed money-transferring business that
allegedly provided money-laundering services to a wide range of clients over a period of
almost a decade.

MONEY LAUNDERING: METHODS AND MARKETS 39

that in the enforcement of the AML regime there is no new set of offend¬
ers, just a new set of charges against the same offenders. Consequently,
the new tools of the AML regime, while they might help increase the effi¬
ciency of law enforcement, would likely bring substantially more modest
gains than has been posited.
For research, the market model may be strained. Price may not be well
defined to most participants because the service is rarely purchased. Risk
may also be hard to observe because it is derivative from participation in
other elements of the crime. Assessing how interventions increase risks
and prices for those transactions that do involve stand-alone launderers
will have only modest value.

Classification of Offenses

Offenses can be classified into five categories for purposes of understand¬

ing the effects of specific money-laundering controls: drug distribution,
other "blue-collar" crime, white-collar crime, bribery and corruption, and
terrorism. The categories are more homogeneous with respect to the effects
of interventions and the seriousness and distribution of the harm caused
by particular offenses to society, but they also differ from each other in
these dimensions. It can be conjectured, for example, that the response of
white-collar offenders to increased scrutiny of, say, casinos, is likely to be
different from the response of those who launder money on behalf of drug
dealers. Similarly, the benefits from reducing white-collar crime by 1 per¬
cent might be seen as substantially less than those associated with a simi¬
lar reduction in drug trafficking. The distribution of benefits from reducing
either of the two types of offenses may also be quite different: those who
are harmed by drug trafficking are disproportionately from poor and
minority urban populations, while the costs of white-collar crime are borne
far more broadly across society.
The five-part classification is offered here as a preliminary typology.
Further research may show that some categories can be collapsed or that
others may need to be expanded. For example, research might eventually
demonstrate a need to distinguish between white-collar crimes in which
the proximate victim is a corporation and those in which the victim is a set
of individuals. It may indeed even be that the characteristic of the fruits of
the crime matter; certainly cash is different from other forms of proceeds.
At this stage, the classification is useful for conceptual purposes and for
suggesting approaches to policy modeling.
Table 3.3 provides hypotheses about the differences among the five cat¬
egories of offenses in four dimensions: reliance on cash, quantities of
money involved, the severity of adverse effects, and whom they affect.
The entries concerning the "severity of harm" and the "most affected
populations" are judgments offered here not as authoritative but simply
40 CHASING DIRTY MONEY
Table 3.3 Taxonomy of predicate crimes for money laundering
Severity Most affected
Scale of
Crime Cash of harm
operations
Exclusively Very large population
Urban minority
Drug dealing Severe

Mostly groups
Small to medium Low to modest ?
Other blue-collar

White-collar Mix Mix Low to modest Broad

Large Developing
Bribery and Sometimes Severe
corruption countries
Terrorism Mix Most severe Broad
Small

to identify dimensions that deserve consideration in policymaking and

research.

Drug Distribution

Major drug traffickers face a unique problem, which is how to regularly

and frequently manage large sums of cash, much of it in small bills. For
example, in Operation Polar Cap in the mid-1980s, US agents acting as dis¬
tributors associated with the Medellin cartel, handled some $1.5 million a

week in currency. Few legitimate establishments — or even illegal ones, for

that matter — operate with such large and steady cash flows.
This distinctive characteristic of drug distribution is particularly impor¬
tant because the current anti-money laundering regime initially was con¬
structed primarily to control drug trafficking, an aspect of the regime that
continues to affect public perceptions of the nature of the money-laundering
problem.

Other Blue-Coliar Crime

Other potential large-scale illegal markets that would seem at first glance
likely candidates for generating a demand for money laundering include
gambling and the smuggling of people. However, as seen in chapter 2,
these crimes in fact generate relatively modest demand for money laun¬
dering simply because they have substantially lower revenues than drug
markets. That is not a historical constant but an observation about the past
two decades in industrial societies.
The amounts of moneyfor any individual operation in these other areas
appear to be much smaller than for drug distribution, in part because total
and unit revenues are smaller and in part because what has to be laundered
is net rather than gross revenues. For example, a bookmaker will receive
from customers and agents only what they owe at the end of the account¬
ing period (perhaps one or two weeks).

MONEY LAUNDERING: METHODS AND MARKETS 41

 White-Collar Crime

The white-collar crime category includes a heterogeneous range of activi¬

ties, such as embezzlement, fraud, and tax evasion. A distinctive feature of
these crimes is that the money laundering is often an integral part of the
offense itself, as illustrated in the Washington Teachers Union case (box 3.2).
The Enron case demonstrates a more complex scheme in which shell cor¬
porations in the Cayman Islands served not only as questionable tax shel¬
ters but also as laundering mechanisms to obscure a trail of fraudulent
behavior. Money-laundering services in such cases often are provided by
the offenders themselves, since the offense requires skills similar to those
involved in money laundering. Indeed, where there are false invoices and
other elements of accounting fraud, such activities often constitute both the
predicate as well as the laundering offense.

Bribery and Corruption

While bribery and corruption can be classified as white-collar crime, they

are distinctive in terms of who benefits (public officials and those who ben¬
efit from their decisions), where they occur (primarily though not exclu¬
sively in poor countries), and the nature of their harm (reduced government
credibility and quality of public services), as well as the almost inherently
international character of the laundering — those corrupted would be well
advised to keep the proceeds out of local banks unless the banks themselves
were complicit or the amounts were small. Money laundering also is often
embedded in the offense itself when the corruption is large-scale.

Terrorism

As has been frequently noted, the distinctive feature of terrorism is that it

takes money both legitimately and criminally generated and converts it
into criminal use. The sums of money involved are said to be modest — tens
or hundreds of thousands of dollars rather than millions. Yet the harm is
unique and enormous.
Table 3.3 summarizes the assessments of the relevant differences between
the five types of offenses categorized in this chapter. There will be near con¬
sensus that terrorism poses a greater threat to social welfare than any of the
other offenses. The harm associated with white-collar crime and non-drug,
blue-collar crimes, on the other hand, may be considered by many to cause
modest harms relative to the others. However, these two categories are very
heterogeneous. For example, major environmental crimes (white-collar)
could well strike some observers as just as harmful as selling cocaine.
The assessment of distributional consequences is intended as a reminder
that benefits of interventions are far from uniform, since these offenses

42 CHASING DIRTY MONEY

affect different parts of society. Indeed, there even are significant differences
across nations; kleptocracy — corruption by high-level officials — is probably
more important for sub-Saharan Africa than any of the other offenses.

Conclusions

This chapter has sketched only a few of the many dimensions of the money¬
laundering business. For example, it has addressed neither the manner in
which laundering is distributed among nations, a matter of great political
interest and controversy, nor the characteristics of those involved (such as
their criminal histories and occupations), about which almost nothing is
known. Rather, the focus has been on important characteristics that have
been little studied in evaluating existing money-laundering controls.
Most striking is the variety of money-laundering methods and the var¬
iegated nature of what generates laundering. Much more is known about
drug dealing, and it probably forms its own submarket, with more reliance
on professional money launderers than other submarkets. So while it may
be useful analytically to consider money laundering as a market, it is
clearly a variegated set of markets at best.
Examining the variety of offenses and their adverse consequences sug¬
gests that the estimates of the total volume of money laundered, as set forth
in chapter 2, have limited value, for a number of reasons. A reduction in
the total amount of money laundering that represented a decline in gam¬
bling or corporate fraud but hid a smaller increase in terrorist finance
would hardly be indicative of progress, given the much greater social harm
caused by terrorism. Similarly, the methods that may prove most effective
in reducing money laundering associated with cash smuggling for cocaine
dealers may be much less useful in controlling money laundering by klep-
tocrats. So while there are certainly commonalities in many dimensions of
money laundering across different offenses, it is also important to track
performance of the AML regime for the individual categories of offenses.

MONEY LAUNDERING: METHODS AND MARKETS 43

 4
The Anti-Money Laundering Regime

The global regime to control money laundering involves three dimensions:

national and international building blocks, a firm legal and enforcement
foundation, and close interaction between the public and private sectors in
order to lower compliance costs and raise the probability of achieving its
objectives. In principle, the global regime should have an agreed-upon
international legal foundation with which national regimes are consistent
in terms of laws and standards. The goal should be uniform enforcement
and seamless cooperation across national jurisdictions. Cooperation and
consultation with the private sector are important because they will con¬
tribute to lower costs, create a level playing field, and promote an accepted
global public good from which the marginal benefit to each participant
exceeds the cost, and where the incentive and scope for free riding are
small.

In practice, this idealization of the global anti-money laundering (AML)

regime is unattainable. The establishment of a robust AML regime is a chal¬
lenge because of differences in institutions, perspectives, and priorities
among countries as well as within them. As a result, compromises driven
by the need to balance competing objectives are made at all levels in all
jurisdictions.
This chapter summarizes the AML regime as it has evolved over some

30 years, and particularly since the mid-1980s.1 It has two basic pillars, pre¬
vention and enforcement. The chapter summarizes the AML regime as it
has evolved in the United States and includes a review of the National

1. See the glossary for thumbnail descriptions of many of the terms and institutions men¬
tioned in this chapter.
45
 Money Laundering Strategies of 1999-2003 in order to illustrate the structure
and goals of the US AML regime. That regime is then briefly compared and
contrasted with those in other countries in the context of multilateral efforts
to establish a uniform, global AML regime. The chapter concludes with a
short section on the costs of anti-money laundering efforts, which includes
an estimate that a reasonable upper bound of the gross financial cost is
$25 per capita.
Three principal conclusions come out of the review of the AML regime.
First, the prevention pillar of the US regime has expanded to include a
growing number of institutions and activities, while the enforcement pil¬
lar covers a growing number of crimes. Second, the past 15 to 20 years have
seen the parallel development of a global AML regime alongside the US
one, in part as a response to US pressure and leadership but also in response
to the political and technological influences of globalization. Just as these
forces have contributed to economic progress, they have provided oppor¬
tunities for economic and financial mischief and have enhanced the scope
for cross-border criminal activity.2 Third, the national and global AML
regimes as they exist today are imperfect because their construction has
involved trade-offs between the actual and perceived benefits and costs of
expanding them, between the cooptation of the private sector and privacy
and human rights concerns, between national and international priorities,
and between national and subnational priorities and structures.

Prevention and Enforcement

The prevention pillar of the AML regime is designed to deter criminals from
using private individuals and institutions to launder the proceeds of their
crimes. Enforcement is designed to punish criminals when, despite pre¬
vention efforts, they have facilitated the successful laundering of those
proceeds.
The prevention pillar has four key elements: customer due diligence
(CDD), reporting, regulation and supervision, and sanctions (figure 4.1,
from bottom to top). CDD is intended to limit criminal access to the finan¬
cial system and to other means of placing the proceeds of crime. Reporting
requirements alert authorities to activities that may involve attempts to
launder those proceeds. Regulations implement anti-money laundering
laws and often specify detailed CDD and reporting requirements, while
supervision ensures compliance with laws and regulations by financial
institutions and nonfinancial businesses and activities. Finally, sanctions
punish individuals and institutions that fail to implement the prevention
regime, in particular with respect to CDD and reporting requirements.

2. Gilmore (1999), Levi (2002), and Wechsler (2001) provide additional background material
on the global AML regime.

46 CHASING DIRTY MONEY

Figure 4.1 Pillars of the anti-money laundering regime

Prevention Enforcement

Sanctions Confiscation

Prosecution and
Regulation and
supervision punishment

Reporting Investigation

Customer due Predicate

crimes
diligence

The enforcement pillar also has four key elements: a list of underlying
offenses or predicate crimes, investigation, prosecution and punishment,
and confiscation (figure 4.1). The list of predicate crimes establishes the legal
basis for criminalizing money laundering. Various detection and investiga¬
tive techniques are used to identify specific instances of money laundering
and link each to predicate crimes. If justified by the investigation, the money
launderer is prosecuted. If convicted, the money launderer is not only fined
or sentenced to serve time, but the criminal proceeds that he was attempt¬
ing to launder may also be confiscated or forfeited after having been initially
blocked or seized.

Consider how this framework interacts with a hypothetical bread-and-

butter money-laundering operation. A drug dealer, having collected $25,000
from the sale of illegal drugs, takes the money to a bank and seeks to open
an account in order to deposit the money (placement) before wiring it to a
bank in Colombia (layering), with the ultimate intention of bringing the
funds back to the United States to invest in a legitimate business (integra¬
tion). An effective AMT regime requires the bank to conduct CDD before

opening this account through a process sometimes referred to as "know¬

ing your customer." What is the true name of the customer? Where does
he live? What is his line of business? Can the bank be reasonably confident
that the money that the customer wants to deposit is not derived from
criminal activity?

THE ANTI-MONEY LAUNDERING REGIME 47

 Assume that the customer succeeds in passing these tests and is allowed
to open an account. The bank is still required to submit a report to the
authorities about the large cash deposit. An employee of the bank may also
be suspicious of the fact that the customer is wiring a large amount of
money to a bank in Colombia, and may decide that it is appropriate to sub¬
mit a suspicious activity report (SAR) to the authorities.
If the bank fails to conduct CDD or to submit one or more reports about
the cash deposit and the wiring of the funds to Colombia, contrary to reg¬
ulations promulgated by the Federal Reserve and other banking super¬
visors, this failure may be uncovered during a supervisory examination
or subsequent criminal investigation. As a consequence, the bank may be
fined or otherwise sanctioned or penalized for not complying with the
regulations.
Turning to the enforcement pillar of the AML regime, laundering (or the
attempted laundering) of the proceeds from drug dealing can be pros¬
ecuted under US anti-money laundering law because drug dealing is a
predicate crime or underlying offense for such a prosecution. The report
of the bank about the large cash deposit or the suspicious transfer of the
funds once they were deposited may lead to the detection of the drug
dealing. Alternatively, law enforcement authorities may have had the
depositor under observation as a suspected drug dealer and might use the
bank reports as part of their investigation and prosecution of the under¬
lying crime. Moreover, if it turns out that the bank deliberately assisted
the criminal in laundering proceeds, or that a bank officer facilitated the
laundering unbeknownst to and in violation of the bank's internal con¬
trols, then the bank or the officer could also be prosecuted for money laun¬
dering. Finally, it is possible that the funds never made it to Colombia,
were seized, and subsequently confiscated as a result of a forfeiture pro¬
ceeding.
In practice, the anti-money laundering regime rarely operates as in the
simplified hypothetical example. The proceeds of the drug sales may be
deposited in many separate branches of the bank, into the existing account
of a legitimate business, or in amounts of less than $10,000 in order to avoid
detection. They may take the form of a check rather than cash because they
are the proceeds of a crime other than drug dealing, such as embezzlement.
The depositor may be a lawyer acting on behalf of a shell corporation set
up for a cigarette smuggler.
The examples of predicate crimes and methods of money laundering
presented in chapter 3 suggest that the number of permutations and com¬
binations of crimes and methods is very large. As a consequence, the pre¬
vention and enforcement pillars of the AML regime have been extended
from banks to other types of financial and nonfinancial businesses and to
individuals such as lawyers and accountants (known as "gatekeepers")
who facilitate access to those institutions and businesses. However, the
basic features of the AML regime remain the same. Prevention combines

48 CHASING DIRTY MONEY

customer due diligence and reporting that is required by regulations under
anti-money laundering laws. Supervision is employed and potential sanc¬
tions are available to ensure that the prevention pillar is firmly in place.
Meanwhile, as criminals gather the proceeds of their predicate crimes, the
investigation, prosecution and punishment, and confiscation elements of
the enforcement pillar are employed to combat the underlying crime as
well as to tighten the screws on the money-laundering process.

Current US Anti-Money Laundering Regime

The US anti-money laundering regime is central to the global regime be¬

cause the central role of the US economy and financial system in the
world today frequently results in the United States being the ultimate
destination, or at least the conduit, for proceeds from crimes that may
have been committed outside the country. Thus, the US AML regime is
often, although not always, a model for other national regimes. The first
column of table 4.1 chronologically lists the major developments in the
US AML regime since 1970. Thumbnail descriptions of the major entries
are in the glossary.
In many respects, the US prevention pillar is more elaborate and has
evolved more than the enforcement pillar. Although the list of US predi¬
cate crimes that can give rise to money-laundering investigations and
prosecutions has expanded substantially. In practice, there may be some
tension between the two pillars, as when, for example, financial super¬
visory authorities are uncomfortable with the techniques of criminal inves¬
tigative authorities.

Prevention

Table 4.2 summarizes the prevention pillar of the current US AML regime,
including changes that have resulted since enactment in October 2001 of
the USA PATRIOT Act and its subsequent implementation. The elements
of the prevention pillar are listed across the top of the table, and three broad
categories of economic actors (along with some subcategories) are listed
down the side. The cells in the table indicate whether or to what extent the
elements of the prevention pillar are applied to the various subcategories
of economic activities.

Core Financial Institutions

The most stringent requirements apply to core financial institutions such as

banks, securities firms, insurance companies, and various combinations of
those institutions. All are required to have comprehensive AML compliance

THE ANTI-MONEY LAUNDERING REGIME 49

 0 <

Globalro
T3
o to to
o to 9 a> >
CO
E 22 0to
■to
O)0 o c u to -Q
_c ■£
o c
Q_ O CO <a o
o <
3 O N o
0 0

Eu
O) to CO
C ^
2 c
0
c 0
o CJ
C0
0
0 o CD
1*:CD c
o _ CD O _cd 3

o00 O0E
0 < CO CD
>2.P
III
0 0
CD _c
8 ji CL _
E 00 oE
V+- CO §»! Pccno 0 -C Q.
H-
O 0
0
0 Li-
- O TD
2 S Eo
0
%is 9 0 O
_g 0 c
Q C/Jo 0 u >2 %
SoS C o 0
i_ 0) c )<
*r- rn m o|-§_ o
0 O ^
< E 9 1!

"0
2; m d o .55 E E O 0
o ~o
tD O CD
£ ®" ro
0 Q_ O
o — « to
globally
I Eo COc O Cj
CD
9 §■
CD a O <<oto y c
^nj OoWto EC0c
_ 0
cc < ■00a o
w O
_£
20 CL
A T30 J+z o
E
£ 0
LL Q
o c o
^ CO
r c
CD w -c *1
« O 7 H Q- c <-2B “oLL
o ^CD
< 0 h-
0 wO •i O
<D
3=CO r-

c
and
Europe,

CD
co .E -I C2
0 LU
-o .9> 0 0 0 tD E
c
3 0 -4=OO O_
0
>
« O
H— C

o
O
o
_
States,

0
< C c O
l »s
Q) t0
0
5
0 SoT I
0
to c 0 ^ t~ 0
c •9 to
S
H" Oo) a 0
it: N O

5 O? o CC5 Q.
05
United

0
c 5 o^
St) CD
c
0 E r* c
C if
c oE
0
CO
CO C LUc | I
0 if CD ^

— CD CO -S
_ '0 *=

I52
—-8
< o o 0 ^
>
0 0O00^
r W 0 ^

o
in the

I?
CD-E O CD v
regime

0" O
^

-'o ^ LU
§COQ.
O

£ o 5 CD
■O

<
<
O
"D
AML

0-
United States

Q. O c
3 LL
o o o o "0 0
o < < 0 CD
of the

—1 O
on

0 o <

m £
Evoluti

< to 0 0
o c
_

0
-p-
c

o o
<§ o O o
0
o CD <
-Q
0
0 c
O
CL
v_ 0
0 <
o c 0 0
i_
>—
O
i_
CD o
C
0 <
c
o
0 CD
0
c
—I 0 o
O
QCD
Table 4.1

CD .£?
0o —
0 0
c C .1 i w E
cq0 cr0
o c 00 i—

6 £ < o

II
_C 00
Year

CO 00 CD
00 00
o CD
CD O o
CD CD
o
CD
§Ji
CD

a>
a>
50 a>
1^

T—

r—
 .g _ "c
CD
m g
0 0
0 <D Q
-n
0 03 < 0 cr C C
p 0 0 0
0 0 0 0
0° o o O *“
0 E03 •— 0 0
(D 73
O
O
0
Q c 0 -C 73w C O
73 =0 c O c 5 03
c C
3

0 81 0 73‘Q)o jS
03 O
0 3 o 0
=3 o °2 E O TD O co°.
E £o§ 0 »-
03 O 0
Q) CO
S 0 O *0
b
> —- -000
o = ®a 0
O) 707 COo O<C CO o c 03
O CO (D 0
_l
0
.g c 0 E co X 0870
3
= 0 8 85
0 ■
0° cq E .2 0 0 0 '43 3 e
C C 0
c ^
O 0 C c ^ C
ct O 0 ii
o.2 S£ o
0 -3

£<0 g c »-
.2 0 c
o) u o
c 0
0 J3
0 73
C ~
'8§
Soqq
c oc)'80 o
<- CD 32 '= c
8 ^
O
O 0 cS o 0 0O
Q o ^ o
g CD 5
s»^| ir ?
V*- O E J o° E ^ O 0—0
0
0 cn -4= 0 0 00
0
o°o©
|° E C
0 <° “S 1 c
3
r- 0 ’0S'g X
c
0 03 0 iS1.2
~o os1 O S o 81 g oi?
c a ra O c
•Bo |8 s? O (-
w0 0 .g
o
0 <D -O o
ip 0
O 0
b 0
J5 cn
O E ^ 5
tr ■B03<
_C o0
O
o jo cn .2 O.
■ s-§
=3 C 0 co3 CD '0§_ 03 ^
CL
0 2 2
-S CC -D H g
cn‘5=
0 0 03 0
•J§ E
O '0 0'S 0 0 ' E8qo

'c
; C*- CO ^ i_
c
Q o o ^
E c
2 CO c

p
y 0 <o
y os
0 c <5 t C 3 Q -0 255
E oo
0 _u.occ CD CD Q 03
-§ CO o3 .o3= < <
'—
Q .= O
LU 0
c o LU Ef « g
0 LL £
LL- c C O
§ “S
if
o — 1 E O O <
c

— I —
o ° ®
off => I-
o O m- CD
3* LU O

>
0
8 ®
73 0 C
E o
■§
0
0
<!■ 6
0 0 >

0
0
E CL %
CL
.Q
0 N E 0
o
0
0 0 O 1?
b 0o> o0
s o>
CO
0
0 0
^
O
O 03 0
LU 0 LU
X3 . 160 4= cC
0
0 ^
c
■0 ^ - £ -
■c -O
0 0
0
§.!
0 0 0 0 CL
_3 0 O
0 E
98
1.1
0

X 03
CL
0 C=3
b
o oo
.2
C LU LLI
0 —
c 3 0
O -b
- 0

Q_ CL FT °-i3
03 0 0
O O "3
i_ i— LU _l
3 3 o
LU LU < O
b ^ o
^2
'0

5:1 E
DC O 0
0
o tr
E c 0
< "i
0
_ Cl
o 0 6
c03 <
O Q.
0 _ 73
0
0
73 c c CO CO 0 73
X3
0 o <2 0
O
1— “
~ -d 3
£ 0
c
3
c 1—
0 C
0
0 O
o 0
0
4=- 0
0
c <
0 0 iz O “ i_ 8
_J cC-2 0 X 0
CL 73
C O
> CO
o
Q_
03 0 O
~ >, c
0 0 0
c 0 >
"S<
DC E 0 C
o 73
C
Ct
0 .O2 CD
CD 1—^ 0
">> c 0 0
c 0
03 O
0 0 .2 ^
a»
0)
c
0 0 C
0C 03
73 ^
c _g 0
C
'n

3 3
0 0
o _J
3
0
0
.E _ 1 0 i_
03
1 >, O
n 13
1 - 0
>
c 0
O 0
0
o c 0 O X CO
CL 5 H
0 0 C
c o>, 03 > O CL 3 O
c -3 0 1— (D 6
L_
X X Q. CO
<
E d O -Q ^
O T3 2 0
CO
in
0 ' ^ 0 ®
0
5
CM o' ID

03
03 03 03
03 03
03
03
03 03
03

51
03

T—
T— h-
 **■
"O

o< 0)Q

0 <D %
0 > .c c
*“ & 0 i_O) c c 0 0
0 O 0 C
o 0 _ 0 C o 0
L_
-Q o c
- c c 9? 5 0 ■ CD
0
CO 0
0
o m
~ a
gl
a) < 4 o 3 cC 0 c C
0
c
o

£I
0 0
„o "D
3
c T3 CD
CD -= ®-Boo0 •*=: 0)0-0 ■E< 0
0
0 O 0
0 .1 o
,E E
C <D
ns c -Q ? « </) O
ECD 8
Q § oa Q_
Cl E
-
H? TD O c Wc
O 0
££ o>Q C/)
3 CO — =3 0 -2
oc 0 LL c
o«2Q p
O
£
c « 3= *” O 0 r- >>
cd a> Q_ 0 0 O N
r- c , t5
® “9 O 0
o) t9
0 ^c O
.E O 0
V •i 0 O
cԤoz
•2 f o S
I£
O 0 C > CL c 0 CD 0
o
c -d£ CO 5 >*- c (- .3? co 3= 3OQ- 0 ^ L_ ^ CD ‘C
E'^

is
O 0 aoig?
aj-2 go
0 0
x: Q- 0
o a

■c
o
c
QC > it= c co ox) o
c o
0 0 0
4= ■— O
0
3 9 ffl
o>;3
O O
^ C
_cC O (D — |_ O a) 0 X) -I -E
wO) -5 0OX C

'
T3 .£■60? CO •DC® CO °
2 O -g c o
<0 C — : 0

III*
>g c
0) <■> co cd.E x) 9- o
_ o o
3 CO .g
0 eo cl C "c CO 0 +3 "O -r«0
0
c 0^0
0 .55 §-
C 0 0 0 ccr-Q o5 "O
o n« £ ■£ 0 0 J— c
O —I
o CO ^
-Q '=3 > ^ § 0
4^ §
®E£E .£8
ru.o.2>S a a.® t O 35
c D§ m 5

111*18
Q. --x i— a3 0o 2
0

(L§ §<
o o o iT o o2® 0 oot^o 2^ § 0 C

|s
o LU Ll
X3 — §
<
aotS
o E
§ LL O)«0 CD

"O
TO
n
c
0 o
•- | e is n3 c go
o o ^ LL
CL
O) 0
' 0
T3
c
oc
0 0 ' ’
•§
TO E
L—
<U o
Q.
O
i_ 0
3 0
o ^
ID 0
2 E
<0
0) §■§
Q_ O
*-
co

+■<
</) CO^C
■O

S0 S
C
a)
'c

tr ^
Q)

d>
“O

'0

JC

"l—
O
~o

a> 0 <
E 0CO
3 0
CO
>,
C 0 ^
o
O)
<u 0
c CO
o c
0 CO
3 O)
0
0
Q
CO
0 c c
‘0

3
d) 0 0
0
J= 0 cCl
C
0 _c
o
>+—
3 C co
0
k:
o
>» O
'£

c -Q

o 0
O 0
o
o
c
0
o
o 0
z
0 c
> 0
ID C 0 LL
o o
CO

LL

d)
o
(0
o> o
o
o> CNJ

>-
52
 c
o S' 2
O 0£

"5
0 0 C o TD
c c 0 c
o © sc 0 CD O
.2 o 0 0 0 0
0 td
0 c
is
0 TD TD c
O TD Q_ 0 0 o
— i_
c 0 CL TD CO1 ^
° C<CD q.2
CDTO£^
Cl o
C C CO >,i= CD O £ o ^
0 3 0 TD 0 | 0
CD C
O) 0 C CO E0
c C ^ 3
E "O
0
O> -£
(X
C £
cf) 8 CD 0
■E , 05 0 C 0 c
0
c o
c
0) ^ C c o
g uj COco O .i=
Q 'qT E eg
CD O
SE ^
< <
0
I -Q g, E0 o
0 o E O
O 0 g c E 13
W -g £
Q=5 ■E—
«-§
CO CD >
0>
0
o 0
CC cn a) <o 0
<3 g. o c
®
»_ f= t- c©
l! o C ® 0
o "o ns 1
— 7 co — _c c
c o E
O 3 GC COo ©^
05
12
L.
c o £ P5 £ t 0
■ § O 5 E LL£ 0 c S 5
05 E
o? £ 2
5 m c c _C o
2 00 o l.1
i o c 0
o ,S .9- c 0 Pa LL
1 M CDg 1; CD 05 < 0
Q-
0
C-D00 0 0 < fE~ TD
.E o)
° >^1 ,2 E < 2
O <w
f) _1 E cn^ O)©^

l|gl
«
c LU O ■*-- 0 ■=: > 0 c3 0 LL TD oC ^0 0
Ll o 0 0 03-b
u .9 u > c <D 3$
2 CD
©% C0 _J |_ 0 O -Q 0
% II ., h— »_ Si5•“
5 O O CL 0 C *- 00^

M ©
©

!?"
2
TD
z o CO

E
8
"c

TD
o 0
0
0
> >
c 0 *-
CD 0
o 2:
TD
0 0
c
TD 0 o
CC ^
TD
0 0
CD 3
CD
C OE
c CD
0 0 0
TD C CL o
C
3
^_
C X <
0
-1
0 OC 0 0 — 1CD 0
TD
0 e
TD u TD
0
o C
0 -o
c C C
0 0 3 2:
0 E TD
0
td
CL
o o 0
11
Q. 0
O
i_ o
o
CO
§1 0
LU
0 ^

o —

0
c
o

0
CD U TD
O C < o
0
c c .£ 0

^ pCD cO
<5
C
_Q

< < tfj

isli
E
3

Qgl
CC CD ®
515 0
Cl- 0 c 0
0
o
Z> <
CO -1
0
0
CO
CD
CO 0
CNJ o
o o
o o
o o
CM
CM CNJ

53
Table 4.2 Prevention pillar of the US anti-money laundering regime
Customer Reporting
due diligence requirements Supervision Sanction
Financial institutions
Yes Yes Yes Yes
Core financial institutions®
Other types of financial Yes
Yes Some Limited
institutions15
Nonfinancial businesses
Casinos Yes Yes Some Yes
Dealers in precious metals
and stones Yes Yes No Yes
Real estate agents No No No No
No
Some No No
Other0
Professions
Lawyers and accountants Limited Limited
Very limited Very limited
Trust and company
services providers Limited Limited
Very limited Very limited
a. Depository institutions, securities firms, insurance companies, and combinations of them.

b. For example, mutual funds commodity trading advisers, and investment advisers.
c. For example, travel agencies, and vehicle sellers.

programs and are traditionally subject to federal as well as state regulation

and supervision.3
With respect to customer due diligence, these institutions must comply
with extensive requirements in setting up new accounts and conducting
transactions.4 The assessment requirements are risk-based in the sense that
the amount of information required depends on the institution's size, loca¬
tion, and customer base; the customer's size, location, and type of business;
and the services offered to the customer. If the institution is unable to reach
a satisfactory finding in the course of its due diligence, it is generally ex¬
pected to decline to open the account or complete the transaction. The insti¬
tution is required as well to retain records of its customer due diligence
activities.
Turning to reporting requirements, institutions are required to submit
suspicious activity reports (SARs) to the US Treasury Department's Fi¬
nancial Crimes Enforcement Network (FinCEN), cash transaction reports
(CTRs) to the Internal Revenue Service (IRS), and Reports of International

3. Stand-alone US insurance companies are primarily supervised at the state level but are cov¬
ered by federal AML laws and subject to federal AML regulations.

4. Current US AML regulations describe the CDD process for various institutions as a cus¬
tomer identification program (CIP). The information required for new customers includes
name, address, date of birth, and taxpayer identification number (or passport number for a
foreign customer). The information required for transactions includes the identity of those
participating, addresses, legal capacity, and beneficial owner of the funds involved.

54 CHASING DIRTY MONEY

Transportation of Currency or Other Monetary Instruments (CMIRs) to
the Customs Service.5 Some types of activity may have a threshold below
which it is not necessary to submit reports, such as suspicious transac¬
tions aggregating less than $5, 000. 6 The normal threshold for CTRs and
CMIRs is $10,000 and covers withdrawals as well as deposits.
One criticism of the US AML regime heard both in the United States and
abroad is that these reporting requirements generate so much data as to
cause an information overload, making it difficult for the recipient agen¬
cies to use the information efficiently in law enforcement and related inves¬
tigatory activities.7 Those who report the data contribute to the problem
because their own procedures may be biased toward submitting unneces¬
sary reports. No entity is penalized for excessive filing, and doing so can
even provide implicit or explicit protection from criticism. The US AML
regime and core financial institutions have also been criticized for apply¬
ing more stringent CDD and reporting requirements to foreign than to
domestic customers and transactions.

Core financial institutions such as banks are subject to substantial super¬

vision that normally includes annual on-site examinations to ensure their
compliance with a wide array of laws and regulations. A significant por¬
tion of the examination covers compliance with anti-money laundering
and Bank Secrecy Act regulations, including reviewing other internal or
external audits and testing institutions' procedures and processes.
If an institution is found to have fallen short of what is required or to be
sloppy in implementing AML regulations, rules, and guidance procedures,
it can be subject to informal or formal administrative actions by the regula¬
tor and, potentially, civil and criminal penalties. Experienced federal officials
note that these examinations primarily serve to reinforce the prevention pil¬
lar of the overall AML regime, and rarely turn up direct evidence of actual
money laundering.
The four elements of the prevention pillar of the US AML regime are

applied comprehensively to US core financial institutions. It is not a zero-

5. US banks have been operating under a de facto order to report suspicious transactions
de jure re¬
since being required to do so by supervisors starting in the mid-1980s. The
quirement came into force in 1992 with the passage of the Annunzio-Wylie Money Launder¬
ing Act.

as a result of
6. Agencies may also lower this limit in certain circumstances. For example,
was lowered to
growing suspicions about check-cashing agencies in New York City, the limit
$1,000 for a period of time.

7. In mid-2004, the American Bankers Association (Byrne 2004) acknowledged that prog¬
that the
ress had been made in reducing the amount of data generated, but recommended
for banks to file CTRs for corporations and businesses be raised from $10,000 to
threshold
$25,000.

THE ANTI-MONEY LAUNDERING REGIME 55

tolerance regime, though, because, beyond the application of specified
minimum elements, institutions are permitted and encouraged to employ
risk-based procedures, depending on the nature of the institution and its
business as well as the characteristics of its customers. Along with the use
of proxy devices such as thresholds on reporting, a risk-based approach
has the potential to let some prohibited customers and transactions slip
through undetected.

l\lon-Core Financial Institutions

A broad range of other types of US financial institutions has been progres¬
sively incorporated into the US AML regime both prior to and as a result
of the passage of the USA PATRIOT Act in October 2001. In effect. Con¬
gress delegated to the US Treasury many of the delicate decisions on
where, as well as how, to draw the lines, reserving to itself the capacity to
criticize Treasury decisions at a later date.8 A major subcategory is money
service businesses, which are now required to register with FinCEN if they
offer such services as money orders, traveler's checks, money transmission,
check cashing, or currency dealing or exchange. However, the catch-all cat¬
egory of other financial institutions also includes entities that may be
engaged broadly in money management activities.9 These US financial
institutions are subject to CDD and reporting requirements that are essen¬
tially the same as those applied to core financial institutions.
A principal difference in the prevention pillar of the US AML regime as
it applies to these institutions is that while they are subject to federal reg¬
ulation, they are not subject to as systematic or comprehensive super¬
vision as are the core financial institutions. For example, money service

8. The USA PATRIOT Act mandated the extension of CDD and associated reporting require¬
ments to certain businesses, such as those engaged in fund transfers as well as all securities
dealers and investment companies. It also authorized, but did not require, extension of the
regime to persons engaged in real estate closings and settlements, futures commission mer¬
chants, commodity trading advisers, and trust companies. FinCEN has more discretion on
whether and to what extent to subject this second group of businesses to the AML regime. To
date, final rules have been issued for broker-dealers (securities firms), money service busi¬
nesses, credit card operators, mutual funds, and futures commission merchants (FinCEN
2003a). Extension of the US AML regime also is being considered for life insurance compa¬
nies, dealers in precious metals, stones, and jewels, commodity trading advisers, unregistered
investment companies (hedge funds), investment advisers, travel agencies, sellers of vehicles,
and real estate agents. An advance notice of proposed rule making has been issued for per¬
sons involved in real estate closings and settlements.

9. FATF's revised Forty Recommendations (2003c) for combating money laundering and ter¬
rorist financing finesses this problem by defining a financial institution as any person or entity
that conducts business activities or operations in one or more of a list of 13 categories of activ¬
ities, some with multiple subparts.

56 CHASING DIRTY MONEY

businesses providing any of the five services listed above are required to
register at the federal level, while many businesses providing other types
of financial services are not required to do so. Those other institutions, in
fact, may be required to register at the state or local level, but they may
not be formally licensed. Licensing normally requires some scrutiny of the
background and other qualifications of the owners and managers. Even
when the institutions are licensed, they generally are not subject to regu¬
lar supervision or inspections. In principle, they can be sanctioned either
criminally or civilly for not complying with the requirements of AML reg¬
ulations, but in practice, the sanctions element of the prevention pillar as
it applies to these financial institutions is more limited and may only come
to light if suspected offenders are already under surveillance, or in the
aftermath of an investigation. As a result, in constructing the AML re¬
gime, historical and institutional differences such as divisions of respon¬
sibility between the federal government and the states have to be taken
into consideration.

Nonfinancial Businesses

The prevention pillar of the US AML regime is even less rigorous for non¬
financial businesses such as casinos, dealers in precious metals and stones,
and real estate agents than it is for non-core financial institutions. With
respect to customer due diligence, casinos are subject to "reasonable proce¬
dures" such as identity checks, record keeping, and determining whether
customers are on lists of known or suspected terrorists. Casinos and card
clubs with more than $1 million in gaming revenues are subject to SAR and
CTR reporting requirements, with special thresholds, ha principle, they are
subject to federal regulation and some degree of state regulation and super¬
vision, but the scope for effective and graduated sanctions is even more lim¬
ited than for non-core financial institutions.10
Dealers in precious metals and stones as well as pawnbrokers are sub¬
ject to general CDD and reporting requirements, but again, supervision of
their compliance and any practical use of sanctions for enforcement are
limited because businesses are licensed in state or local jurisdictions. Aside
from withdrawing licenses as the result of criminal or civil proceedings
against the business, the authorities have little leverage to supervise the
CDD or reporting requirements or to sanction noncompliance.
Real estate agents provide a useful illustration of some of the issues
involved in expanding the AML regime. The USA PATRIOT Act provided

10. Further complicating the process of applying the AML regime to nonfinancial businesses
is the fact that they would prefer, if they are to be covered at all, that there be clear distinc¬
tions between what is required and permitted and what is not. Core financial institutions have
similar preferences, but they are more experienced in living with regulatory ambiguities.

THE ANTI-MONEY LAUNDERING REGIME 57

for potential inclusion in the US AML regime of "persons involved in real
estate closings and settlements," which would include most agents. They
were already subject to CDD with respect to name checks in the course of
doing business — the issue was whether they should be subject to general
CDD and reporting requirements under the AML regime. On April 29 and
November 6, 2002, FinCEN temporarily exempted real estate agents from
the regime, and on April 10, 2003, FinCEN called for assistance in deter¬
mining and designing any necessary regulations. The resulting regime as
applied to real estate agents and other participants in closings and settle¬
ments is likely to be less rigorous and comprehensive — and perhaps
appropriately so — than that applied to casinos.
Publicly available responses that FinCEN had received as of June 9, 2003,
to its request for assistance addressed four main points:

■ Participants in real estate transactions are already covered because

they are required to file CTRs, which can be designated SARs by the
filer, and because they can be prosecuted for knowingly assisting or
participating in money laundering committed by a client.
■ There is little evidence of money laundering in this area of economic
activity.

■ The costs to those involved in the real estate industry (many are small
businesses) imposed by requirements to establish full-fledged anti¬
money laundering programs would be high.
■ Financial institutions are involved in most real estate transactions and
should be responsible for AML aspects of those transactions.

The American Bar Association (ABA Task Force on Gatekeeper Reg¬

ulation and the Profession 2003b, 5) made an additional point on the pos¬
sibility of regulation in the area of real estate: "Not unlike the constitutional
protection against compelled self-incrimination, the [legal] ethics rules
[preserving the independence of the bar from government enforcement
agencies] reflect an informed and time-tested decision to protect overarching
principles critical to our system of justice, even if it means that government
enforcement agencies must use their own devices (and not independent
lawyers assisting private citizens) to advance certain investigative objec¬
tives (in this case, in the area of money laundering)."
The ABA's point illustrates an underlying feature of the AML regime,
especially as its coverage expands. The public has an interest to deal with
an identified problem. In response, the government in effect co-opts the
private sector to perform or assist at the financial and even ethical expense
of the private sector.
When FinCEN issues final AML regulations for CDD and reporting cov¬
ering real estate — assuming that even happens — the scope to supervise
compliance with regulations currently on the books will be very limited. In

58 CHASING DIRTY MONEY

principle, FinCEN has such authority. It was one of the first national finan¬
cial intelligence units in the world, and its role has subsequently been sub¬
stantially expanded through such legislation as the USA PATRIOT Act to
become the residual federal-level regulator and supervisor of various enti¬
ties and activities for which there is no existing federal supervisor, such as
trust companies. However, with a budget in fiscal year 2004 of $57 million
and a staff of only 277 and a wide range of responsibilities, the unit lacks
resources. Federal authority in this area is at most residual, because the
basic authority lies with state and local jurisdictions. Sanctions available at
the federal level to enforce uniform compliance are scarce.
Table 4.2 singles out three categories of nonfinancial businesses (casinos,
dealers in precious metals, and real estate brokers) because they are specif¬
ically identified in the FATF's 2003 revision of its Forty Recommendations
(FATF 2003c). Recommendation 20 also states: "Countries should consider
applying the FATF Recommendations to businesses and professions, other
than designated nonfinancial businesses and professions, that pose a money
laundering or terrorist financing risk."
A reasonable question is what evidence is required to substantiate an
assertion that a particular type of economic activity poses a risk of money
laundering or the financing of terrorism. A description of a particular inci¬
dent in a FATF typologies report would be insufficient evidence to con¬
vince many observers.
Other nonfinancial businesses covered to some degree by the US AML
regime include travel agents as well as pawnbrokers, telegraph operators,
and businesses involved in vehicle, boat, auto, and airplane sales. The re¬
gime does not currently cover other businesses sometimes involved in
high-value transactions, such as stamp dealers. The line has to be drawn at
some point, even if it is moved later.

Professions

FATF's 2003 revision of its Forty Recommendations called for extending the
prevention pillar of the global AML regime to lawyers, notaries, other inde¬
pendent legal professionals, accountants, and trust and company service
providers, insofar as they are engaged in specified activities.11 The recom-

11. Recommendation 12d of FATF (2003c) calls for CDD by lawyers, notaries, other legal pro¬
fessionals, and accountants when they assist clients with such activities as buying and selling
real estate, managing assets, managing accounts, organizing contributions to create, operate,
or manage companies, legal persons, or arrangements, and buying and selling business.
Recommendation 12f calls for trust and company service providers to use CDD when they act
as a formation agent for a legal person, a director of a company, a trustee, or a nominee share¬
holder, or when they facilitate the process by providing, for example, a registered office or
address. Lawyers are subject to reporting requirements when engaging with or for their client
in financial transactions in these areas of activity, and countries also are "encouraged" to
extend reporting requirements to the remaining professional activities of accountants and
auditors (recommendation 16a).

THE ANTI-MONEY LAUNDERING REGIME 59

mendations were an outgrowth of the so-called "Gatekeepers Initiative
agreed to at the G-8 Moscow Ministerial Conference in October 1999. The
objective was to expand the scope of the prevention pillar by placing
responsibility on professionals involved in facilitating financial transac¬
tions. However, there is a broad exemption where these professionals are

subject to "professional secrecy or legal professional privilege." Trust and

company service providers are required to report suspicious transactions
for or on behalf of a client in the indicated areas (recommendation 15c).
The United States has no CDD or AML reporting requirements that
apply to these professionals at present beyond CTR requirements and the
penalties that professionals incur if they aid in money laundering. How¬
ever, this type of sanction is part of the enforcement rather than the pre¬
vention pillar of the AML regime. US officials have told us that they can get
the information they need from lawyers if they want it. On the other hand,
in their critique of the levelness of the international playing field on
anti-money laundering, Mark Pieth and Gemma Aiolfi (2003, 27) comment
that "it would rather stretch the general meaning of the words self-regula¬
tion or 'risk-based approach'" to subject attorneys, notaries, and unregu¬
lated fiduciaries to this type of ad hoc regulation.
While supporting domestic and international AML efforts, US legal
groups have resisted the application of AML requirements to lawyers as
well as to trust and company service providers, which are often lawyers.
For example, although lawyers are required to submit a CTR, legal groups
argue that, on confidentiality grounds, they should decline to check the box
indicating that the transaction may be suspicious. An ABA resolution in
February 2002 urged "that the United States government seek to protect
and uphold the attorney /client relationship, including the attorney/ client

privilege, in dealing with international money laundering" (ABA Task

Force on Gatekeeper Regulation and the Profession 2003a, 2). In another
resolution a year later the ABA stated that it "opposes any law or regula¬
tion that, while taking action to combat money laundering or terrorist
financing, would compel lawyers to disclose confidential information to
government officials or otherwise compromise the lawyer-client relation¬
ship or the independence of the bar" (ABA Task Force on Gatekeeper
Regulation and the Profession 2003a, 1).
US lawyers also argue that they have their codes of ethics, are subject to
disbarment by state or federal courts for misconduct, and are licensed by
the states. Further, as one argument goes, federal involvement in lawyers'
activities raises constitutional (states' rights) issues. In submissions to the
FATF when it was considering revisions to its Forty Recommendations,
lawyers argued that they should be subject to the AML regime only when
acting as financial intermediaries. They noted similar views expressed by
groups of legal professionals around the world, and in a Joint Statement to
the FATF on April 3, 2003, they complained that research did not support
the inclusion of any recommendations pertaining to lawyers in the Forty

60 CHASING DIRTY MONEY

Recommendations, and that consultation with them by FATF (on two sep¬
arate occasions) had been inadequate (CCBE 2003).
As if reaching consensus on how to expand the AML regime to the legal
profession were not difficult enough, efforts to regulate accountants and
auditors in the United States and other countries further illustrate the dif¬
ficulties in widening the AML net (box 4.1). Without specific CDD and
SAR reporting requirements, the prospects for effective coverage of law¬
yers, accountants, or trust and company service providers within the pre¬
vention pillar of the US AML regime are thus extremely limited at this
point. Moreover, the United States has no countrywide mechanism read¬
ily available to conduct supervision of these professions, so sanctions at
present are left to the enforcement pillar, although recommendation 24b
of the LATL (2003c) envisages supervision and sanctions that could in¬
volve self-regulatory as well as governmental organizations, as long as
they can ensure compliance. Lor the moment, however, compliance with
ethical standards offers no protection from prosecution for participation
in money laundering. So although these groups of professionals were
included in the new LATL Lorty Recommendations with respect to CDD,
reporting, supervision, and ensuring compliance, it remains to be seen
whether and to what extent jurisdictions comply in the near term. More¬
over, the recommendations themselves contain a large loophole exempt¬
ing reporting on compliance when it conflicts with professional ethics
regarding secrecy.
On the other hand, it is to be noted that in the related area of corporate
governance, the American Bar Association in 2003 reluctantly adopted a
resolution that permitted, but did not compel, disclosure of "confidential
client information if the client is using the lawyer's services to commit a
crime or fraud that would cause financial harm to others." However, this
was a weak response to a crisis of confidence in the legal profession asso¬
ciated with an outbreak of major corporate scandals.
The USA PATRIOT Act (Section 314) calls for increased cooperation
between financial institutions, regulatory authorities, and law enforcement
authorities in operating the prevention pillar of the US AML regime. Such
cooperation would recognize that information sharing improves information
content. Public-private cooperation has been a formal part of the US AML
regime since 1992, when the Annunzio-Wyhe Money Laundering Act autho¬
rized the US Treasury to create a Bank Secrecy Act Advisory Committee with
members of the government and representatives of the banking industry.12
The US banking industry has been critical of the framework that has
evolved under the USA PATRIOT Act, commenting through the American
Bankers Association to the US Treasury that its initial proposals amounted
to a one-way street in which the government placed additional demands

12. The committee was not actually created until 1994.

THE ANTI-MONEY LAUNDERING REGIME 61

 Box 4.1 Accountants, auditors, and anti-money laundering
regimes
The extent to which the global anti-money laundering regime ultimately will affect accoun¬
tants and auditors is not yet clear, even in the United States. The nature of the duties car¬
ried out by these professionals puts them near the top of government lists in terms of those
who can identify and report money-laundering activities. Not only are auditors and accoun¬
tants regularly exposed to companies’ financial records, but they also have expertise in the
design, maintenance, and control of various internal operations.
In congressional testimony before the Committee on Banking and Financial Services,

then Deputy Treasury Secretary Stuart Eizenstat (2000) stated: “We are considering how
existing accounting standards on such subjects as illegal acts of clients . . . can incorpo¬
rate money laundering safeguards." Although the USA PATRIOT Act of 2001 authorized
the Treasury to extend existing anti-money laundering rules or enact new ones for audi¬
tors and accountants, US regulations have not yet incorporated these professions into
the national AML regime.
Under the current US system, accountants and auditors generally play only a passive
role in the detection of money laundering and are not directly involved in identifying illegal
or suspicious transactions at any stage of their relationships with clients. While they must
be aware of the possibility of encountering money laundering or suspicious transactions,
they are not required to set up a program for their detection.
What responsibilities that accountants and auditors do have vary slightly according

to which profession is involved. Accountants in charge of keeping companies’ financial

records are not required by any applicable accounting standards to detect money laun¬
dering.1 Their responsibility is limited to reporting illegal acts that have both direct and
material effect on clients’ financial statements.2 However, most of the time money laun¬
dering has only an indirect effect on financial statements, as in the case of a contingent
liability such as fines in connection with crimes that have been committed. Furthermore,
even if accountants happen to encounter possible money-laundering activities, current
standards do not unequivocally require them to report actual or suspected transactions
to the authorities. Their only obligation is to report to the board of the corporation
involved.

Auditors have different requirements because they must certify the true representa¬
tion of a company's activities through its records. Auditors must directly notify the gov¬
ernment of illicit or suspected transactions in the event that management does not take
satisfactory steps to address the matter. However, they are only required to report
transactions that may have a direct negative effect on financial statements. Once a sus¬
picious money-laundering activity that modifies the true representation of a company’s
activity is identified, it is the auditors’ responsibility to assess the legitimacy of the trans¬
action. If auditors find an illegal activity by a publicly traded company, they are required
to report it to the Securities and Exchange Commission either through the corporate
board or directly if the board is uncooperative. They must also provide an opinion on
the financial statement.3 Therefore, like accountants, auditors do not have to design
their audit procedures to identify and track illegal activities, since they are simply
required to be aware of the possibility that illegal activities may take place.
For the moment, the only stick-and-carrot mechanism available to the government is
the standard of “willful blindness” that the government can use during a prosecution. In
addition, the Supreme Court in 1984 in United States vs. Arthur Young & Co. held that
“no confidential accountant-client privilege exists” (465 US 805, 818, 1984).
(box continues next page)

62 CHASING DIRTY MONEY

 Box 4.1 (continued)
However, while there is no binding requirement on accountants or auditors to practice
customer due diligence or to report suspicious transactions, they are encouraged to follow

precautions and a risk-assessment approach when dealing with clients’ activities. The
issue is whether the government will impose more stringent requirements in the future. One
possibility with regard to reporting suspicious activity would be to follow the example of
other countries such as the United Kingdom, Switzerland, and Belgium, which require
accountants to report all instances of suspected money-laundering activity by clients.
UK legislation covering accountants, auditors, and tax consultants is among the most

comprehensive and follows the pattern applied to that country’s financial industry. The
UK Proceeds of Crime Act passed in 2002, along with associated money-laundering reg¬
ulations that came into effect in March 2004, set a new and higher standard for accoun¬
tants, auditors, and tax consultants. These professions are required to perform CDD and
maintain evidence of client identification and transactions. In addition, they must set up
anti-money laundering procedures, appoint a reporting officer to collect reports on all
suspicious transactions, and submit those reports to the National Criminal Intelligence
Service. The legislation authorizes criminal penalties for failure to abide by regulations
or report suspicious activities.
Anti-money laundering regimes elsewhere in Europe are still in the process of devel¬
oping procedures with respect to accountants and auditors. The European Federation of
Accountants (FEE 1999) has proposed a monitored self-regulation regime controlled by
accountant and auditor associations, consistent with the guidelines and principles in the
money-laundering directive issued by the European Union. This position is also consis¬
tent with that of the 113-nation International Federation of Accountants, which argues:
“Money laundering is far less likely to affect financial statements than are such types of
fraud and misappropriations. Consequently, it is unlikely to be detected in a financial
statement audit. Nevertheless, money-laundering activities may have indirect effects on
an entity’s financial statements and, thus, are of concern to external auditors” (IFAC
2002). Both the IFAC and FEE (2003) have expressed concern that extending the
anti-money laundering regime to the accounting and auditing professions could increase
the fees charged to clients and divert resources away from the principal duties and
responsibilities of these professions.

1. See Private Securities Tort Reform Act of 1995 (Public Law No. 104-67, 109 Statute
737) and the corresponding SAS No. 54, Illegal Acts by Clients (American Institute of
Certified Public Accountants, Professional Standards, vol. 1, AU sec. 317).
2. Securities Exchange Act of 1934 (15 U.S.C. sec. 78j, 1994).
3. SAS 54, Illegal Acts by Clients.

and requirements on financial institutions, rather than the type of network

that they thought was the intent of Section 314.13 Similar criticism can be
found in a report by the Council on Foreign Relations (2004), which focuses
principally on the financing of terrorism. A group of global financial insti¬
tutions has established Regulatory DataCorp as a for-profit enterprise to

13. Byrne (2004) testified that administration of the USA PATRIOT Act's Section 314 "demand
process" had been improved by implementation of procedural changes that reduced burdens
on banks.

THE ANTI-MONEY LAUNDERING REGIME 63

provide information to subscribers to aid them in their customer due dili¬
gence and to manage their legal, regulatory, and reputation risks. In pri¬
vate conversations, representatives of some of the sponsoring institutions
have been critical of the degree of cooperation by the US government with
this enterprise. Such public-private tension is inherent in attempts to estab¬
lish and refine the AML regime.
A summary numerical description of the current US AML regime can be
used to measure the extent of coverage by the prevention pillar of the seven
categories of financial institutions, nonfinancial businesses, and profes¬
sions listed in table 4.2. With full coverage receiving 4.0, a full point for each
element of the pillar, coverage of the categories could be measured as
follows:

■ Coverage of core financial institutions would be scaled at close to 4.0,

unless there was a small deduction for the risk-based nature of the CDD
regime, the presence of thresholds in some reporting requirements, and
the dispersed (nonfederal) supervision of insurance companies.

■ Other types of financial institutions, on average, would be at 2.75, with

credit of a full point each for the CDD and reporting elements, but half
a point for supervision, and a quarter point for the limited scope of
sanctions for noncompliance.

■ Casinos would be at 2.0 overall, consisting of three-quarters of a point

each for CDD and reporting because of exemptions and reporting
thresholds, and a quarter point each for supervision and sanctions.

■ Dealers in precious metals and stones would be at 1.50, because while

they are almost at the same level in CDD and reporting requirements
as casinos, they fall short in supervision and sanctions.

■ Real estate agents would be at 0.50 because they are subject at least to
CTR reporting requirements and the threat of the enforcement pillar.

■ The two categories of professionals — lawyers, notaries and accountants,

and trust and company service providers — might receive a slightly
higher 0.75 because there is a somewhat more developed mechanism
to police behavior and withdraw licenses to practice, at least after an
offense has been detected and proved.

Although this numerical description of the scope of the prevention pil¬

lar suggests it is currently incomplete, such a description would have been
very much lower before passage of the US Money Laundering Control Act
in 1986. Core financial institutions would have registered about 2.0 on
average because of some CDD and reporting requirements (for banks in
particular), although the extent of the supervision and sanction mecha¬
nisms was limited. The Bank Secrecy Act of 1970 required banks and cer-

64 CHASING DIRTY MONEY

tain other financial institutions to retain records to facilitate subsequent
tracing of financial transactions and to submit CTRs and CIMRs. The sub¬
mission of SARs was not mandated until the Annunzio-Wylie Money
Laundering Act in 1992 and not implemented until 1996. 14 Moreover, secu¬
rities and insurance institutions were generally not covered, and neither
were most other types of financial institutions. Broker-dealers unaffiliated
with banks were not subject to SAR reporting requirements until passage
of the USA PATRIOT Act in 2001, and the AML regime essentially did not
apply at all to nonfinancial businesses, let alone professions.
Chapter 8 will return to the significance of how this pattern of regime
expansion has evolved. For the moment, the description helps to illustrate
the political, technical, and institutional compromises and trade-offs that
are required to establish the prevention pillar of the US AML regime. At
the same time, it tells us little about actual compliance.

Enforcement

The enforcement pillar of the US AML regime has expanded over the past
15 to 20 years, although less dramatically than the prevention pillar. The
number of predicate crimes has increased during the period. New tools
have been developed, and new mechanisms offer the promise of greater
efficiency of the enforcement pillar, but their use has not been substantial.
Money laundering was not criminalized in the United States until pas¬
sage of the Money Laundering Control Act (MLCA) in 1986. 15 Of course,
money laundering existed before it became a criminal offense, and law
enforcement authorities have long known to "follow the money" when
investigating crimes that generate proceeds. Correspondingly, criminals
worked hard to break or obscure the connection between their crimes and
the proceeds from them. The Bank Secrecy Act of 1970 fully recognized the
links between money and crime, including securities fraud, as well as the
international dimensions of the phenomenon (Eldridge 1986).

14. The Annunzio-Wylie legislation was the first to require that banks and certain other finan¬
cial institutions have AML programs, effectively mandating internal control procedures that
were subject to outside scrutiny and supervision.

15. The MLCA defines money laundering as conducting or attempting to conduct a financial
transaction "knowing that the property [or monetary instrument] involved in [the] financial
transaction represents the proceeds of some form of unlawful activity with tire intent to pro¬
mote the carrying on of specified unlawful activity or knowing that the transaction is de¬
signed in whole or in part to conceal or disguise the nature, the location, the source, the
ownership, or the control of the proceeds of specified unlawful activity or to avoid a trans¬
action reporting requirement under State or Federal law" (Section 1956, laundering of mon¬
etary instruments).

THE ANTI-MONEY LAUNDERING REGIME 65

Predicate Crimes

The list of predicate crimes or underlying offenses that could lead to a con¬
viction for money laundering was relatively short in 1986, with the primary
focus on drugs and drug-related criminal activity. The list has been ex¬
panded considerably in subsequent AML legislation, and includes more
than 150 offenses covering almost everything that might be considered a
serious crime, from environmental violations to health insurance fraud.
The US list of predicate crimes conspicuously does not include tax eva¬
sion. The ironically titled Bank Secrecy Act (because it was about breaking
down bank secrecy) was prompted in part by tax evasion considerations in
terms of identifying the parties with underlying financial interests, address¬
ing concerns about tax havens and the role of financial services providers
in facilitating access to them, and dealing with international flows of funds.
Moreover, the Anti-Drug Abuse Act of 1988 made it a criminal offense to
evade taxes on the proceeds of an unlawful activity. This provision was
added to the enforcement pillar to allow the IRS to use its expertise to de¬
velop anti-money laundering cases. Thus, US enforcement of anti-money
laundering de facto is closely tied to the enforcement of tax evasion,16 even
though the crime of tax evasion per se does not normally lead to money¬
laundering charges, except to the extent that the criminals have evaded
paying taxes on the proceeds of their crimes.17
Although the absence of tax evasion from the list of predicate offenses is
not a particular enforcement problem for the United States, and it conve¬
niently sidesteps domestic political sensitivities concerning privacy and
how US tax laws are enforced, the omission adversely affects global coop¬
eration. A former high-level Latin American official commented to one of
the authors that if the United States wants other countries to cooperate
more on countering the financing of terrorism and money laundering, it
should cooperate more aggressively in dealing with Latin Americans who
evade taxes on investments in the United States.
The United States has been somewhat responsive to these concerns,
which was one reason for US support of a European initiative to address
harmful tax competition. The Organization for Economic Cooperation and
Development (OECD 1998) released a report on the subject, laying out
principles for the identification of harmful tax competition, which was
directed primarily at tax havens and their low tax rates, solicitation of

16. For example, in July 2003, a stockbroker (Adam Klein) pleaded guilty to money launder¬
ing and admitted that he had also evaded taxes on the proceeds of the underlying crime.

17. Barrons (October 13, 2003, F5) reports some new links between the tax enforcement and
AML regimes. Multiple payments of cash equivalents adding up to more than $10,000 to fund
managers require the filing of a form 8300 that used to go only to the tax authorities but now
is reviewed by AML authorities as well. The investor is also notified of these reports, in con¬
trast with SARs or CTRs.

66 CHASING DIRTY MONEY

investment funds, and reluctance or inability to provide information to the
tax authorities of other countries.18
The OECD report stirred a furor not only among those it targeted —
so-called tax havens, many of which saw one of their principal sources of
foreign exchange earnings under attack — but also within some OECD coun¬
tries whose governments indirectly sponsored the report. Tax evasion and
the international exchange of tax information is politically explosive in
many countries, not just the United States.19 The report led to designating
35 jurisdictions as having harmful tax practices, principally offshore finan¬
cial centers (OECD 2000). Thirty of these jurisdictions subsequently made a
commitment to increase transparency and exchanges of information, so that
by December 2003 only five — Andorra, Liechtenstein, Liberia, Monaco, and
the Marshall Islands — remained "uncooperative tax havens."
The US Treasury under Secretary Paul O'Neill, during the Bush admin¬
istration, initially was quite critical of the OECD tax haven initiative. The
US eventually agreed with its G-7 colleagues on a joint statement of sup¬
port in June 2002: "We agree that the administration and enforcement of
tax laws depend increasingly on transparency and effective international
exchange of information. We call on all countries to permit access to, and
exchange, bank and other information for tax purposes; OECD countries
should lead by example." Elowever, the OECD initiative has been weak¬
ened by less progress than had been expected with respect to intra-OECD,
in particular intra-European, tax cooperation.
Treatment of offenses abroad is another feature of the predicate crime ele¬
ment of the enforcement pillar of the US AML regime. Although many
domestic crimes qualify as predicate offenses under US AML law, only a
subset of crimes committed abroad can lead to money-laundering prosecu¬
tions in the United States. This reduces the scope for the United States to
cooperate in the global AML regime, since nonrecognition under US law of
a money-laundering offense abroad impedes the law enforcement process in
other countries. For example, until passage of the USA PATRIOT Act in 2001,
foreign corruption was not a predicate crime for a money-laundering offense
in the United States. In addition, trafficking in human beings, counterfeiting

18. The principles involved the combination of (1) low or no taxes (or withholding) with a
lack of transparency, (2) ineffective information exchange, and (3) facilitation of tax evasion
by applying different tax regimes to foreign rather than domestic investors, such as by not
requiring a physical presence or substantial activities in the jurisdiction (OECD 1998). The
third criterion was later set aside at the insistence of the United States on grounds that it is dif¬
ficult to articulate clearly.

19. US opposition is based primarily on privacy concerns, along with the view that tax com¬
petition leads to lower taxes. In the context of countering terrorism financing and money laun¬
dering, the two views have led to recommendations to promote the international exchange of
information, but to not allow under any circumstances that such information be used for tax
purposes (Rahn and de Rugy 2003).

THE ANTI-MONEY LAUNDERING REGIME 67

of currency, and forgery can be predicate offenses for money-laundering
charges if committed in the United States, but not if committed abroad.20

Investigation

Turning to the investigation element of the enforcement pillar, reporting

through such mechanisms as SARs is one of the links between the AML
prevention and enforcement pillars. However, consistent with complaints
about the US government's failure to open a two-way street with the pri¬
vate sector, critics often view SARs as a black hole with little if any feed¬
back as to their relative effect (James 2002).
One feature of the investigative element that has received better marks
is the explicit authorization via the 1988 US Anti-Drug Abuse Act to use
sting operations to obtain evidence to convict money launderers, which
has proved to be an important enforcement tool. For example. Operation
Highbind in 1993 and 1994 enabled FBI and Immigration and Naturaliza¬
tion Service agents to pose as drug dealers with dirty money and meet
with members of Chinese- American fraternal organizations that func¬
tioned in part as laundering rings investing criminal proceeds in illegal
gambling enterprises.
Federal agents will often masquerade not as customers but as providers
of money-laundering services. Such was the case in Operation Juno in 1996,
when IRS and Drug Enforcement Administration (DEA) agents ran a stock
brokerage that laundered $14.5 million in drug profits through the black
market peso exchange.21 One of the most prominent money-laundering
sting operations was Operation Dinero, when DEA agents opened an off¬
shore bank to provide services to the Cali drug cartel.22 Over the course of
six months, they laundered $52 million, recovered nine tons of cocaine, and
uncovered ties between the cartel and Italian organized crime.

Prosecution and Punishment

Little is known about how prosecutors make resource allocation and pros¬
ecutorial decisions in money laundering or other cases. Mariano-Florentino

20. Seven of the 20 categories of crimes explicitly cited by the FATF for coverage under
national AML legislation are not considered for this purpose to be crimes in the United States
if the offense is committed outside the country. The other four categories are sexual exploita¬
tion, counterfeiting of products, environmental crime, and insider trading and market
manipulation. The last is a particularly sensitive area because the US approach in this area
differs substantially from that of many other jurisdictions in that it is not grounded on a def¬
inition of the underlying offense.

21. See www.usdoj.gov/dea/major/juno.htm (accessed November 12, 2003).

22. See www.usdoj.gov/dea/major/dinero.htm (accessed November 12, 2003).

68 CHASING DIRTY MONEY

Cuellar (2003) is critical of the failure of the law enforcement community to
use the AML enforcement pillar more actively to disrupt the financial infra¬
structure of criminal activities. Sentencing guidelines, which may or may
not be followed closely, govern subsequent punishments (see chapter 5).
R. E. Bell (2001), writing from a British perspective, lays out a useful two-
by-two taxonomy of prosecutorial approaches involving the nature of the
offense and the prosecution. He asserts that most common are prosecutions
in which the charge of money laundering is linked with the charge for the
underlying offense, of which money laundering was an integral part. Less
common are prosecutions in which the money laundering was outsourced,
but the prosecution of the money launderer is integrated with the prosecu¬
tion of the person who committed the crime. Integrated money laundering
in stand-alone prosecutions, according to Bell, is reserved for big interna¬
tional cases where the crime is committed in another jurisdiction. Least
common are cases involving outsourced laundering in stand-alone prose¬
cutions, because they are the most difficult and resource intensive to
develop. Unfortunately, there are no data publicly available to test these
plausible assertions about relative incidence.

Confiscation

Confiscation of the proceeds of crime through seizure and forfeiting pro¬

cedures is a powerful element of the US enforcement pillar. However,
criminal forfeiture was outlawed by Congress in 1790 and not reintroduced
until passage in 1970 of two major pieces of legislation directed at combat¬
ing organized crime: the Racketeer Influenced and Corrupt Organization
(RICO) and the Continuing Criminal Enterprise (CCE) Acts (Truman 1995).
Not only does confiscation serve as a deterrent to criminal activity and
deprive criminal organizations of resources, but it also helps to fund law
enforcement activities. Such tied funding violates recommended best prac¬
tices in public finance, since the revenue side of government finance should
be separated from the expenditure side. Moreover, confiscation and for¬
feiture can create odd incentives. In a number of states, including Indiana,
Missouri, North Carolina, Washington, West Virginia, and Wisconsin,
forfeited property under state law goes to education, but when the prop¬
erty is seized and confiscated in collaboration with the federal authorities,
the state law enforcement authorities can keep what is returned by the fed¬
eral authorities for their own use, which is 80 percent of the total amount
confiscated.

Such arrangements may distort law enforcement in the direction of purs¬

ing cases that have larger expected value in the forfeiture dimension, which
may not be the same as those with larger expected value on the scale of
other enforcement priorities, such as addressing the most heinous crimes.
As Rep. Bob Barr (R-Georgia) said in 1999, "In many jurisdictions, it [con¬
fiscation] has become a monetary tail wagging the law enforcement dog".

THE ANTI-MONEY LAUNDERING REGIME 69

Senator John Kerry (1997, 176), reflecting on his experience as chairman
and ranking member of the Senate Subcommittee on Terrorism, Narcotics,
and International Operations, warned against the "dark and dangerous
underside" of asset seizure in which defendants buy lighter sentences by
bargaining with hidden property. Nevertheless, the tool is widely used in
the United States at the federal as well as the state and local levels.
As noted, one important aspect of the confiscation element is the sharing
of the proceeds with other jurisdictions. A longstanding program for the
equitable sharing of forfeited property with other federal, state and local
jurisdictions is governed by published guidelines.23 In addition, since US
ratification of the 1988 UN Drug Convention, the United States has had a
program of equitable sharing of forfeited assets with foreign governments
that cooperate and assist in investigations. From 1989 to March 2002, the
international program shared 44 percent ($171.5 million) of eligible for¬
feited assets with 26 foreign governments (US Treasury 2002, 61).
A final noteworthy aspect of the enforcement pillar is related to one of
the reasons behind asset sharing, which is to encourage cooperation and
the sharing of information; this is important not only between govern¬
ments and levels of government but within a given level of government.
A significant step forward in this area was the establishment in 1990 of
FinCEN in the US Treasury Department as a central repository of infor¬
mation, with the additional mandate to analyze the information from such
sources as SARs.

US National Money-Laundering Strategies

The US Congress passed the Money Laundering and Financial Crimes Act

in 1998 in part to improve coordination of the nation's anti-money laun¬

dering efforts. The act mandated the Secretary of the Treasury, in consul¬
tation with the Attorney General, to develop an annual National Money
Laundering Strategy (NMLS) for presentation to Congress over the five
years from 1999 to 2003 in Lebruary of each year.
The intention of the reports was to cover research-based goals, objec¬
tives, and priorities; coordinated prevention measures; detection and pros¬
ecution initiatives; and proposals for partnerships with the private sector
and cooperation between federal, state, and local officials.24 The Treasury
also was required to submit an evaluation of the effectiveness of US AML
policies, an additional task that the US Treasury argued was fulfilled by the

23. The program began in October 1986 with an amendment to the Tariff Act of 1930.

24. The act also mandated identification of "high-intensity financial crimes areas" as part of
a stepped-up enforcement effort on which the NMLS was to report. Most of the progress in
this area reflected efforts that were already under way in 1998.

70 CHASING DIRTY MONEY

NMLS itself (GAO 2003a, 67). The 1999 NMLS was completed and deliv¬
ered on September 23, 1999 .25
In many respects, the legislation requiring the NMLS reports reflected the
same concerns as those behind the National Drug Control Strategy, an
annual report mandated since 1988. Many agencies are involved in dealing
with the same problem and need to find meaningful progress indicators. A
review of the five NMLS reports — using the framework for the AML regime
that involves three broad types of AML goals, the two-pillar regime, ap¬
plication of the market model to money laundering, and evolution of the
regime — illuminates the structure and goals of the US AML regime as well
as some of the trade-offs and constraints it has faced.
Each NMLS included on average 55 action items (from a high of 66 in 1999
to a low of 48 in 2003) grouped under three to six major objectives. The
action items covered a wide range of issues and were much too numerous
to constitute a particularly pointed strategy or even to be completed in a
period as short as a year. It was difficult to identify the highest-priority
items because they were all presented with roughly equal emphasis. The
apparent aim of the NMLS was to present a broad strategy rather than spec¬
ify precise and well-defined objectives. One consequence of this approach
was that many items were implicitly carried over from one year to the next,
even while new items were added identifying new initiatives or reflecting
changing circumstances. The most notable example was a shift in emphasis
to countering terrorist financing following the attacks of September 11, 2001.
The NMLS was thus largely an annual report on major aspects of on¬
going AML initiatives by various agencies of the US government. Little
effort was made explicitly to follow up on issues raised in the previous

year's report, even though the Money Laundering and Financial Crimes
Act required just such follow-up. The result was a pattern of promises but
with limited public disclosure of results. For example, an interagency
report on US policy toward foreign government officials (kleptocrats) who
use the international financial system to convert public assets to personal
use was completed in November 2000, but the results were neither pub¬
lished nor mentioned in the next year's NMLS.26 Another example was a
study completed in 2001 by the Customs Service, summarized briefly in the
2002 NMLS but not released, on percentage commissions charged to laun¬
der money in narcotics cases, based on undercover cases. Despite a stated
intention of posting the entire study — which included such important in¬
formation as the annual costs of compliance with the Bank Secrecy Act —

25. As is frequently the case with mandated reports to Congress, only one (the 2000 NMLS)
arrived roughly on time (March). The 2001 report was released on September 12; the 2002
report in July; the 2003 report in November.
26. In the interests of full disclosure, Edwin Truman was involved with this project. One
innovation in the 2000 NMLS that generally was carried forward was the identification of
responsible officials (by position) or groups of officials for each action item.

THE ANTI-MONEY LAUNDERING REGIME 71

only a summary of a related report prepared by FinCEN and Deloitte &
Touche on the costs of SAR preparation was ever posted (FinCEN 2002).
There were some exceptions in terms of adequate follow-up of previous
proposals. In the 2002 NMLS, an important action item from 2001, to study
the use of the Internet to raise and move funds to terrorist groups, resulted
in a thorough report entitled Terrorist Financing Online, included in the 2003
NMLS (US Treasury 2003, appendix H) with supporting analysis of actual
and potential policy responses. (See box 4.2 for a more general discussion
of money laundering and electronic finance.)
Notwithstanding their shortcomings, the five NMLS reports provide
useful illustrative material on the objectives, structure, and evolution of
the US AML regime. Table 4.3 presents the distribution of action items in
the five strategies across the three broad goals of the US AML regime to
reduce predicate crime, promote the integrity of the financial system, and
address the global, "public bads" of terrorism, corruption, and failed states
(chapter 1). For the five-year period of the strategies, about 65 percent of
the action items concerned predicate crime and about the same percentage
involved integrity of the financial system. The pattern was quite uniform
until the final year, when the share of items focused on crime fell off and
those focused on the financial system rose. The three global "public bads"
generally received much less attention — even in the post-September 11
period covered by the 2002 and 2003 NMLS reports, less than 30 percent of
the action items addressed this AML regime objective. The few action items
classified as addressing failed states, for example, were aimed at countries
that failed to adopt or live up to international AML standards.
Table 4.4 shows the distribution of the same action items across the ele¬
ments of the prevention and enforcement pillars of the US AML regime.
Each of the four elements in each of the two pillars was addressed to some
extent over the five-year period, but in general prevention received more
emphasis than enforcement.
Within the prevention pillar, the reporting element (e.g., extension of
SAR requirements) generally received the most emphasis, but not in all
years. Customer due diligence was the next most frequently addressed ele¬
ment, but in the 2003 NMLS the largest number of action items matched up
with the supervision element, in part because of enhanced efforts to com¬
bat terrorist financing.
Within the enforcement pillar, the investigation element received the most
attention in every year, accounting for about 40 percent of all action items
over the period as a whole. Prosecution and punishment were emphasized
substantially less but were second, followed by confiscation and augmenta¬
tion of the list of predicate crimes.
The element of identification of predicate crimes received relatively more
attention in the 1999 and 2000 NMLS reports, when the Clinton adminis¬
tration sought to expand the enforcement scope of the AML regime by pass¬
ing new legislation, an effort that failed in both years. The 2001 NMLS, the

72 CHASING DIRTY MONEY

Box 4.2 Electronic finance and money laundering

Electronic finance has opened a new frontier for the global anti-money laundering
regime, as well as for the criminals it aims to apprehend. The interaction of technical
change with finance and business can take the form of electronic banking, Internet pay¬
ment systems, or electronic debit cards such as smartcards. The lightly regulated Internet,
which combines considerable anonymity with a global scope and electronic speed, is a
major concern with respect to money-laundering techniques and the financing of terror¬
ism. Such mechanisms open up opportunities for new types of crime such as cyber fraud
that can drain funds from a bank account in Pittsburgh and transfer them to an account
in Dubai.
Although such impersonal interactions suggest new ways to disguise the movement
of funds, the basic challenges they present in terms of prevention and enforcement are
fairly familiar. The opportunities provided by electronic finance to move and launder sub¬
stantial amounts without interacting with the core financial system are severely limited.
For example, drug dealers cannot deposit actual cash in an electronic bank, as the cash
first has to be physically deposited in an institution. Electronic cash has to be uploaded
to a smartcard from an account in some type of financial institution located somewhere
that can be required to keep records of such transactions.
Consequently, the elements of the AML prevention pillar as they are applied to insti¬
tutions in the core financial system can be brought to bear on electronic finance. The sur¬
face anonymity of the interactions may make customer due diligence and reporting
suspicious transactions more difficult, but banks and other institutions in the core finan¬
cial system have the same anti-money laundering obligations relative to their electronic
customers as they do relative to their flesh and blood customers. Moreover, in a number
of major jurisdictions, those obligations can be and have been imposed on the virtual
institutions of electronic finance.
The elements of the enforcement pillar also are available to investigate and prosecute
crimes involving electronic finance. Such investigations may involve different skills, but in
some cases trails may actually be easier to follow through the Internet, for example,
because Internet messages normally leave records behind. Techniques are available to

disguise Internet tracks by passing messages through sites called “anonymizers,” but deci¬
phering disguises and following leads that go cold is not new to 21st century law enforce¬
ment. A more significant challenge, as with other aspects of the global AML regime, is to
improve international cooperation associated with different priorities, procedures, laws,
and regulations.
As noted in the 2003 National Money Laundering Strategy, the Internet poses unique
additional challenges with respect to the specific area of terrorist financing (US T reasury
2003, appendix H). However, most of those challenges are not associated with the actual
movement of funds, aside from the familiar fact that the amounts are small and the ter¬
rorist act occurs after the funds have been used. The principal additional challenges
come from the fact that terrorists use the Internet both to communicate and to raise funds
through such institutions as charities.
In sum, while electronic finance may be a new frontier for the global anti-money laun¬
dering regime, the frontier is located not in another part of the universe but in the same
world where ground rules of economics and finance apply. To be effective, the global
regime must build on what it already has started to do, which is to address the challenges
posed by electronic finance on an ongoing basis.

THE ANTI-MONEY LAUNDERING REGIME 73

Table 4.3 Distribution of US National Money Laundering Strategy
action items across AML goals, 1999-2003
(percent of all action items)
Five-year
Goal 1999 2000 2001 2003
2002 10
67 90 83 78 66
Reducing predicate crime 48
Promoting integrity of 13
financial system 65 63 to63
tal3
7 576 2 28
81
Reducing global “public bads” 48
Memorandum: 24
Total number of action items 66 58 41 50 263

a. Number of action items affecting each anti-money laundering goal as a percentage of total
number of action items over five years.

Note: Some action items addressed more than one anti-money laundering goal.

Source: National Money Laundering Strategies (US Treasury 1 999-2003).

first strategy issued by the Bush administration, which was drafted be¬
fore September 11 and coincidentally issued the next day, included only one
item that matched up with the predicate crimes element, a general commit¬
ment to submit a "money laundering bill which will address [unspecified]
deficiencies" in the then current statutes.
It took the tragedy of September 11 — an example of how the AML regime
has been shaped by events — for the Bush administration to embrace the
Clinton administration's legislative proposals and move them through the
Congress. Most were subsequently incorporated into the USA PATRIOT
Act.27 For example, as proposed by the Clinton administration and at the
urging of other OECD countries, the post-September 11 legislation made
foreign official corruption an offense that can be prosecuted in the United
States; criminalized bulk cash smuggling across the US border; and pro¬
vided the executive branch with additional authority to crack down on for¬
eign jurisdictions, institutions, and classes of transactions thought to pose
a serious money-laundering threat.
We also classified the 263 action items in the five NMLS reports on the
basis of whether they could implicitly or explicitly be fit into an analytic
framework of a market for money-laundering services. An example would
be whether the item was intended to tighten the supply of those services

27. The incorporation into the USA PATRIOT Act of Title III (International Money Launder¬
ing Abatement and Anti-Terrorist Financing Act of 2001) was largely at the insistence of
Democratic Senate leaders, with support from influential Republican senators. The Republican
House leadership initially argued for stand-alone AML legislation, which would have met
substantial resistance in the House. The final legislation incorporated not only proposals to
strengthen and expand the US AML regime, but also important provisions on correspondent
banking and private banking. This substantially expanded the US AML regime in directions
that had not yet been accepted globally.

74 CHASING DIRTY MONEY

Table 4.4 Distribution of US National Money Laundering Strategy
action items across AML prevention and enforcement
elements, 1999-2003 (percent of all action items)
Five-year
1999 2000 2001 2002 2003
Prevention 10
28
46 38 16
Customer due diligence 265 12 12 13
Reporting tal3
to35
Supervision 8 503 5 2 20 8
282
Regulation and sanctions 185 0 2
70
4 3
General6 5 44 22
55 74 76
24 68
Total c
Enforcement 81
Predicate crimes 6 459 2 0 2 4
Investigation 55 30 28 40
Prosecution and punishment 16 12
32 108 22 14
Confiscation 115 3 12 2 6
General6 0 3 15 12 31 11
61 52 48
69 60
Total c 62 64
Memorandum: 263
Total number of action items 66 41 50
58
a. Number of action items affecting each pillar/element as a percentage of total number of
action items over five years.
b. Items related to the specified pillar but not to any particular element or elements.
c. Excluding items that may involve more than one element of the pillar.
Note: Some action items are classified in more than one pillar or more than one element of
the pillar.

Source: National Money Laundering Strategies (US Treasury, 1999-2003).

by limiting certain types of laundering channels, or to reduce the demand

for money-laundering services by increasing the probability of confisca¬
tion. On average, only about a third of the action items in each area could
be interpreted in this manner.
Almost a quarter of the annual action items in the five strategies called
for further study or research with regard to money laundering and the
AML regime. The incidence was considerably higher in the 2000 and 2001
NMLS reports, more than 30 percent, and about average in the 1999 NMLS,
but substantially below average in the 2002 and 2003 reports. Note that the
sharp break in this tendency is not between the Clinton and Bush admin¬
istrations but with the two strategies prepared after September 11.
One carryover from the Clinton to the Bush administration was an effort
to review the federal government financial resources devoted to anti¬
money laundering at the federal level, with a view to eventually ensuring
the appropriate and effective allocation of resources. The 2000 NMLS
included such a review along with a "rough cut" set of estimates, but the
initiative apparently did not reach fruition because in the 2001 NMLS a new
action item was included to review the costs and resources devoted to

anti-money laundering efforts, this time to allow for more informed bud-

THE ANTI-MONEY LAUNDERING REGIME 75

get allocations. A high-level interagency working group was formed to
accomplish this task. The 2002 NMLS, which, as noted, was the only one of
the five strategies systematically to follow up the action items in the previ¬
ous report, stated that analytical disagreements between agencies pre¬
vented fuller development of the material submitted. Nevertheless, the
work was to continue during 2002 and incorporated the element of
resources devoted to stopping the financing of terrorist entities. The effort
apparently has not been successful to date, since the 2003 NMLS did not
mention its continuation.
Such limited follow-up and disclosure not only hindered the overall
success of the NMLS but also undercut the effectiveness of the interagency
cooperation that the NMLS approach was intended to encourage. Such
cooperation would have been strengthened through public commitments
to build on the progress that had been achieved, and with more consistent
identification of those to hold responsible for that progress. For example,
despite numerous statements of good intentions to develop comprehen¬
sive databases on money-laundering cases, as called for in the 2001 NMLS,
the NMLS of the following year (US Treasury 2002, 10) reported that "the
cost involved in taking any one system used by a federal law enforcement
agency as the relevant model outweighed the potential benefit, since dif¬
ferent investigative agencies have different goals, missions and perfor¬
mance measures." The 2003 NMLS, in turn, made no mention at all of
developing uniform databases, leaving one to assume that the idea was
abandoned as attention shifted to countering terrorist financing.
Touching on another important area of cooperation, more than 10 per¬
cent of the action items in the five NMLS reports called for consultation
with the private sector, with no significant difference in the incidence of
such items between the first two and the last three strategies. Action items
in the NMLS for 1999, 2000, and 2003 explicitly although briefly recognized
privacy concerns.
A prominent element of the first three NMLS reports was interagency and
intra-agency studies of the role of "gatekeeper" professions (lawyers,
accountants, auditors) in efforts to combat money laundering, including
ways to better inform them of their responsibilities, and how, or even
whether, to best incorporate coverage of their activities within the AML
regime prevention pillar. By the 2001 NMLS, incorporation of these pro¬
fessional categories was directed toward formulating the US position on
the revision of the FATF Forty Recommendations in this area. The 2001
NMLS (p. ix) quoted President Bush on the relevance of the professions to
the AML regime: "We will aggressively enforce our money laundering
laws with accountability and coordination at the Federal, State and inter¬
national levels. Our goal is to disrupt and dismantle large-scale criminal
enterprises and prosecute professional money launderers, including cor¬
rupt lawyers, bankers, and accountants." Although this statement does not
explicitly recommend extending the prevention pillar to cover lawyers and

76 CHASING DIRTY MONEY

accountants, it certainly is consistent with such a reading. Nevertheless, to
date those professionals have resisted being included in the prevention
pillar of the US regime.
Also prominent in the first two NMLS reports was a focus on taxation,
particularly in terms of promoting effective fiscal enforcement, addressing
harmful tax competition problems that had been flagged in the 1998 OECD
study, examining tax havens, and considering expansion of the list of US
predicate crimes to "selected tax crimes." This topic was not included in
the 2001 and 2003 strategies, but surfaced in the 2002 NMLS as a call for
improving the international exchange of tax information. Raising taxation
issues illustrates how the early NMLS process was responding to inter¬
national concerns about harmful tax practices, including criticism of the US
AML regime for not treating foreign tax evasion as a predicate crime with
respect to US money-laundering prosecutions or legal assistance.
The taxation issue also illustrates the influence of domestic politics on
the AML regime. Although the positions of the Clinton and Bush admin¬
istrations on the basic issue of harmful tax practices were not substantively
different — both favored entrancing information exchange and enforcing
international financial standards — the accompanying rhetoric was quite
different. The Clinton administration was more receptive to concerns of
other countries about low- or zero-tax jurisdictions in the context of tax
competition, while the Bush administration, with its antitax orientation,
was much less receptive, particularly with respect to the treatment of cer¬
tain Caribbean jurisdictions. The privacy issue resonated more within the
Bush administration, notwithstanding the fact that it was mentioned at
least as often in the first two NMLS reports prepared in the Clinton admin¬
istration as in the last three NMLS prepared in the Bush administration.
A review of the National Money Laundering Strategies illustrates sev¬
eral important points. Lirst, it confirms the relevance of the three broad
types of goals for the AML regime: reducing predicate crime, maintaining
the integrity of the financial system, and combating global "public bads."
Second, it supports the view of the AML regime as having prevention and
enforcement pillars with multiple elements. Third, it provides limited but
weak support for the market model of the AML regime. Lourth, more
broadly speaking, it provides evidence that the AML regime is not an
abstraction — its evolution reflects shifting priorities, compromises, and
trade-offs. We return in chapter 8 to the future of the US NMLS and our
recommendations for it.

The Global Anti-Money Laundering Regime

Globalization in the form of an increase in the volume and speed of flows

of international as well as national goods, funds, and information has
played a major part in conditioning the evolution of the global AML re-

THE ANTI-MONEY LAUNDERING REGIME 77

gime. Lawrence Summers, then Deputy Secretary of the US Treasury, told
the Financial Action Task Force plenary on June 26, 1996: "At the very
moment that the world economy is expanding and integrating, creating
vast new opportunities for business, so the technology and capacity at the
disposal of criminals is greater than before."
Four years later, at a meeting of the IMF's International Monetary and
Financial Committee, Summers, by then secretary of the US Treasury,
ratcheted up his rhetoric several notches. "Abuse of the global financial
system is a clear case of a 'global public bad' — indeed, it is the dark side of
international capital mobility," he said. "The international community has
begun to take action against financial system abuse, including the public
release of three lists of uncooperative or problematic jurisdictions [with
respect to money laundering, practitioners of harmful tax practices, and lax
financial supervision], and has called on the international financial institu¬
tions to join this effort. Assisting this effort should be seen as an integral
part of the international financial institutions' mandate to protect the in¬
tegrity of the international financial system. Money laundering activities
have the potential to cause serious macroeconomic distortions, misallocate
capital and resources, increase the risks to a country's financial sector, and
hurt the credibility [or] integrity of the international financial system."
Summers was by no means alone in his assessment. Prior to the 2000 G-7
summit in Japan, the G-7 finance ministers stated in a report entitled
"Actions Against Abuse of the Global Financial System": "Financial crime is
increasingly a key concern in today's open and global financial world, which
is characterized by the high mobility of funds and the rapid development of
new payment tools." The actual G-7 summit statement on the topic read: "To
secure the benefits of the globalised financial system, we need to ensure that
its credibility and integrity are not undermined by money laundering, harm¬
ful tax competition, and poor regulatory standards."28
These statements illustrate two points about the interaction of money
laundering with the globalization phenomenon and the stability of the
financial system. First, globalization, and the deregulation that helps pro¬
mote it, assists money launderers in their criminal activities. Second, the
global spread of criminal activity, and the associated phenomenon of money
laundering that fuels that process and facilitates the amassing of substan¬
tial financial resources, undercut public support for globalization. Vigorous
efforts to deal with these problems are understandably limited by consid¬
erations of privacy and human rights, and by concerns about the arbitrary
use of state power. However, these concerns can also constrain the capac¬
ity of governments to deal expeditiously and effectively with money laun¬
dering and other criminal activities, which in turn further undercuts
support for globalization. The global "public bads" linked to money laun-

28. See Wechsler (2001) for a description of the Summers strategy on these issues.

78 CHASING DIRTY MONEY

dering are reckoned by some to outweigh the global "public goods" asso¬
ciated with globalization.
Differences between the US AML regime and those of other nations and
regions also continue to impede multilateral efforts to achieve a uniform
global regime. While the current global regime has been shaped and prod¬
ded to a considerable extent by US developments and initiatives, different
forces have at times affected the structure and evolution of other national
regimes. For example, the principal concern that prompted establishment
of the Australian AML regime in 1990 was tax evasion rather than drugs. In
the United Kingdom, concerns about drugs and terrorism dominated in the
1980s, even though comprehensive money-laundering regulations includ¬
ing customer due diligence and reporting requirements for the financial sec¬
tor did not go into effect until 1994. (Table 4.1 provides a chronological list
of the major developments in US, European, and global or multilateral
anti-money laundering regimes. The glossary provides thumbnail descrip¬
tions of the major entries.)
Despite differences across regimes, and notwithstanding the enormous
complications posed by globalization itself to combating criminal activity,
the prevention and enforcement pillars of the global anti-money launder¬
ing regime have evolved rapidly over the past 15 years. At the very least,
as William Gilmore (1999, 204) concluded, while "the strategy will not
eradicate international drug trafficking or international organized crime, it
will . . . create an increasingly hostile and inhospitable environment for the
money launderer and others involved in highly lucrative forms of criminal
behavior and afford new elements of protection to economic and political

systems. To achieve this is to achieve something of real and lasting value."

Prevention

Customer due diligence is an area where the US prodded other jurisdic¬

tions to adopt a parallel component in their AML regimes. The 1986 US
MLCA legislation required the chairman of the Federal Reserve Board,
who was then Paul Volcker, to consult with his fellow Group of Ten (G-10)
central bank governors at the Bank for International Settlements (BIS)
about money laundering and bank efforts to control the activity. The leg¬
islation pointed out that money laundering is a global phenomenon and
that if one country's banks and financial institutions are required to imple¬
ment AML measures — increasing their costs and turning away their cus¬
tomers — then the same standards should be applied globally to reduce
regulatory arbitrage and level the playing field.
The initial reaction of Volcker's central bank colleagues was horror —
they maintained that central bankers and bank supervisors should be con¬
cerned more about the safety and soundness of individual banks than
about the stability of the banking system. Under this view, banking super-

THE ANTI-MONEY LAUNDERING REGIME 79

visors should not become involved in law enforcement, which is the re¬
sponsibility of the judiciary and the police. Calmer heads prevailed, but
this episode illustrates the tension between the prevention and enforce¬
ment pillars of AML regimes.
The principal result of the US initiative was that the Basel Committee on
Banking Supervision in 1988 issued a statement of principles regarding the
obligations of banks to know their customers, avoid suspicious transactions,
and cooperate with law enforcement authorities. An associated working
paper acknowledged that differences in roles and responsibilities of national
bank supervisory agencies in the suppression of money laundering at the
time reflected "the role of banking supervision, the primary function of which
is to maintain the overall financial stability and soundness of banks rather
than to ensure that individual transactions conducted by bank customers are
legitimate" (Basel Committee on Banking Supervision 1988, paragraph 3).
However, the report added that "despite the limits in some countries on their
specific responsibility, all members of the [Basel] Committee firmly believe
that supervisors cannot be indifferent to the use made of banks by criminals. "
The involvement of the Basel Committee in the anti-money laundering
campaign marked a significant step not only because it recognized the risks
of regulatory arbitrage and the need for a level global playing field, but
because it involved the setting of international standards. Following the
Basel Committee's lead, the International Organization of Securities
Commissions (IOSCO) issued a resolution on money laundering in 1992.
In 2002, the International Association of Insurance Supervisors (IAIS)
issued Anti-Money Laundering Guidance Notes for Insurance Supervisors
and Insurance Entities.

Switzerland is an example of a national AML regime that evolved quite dif¬

ferently than the US regime. The Swiss trace their concern with money laun¬
dering to adoption of a code of conduct by the Swiss Bankers Association in
1977 in the wake of the Chiasso banking scandals, which began as simple
fraud in the early 1960s and ended up as major financial and embarrassment
for Credit Suisse and much of the rest of the Swiss banking system. The Swiss
approach places heavy emphasis on deep knowledge of customers, which is
well justified in a banking system principally oriented toward private bank¬
ing and investment management rather than retail banking.
In addition, in Switzerland as in Germany and France, suspicious activ¬
ity reports are commonly based more on strong evidence than on hunches,
and they more often lead to criminal investigations than in other countries.
The Swiss (Pieth and Aiolfi 2003) argue that their AML regime is more rig¬
orous than the US AML regime; US observers such as former treasury offi¬
cial William Wechsler (2001) think they have a point.
The Swiss system also relies heavily on the integrity and responsibility
of financial institutions to ensure compliance with national AML laws and
regulations. In contrast, the US and UK AML regimes operate in financial
systems where retail transactions are at least as important as wholesale

80 CHASING DIRTY MONEY

transactions and asset management relationships.29 Pieth and Aiolfi (2003)
have characterized the US and UK AML regimes as emphasizing the col¬
lection and submission of data to national authorities as part of an "early
warning system" that may produce little more than information overload.
Financial Action Task Force

At the Paris Economic Summit of the Group of Seven (G-7) in 1989, France
and the United States proposed an initiative that led to establishment of the
Financial Action Task Force on Money Laundering (FATF) as a temporary
body housed at the OECD but separate from that organization. However,
establishing the FATF involved an agreement that it would not address tax
issues.

The principal initial motivation for the establishment of the FATF was to
combat drug abuse and the financial power of drug traffickers and other
organized crime groups whose activities are facilitated by money launder¬
ing. Public concern about illegal drugs in the United States had reached
extraordinary levels in 1989. The FATF delegations include supervisors,
officials from finance ministries, and representatives of ministries charged
with law enforcement (in the US case, the Justice Department). This inter¬
disciplinary character has contributed to an impressive amount of intra-
governmental cooperation as a positive by-product. The FATF's initial
five-year mandate was to assess the results of cooperative efforts and sug¬
gest additional preventive steps. That mandate was extended in 1994 and
1999 and extended for a record further eight years on May 24, 2004.

In 1990, FATF promulgated its initial "Forty Recommendations" that

provided a general AML framework, starting with ratification and imple¬
mentation of the 1988 Vienna Convention, outlining the roles of national
legal and financial systems and regulators in combating money launder¬
ing, and setting forth certain principles of international cooperation.1'1 The
Forty Recommendations were revised slightly in 1996 and then revised

comprehensively in 2003 based on analysis, in light of FATF's review of

trends in money laundering, of how far the recommendations should be
extended to cover the financial and nonfinancial sectors as well as various

29. So important are retail transactions to the UK financial system that the AML regime there
is sensitive to the charge that the regime itself may impede access to retail financial services.
For example, the UK CDD regulations contain a subsection providing guidance about appli¬
cation of the regulations to limit the risk of financial exclusion.

30. In 1991, the European Community adopted its first directive on money laundering that
sought to establish minimum standards throughout what is now known as the European
Union. Stessens (2000) maintains that the action was motivated in part by other global
attempts to address the money-laundering phenomenon, and also by concerns that money
launderers or criminals would take advantage of the increasingly free flow of capital and
financial services throughout the European Union. The need to establish a level playing field
in Europe also was a concern. Gilmore (1999) stresses the particular challenge that human
rights concerns have posed to establishing an AML regime in Europe.

THE ANTI-MONEY LAUNDERING REGIME 81

"gatekeeper" professions through such methods as due diligence, report¬
ing, regulation and supervision, and international cooperation.
A detailed report prepared in October 2001 by the Basel Committee Work¬
ing Group on Cross-Border Banking and the Offshore Group of Banking
Supervisors on Customer Due Diligence for Banks expanded on the obliga¬
tions of banks to know their customers and upgrade record keeping to
include more extensive due diligence for higher-risk accounts, ongoing mon¬
itoring, and proactive account management. The report (Basel Committee on
Banking Supervision 2001) became the basis for recommendations 5 through
12 of the 2003 FATF Forty Recommendations on customer due diligence for
the full set of financial, nonfinancial, and professional categories listed in
table 4.2.

Neither the US AML regime nor the regimes of many other major coun¬
tries today fully conform to the FATF recommendations in this area. For
example, the United Kingdom does not apply the regime to insurance com¬
panies because UK authorities to date have judged the risk of money laun¬
dering through general insurance business as low (IMF 2003d, 108). The
United Kingdom has applied its CDD and reporting requirements to
lawyers under its Proceeds of Crime Act of 2002, although compliance has
been limited. A 2004 survey by Coleman Parkes Research of companies
selling high-value goods, such as car dealers or estate agents, found that
two-thirds of such UK businesses do not comply with anti-money laun¬
dering regulations with respect to CDD and reporting requirements.
Canada was ahead of other jurisdictions in applying AML regulations to
lawyers, but repealed the regulations after they were successfully chal¬
lenged in a number of provincial courts on the grounds that they eroded
the right of Canadians to independent counsel and to confidentiality with
lawyers. Canada now leaves to the provinces the responsibility for the
supervision of lawyers.
In contrast, the Australian minister for justice and customs announced
in December 2003 his government's intention to fully comply with the
2003 FATF Forty Recommendations in order to ensure that Australia's
anti-money laundering system "continues to model best practice."31 At
the same time, the government released a full set of consultation docu¬
ments based on the new recommendations.

With respect to reporting requirements, some countries' AML regimes

differ from those of the United States. For example, most other jurisdictions
use suspicious transaction reports (STRs), which differ from suspicious
activity reports (SARs) in that there has to be an actual transaction in¬
volved. A SAR can be about a transaction that may merely have been dis-

31. Press release at www.ag.gov.au/ www/justiceministerhome.nsf.

82 CHASING DIRTY MONEY

cussed or attempted but not consummated. This may make the STRs more
useful because they are more focused, but the distinction, as in the earlier
comparison with Switzerland, illustrates differences in AML regimes. In
Australia, any overseas remittance is entered into a database for scrutiny,

a feature of the AML regime that relies on that country's relatively recent
abolition of most formal capital controls. In addition, many non-US juris¬
dictions do not require CTRs or CIMRs or the equivalent, except to the
extent that the underlying activity or transaction might be captured in an
SAR/STR.32
In the area of supervision and sanctioning of noncompliance, recom¬
mendation 29 of the 2003 FATF Forty Recommendations calls for supervi¬
sors to have "adequate powers to monitor and ensure compliance by
financial institutions" and explicitly includes the authority to conduct
inspections to ensure compliance.33 Some jurisdictions do not allow on-site
examinations of financial institutions except under special circumstances,
relying instead on the audit process. This in turn means that such jurisdic¬
tions may be more amenable to including accountants in their role as audi¬
tors as participants in the preventive aspects of the AML regime (box 4.1).
A particular challenge to the global AML regime is the existence of infor¬
mal funds transfer (IFT) systems, such as hawalas, that operate on the
unregulated side of the international financial system. Although hawalas
are found throughout both the developed and developing world, their
widespread use in the Middle East and South Asia prompted increased
attention after September 11, 2001, despite the fact that those terrorists do
not appear to have used an IFT mechanism to fund that operation.
Some argue that hawalas or other forms of IFT do not pose a unique money¬
laundering threat because other mechanisms provide the same opportuni¬
ties (Passas 2000). Others stress that IFT systems often provide low-cost.

32. Recommendation 19 of the 2003 FATF Forty Recommendations calls upon countries only
to "consider" adopting cash-reporting measures. SAR/STR may also be subject to different
minimum reporting requirements in different jurisdictions (for example, 15,000 euros in the
European Union and generally $10,000 in the United States, roughly 50 percent lower) and
differ with respect to type of financial and nonfinancial business required to submit reports.

33. Recommendation 24b applies to nonfinancial businesses and professions, and sets a lower
standard for "effective systems for monitoring and ensuring their compliance with require¬
ments to combat money laundering and terrorist financing," including a potential role for self-
regulatory organizations in this area. Countries with dual (federal-state) or only local-level
approaches to the regulation and supervision of financial and nonfinancial businesses and
professions will find it more difficult to meet this standard. For example, the regulation and
supervision of the insurance industry in the United States is largely at the state level although
it was reported in the New York Times (December 26, 2003) that some large insurance compa¬
nies are pushing for establishment of a federal regulator to level the playing field in US insur¬
ance regulation and to influence federal legislation and regulations that affect insurance firms,
including elements of the USA PATRIOT Act.

THE ANTI-MONEY LAUNDERING REGIME 83

safe, and convenient services that are not otherwise available to the pub¬
lic (Buencamino and Gorbunov 2002). Still others stress that IFT systems
thrive where the formal banking system is weak or nonexistent, or where
there are significant distortions or controls in the payments or foreign
exchange systems. Mohammed El Qorchi, Samuel Maimbo, and John
Wilson (2003) have noted the potential adverse statistical, fiscal, and bal-
ance-of-payments implications of such operations.
What is clear is that bringing IFT mechanisms into the global AML regime
presents complex problems for both recipient and remitting countries
because of the very incentives that have allowed these systems to survive
and thrive for generations. Regulations alone will not suffice — hawalas are
essentially illegal in Pakistan and India, yet they thrive in those countries.
High-cost regulations requiring registration, transparency, record keeping,
and reporting are not likely to be fully effective because the institutions that
can incur the costs will not be able to pass them on to their customers, who
will continue to demand low-cost financial services. Reasonably effective
incorporation of IFT systems into the global AML regime requires balanc¬
ing many considerations within individual countries and the international
financial system as a whole.
The repeated references in this chapter to the FATF evidence its key
role as a standard setter in developing the global AML regime, particu¬
larly the prevention pillar. The FATF has limited membership (33 nations
or jurisdictions)34 and operates by consensus — potential constraints that
it has addressed by maintaining high standards for its members, and by
directly or indirectly sponsoring a number of regional clones, a move that
reflects its recognition of the economic and political implications of glob¬
alization. The Caribbean Financial Action Task Force (CFATF) was estab¬
lished in 1996, the Asia-Pacific Group on Money Laundering (APG) in
1998, the Eastern and Southern Africa Anti-Money Laundering Group
(ESAAMLG) in 1999, and the Financial Action Task Force on Money
Laundering in Latin America (GAFISUD) in 2000. In addition, the Council
of Foreign Ministers of the Commonwealth of Independent States (CIS),
the former Soviet Union, announced in March 2004 that it was consider¬
ing setting up a regional task force on money laundering. Conspicuous in
their absence from the regional list are the Middle East and Central Asia,
although by the fall of 2004 plans were under way to create a Middle
East-North Africa FATF.

34. Current FATF members are Argentina, Australia, Austria, Belgium, Brazil, Canada,
Denmark, the European Commission, Finland, France, Germany, Greece, the Gulf Cooperation
Council, Flong Kong, Iceland, Ireland, Italy, Japan, Luxembourg, Mexico, the Netherlands, New
Zealand, Norway, Portugal, Russia, Singapore, South Africa, Spain, Sweden, Switzerland,
Turkey, the United Kingdom, and the United States.

84 CHASING DIRTY MONEY

 The FATF is without significant enforcement power over jurisdictions
that do not live up to its standards. It sponsors self-evaluations and peer
reviews of its members, thus exercising global-level supervision, but has
limited scope to sanction noncompliance. As of June 2003, even the United
States was in full compliance with only 19 of the relevant 28 former FATF
recommendations based on a self-assessment of its AML regime. Its short¬
falls concerned measures applied to insurance companies and money ex¬
change and transmission operations. Since passage of the USA PATRIOT
Act, efforts have been under way to correct these problems.
Starting with the Seychelles in 1996, the FATF also has periodically issued
statements critical of the actions or inactions by non-FATF jurisdictions.35
The FATF's success in forcing the Seychelles government to impose rules
on its financial businesses inspired a more comprehensive effort to iden¬
tify other countries and territories in the world whose regimes to prevent,
detect, and punish money laundering did not meet internationally recog¬
nized standards (box 4.3). This "name and shame" initiative was launched
officially in 2000 with the publication of 25 criteria, based on the Forty
Recommendations, for identifying noncooperative countries and territo¬
ries falling short in their AML regimes (FATF 2000). While most jurisdic¬
tions on the list were small, it included some major nations such as Egypt,
Indonesia, Nigeria, the Philippines, and Russia.
As of July 2004, six countries remained on the list. One of them, Nauru, a
Pacific island with a population of 12,600, has been subject since November
2001 to countermeasures that include the application of old FATF recom¬
mendation 21 (see footnote 35) as well as enhanced surveillance and report¬
ing of financial and other relevant actions involving that jurisdiction.36 The
countermeasures require financial institutions to report transactions or
attempted transactions that involve the jurisdiction or entities known to be
incorporated in the jurisdiction. More recently, such measures were applied
to Myanmar /Burma.
Liechtenstein, which was on the initial list, is credited with having accom¬
plished one of the most rapid and impressive reconstructions of its AML
regime, but it is left with the historical residue of more than 80,000 shell cor¬
porations, some with known unsavory links (including Saddam Hussein).

35. Regarding the Seychelles, the FATF invoked old recommendation 21 to scrutinize closely
business relations and transactions with persons, companies, and financial institutions from
countries that do not or insufficiently apply the Forty Recommendations. This aim was to force
the Seychelles to repeal its Economic Development Act, which was designed, in the words of
US Treasury official Ronald Noble, then president of the FATF, "to attract capital by permit¬
ting international criminal enterprises to shelter both themselves and their illicitly gained
wealth from pursuit by legal authorities." In the same year, the FATF also applied recom¬
mendation 21 to one of its own members to induce Turkey to pass adequate AML legislation.

36. See chapter 7 for a more detailed account of the Nauru case.

THE ANTI-MONEY LAUNDERING REGIME 85

 Box 4.3 The FATF Non-Cooperative Countries and Territories
Initiative
The Financial Action Task Force (FATF) undertook an initiative in 1999 to identify juris¬
dictions that had inadequate AML regimes and were not cooperating sufficiently with the
global AML effort. The FATF issued its first report (FATF 2000) the next year on the basis
of 25 assessment criteria applied to four areas: (i) financial regulations, including customer
identification, excessive financial secrecy provisions, and lack of a suspicious transactions
reporting system; (ii) other regulatory impediments, including registration requirements for
certain types of businesses and their beneficial owners; (iii) obstacles to international
cooperation, including administrative and judicial constraints; and (iv) inadequate bud¬
getary outlays for anti-money laundering activities and the lack of a financial intelligence
unit (FIU) or similar entity.
During two rounds in 2000 and 2001 , the FATF reviewed 46 countries or territories as

part of what has been called its “name and shame” process. The 15 territories included
dependent territories such as Bermuda and autonomous territories such as Aruba. In
2002, the FATF suspended new reviews while cooperating with the IMF and World Bank
on their reviews of compliance with global AML standards, but FATF continues to moni¬
tor jurisdictions that it previously reviewed. On the first or second round of its reviews, the

FATF “passed” 23 and “failed” 23 jurisdictions. Of the latter group, 17 later “passed” as
the result of subsequent reviews. The AML regimes of six jurisdictions have not yet been
passed by the FATF: Cook Islands, Indonesia, Myanmar/Burma, Nauru, Nigeria, and the
Philippines.
As part of this process, the FATF has threatened to endorse the application of counter¬
measures on jurisdictions that do not make adequate progress to improve their AML
regimes. Under the original FATF recommendation 21, these countermeasures were
envisaged as being applied to jurisdictions with serious deficiencies in their AML regimes.
The measures may include more stringent customer due diligence requirements,
enhanced reporting requirements, limits on establishing financial institutions in FATF
countries, and warnings to nonfinancial-sector businesses with respect to dealings with
entities in those jurisdictions. The threat of countermeasures was applied to Nauru, the
Philippines, and Russia in 2001, Nigeria and Ukraine in 2002, and Myanmar/Burma in
2003. In the cases of the Philippines, Nigeria, and Ukraine, the threat was later lifted
because those countries made some progress in improving their AML regimes. In the
case of Russia, sufficient progress was made that it was admitted to the FATF as a full
member. The countermeasures were applied to Nauru and Burma/Myanmar.
The FATF has also conducted mutual assessment reviews of the AML regimes of its
members, which currently number 33. Two members are regional institutions: the
European Commission and the Gulf Cooperation Council. Except for Russia, none of the

other 31 members has “failed” a FATF assessment. Thus, the 30 other FATF jurisdic¬
tions, including Hong Kong, which is an autonomous territory of China, can be said to
have “passed” their mutual assessment FATF reviews.

If Liechtenstein is to move on to clean up this residue, it will require the

cooperation of other jurisdictions.
The FATF, International Monetary Fund (IMF), and World Bank agreed
in the fall of 2002 on a one-year pilot project to assess compliance with mea¬
sures to combat money laundering and the financing of terrorism. The proj¬
ect in effect transferred most of the peer reviews and evaluations of FATF
members and nonmembers to the other two organizations as a step toward

86 CHASING DIRTY MONEY

a global supervision system.37 Drawing largely on national experts, the IMF
and World Bank organized overall assessments in cooperation with the
FATF and FATF-style regional bodies.
Guidelines also were established for the use of independent experts with
regard to criminal law enforcement matters and nonprudentially regulated
financial activities of such professionals as lawyers, accountants, and real
estate dealers. Their portions of the reports were printed in italics. This was
done out of concern about so-called mission creep in the IFIs and out of
concern that greater focus on this area would distract them from their core
missions.
Participation in an IMF/World Bank review is entirely voluntary, and
the country being reviewed itself decides whether the resulting report is
published, a procedure that has the potential to negate the "name and
shame" mechanism used to press countries with shortcomings.38 The IMF
and World Bank have no powers to sanction, and limited scope to promote
compliance through their lending and technical assistance programs.
Under the 2002-03 pilot project, 41 reviews have been conducted (IMF
2004a, 2004c), 33 of them by the IMF and/or World Bank with the help of
independent experts, and the remainder done by regional FATFs. As of
April 2004, IMF/World Bank reports on 15 of the 19 country reviews that
had been completed had been published in whole or in part.39
Notwithstanding the drawbacks to IMF/World Bank monitoring of
compliance with international standards on combating money laundering
and terrorism financing, the G-7 and the FATF called for continuation of the
assessment program on a permanent basis, and the executive boards of the
IMF and the World Bank agreed in March 2004. They also agreed to fully
integrate the treatment of criminal law enforcement and nonprudentially
regulated financial activities into a single assessment document, dropping
the italics.

Offshore Financial Centers

Following a recommendation of a working group of the Financial Stability
Forum (FSF 2000), the IMF also reluctantly undertook a program in 2000 to
assess 44 jurisdictions identified as offshore financial centers (OFCs). Part of
the IMF's reluctance to become involved in this area related to the fact that

37. The IMF had already accepted FATF's standards to combat money laundering and ter¬
rorism financing as one of its 12 internationally recognized standards and codes.

38. The country reviews are conducted as part of the Reports on Observance of Standards
and Codes (ROSCs), which in turn are part of the IMF's more comprehensive Financial Sector
Assessment Program (FSAP) that sometimes involves an in-depth Financial System Stability
Assessment (FSSA). ROSCs discuss compliance in a general context, while full assessments
also include detailed ratings of compliance.

39. The reviews for Bangladesh, Honduras, Israel, and Tanzania had not been published.

THE ANTI-MONEY LAUNDERING REGIME 87

nearly half of the jurisdictions were not Fund members. Several were de¬
pendencies of members, such as Montserrat, which is an overseas territory
of the United Kingdom. In addition, some Fund members perceived the
assessment task as being only loosely related to the institution's core re¬
sponsibility to ensure macroeconomic and financial stability. Their concern
was that the program could take resources and focus away from the Fund's
principal tasks. After the assessment program was up and running, and
perhaps in response to those sentiments, program director Barry Johnson
stated: "First, let me be clear that the IMF does not chase criminals. We help
jurisdictions set up the necessary legal and financial infrastructure, provide
technical assistance to draft laws and regulations, work with the authori¬
ties to develop the required expertise and set up the financial intelligence
units needed to gather information from the financial services industry,
and ensure that there are trained staff to implement laws and regulations"
(IMF Survey, February 16, 2004, 39).
As of November 2003, the IMF had conducted some type of assessment of
40 of the 44 offshore financial centers. The program is regarded as generally
successful based on a review by the IMF (2003c) and the Financial Stability
Forum (FSF 2004). It is difficult to reach an independent judgment because
by mid-2004 reports had been published for only 16 assessments, with pub¬
lication of another 10 in prospect. The IMF (2003c) review of the program
reached four broad conclusions. First, the supervisory deficiencies in off¬
shore financial centers are similar to those in the most other jurisdictions
where the IMF has conducted assessments. Second, the wealthier the juris¬
diction, the closer it comes to meeting high supervisory standards. Third, the
offshore financial center assessments have led to improvements in many
supervisory and regulatory systems in the jurisdictions evaluated. Fourth,
there is lax supervision and regulation of the nonbanking sector, which is par
for the course as AML regimes are established and implemented.
One concern of the offshore financial centers themselves had been that
they would lose legitimate business and associated revenues as a result of the
assessment process.40 Figure 4.2 suggests that such concerns may have been
misplaced at least with respect to business in 10 offshore financial centers.
The share of total cross-border financial assets accounted for by the five juris¬
dictions normally identified as traditional offshore centers — the Bahamas,
Bahrain, Cayman Islands, Luxembourg, and the Netherlands Antilles — has
been essentially unchanged for a decade, including since 1999, when OFCs
began to receive increased attention. By contrast, the shares of the other juris¬
dictions in the 10 OFCs — Japan, the United Kingdom, the United States,
Hong Kong, and Singapore — have been in a rather steady decline.
Of course, these data are only suggestive. Illegitimate business is proba¬
bly only a small portion of total business in these centers. Offshore finance

40. See Suss, Williams, and Mendis (2002) for a discussion of the economic motivation for
becoming an offshore financial center and an early assessment of the impact of FSF and FATF
initiatives on such jurisdictions in the Caribbean.

88 CHASING DIRTY MONEY

Figure 4.2 Share of cross-border assets of offshore financial
centers, 1992-2003 (percent)

percent

Note: Offshore financial centers’ cross-border assets are equal to external positions of banks
in individual reporting countries minus local positions in foreign currency of banks in those
countries.

a. Data are for the third quarter of 2003.

Source: BIS (2004).

could have attracted more legitimate business to some of these jurisdictions,

or the program and associated improvements related to global standards
may have had no effect on financial flows. Moreover, smaller offshore finan¬
cial centers may have lost more business. Esther Suss, Oral Williams, and
Chandima Mendis (2002) report that the number of licensed banks in
Antigua and Barbuda declined from 58 in 1997 to 21 at the end of 2001 as a
result of the attention focused on their offshore financial center activities

by and Territories
the FSF, and concerns inraised
because ofInitiative by the FATF's Non-Cooperative
Countries 2000.

Enforcement

The enforcement pillar of the global AMT regime relies heavily on a num¬
ber of regional and United Nations conventions (table 4.1). Particularly
important are the 1988 UN Convention Against Illicit Traffic in Narcotic
Drugs and Psychotropic Substances (Vienna), the 1999 OECD Convention
on Combating Bribery of Foreign Officials in International Business Trans¬
actions the 2000 UN Convention Against Transnational Organized Crime
(Palermo), and the 2003 UN Convention Against Corruption.

THE ANTI-MONEY LAUNDERING REGIME 89

 While these conventions and agreements have succeeded on paper in
establishing a common AML framework, in practice there are important
differences.
An initial European Commission directive on money laundering issued
in 1991 was principally concerned with laundering of drug proceeds and
applied only to the financial sector. A second directive in 2001 widened the
focus to all "serious offenses" and required members to bring within the
coverage of the AML regime auditors, accountants, tax advisers, notaries,
other legal professionals, real estate agents, casinos, and dealers in high¬
valued goods. In June 2004, the European Commission proposed a third
AML directive incorporating terrorism into the AML framework and con¬
forming the EU AML regime to the 2003 FATF Forty Recommendations.
There are two basic approaches to criminalizing money-laundering
predicate offenses. Under the US approach (used in Japan as well), specific
crimes are listed, which has the advantage of pinpointing those types of
activities of substantial concern.
The alternative approach, found in most other jurisdictions, criminalizes
money laundering in connection with all offenses involving proceeds from
the crimes or in connection with all "serious offenses," conventionally
defined in terms of a punishment threshold such as a maximum or a min¬
imum imprisonment. This approach has the advantage of capturing what
is regarded as serious by the local authorities,41 but it may exclude cate¬
gories important to other authorities, thereby undermining the quid pro quo
mutual reinforcement of national AML regimes. In general, law enforcement
has every reason to support adding concerns ranging from narcotics to ter¬
rorism to the AML agenda.
Legislative implementation of enforcement measures around the world
has been spotty. For example, according to the IMF/World Bank 2004
assessment report on the Czech Republic's AML regime, that country's
definition of money laundering as an offense generally meets the inter¬
national standard, but the implementing criminal code fails to cover the
acquisition, possession, or use of assets with the knowledge that such
assets originate from a criminal activity, and sets a higher burden of proof
than is consistent with the international standard. Similarly, as of 1993,
Greece was a party to various international agreements but had not yet
enacted any implementing legislation, and France, Luxembourg, Portugal,
and Spain had criminalized money laundering only in connection with
drug offenses.
The 2003 FATF Forty Recommendations sought to resolve this dilemma
by linking the two approaches. It establishes a definition of "serious

41. Even using a definition of a "serious crime" that specifies either a maximum or minimum
does not guarantee uniformity of treatment. Some serious crimes are not "caught" by either
test, and in other cases the activity is not a crime.

90 CHASING DIRTY MONEY

offenses"42 and also specifically designates 20 broad categories of predicate
offenses for money laundering that include corruption, bribery, market
manipulation, and environmental crimes.
However, the designated predicate money-laundering crimes in the
2003 FATF Forty Recommendations do not include tax evasion, consistent
with the FATF's mandate to avoid tax evasion. As discussed earlier, this
omission not only reveals a lack of uniformity in the global AML regime
but can also hinder cooperation. Some countries such as France include
tax evasion as a money-laundering offense in their national legislation, but
undercut the inclusion because the offense is not reportable by financial
institutions.
An Oxfam global report on tax havens in 2000 claimed that developing
countries as a group may be forgoing $50 billion in annual tax revenues as
a result of tax competition and the use of tax havens. The report does not
provide background on how the estimate was derived, so it is impossible
to evaluate the figure. However, the figure itself conveys the criticism that
the playing field in international finance as regards tax issues is uneven. In
August 2003, Secretary-General Kofi Annan issued a report to the UN
General Assembly calling for increased international cooperation on tax
issues involving money laundering, transnational crime, international ter¬
rorism, tax evasion, and tax incentives for investment by competing coun¬
tries.

With respect to the investigation element of enforcement, some juris¬

dictions such as the United Kingdom do not condone sting operations.
Recommendation 27 of FATF (2003c) endorses investigative techniques
such as "controlled delivery, undercover operations and other relevant
techniques," but not explicitly sting operations. Jurisdictions also differ in
their standards and procedures to enforce due process, human rights, and
privacy. The US Casablanca undercover operation in 1998 was regarded
as a resounding success in the United States, but the operation and its
aftermath created tensions with Mexico. In February 2004, four bankers
convicted in Mexican courts in connection with the operation were re¬
leased when the US sting operation was declared unconstitutional under
Mexican law.
Approaches to the structure of the AML regime as well as prosecutions for
money-lamadering offenses differ across jurisdictions depending in part on
fundamental legal principles, such as whether they are based on common
law or civil law. For example, under civil law principles it is more difficult to
establish criminal liability for legal persons. Common law principles are
generally credited with supporting better economic performance (Caprio
and Honohan 2001), but the better protection they afford property rights

42. The FATF definition requires a maximum penalty of more than one year of imprisonment
or a minimum of more than six months.

THE ANTI-MONEY LAUNDERING REGIME 91

tends to get in the way of a tight AML regime in which everything is regu¬
lated. These differences and impediments generally are not regarded as
insurmountable, but they complicate the establishment of a uniform global
AML regime.
Finally, in the area of confiscation, national practices again differ. For
example, the 2003 FATF Forty Recommendation 3, citing the Vienna and
Palermo UN Conventions, carefully calls upon countries to adopt confis¬
cation measures, outlines the desirable content (identification and freezing
of property, preserving the state's ability to recover the property, and
investigation), and suggests that countries consider measures to allow con¬
fiscation without a criminal conviction, or through procedures that shift the
burden of proof to the defender to demonstrate the lawful origin of the
property. Cross-country practices are far from uniform in this area, which
tends to impede international cooperation by, inter alia, limiting the scope
to share assets that might be seized.
On the other hand, most countries have financial intelligence units (FIUs)
to combat money laundering, and the 2003 FATF Forty Recommendations
include nine (26 through 34) that primarily relate to enforcement issues,
including the establishment of FIUs, along with six (35 through 40) that
relate to international cooperation on anti-money laundering. The powers
of such national FIUs differ, however. For example, the Czech FIU cannot
share information with its foreign counterparts in the absence of treaties or
conventions. In addition, where an FIU is situated within governmental
structures may affect its ability to coordinate information flows and inter¬
act with those who are conducting criminal investigations.
The Egmont Group of Financial Intelligence Units established in 1995
now includes more than 80 entities that coordinate their work on money¬
laundering issues. The aim is to support national programs through the
exchange of information (under closely specified conditions), enhance ex¬
pertise, and foster better communication. Although such organizations as
the Egmont Group are not centrally involved in international enforcement —
which continues to be handled through traditional police and justice chan¬
nels governed largely by bilateral and multilateral treaties, conventions,
and arrangements (such as mutual legal assistance treaties in criminal mat¬
ters in the case of the United States) — they represent a significant recogni¬
tion that enforcement is an important dimension of the global AML regime.
In Europe, the mandate of the Europol Drugs Unit (EDU) established
early in 1995 to coordinate actions on drug crimes was eventually ex¬
panded to cover criminal activities associated with money laundering.
Later in 1995, Europol itself was created and it absorbed the EDU in 1999.
The European Union has been faced with more intense coordination and
enforcement issues as it has developed and deepened the single market,
with a greater need for common rules and area wide enforcement actions.
At the global level, the international police coordination organization
Interpol established a Financial and High Tech (FHT) Crimes Sub-Directorate

92 CHASING DIRTY MONEY

under its Specialized Crimes Directorate in September 2001. The FHT in¬
vestigates funds derived from criminal activities, as well as currency coun¬
terfeiting and intellectual property rights offenses. It focuses on providing
expertise to law enforcement agencies and increasing interjurisdictional
communication and cooperation (Interpol 2003).
National and global AML regimes initially exist only in agreements and
laws. They are effective only if there is compliance and implementation.
For example, in early February 2004, the European Commission notified
France, Greece, Italy, Luxembourg, Portugal, and Sweden that they had
failed to implement changes in their national laws mandated by the 2001
EC Money Laundering Directive.43 On the other side of the world, the story
is told that the government of Vietnam is unconvinced that it needs to
apply CDD to its banks if it owns them.
The World Bank (2003a) makes the case for how money laundering ad¬
versely affects developing countries because they are often small and more
susceptible to disruption. But it acknowledges that the magnitude of these
consequences is difficult to establish because the impact cannot be precisely
quantified either in general or for a specific country. In developing countries
with limited governmental resources, and where anti-money laundering
may, correctly or incorrectly, be regarded as a luxury good, one option for
strengthening the global AML regime is to provide technical and financial
assistance bilaterally or through international financial institutions.

Costs of the US Anti-Money Laundering Regime

In principle, the gross financial costs of the AML regime should be weighed
against the benefits of efforts to prevent and combat money laundering, to
which we turn in chapters 5 through 7. Systematic quantitative informa¬
tion on the costs of the AML regime is scarce, so this section draws on
scraps of US and British information in order to make a ballpark estimate
of the gross financial costs of the US AML regime. That estimate for 2003 is
on the order of $7 billion, or about $25 per capita.
An anti-money laundering regime imposes both direct and indirect as
well as financial and nonfinancial costs that take three forms: (1) costs

incurred by the government or public sector in establishing and adminis¬

tering the regime, (2) costs incurred by the private sector in carrying out
the requirements of government, and (3) costs borne by the general public.
Were it not for these costs, the standard of zero tolerance for money laun¬
dering could be applied to measuring the success of the AML regime in
achieving all of its goals. In other words, any dollar of proceeds of crime or

43. The European Union also could have made a major contribution to the global AML
regime by limiting the largest new euro note to €100, comparable to the largest US note of
$100; instead the largest note is €500.

THE ANTI-MONEY LAUNDERING REGIME 93

some other form of those proceeds that was laundered would be proof that
the system was flawed.
In addition to the gross financial costs of prevention and enforcement,
the potential for information overload, in practice, should also be consid¬
ered. The government may collect so much information because it per¬
ceives the costs of doing so as being negligible, but this approach runs the
risk that the massive amount of irrelevant information might drown out
information that is of value.44 The result may be a point on a "Laffer curve"
for AML information beyond which additional filings reduce the inves¬
tigative capability of the government. Pieth and Aiolfi (2003) as well as sev¬
eral reports by the government itself have voiced concern about the
collection of excessive information by the US AML regime. The General
Accounting Office (GAO 1996) argued specifically that too many CTRs
were being filed, imposing unnecessary costs on both the private and pub¬
lic sectors, and undercutting the usefulness of the reporting system. Seven
years later, FinCEN (2003a) set a goal to reduce the number of CTRs (now
numbering over 12 million annually) by 30 percent. Although the stated
aim was to reduce the submission of CTRs with little or no value to law
enforcement, and which thus imposed costs for industry, an implicit moti¬
vation was to lower costs and increase the effectiveness of CTRs for the US
government.
Evidence on the issue of information overload in the United Kingdom
comes from KPMG (2003, 16): "The nature of the process at ECB [Economic
Crime Branch of the National Crime Intelligence Service], exacerbated by
the significant number of poor quality SARs provided by disclosing enti¬
ties, has led to significant delays in disseminating SARs which have not
been fast-tracked." Michael Gold and Michael Levi (1994) reached a simi¬
lar conclusion.
Problems of incidence and interpretation also are involved. Consider each
of the three areas in which the AML regime imposes costs: the government,
private-sector institutions, and the general public. The government incurs
direct costs through prevention and enforcement, such as operating the SAR
and similar databases and investigating specific transactions.
The government uses laws and regulations to shift some of its costs onto
financial and nonfinancial institutions and, potentially, independent pro¬
fessions.45 Banks incur direct costs when they are required to set up inter-

44. Generalizations in this area are difficult in the absence of carefully assembled data. For
example, it is possible that the costs of compliance are unknown. The government may bear only
a small share of those unknown costs. It may not be in a position to judge costs and benefits for
either the private or public sector. It may be unaware of the issue, though that would be unlikely
in the often contentious environment of US financial regulations. Finally, the government may
not recognize that one of the costs of the AML regime is the less efficient use of any information
because of the costs of searching for it.

45. Reinicke (1998) refers to this process of increasingly assigning and delegating informa¬
tion and analysis responsibilities to the private sector as "horizontal subsidiarity."

94 CHASING DIRTY MONEY

nal control systems to detect money laundering or systems to generate the
data sought by the government. Costs borne by financial and other insti¬
tutions and businesses may be shifted to customers. Of course, the indi¬
vidual institutions consider the costs net of any direct or indirect financial
benefits. Representatives of a number of the institutions interviewed for
this study emphasized that they integrate the requirements of the AML
regime into their overall risk management systems, and that they see rep¬
utation or "image" advantages and other gains from running "squeaky
clean" operations.
Also relevant are distinctions between the institution's long- and short-
run costs as well as the general level of competition it is facing. Never¬
theless, to the extent that the net effective financial costs are positive, the
price of the services provided by individual financial institutions and busi¬
nesses go up; if the institutions use a cost-plus-markup approach to pric¬
ing decisions, the general public pays some of the costs. For example, banks
raise the price of services to the customer, either directly via monthly fees
or indirectly via higher minimum balance requirements to recoup some or
all of their net AML costs. A potential indirect effect of this cost shifting is
that these price increases may exclude some customers from access to stan¬
dard banking services such as electronic deposits or check cashing.
Customers also bear opportunity and transaction costs, since they must
provide additional documentation to satisfy the regulations and may expe¬
rience as a consequence delays and other types of transaction costs. A less
tangible b)t often no less important cost to some customers is the increased
intrusion, or loss of "privacy." Intrusiveness also can contribute to a will-
not-pay or will-not-play posture on the part of customers who either defy
the rules or withdraw from participation in the financial system. Either
way, the overall effectiveness of the AML regime may be weakened be¬
cause of a loss of information. Similar compliance problems can arise if
AML requirements are perceived to be applied unevenly among customers
(e.g., more harshly to individuals than to corporations).

Costs to the Government

No systematic estimates are available for the costs of the various elements
that make up the AML regime in the United States or any other country.46
Moreover, developing such estimates poses complex conceptual chal¬
lenges. For example, the government's costs to operate SAR and CTR data¬
bases can be estimated, but those costs are a small part of total government
AML costs and do not include the enforcement costs associated with inves-

46. The best analyses are by the UK authorities, and we rely on them quite heavily. Although
the 2000 National Money Laundering Strategy initiated a tally of resources that the US govern¬
ment devotes to combating money laundering, the findings were never made public.

THE ANTI-MONEY LAUNDERING REGIME 95

Figure 4.3 Trends in US federal prevention and enforcement real
outlays, fiscal 1985-2004 (in millions of 1995 dollars)
enforcement
prevention3
40,000

35,000

30,000

25,000

20,000

15,000

10,000

5,000
0

e = budget estimate
r = budget request
a. Includes outlays for US Treasury and non-Treasury agencies involved in the prevention
process. US T reasury agencies are the Office of Comptroller of Currency and the Office of Thrift
Supervision. Non-Treasury agencies are the Commodities Futures Trading Commission, Federal
Reserve System, National Credit Union Administration, Federal Deposit Insurance Corporation,
and Securities and Exchange Commission.
b. Includes outlays in Function 750 (Administration of Justice: federal law enforcement, federal
litigative, judicial, correctional activities and criminal justice assistance) of the US federal budget.
FinCEN is also included in Function 750.

Source: US Federal Budget, Annual Report: Budget Review (fiscal 1985 to 2004). Board of
Governors of the Federal Reserve System, fiscal 1985 to 2004.

tigative and regulatory activities employing these databases. Those latter

costs are difficult to estimate, in particular the marginal costs, because few
investigations focus solely on money laundering. Rather, the AML regime
generates data that government agencies use along with other investiga¬
tive materials in pursuing criminal investigations and developing prose¬
cutions.
Despite such constraints, it is possible to put together ballpark estimates
of total US annual (fiscal year) federal outlays for all types of prevention
(principally by financial regulators) and enforcement activities, using as a
starting point the "750" account of the federal budget covering the admin¬
istration of justice. Figure 4.3 displays the trends in real (1995) dollars from
1985 (prior to enactment of the 1986 Money Laundering Control Act) to

96 CHASING DIRTY MONEY

2004. The data include expenditures associated with the AML regime as
well as expenditures associated with other regulatory or law enforcement
activities. In real terms, the average annual increase over the period was
6.1 percent for total prevention expenditures, 7.4 percent for enforcement
expenditures (almost an order of magnitude larger than prevention), and
7.3 percent for the combined total.
Not a great deal can be said about expenditures on the actual AML
regime, although one can ask whether the passage of the four major pieces
of AML legislation in the years indicated in figure 4.3 by the vertical lines
had a marked impact on our proxies for aggregate prevention and enforce¬
ment expenditures.
Over the period measured, the average three-year increase in prevention
expenditures was about 13 percent and in enforcement expenditures about
26 percent. These averages can be compared with the three years sur¬
rounding the passage of the four pieces of US AML legislation from one
year prior to passage, as the base, until two years after passage. For pre¬
vention, the comparison reveals that the only above-average increase was
surrounding the passage of the MLCA in 1986, and the excess was less
than 10 percent; for the other three dates the increase was the same as or
less than the average. For enforcement, a similar above-average increase
occurred from the base in 1985 to 1988, but the increase from 2000 to 2003
(surrounding the passage of the USA PATRIOT Act) was substantially
larger, about 20 percent more than average. Thus, there may have been
marked increases in expenditures by the federal government associated
with these four major adjustments to the US AML regime, but they appear
to have been largely washed out in the aggregate data by other factors
affecting regulatory and supervisory expenditures.
A rough rule of thumb is that perhaps 2 to 3 percent of total US govern¬
ment regulatory and enforcement expenditures at present might be attrib¬
utable to the AML regime.47 On this basis, the federal government in fiscal
year 2003 may have spent something like $1.5 billion on the AML regime,
which would be 0.014 percent of nominal GDP in 2003.48 For the public sec-

47. The costs of AML regime prevention and enforcement are combined with data on other
aspects of regulation and law enforcement that have nothing to do with money laundering.
as
This is easiest to appreciate on the law enforcement side, which includes activities (such
US
Secret Service protection) that have negligible or zero money-laundering components.
generally opt for the lower estimate (2 percent), which is consistent with
government officials
the "rough cut" estimate of the federal budgetary costs of the US AML regime of $1 billion in
fiscal year 2001 presented in the 2000 NMLS (appendix 6). The higher estimate (3 percent) is
consistent with the data assembled for this study, which included expenditures by the Federal
Reserve System that were excluded from the 2000 NMLS estimates.

that
48. Suskind (2004) reports that US Treasury Secretary Paul O'Neill subsequently decided
to
the $1 billion estimate for budget expenditures in fiscal year 2001 should have been closer
the original esti¬
$700 million. We were unable to find the basis for his revision, and, as noted,
mate excluded costs incurred by the Federal Reserve System.

THE ANTI-MONEY LAUNDERING REGIME 97

tor as a whole, including state and local expenditures, the total may have
been twice as large, about $3 billion as an upper-bound estimate.49

Costs to Private-Sector Institutions

Estimates of AML regime costs to private-sector institutions are similarly

elusive. In the case of large banks it is possible to separate out the costs of
units explicitly set up to comply with AML regulations and reduce a bank's
vulnerability to it. However, internal costs associated with filling out forms
and examining customer records for AML compliance cannot be readily
separated from other reporting requirements related to the Bank Secrecy
Act. Interviews with officials in a number of large financial institutions sug¬
gest that it is these dispersed costs that are the principal expense.
The literature on the costs of bank regulation is small, but a review by
Gregory Elliehausen (1998) of what is available helps to obtain some sense
of the order of magnitude of costs for banks and, by extension, other finan¬
cial institutions and nonfinancial businesses subject to AML regulations.
Elliehausen distinguishes four concepts of costs: opportunity costs when reg¬
ulation prevents an institution from engaging in a profitable activity; oper¬
ating costs, divided between start-up and ongoing costs; total costs of the
activity, even if not required by law or regulation; and incremental costs
specifically associated with the required activity.
Most useful for purposes here are estimates cited by Elliehausen drawing
on the work of the accounting firm Grant Thornton (1992, 1993) for the
Independent Bankers Association of America. Elliehausen is relatively com¬
fortable with the methodology in these studies, which provide estimates of
ongoing operating costs of complying with Bank Secrecy Act regulations as
of 1991, including a large component of AML regulations as well as other
matters. One Grant Thornton study estimates that compliance with the
Bank Secrecy Act accounted for 0.2 percent of individual banks' noninter¬
est expenses, while the other said 0.5 percent. This range translates into
between $240 million and $600 million for the entire banking system. Using
the larger number and scaling it by the growth of nominal GDP between
1991 and 2003 produces a figure of about $1 billion. A ballpark upper-end
estimate of the costs of AML regime compliance for banks would be $1.5 bil¬
lion in 2003, recognizing that AML requirements have increased over that

49. We did not find even roughly comparable data on the AML
prevention and enforcement
expenditures by state and local governments, so we were forced to rely on some ready reck¬
oning. Government consumption expenditures at the state and local levels are about 1.7 times
those at the federal level, which might suggest applying a larger multiplier to estimated fed¬
eral expenditures to come up with total government expenditures on anti-money launder¬
ing. However, a disproportionate amount of the AML activity is at the federal level, which
suggests doubling the estimate for the federal government should produce a reasonable
upper-bound estimate of total government financial costs of the US AML regime.

98 CHASING DIRTY MONEY

period. The figure might reasonably be doubled to obtain an upper-end
guesstimate of $3 billion for all private-sector institutions.50
Studies by KPMG (2003) and Pricewaterhouse Coopers (2003) for the UK
National Criminal Intelligence Service and the Financial Services Authority
(FSA), respectively, provide a partial cross-check for these estimates. The
KPMG study provides a "rough estimate" of the current cost of the UK SARs
regime for reporting entities of £90 million. If we scale this figure by the size
of the US economy relative to the size of the UK economy and convert it
from pounds sterling to dollars, the corresponding estimate is $1.1 billion.
Since reporting of SARs is an important but only one element of the AML
regime, this rough estimate suggests that a figure of $3 billion for the entire
US regime is not unreasonable as an upper-bound estimate.
The Pricewaterhouse Coopers study for the FSA reports on the one-time
compliance costs of extending customer due diligence to existing cus¬
tomers who had not previously been subjected to CDD because they were
customers of the institutions prior to the introduction of those require¬
ments in the United Kingdom in 1994. The study suggests that the costs
for banks are slightly more than half those for all major financial institu¬
tions.51 Pricewaterhouse Coopers also compares the costs of the two pro¬
posed approaches with the overall AML compliance costs in the United
Kingdom. Using the arithmetic outlined above, the comparable total for
compliance costs to financial institutions in the United States would be
about $2.1 billion, which is in the same ballpark as our estimate of $3 billion.52
A 2002 study by Celent Communications, a Boston financial-research
organization, reportedly concluded that AML spending by US financial
institutions in the wake of the passage of the USA PATRIOT Act would
total $10.9 billion over 2003-2005, or about $3.5 billion a year. Allowing
for the fact that some of these would be one-time costs and Celent might
have had an institutional interest in generating a larger rather than a

50. Excluding real estate from the "finance, insurance, and real estate" category of US GDP
by industry, depository institutions account for about 40 percent of the rest of the category,
which also includes nondepository institutions, security and commodity brokers, and insur¬
ance. On the other hand, the AML prevention regime is currently applied with much greater
force to depository institutions, which suggests that doubling the estimate for banks should
produce a reasonable upper-bound estimate.

51. Pricewaterhouse Coopers examined two approaches to implementing the AML regime.
The finding cited in the text combines estimates of the costs of the more expensive approach
with estimates of the costs for six major banks that are already implementing something close
to the more expensive approach.

52. The May 9, 2003, Federal Register (page 25108) notice of the incremental paperwork bur¬
den associated with revisions to the US CDD regulations applied to 22,057 US financial insti¬
tutions covered by the USA PATRIOT Act. Each institution was expected to incur 11 hours
per respondent on average. Assuming that these hours would be divided equally among
tellers, bookkeepers, and financial officers on the basis of mean hourly earnings as reported
by the US Bureau of Labor Statistics, augmented by 30 percent for benefits, the total cost of
this adjustment to the US AML regime was a modest $5.7 million.

THE ANTI-MONEY LAUNDERING REGIME 99

smaller estimate, the Celent estimate would appear to be in the same ball¬
park as the aforementioned estimates.53 Recall, however, that these are
estimates of gross financial costs and do not take into account any finan¬
cial or nonfinancial benefits of complying with the AML regime.
The two studies for the UK authorities also looked at AML costs to the
government. The KPMG study of the SAR regime estimates these costs at
only about 12 percent of the costs to reporting institutions. However, that fig¬
ure should not undercut our guesstimate of the rough equality between the
overall costs of the US AML regime to the government and private-sector
institutions. Many government costs with respect to prevention, such as
drafting regulations and conducting examinations, are unrelated to the
actual management of information flows. The United States is also far more
active than the United Kingdom in prosecutions (chapter 5). Moreover, our
estimate of total US federal prevention costs (figure 4.3) is only about one-
fifth of the total prevention plus enforcement costs.54
Elliehausen's summary of the research in this area establishes a strong case
for the presence of economies of scale with respect to the costs of AML com¬
pliance, where assets or some more refined definition of a bank's output
measure scale.55 Fixed costs of compliance reflect such items as computer sys¬
tems, which account for a large component of start-up costs for AML
regimes.56 However, variable costs may also be lumpy for small institutions;
ongoing compliance involves larger proportions of labor costs, but special
skills (human capital) are involved. To the extent that a large component of
the costs imposed on firms are fixed costs, it is understandable that there is
resistance by casinos, the real estate industry, lawyers, and accountants to
expanding the AML regime into their lines of business.57 The AML regime

53. Celent Communications was cited in Business Week (December 1, 2003, 102) as the source
for an estimate that more than 13,000 US financial institutions had yet to implement basic
watch lists to screen new customers. The firm argued that many small institutions would
rather pay a fine than install costly new technology for customer screening.

54. The Pricewaterhouse Coopers study estimates that the cost to the UK government of the
expanded retrospective CDD would be a minuscule amount (0.3 percentage points) of the total
cost to firms. This low figure is understandable because the change in the AML regime applies
principally to the reporters and the cost to the government only involves receiving additional
reports but not any subsequent action using the reports.

55. Pricewaterhouse Coopers (2003, 88) reports that large banks benefit from economies of
scale in outsourcing some aspects of customer due diligence to credit bureaus.

56. KPMG (2003, 48) estimates that the capital costs of the UK SAR regime are roughly
two-thirds of annual compliance costs. On the other hand, Celent (www.celent.net/japanese/
PressReleases/ 20020927/ AntiMoneyLaundering.htm [accessed March 8, 2004]) reported that
hardware costs were only 6 percent of the estimated total of three-year AML regime expen¬
ditures by US financial institutions.

57. HM Treasury (2002) estimates that the cost of extending the UK AML regime to lawyers,
accountants, and real estate dealers, converted into dollars and scaled to US GDP, is $2.1 bil¬
lion. However, it should be noted that the US AML regime does not as yet apply extensively
to these professions.

100 CHASING DIRTY MONEY

imposes additional one-time costs for continuing in business, as well as sub¬
sequent costs resulting from frequent changes in regulations.
A report to Congress on the use of currency transaction reports illus¬
trates some of the complexities of AML regime costs for banks, and pre¬
sumably for other reporters as well. A survey by FinCEN (2002) of the
reasons why banks did not take advantage of the potential exemptions
from CTR reporting detected little difference between the responses of so-
called megabanks, each with more than $32 billion in assets, and all other
financial institutions. Among the more prominent reasons for not using
the exemption was that it is not cost effective to change systems once they
are in place, suggesting a large role for fixed costs. Other reasons included
the additional due diligence costs involved and fear of regulatory action if
an exemption were to turn out to be wrong. The survey findings suggest
some of the reasons for the potential information overload in the reporting
element in the US prevention pillar.

Costs to the General Public

Notwithstanding the economic and social benefits of the AML regime, the
general public incurs costs from the increased regulation in the form of
reduced efficiency and higher charges. While there is little to go on to esti¬
mate out-of-pocket or opportunity costs of the AML regime, comprehen¬
sive cross-country research by the World Bank (2004) shows that more
regulation in general is associated with lower labor productivity, greater
use of the informal economy, increased corruption, and higher costs. There
is no reason to believe that such effects of regulation would be any differ¬
ent in the case of the AML regime — the only question is the magnitude of
the adverse effects. A topic for future research might be to adapt the World
Bank's survey approach in connection with estimating AML regime costs
to the general public.
The Pricewaterhouse Coopers report (2003, 75) on expanded CDD in the
United Kingdom stated that "a new regulation, which makes it more
expensive to enter a market, might be expected to soften the level of com¬
petition, to the potential detriment of consumers." In addition, Elliehausen
(1998) argues that the policy implications of economies of scale in the costs
of regulation are that they inhibit entry of new firms into banking, limit
interinstitutional competition, undermine deregulation efforts across the
financial sector, and reduce incentives for financial innovation because reg¬
ulatory costs are relatively high at low levels of output that are associated
with the early phase of regulation.
Pricewaterhouse Coopers provides some quantitative insight into the
costs of the AML regime to the UK general public. The report estimates a
cost per customer of between 7 and 12 percent of the overall cost of imple¬
menting the one-time adjustment in the UK AML regime with respect to

THE ANTI-MONEY LAUNDERING REGIME 101

CDD, depending on which approach would be chosen. The estimate is
based on the customer's time required, the opportunity cost of that time,
and any materials. In addition, there may be other transaction costs asso¬
ciated with routine business operations.
What might be a plausible upper-end estimate of the out-of-pocket costs
of the AML regime to the US general public? A rough guess is that private-
sector institutions are able to shift up to a third of their gross financial costs
to consumers, perhaps something on the order of $1 billion per year on the
basis of the earlier guesstimate of the $3 billion annual cost of the AML
regime to those institutions as a group. We have no empirical basis for this
figure, but in the context of our estimate of the costs of the overall AML
regime, these costs are not additional to those borne by those institutions.
Perhaps a similar figure of $1 billion, which amounts to one-third of the esti¬
mated gross costs to financial and nonfinancial institutions and businesses,
is reasonable as an upper-end estimate of current additional costs of the
AML regime to the US general public. It may be on the high side given the
Pricewaterhouse Coopers estimate cited in the previous paragraph, but that
estimate appears to be based on a rather narrow conceptual base.
In addition to the general costs of regulation to the public, certain aspects
of the AML regime may impose a disproportionate burden on poorer
households. For example, requirements to present multiple forms of iden¬
tification to establish a bank account have created problems in low-income
(legal and illegal) immigrant communities. Indeed, as noted above, the
United Kingdom provides an exception it its requirements for CDD in part
so as not to make access to the financial system too difficult for low-income
and immigrant groups. In the United States, the risk-based approach to
customer due diligence raises social concerns about possible profiling by
country of origin or by race. Moreover, increased regulation in the formal
economy is likely to increase reliance on the informal economy. When it
comes to money laundering, this means reliance on informal means of
transferring funds across borders, which is particularly relevant for legal as
well as illegal immigrant workers who transmit a large portion of their earn¬
ings back to their countries of origin.58 One consequence of customers lack¬
ing low-cost access to the traditional banking system is a less effective AML
regime, because their exclusion makes tracing money that much more dif¬
ficult. The larger the flow of legitimate funds through unregulated channels,
the harder it is to find money laundering through the same mechanisms.
In summary, adding together our crude estimates of the gross financial
costs of the US AML regime to the government, private-sector institutions,
and the general public produces a rough upper-bound total of $7 billion.
The figure includes $3 billion for the government, $3 billion for private-
sector institutions (of which $1 billion might be shifted to the general pub-

58. A task force report by the Inter- American Dialogue (2004) estimates total remittances to
Latin America in 2002 were $32 billion.

102 CHASING DIRTY MONEY

lie), and $1 billion in additional costs for the private sector. Assuming this
figure is in the ballpark, it represents about 0.06 percent of US GDP in 2003,
or about $25 per capita.
The costs of a given AML regime differ across countries. The United
States and some other wealthy industrial countries can establish a com¬
prehensive AML regime at modest cost relative to GDP because they have
efficient banks and other institutions with sophisticated internal control
systems that make it relatively easy to add AML dimensions. Institutions
in poorer nations may have more difficulty in complying with require¬
ments such as the FATF's Forty Recommendations. Similarly, large insti¬
tutions may enjoy competitive advantages within wealthy nations to the
extent that setting up an effective AML system within an institution has
substantial fixed costs. Interviews with officials of financial institutions

suggest that these are very much second-order issues, but too little is
known about them to dismiss them summarily.
Much more attention deserves to be given to the costs of the AML re¬
gime, but it would seem that the biggest issue is not necessarily the finan¬
cial costs. Of course, views may differ with respect to nonfinancial costs,
such as intrusions on privacy. In the wake of September 11, the judgment
was that additional nonfinancial costs should be absorbed, and the advo¬
cates of increased prevention and enforcement received new powers. More
recently, the pendulum may have swung back some, but in the United
States this is a familiar story in terms of the AML regime, where it has not
been at all uncommon over the years for any given piece of legislation actu¬
ally to relax provisions mandated in previous legislation. The principal
example is the Money Laundering Suppression Act of 1994, which autho¬
rized the liberalization of regulations for the granting of exemptions for
cash transaction reports. As noted by FinCEN (2002), the relaxation of
those particular regulations has been only a partial success. The legislation
authorized the Treasury to designate a single agency to receive SARs at the
same time it extended the AML regime to money transmission businesses
and to casinos. In other words, relaxing the regulations was part of the
political quid pro quo for expanding coverage of the AML regime.
Even if the estimated total financial cost of the US AML regime is con¬
sidered to be small — as stated, $7 billion is a mere 0.06 percent of 2003
GDP — the cost needs to be justified, particularly if the prevention pillar is
to be expanded or the enforcement pillar strengthened. This suggests a
need for better data and fresh approaches to the analysis of these issues in
the years ahead.

THE ANTI-MONEY LAUNDERING REGIME 103

Combating Predicate Crimes Involved
in Money Laundering

How successful have the anti-money laundering (AML) regimes of the

United States and other countries been in controlling crime, particularly
''predicate" offenses that give rise to money-laundering activities? While
previous chapters have primarily focused on describing the phenomenon
of money laundering and the structure of regimes to combat it, this chap¬
ter moves to assessing those regimes, starting with a look at implementa¬
tion of the enforcement pillar in the United States and other industrialized
nations. The emphasis is on intermediate measures of performance such
as the number of suspicious activity reports, prosecutions, convictions
and incarcerations, seizures, and forfeitures. A model of the market for
money-laundering services is used to link interventions to outcomes,
although conceptual and empirical complexities limit the utility of the
model, particularly in generating performance measures. The chapter ends
with some indicative calculations that suggest that enforcement efforts
have yet to make money laundering a particularly risky business in the
United States.

Enforcement and Predicate Crimes

Until September 11, 2001, the primary emphasis of money-laundering

enforcement in the United States and elsewhere was on reducing predicate
offenses — initially the sale of illegal drugs, but more recently a wide array
of crimes. Critical to the prevention pillar of those efforts is the flow of re¬
quired reports that are intended to generate investigative information.
More important are the number and characteristics of criminal prosecu-
105
tions and convictions for money laundering, and the application of associ¬
ated penalties such as seizures and forfeitures.

Suspicious Activity Reports

In the United States, SARs have replaced currency transaction reports as

the primary source of information from financial institutions and other
reporting sources (such as casinos and currency exchange bureaus) to
assist anti-money laundering efforts. Although CTRs still play a role —
more than 13 million of them were filed in 2001 and 1.5 million were at
some point identified in the course of a criminal investigation (FinCEN
2002) — SARs at this point are viewed by professionals in the enforcement
field as more informative.
According to the US Financial Crime Enforcement Network (FinCEN
2003b), the number of SARs filed in the United States increased from 52,000
in 1996 to 288,000 in 2003 (figure 5.1). Little information is available on
the underlying suspected activity, as 48 percent of SARs filed between
April 1, 1996, and June 30, 2003, were characterized only as "violations of
the Bank Secrecy Act (BSA)/ Structuring /Money Laundering," which is
not very informative. The only other large category was check fraud,
which accounted for 12 percent of the filed reports.
The number of SARs related to terrorism predictably increased sharply
following September 11. Whereas in September 2001, 27 SARs mentioned
possible terrorism, another 1,342 were filed in the following six months.
However, this number had decreased again within another year. In the six
months beginning October 2002, only 290 terrorism SARs were filed.
An encouraging trend has emerged with regard to the party initiating
terrorism-related SARs. In the six months after September 2001, 85 percent
of these SARs were filed due to apparent matches with the names of indi¬
viduals or entities provided to institutions by government agencies. But
from October 2002 to March 2003, most such SARs were a result of due dili¬
gence processes of financial institutions themselves, independent of any
government-published lists. Several banks created internal watch lists to
alert tellers and other employees to customers' previous suspicious behavior.
Though 69 financial institutions have filed SARs related to terrorism, over
half of the reports from October 2002 to March 2003 came from just three
banks. Sixty-eight terrorism SARs (23.4 percent) were reported directly to
law enforcement authorities, meaning that the violation was ongoing and
required immediate attention.
The number of SARs filed, like most other criminal justice outputs, is an
inherently ambiguous indicator of changes in enforcement. A rise in the
number of SARs may reflect either an increase in money laundering or
increased stringency of the AML regime. The rate of increase in recent
years is so large that, with a caveat as to quality, there is good reason to
believe that it is the stringency of the regime that has intensified. There are

106 CHASING DIRTY MONEY

 Figure 5.1 Suspicious activity reports filed,
1996-2003
thousands

Source: Financial Crime Enforcement Network (FinCEN 2004).

no events that would explain a comparable increase in the incidence of

money laundering.
SARs are of variable quality. An enforcement official described one bank
as encouraging the filing of SARs with a low threshold but providing min¬
imal information about the transaction — in other words, the purpose was
to protect the bank against charges of violating reporting requirements,
with little focus on assisting the government. If banks and other regulated
firms feel a greater need to protect themselves against government sanc¬
tions by filing reports, the increase in numbers may not indicate improved
diligence. Moreover, the increase may be weakening the effectiveness of the
regime in the process by lowering the signal-to-noise ratio. On the other
hand, another bank was described as having invested in training its staff to
file reports only when there was indeed reasonable suspicion and to make
the reports informative. Such discrepancies in efforts point to the need to
examine not merely the number of filings but the extent to which they have
resulted in detection and punishment of money-laundering offenses.
For the 6V2-yeav period ending October 31, 2002, 940,000 SARs produced
70,000 direct referrals to federal law enforcement agencies, of which almost
half were to the FBI. Unfortunately, there is no information on how many
resulted in or contributed to cases.
The US General Accounting Office (GAO) has periodically attempted to
ascertain the results of the SAR filings in terms of prosecutions and convic¬
tions. An early study (GAO 1998) found that state officials reported "lim¬
ited or no investigative actions" from materials supplied by FinCEN. Even
today, the extent to which information provided by SARs has helped to
make criminal cases is unclear. Another GAO study (2002) that examined

COMBATING PREDICATE CRIMES 107

 Box 5.1 Detecting money laundering through failures to file
suspicious activity reports
Two relatively small US banks were fined in 2002 for failure to file suspicious activity
reports (SARs). In September, the US Financial Crime Enforcement Network (FinCEN)
imposed a $100,000 civil penalty against the Great Eastern Bank of Florida, which had
total deposits of $55 million. Great Eastern failed to alert authorities about the question¬
able activities of at least 20 different customers. Among the unreported activities were
structured deposits, transfers immediately following large deposits, and the deposit of

large numbers of sequentially ordered money orders and travelers’ checks. In some
cases, when Great Eastern did file SARs, the reports lacked the relevant information that
would have provided assistance in an investigation. FinCEN determined that Great
Eastern’s violations were willful because the bank had been on notice to improve its com¬
pliance with the anti-money laundering (AML) rules after the Federal Deposit Insurance

Corporation (FDIC) repeatedly found material deficiencies in the bank’s AML program.
Later that year, Broadway National Bank (BNB) of New York City pled guilty to charges
of failing to file SARs, failing to have an anti-money laundering program in place, and per¬
mitting the structuring of transactions. According to convicted money launderer Joseph
Vershish, the laxness of the BNB staff was well known among those of his profession.

Management did not question the origins of cash or ask to know their customers’ business
locations, and would readily authorize transfers just a few days after large or structured

deposits. Security guards would even warn “smurfs” (couriers) of the presence of govern¬
ment agents. The bank’s biggest client was convicted drug dealer Alfred Dauber, who alone
laundered millions of dollars through BNB. The bank paid a $4 million fine, a significant but
not crippling amount for a bank with $89 million in total deposits. Since 1 998, BNB has been
operating under a cease-and-desist order by the Office of the Comptroller of Currency and
has established and implemented an effective AML program, according to that agency.

all SARs involving credit cards during a two-year period from October 1,
1999, to September 30, 2001, found 499 such filings, of which 70 were
referred to law enforcement agencies (39 federal, 31 state or local). But the
GAO noted that FinCEN was unable to report whether any of these refer¬
rals resulted in criminal prosecutions.
The requirement itself to file SARs can indirectly generate useful in¬
formation. Box 5.1 describes two instances in which it was the failure to
file by two small banks that provided information that led to penalties
against them.

Prosecutions and Convictions

Available data on money-laundering charges in the United States — which

come from judicial sources for the federal level and from surveys of inmates
in federal and state prisons — cover the number of persons charged, con¬
victed, and imprisoned in federal courts.1 Charges can be brought against

1. Judicial data on state court convictions are not available, although the inmate survey pre¬
sented in more detail below reinforces a general impression that there are few convictions in
state court on such charges.

108 CHASING DIRTY MONEY

Table 5.1 Defendants charged with money laundering, 1994-2001
Percent with
Money laundering money laundering
Money laundering as primary or as primary or
Year as any charge secondary charge secondary
70 charge

1994
1,907 1,341 73
1995 2,138 70
1996 1,487
1,994 1,457
1997 68
2,376 1,619
1998
2,719
1999 2,656 1,831
1,885
2000 67
2,503 71
2001 1,771 71
70
2,110 1,480
Source: Bureau of Justice Statistics (2003).

both the customer who seeks to have money laundered and the provider
of the service, although the data do not distinguish between these two
types of offenders.

Federal Prosecutions and Convictions

Data on prosecutions and convictions in federal cases are available from the
Administrative Office of the Courts (AOC) and the US Sentencing Com¬
mission. The data sets are different because the US Sentencing Commission
data reflect only those who were sentenced in a given year,2 while the AOC
data reflect all matters related to a criminal charge that were conducted in
a given year. For example, some of those convicted in 1999 were not sen¬
tenced until 2000. The analysis presented here refers primarily to the AOC
findings (Bureau of Justice Statistics 2003), although US Sentencing
Commission data are also used to provide some additional insights.
Table 5.1 shows that the total number of defendants charged with money
laundering rose from 1994 to 1998 and then fell sharply through 2001.
Slightly more than 2,100 persons were charged with money laundering
offenses in 2001, compared with more than 2,700 in 1998. Only 22 busi¬
nesses were criminally convicted in 2001, and for about 30 percent of those
charged with money laundering, the offense was not one of the two most
serious charges. Table 5.2 shows the number of convictions for which
money laundering was the lead offense, which is not necessarily the
offense with the highest statutory penalty but normally the one that gen¬
erated the investigation. In the vast majority of these cases (81 to 88 per¬
cent, depending on the year), those charged were convicted.
What predicate crimes generate money-laundering convictions? In this
respect, there is a significant difference between those charged with money

2. These data were accessed online at the National Archive of Criminal Justice Data of the
Inter-University Consortium for Political and Social Research, www.icpsr.umich.edu/
NACJD/ index.html.

COMBATING PREDICATE CRIMES 109

 Table 5.2 Adjudications and convictions in cases with money
laundering as the most serious offense, 1994-2001
Year Adjudications Convictions Percent convicted
1994 933
1,159
1995 906
1,073 81
1996
1,241 1,080 84
1997 89
1,245 1,108 87
1998 88
1,370 1,199
1999 87
1,571 1,371
2000
1,527 1,329
2001 88
1,420 1,243 87

Source: Bureau of Justice Statistics (2003).

laundering as the lead offense and those for whom it was a secondary
offense. For about 60 percent of the first group (which constitutes two-thirds
of the total), a property offense (embezzlement or fraud) was the predicate
crime and for only one in six was the predicate crime a drug offense.
However, among the smaller group whose lead charge was not money
laundering, about 90 percent were charged with drug trafficking. That is
probably the consequence of differences in maximum statutory sentences.
Drug offenders face longer sentences than those convicted of money laun¬
dering, so drug-money launderers are more likely to be charged with the
drug offense if they had any involvement beyond pure money laundering.
The vast majority (84 percent) of those charged with money laundering
in 2001 were charged under Title 18 sections 1956 and 1957, which cover
the transfer or transportation of criminally derived money or property
with the intent to conceal or disguise its illicit nature or origin. The other
16 percent were charged under sections of Title 31 that address monetary
reporting/ recording offenses such as cash smuggling, structured transac¬
tions, and failure to file required reports.
US Sentencing Commission data in table 5.3 provide another view on the
same matter, since they include other charges that resulted in convictions.
Crimes identified in these data are different from the predicate crimes, in
that they may involve an individual who laundered money from a fraud
committed by someone else but was also convicted of embezzlement in his
or her own right.3 Of the 1,543 defendants sentenced under one or more
money-laundering statutes in 2000, 828 were also convicted of one or more
other criminal offenses, and 715 were sentenced only for money laundering.
Of the latter, 125 individuals were also charged either with a conspiracy,
which might have included a crime other than money laundering, or with
being the principal offender. Thus, nearly half of those convicted may have
been involved only in the laundering and not in other aspects of the crimes,

3. Data are not available to compare predicate offenses and these other charges.

110 CHASING DIRTY MONEY

Table 5.3 Other offenses for which convicted federal
money launderers were sentenced, 1995, 2000
(numbers in parentheses indicate percent of total cases)
Offense 1995 2000

Number of cases involving money-laundering statutes3 467 (30)

351 (30)
1,155 1,543
Also charged with drug trafficking 151 (13) 59 (17)
258 (4)
32 (3)
Also charged with fraud 37 (2)
31 (3)
Also charged with smuggling (non-drug) 22 (2) 25 (2)
15(1) 23(2)
Also charged with racketeering
Also charged with embezzlement
Also charged with tax evasion 568 (49) 715 (46)
Charged with no other crimes
Percent of drug cases also charged with money laundering 2 2
Percent of fraud cases also charged with money laundering 2 1
Percent of smuggling cases also charged with money laundering 1 2
Percent of racketeering cases also charged with
money laundering 4 4
Percent of embezzlement cases also charged with
money laundering 1 1
Percent of tax evasion cases also charged with
money laundering 3 6

a. Includes Title 18 sections 1956 and 1957 (money laundering), and Title 31 sections 5316
(cash smuggling), 5324 (structuring transactions to evade reporting requirements), and 5313
(failure to file currency transaction report).
Source: US Sentencing Commission (2003).

although there is no way to know whether they were customers or providers

of money laundering. More than two-thirds of these "pure" money laun¬
derers were sentenced under Title 18 statutes as opposed to Title 31.
Table 5.3 shows that for both 1995 and 2000, 30 percent of all defendants
convicted of money laundering were also convicted of drug offenses. Of
255 cases in 2000 where a Title 31 offense was one of the charges, 223 (87
percent) had no non-money laundering charges, while 218 (85 percent)
had nothing besides Title 31 charges.
Interestingly, of all persons convicted of drug offenses in federal court,
only 2 percent were also convicted of money laundering. This figure was not
much different from that for those convicted of fraud who were also con¬
victed of money laundering (2 percent in 1995 and 1 percent in 2000). The
dominance of drugs among secondary charges for money laundering reflects
the dominance of drug offenses in the federal criminal justice system. About
60 percent of federal prison inmates have been convicted of drug offenses.
Money-laundering sentences averaged about 36 months, substantially less
than the average of approximately 48 months for all those convicted in fed¬
eral court (Bureau of Justice Statistics 2003). Seventeen percent of those sen¬
tenced under the guidelines were given longer sentences because they had
leadership roles. About 20 percent of the cases involved more than $1 million

COMBATING PREDICATE CRIMES 111

in funds (US Treasury 2002, 5). The mean sentence for cash/ monetary instru¬
ment smuggling (Title 31) was 19.6 months; for structuring transactions,
13.4 months; and for failure to report currency transactions, 8.5 months.
These figures are consistent with the conjecture that most of these cases are
pure money laundering, usually with a low-level offender who does noth¬
ing else but some illegal legwork for the "predicate criminal"; once again
though it must be emphasized that the role may be delivering the funds to
the launderer rather than providing the actual service.

Inmate Survey
The federal court data reviewed above have significant limitations because
not every successful money-laundering investigation results in a convic¬
tion for money laundering, as opposed to some other offense. For example,
the prosecutor may drop the money-laundering charge in return for a plea
to another charge related to the predicate offense itself. The fact that
money-laundering charges usually result from investigations that began
with another crime (Joseph 2001) reinforces the concern about the com¬
prehensiveness of the figures. Fortunately, some other data throw light on
how many money launderers are in prison, regardless of the offense asso¬
ciated with the inmate's conviction.
Approximately every five years, the Bureau of Justice Statistics interviews
a large sample (about 18,000 in 1997) of inmates in both federal and state
prisons. The questionnaire includes items on their criminal activities, not
restricted to those for which they were convicted. These data provide an
important supplement to the administrative data. Questions in the most
recent (1997) survey concerning money laundering have been analyzed by
Jonathan Caulkins and Eric Sevigny (of Camegie-Mellon University and the
University of Pittsburgh, respectively), who reported the relevant results of
their work in personal communication with the authors. Note that although
these data are not directly comparable to any year of court data, since most
of those incarcerated in 1997 were convicted in an earlier year, table 5.3
showed little change in the pattern of convictions from 1995 to 2000.
Among federal prison inmates, 3,030 (2.8 percent of the total population)
reported that they were serving time for a money-laundering conviction.
Two-thirds of those had some drug involvement and another 18 percent
reported forgery /fraud convictions.4 Including those who said that they
laundered drug money but were not convicted on that charge, federal pris¬
ons in 1997 contained an estimated 4,416 money launderers (5 percent of

4. Some of the discrepancies between the inmate survey and the court data may reflect the
longer sentences of drug offenders; thus the prison population of money launderers will
be richer in drug-money launderers than the population entering prison.

112 CHASING DIRTY MONEY

the total population). Among those who said that they laundered drug
money, only about one-sixth (467) were estimated to have had no other
involvement with drugs.
None of the state prison inmates reported that they were currently serv¬
ing time for a money-laundering offense. However, an estimated 6,368 state
prison inmates (0.6 percent of the total population) self-reported that they
did launder money, and in every case they reported that the money in¬
volved came from drug offenses. This finding that only drug offenders have
laundered money is an artifact of the questionnaire, since only inmates
reporting drug convictions were asked about money-laundering activities.
Only about 100 reported that they were money launderers exclusively; the
others said they were also drug dealers. Again, this response may be a con¬
sequence of the way the questions were asked.
Many persons who launder money are in prison for other offenses, par¬
ticularly for drug offenses. Thus, though there appears to be a negligible
number of state-level convictions for money laundering, the self-reported
inmate data suggest there are actually more money launderers in state
prison than in federal prison; money laundering may have been a minor
part of their drug-dealing activities and they may have been customers
rather than providers of money-laundering services.
In the federal system, the court statistics (both from the AOC and the US
Sentencing Commission) do not suggest a dominant role for drugs. How¬
ever, the inmate survey suggests that most of those in prison on money¬
laundering charges were in fact involved in drug dealing. There do not
appear to be many stand-alone money launderers in the prison system.

Financial Penalties

Given the nature of money laundering as an offense, prosecution of it, un¬

like prosecution of a violent crime, can be expected to generate substantial
financial penalties. The government may seize the laundered money and
other assets of those charged and seek forfeiture upon their conviction. In
some cases, these seizures and forfeitures can generate very substantial
amounts — a prominent case involving a Bank of New York official in 1999
resulted in criminal forfeitures of $8.1 million (US Treasury 2002, A-15).
In 2001, the only year for which comprehensive data are available, total
seizures and forfeitures were $627 million (US Treasury 2002).
The federal government also levies other financial penalties. Figure 5.2
shows a substantial increase in the total amounts of fines and restitution,
from $100 million in 1996 to $665 million in 2001. The data, provided by
the US Sentencing Commission, reflect the growth in the size of the aver¬
age penalty rather than the number.

COMBATING PREDICATE CRIMES 113

 Figure 5.2 Total fines and restitution for money
laundering in the United States, 1990-2001
millions of dollars

The relevant metric to assess these figures is the total volume of funds
laundered. Choosing an estimated total figure toward the lower end of the
usual range, such as $300 billion, implies that the current level of penal¬
ties — seizures, forfeitures, fines, and restitutions — is almost trivial, only
four-tenths of 1 percent. However, if the total figure is only a few tens of
billions — or at least if the forms of money laundering of greatest social con¬
cern are only a few tens of billions — then the level of penalties might be
1 to 3 percent, perhaps enough to have a modest deterrent effect on those
tempted to commit the predicate crimes.

Enforcement in the United Kingdom

and Other Nations

We treat the United Kingdom separately not so much because of its impor¬
tance in the international financial system but because it has been more
advanced than any other nation, including the United States, in the analy¬
sis of its own AML regime. A study of the country's system for suspicious
activity reports, funded by the British government (KPMG 2003), found an
extraordinary increase in the number of SAR filings since 2000: from 20,000
in 2000 to a projected 100,000 in 2003. According to the study, the increase
reflects the extension of AML requirements to lawyers and real estate
agents. At least 6 percent of a sample of SARs disseminated by the UK
National Crime Intelligence Service resulted in "a positive law enforce-

114 CHASING DIRTY MONEY

merit outcome (i.e., prosecution, confiscation, cash seizure, etc.)," while
another 5 percent were still being used in an active investigation.5 The
study also noted, however, that there was little feedback from law enforce¬
ment agencies to the filing institutions.
In England and Wales, there were only 357 prosecutions for violation of
money-laundering statutes in the 12 years from 1987-98 (KPMG 2003). 6
Though the UK Financial Services Authority has made AML a major pri¬
ority, the sense of strategy in criminal investigations and prosecutions is
lacking. That view was reinforced by a recent review of the AML regime in
the United Kingdom by the International Monetary Fund, which found
that enforcement was limited even though the structure and laws for it
were in place (IMF 2003d).7 The report went on to state: "Cases are gener¬
ally considered for enforcement only when there is little chance that they
will be seriously contested or complicated. . . . The Crown Prosecution
does not prosecute any matters other than narcotic money laundering." In
2000 and the first half of 2001, the report continued, there were only 18
"skilled-person" visits to financial institutions focusing on AML issues
(IMF 2003d, 100, 102).
Finally, the IMF noted many financial and professional entities effec¬
tively out of the sight of regulators. Money service businesses in the UK
had always been subject to the AML regime, but until November 2001 no
regulatory agency had responsibility for assuring compliance. Even now,
the UK Financial Services Authority has been mandated to provide only a
"light touch" supervisory regime for money service businesses.
A report by the Performance and Innovation Unit (2000) of the UK
Cabinet Office on the use of confiscation orders also found that the system
performed poorly. For non-drug crimes in 1998, for example, only 136 con¬
fiscations were ordered and 6 million pounds collected — less than half of
what was ordered to be confiscated. For drug cases the numbers were
larger, but only 10 million pounds were collected in a market estimated to
total some billions of pounds in revenues.

Other Nations

The US Department of State (2003) provides the numbers of suspicious

activity reports filed in other nations as reported by financial intelligence
units in those countries. Table 5.4 shows the number of SARs for seven

5. Given that the researchers were unable to track the ultimate use of most of the sample of
SARs, this is the least favorable presentation of the data. By the most favorable analysis, one-
third of SARs resulted in a law enforcement success, mostly as intelligence rather than evi¬
dence for prosecution.
6. Data were not available for Scotland and Northern Ireland.

7. The United Kingdom was one of the first jurisdictions to permit publication of the detailed
findings of the IMF/ World Bank review team.

COMBATING PREDICATE CRIMES 115

 10

CD co
CM co TD ■5 «
0 cn _vr ■D

1,000 1,000
LO

SARs/ /
E. 1 0 0 0 .E o
-C o
ID
ir i2

States United
co ID 0 0
d d d d d d d co
LO O) eg tr
O)*oo
CO
<0 co
CD OJ CvJ C\J ll C
JD
O) 0 • -
£2
0 co O eg
CC IT)
00 CD
CM ^
< CD
C\J
LO OJ
c
0 CO |
0 ^ -2
</) LO CD 0 23
-C
0 ^

Kingdom
■J— I- ^ eg eg
U_
00
eg CM CM CO D <Cl
E
eg eg
D C

SARs
CD
Q. o r- CM CD Cr- ^
0 c
o
O 0 d d d d d lO
d d
Q. eo o

1,000
c c .b
3

/
CM CM 3
-T—
United
0 0 O 0 O O 0 CD IT
0
CC 0 0 O
CO
0 r- 0
o 03 CC
0
<(/) ”
c
0
cm
CDt- 0 _ 5 O
CM
Switzerland

o'
T— O)
c
SARs

O O 0 0
h- d d d d cc c ”
‘■4— > rt
coi
1,000

o
CO c c
/

T— _ c|
1— CD 0 0
CO
i— 0
CO c
O £0 £
1 —10” 0
LO 0 0-
1— ” ^
coT-
Netherlands

CD o c =>o
00 ” T-|
CMc1d— LO
O0 =O
CD CD <0 3 0 4-
CO CM T—
CM O
< W
SARs

O)
0
co O— >
CM CO O- c

d CD
d d d d d d
CO CM o £ O^F
1,000

in CO CO CO ®■2CO
CM is
/

00 CM
00 <D "
LO O 0
CM
c o ~o « 2Q
CO - J) W 2' o
0 O CD 0 0
g S§>O 2 <d
CM o C 03
0 CD
■^T
2

o
-200

1—
Japan

0
T~ T— 1— CO
c
co o — ’
CD
h- 0
CO
.2
2 OT
-2 £co-D2 c
o>
co OE
C
8 t.9
SARs

t <
1994

O
1 I O O
1 1 LOl 1 O LO
0 O
— T-
Th- 0 CD
CM 00 3 21
1

d d d
CM d d £
co (- o>
SARs/

CD CO tc
3 0£3 OI
,000

CM
(0 CD CD CT— T—
8 x: c
ry,

CD P 2
CC o” < 8
France

< c CD
o” CD^ T~~
O .j? Z|
count

0) c o
"■4-1 m”
c T~ CL ■2 2
CO ” 0
co
0”

c
0 I I LO
0 O 1
co
”
’c 03
— 03
r <u ^ 0
Is-1

O”
|
1 1 1 1 1 T3 c ~
0 d LO
d 7— 0 -2 O l O
o eS ■ <Q
1/

i—
CD co T” 0 d- ^ ”
LO
obryts
,000 SARs 1

C CD
0 0 0 O
CC 1 I 1 I | 1— CD
1 1 r: *- -r—
Belgium

1 1 1 (D 1 1 r03.b ffl
y rep

< C\J
0) t3‘ i2 LL tr
0 0

CD
CD (D
Q_ co 3 o
c
CO oCO
o
0 I r 0 CO
tivit

0u O LO cr
O CO LO
8 2
q ”C\J
T- -” =O O
LO |
0 d d d d d d d Ios
cioaucs

C
yJ) CD CD 00 3 C3
0
<D
SARs/

00 ”
CO
r- JZ cm
,000

Tf Is- .b 0
CM
CD 0 Is- CM ”
0 co O
c co
CC o
CD CD
Suspi

LO 1 E
Australia

< 1 y= o CD .E
O c CJ
(/) o
T_
3
CO co
CO co
c 0 O
0 Is- LO
co Q- >, 00)1—=
0 0)
03
03
1 1I 1| 0 0 0-
0 Tn 0
0 d d d d d tc 0” “cm 0
00 0 o m
cO 0
z
00 Is- o
_Q
1— CM 0 1—
0 CM 2 c 3
Is- 1— o
5.4

CC I 1 1 1 0 0
m Is-
0 > 0 0
cn 1 1
< LO
CD co
” 0 3 0
o TD
o
c c0 QC O
c
o
“co
CD co 0
Table

Is- O 3
CD CD
Is-
CD c
|6
”

CD 0 CM — b 0
h-

Is-
0 CD CD Is-
CD CD CD 0 O O ” 0”
I-. 0
Q.
3 O CO .E LL
0 CD CD CT) 0 O O 0 C
0
CM CM CM 0 3
”
cd
O) o
^
116 1— h-
S
r-

T—
>- 1— ”
T— LO
nations and the United States from 1994-2001. The numbers have such a
range as to suggest that very different processes generate them in different
nations. For example, Mexico reported 500,000 SARs in 2002, while
Argentina reports only 200, yet no one would assert that the Mexican bank¬
ing system is three orders of magnitude more aggressive in this respect or
that money laundering is so much rarer an event in Argentina.8
Nations do indeed differ in their approach to the reporting of suspicious
activities. For example, in Switzerland the emphasis is on the customer
rather than the transaction, and great discretion is given to the reporting
institution, in sharp contrast to the United States (Pieth and Aiolfi 2003).
The Netherlands saw an 81 percent increase in 2002 over the previous year
in the number of "subjective unusual transaction reports," largely due to
the increase in the reporting of international money transfers. Suspicious
transaction reports in that country rose 25 percent over the same one-year
period, from roughly 20,000 to around 25,000. Unlike in other nations,
Dutch financial institutions (as well as casinos and credit card companies)
are required to file "subjective unusual transaction reports" rather than
SARs with their financial intelligence unit, the Meldpunt Ongebruikelijke
Transacties (MOT — Office for the Disclosure of Unusual Transactions). The
MOT in turn decides whether to forward these cases in the form of suspi¬
cious transaction reports to law enforcement authorities. The strategy
appears to be designed to encourage institutions to file a report if they have
any doubt at all about a transaction, and let MOT sort the wheat from the
chaff. In 2000, a Suspicious Transactions Intranet went into operation in
the Netherlands, allowing police to view all such reports since 1997.
In February 2000, Japan expanded the scope of predicate offenses to
include all serious crimes. It also overhauled its suspicious transaction
reporting system by creating a financial intelligence unit to analyze and
disseminate SARs, cooperate with law enforcement officials, and exchange
information with units in other jurisdictions. Previously, the Ministry of
Finance received the reports but was not required to take any of the afore¬
mentioned actions. With these changes, the number of SARs in Japan rose
from 13 in 1998 to nearly 19,000 in 2002, more than 12,000 of which were
deemed useful in law enforcement investigations.9

8. These countries are not included in table 5.4 but are cited in US State Department (2003).

9. The latter figure has to be taken with some skepticism, since both the share of all SARs and
the total number useful to law enforcement seem extraordinarily large. The actual figure sug¬
gests a reporting system that, though new, is effective at identifying only questionable trans¬
actions, and an enforcement system that is unusually energetic in pursuing these reports. On

the other hand, a 2004 report on Japan by the IMF/ World Bank (IMF 2004b) noted "the effec¬
tive application of legal powers appears to be limited, as evidenced by the low numbers of
investigations and prosecutions, which may be due, inter alia, to the limited resources allo¬
cated and to the insufficient level of coordination among the different agencies involved in

anti-money laundering and combating the financing of terrorism."

COMBATING PREDICATE CRIMES 117

 In Australia (population of 20 million), where every international trans¬
action must be reported, the country's financial intelligence unit — the
Australian Transaction Report and Analysis Center (AUSTRAC) — reported
7 million such transactions in 2001, among them 1.7 million deemed by
AUSTRAC as "significant," a broader notion of "suspicious" than that used
in the United States. Actual SARs in 2001 numbered only 7,000.
Data on money-laundering convictions and sentences are available for
only a few countries. Petrus van Duyne (2003) reports that over 1993-2000
the Dutch police identified 159 criminal cases involving more than $500,000
(about 1 million guilders at the time), an average of about 20 cases annu¬
ally; the population of the Netherlands is about 15 million.10 Arie Freiberg
and Richard Fox (2000) report that Australia's money laundering control
efforts generate only about 20 convictions on such charges annually, and a
German criminologist, Michael Kilchling (personal communication),
reports a similarly small figure for Germany.
In summary, it appears that no other nation prosecutes money-laundering
offenses as aggressively as the United States, which is a recurrent com¬
plaint of US officials involved in international money-laundering matters.
Even with the creation of systems that generate large numbers of reports,
there is little evidence of substantial criminal investigations that are con¬
sistently pursuing substantial cases against violators. The Netherlands,
which has relatively sophisticated capability in criminal intelligence and
investigation of organized crime, may be an exception, but even there the
numbers for major cases remain small compared with the United States.
What might explain this apparent difference in anti-money laundering
efforts between the United States and other wealthy nations with sophisti¬
cated financial and judicial systems? First, drug trafficking, central to the
creation of the AML regime, has been a more significant problem in the
United States than in any other industrial nation. Second, the United States
launched a successful prosecutory campaign against the Mafia in the 1970s
and 1980s that developed many of the legal tools and much of the organi¬
zational expertise for money-laundering prosecutions. Only in a very few
other nations (notably Italy) has organized crime prosecution been promi¬
nent. Finally, the United States has a more aggressive law enforcement cul¬
ture generally than do most other nations.
These differences between the United States and the rest of the world
could be either exacerbated or reduced by the new concern with terrorist
financing. While Western European nations have been generally responsive
to the new terrorist threat and have collaborated with US agencies, the issue
remains more prominent in the United States, which arguably has suffered
from terrorism more directly than most other nations, with the most recent
tragic exception of Spain in March 2004. However, a 2002 Eurobarometer

10. A comparable number of cases in the United States on a per capita basis would be 400,
rather than the actual figure of about 2,000.

118 CHASING DIRTY MONEY

poll suggested similarly high levels of concern about terrorism in the Euro¬
pean Union and the United States, with 82 percent of Western Europeans
fearing a terrorism incident (European Commission 2003, table 1.13).

Outcome Measures and Analytic Frameworks

The intermediate performance measures mentioned so far in this chapter do

not provide a basis for assessing the effectiveness of the AML regime.
Considering the many ways and means to launder the proceeds of crime in
the United States and globally, it can be reasonably assumed that no AML
regime, consistent with smooth operation and low costs for the economy
and financial system, can make laundering impossible. Zero tolerance may
be a strategy, but the elimination of money laundering is not feasible, and
subtler outcome measures are needed. This section deals with the problem
of finding such measures of the effectiveness of the AML in reducing crimes
other than terrorism and bribery/kleptocracy, since the bulk of AML activ¬
ities have been devoted to such criminal activities as drugs, other illegal
markets, and white-collar crimes.
Two broad mechanisms of the AML regime help reduce crime. Lirst, cus¬
tomer due diligence and other elements of the prevention pillar can make
it more difficult for offenders either to carry out the crime (for example, pay
their suppliers) or obtain its full benefits. In this respect, the controls have
a prospective effect. Second, suspicious activity reports and other back-end
activities (such as those associated with reporting) can generate evidence
of a crime, and SARs can also become evidence for investigations that have
other origins. In this sense, SARs act retrospectively in that they not only
increase the risk of criminal sanction but also provide the basis for seizure
of criminal proceeds.
In terms of crime control, the AML regime may generate two other ben¬
efits. Lirst, it produces a form of condign punishment. Seizure of funds and
incarceration of those who conspire to make the profits legitimate gener¬
ates revenue for the government and punishes senior offenders. In some
instances the only way to apprehend the principal offenders, who separate
themselves from the predicate offenses, is to convict them of money¬
laundering offenses associated with predicate crimes that have been com¬
mitted by others.11 Such cases show that the law with respect to a wide
range of predicate crimes applies to everyone. Second, AML systems may
improve the efficiency of law enforcement, an effect distinct from reducing

11. In a 1997 speech, then US Treasury Secretary Robert Rubin described money laundering
as "the 'Achilles heel,' as it gives us a way to attack the leaders of criminal organizations.
While the drug kingpins and other bosses of organized crime may be able to separate them¬
selves from street-level criminal activity, they cannot separate themselves from the profits of
that activity." www.usdoj.gov/ criminal/ press/ VIcount.htm.

COMBATING PREDICATE CRIMES 119

predicate crime. Even if they do not necessarily result in the government
apprehending more offenders, the existence and tools of the AML system
may permit the same number of offenders to be captured and convicted at
lower cost.

Market Model

A starting point for an assessment of the relationship between the AML

regime and the reduction in predicate crime is to view money laundering
as a market, with customers for, and suppliers of, money-laundering ser¬
vices. Specific AML interventions can then be linked to predicate crime
reductions by how they affect the money-laundering market, particularly
the price of services and, thus, the returns from crime, as illustrated by the
analysis in chapter 4 of the five US National Money Laundering Strategies.
For example, if money laundering is more difficult (expensive), then
drug dealers will face higher costs and charge higher prices for their ser¬
vices, thus reducing the consumption of drugs. Assume that prior to the
creation of an AML regime, a high-level drug dealer charged his customers
(themselves lower-level dealers) $10,000 per kilogram of cocaine and re¬
ceived $10 million annually in gross revenues. As a consequence of the bar¬
riers imposed by the AML regime, assume he now has to pay 10 percent of
the proceeds to the money launderer and hence receives only $9 million.
Assuming competition between drug dealers, which seems a reasonable
characterization of such a market, the $10 million previously just compen¬
sated the dealer for risks (legal and otherwise) and other costs. In the face
of reduced net returns, the dealer will raise prices to customers, and thus
increase the retail price of cocaine, making the reduction in cocaine con¬
sumption a function of the elasticity of demand.
While combating drugs has been particularly prominent in the develop¬
ment of the AML regime, the same logic applies to other income-generating
offenses such as fraud and gambling. By creating a probability of detection
and punishment, the AML regime makes money laundering more risky for
both customers and providers. It raises the price of the service and/ or the
costs of searching for the service (customers finding suppliers), which in
turn reduces the return from the predicate crime and thus the quantity of
these offenses.

What determines the demand for, and supply of, money-laundering ser¬
vices? The demand for money-laundering services can be thought of as a
function of the following:

■ The volume of criminal revenues. Since scrutiny of sources of criminal

earnings for low earners is limited, and because low earners do not
need to transform (launder) the money they make, it is probably only
criminal incomes of more than perhaps $50,000 annually that create a
need for concealing the source.

120 CHASING DIRTY MONEY

h The price of these particular money-laundering services relative to other meth¬
ods of obtaining the benefit of the unimpeded use of these funds. At very high
prices, potential customers may be content to keep the money in mat¬
tresses (metaphorically) or transfer it in small and inconvenient bun¬
dles of cash across borders to locations with less stringent controls.
Those offenders may be worse off than they would be with untram¬
meled access to money-laundering services, but they can continue to
function, albeit at a lower level.

■ Other costs of laundering money, such as the time it takes to find a supplier
and the risk of the search. Both are influenced by the intensity of enforce¬
ment of the AML regime. Money-laundering customers face a risk of
legal and financial penalties if they transact with unreliable suppliers.
Some potential penalties are derivative of supplier-oriented risk; the
launderer can mitigate the penalty by turning in the customer who has
committed the predicate offense. In addition, the continuing presence
of sting operations, in which the government simulates the behavior of
a launderer, poses a separate risk to the customer. That risk may be
manifested in the time it takes for customers to find a provider to
whom they assign a low probability of being a government agent.

The supply of money-laundering services is determined primarily by the

stringency of the AML regime. Absent the prevention pillar of the AML
regime and with only a rudimentary enforcement pillar associated with
tracing the proceeds of crime back to their source, the cost of laundering
those proceeds would surely be low, but not zero. Prior to the 1970 Bank
Secrecy Act, criminals could deposit the cash proceeds (no matter how
large) from their crimes in local banks with no questions asked by the bank
(placement), move them around the world (layering), and enjoy them at
their leisure (integration).12 Criminals and their banks were subject to ex
post investigation, but in the absence of authorities having seizure and con¬
fiscation powers, and once the proceeds were safely outside the jurisdic¬
tion, risks would be minimal. Meyer Lansky, a well-known associate of the
US mafia in its heyday, is only the most prominent of criminals who earned
money by finding ways to launder earnings from gambling and other ille¬
gal markets so as to reduce the risk of the money being traced to its crimi¬
nal source. He might be charged as an accomplice in the criminal act, but
the actual money laundering was not an offense.
Today, it is reasonable to assume that money launderers face no costs
other than those posed by law enforcement such as seizure of assets and
incarceration. The time involved in actually laundering funds is minimal.

12. Banks may have nonetheless preferred not to deal with criminal offenders, since a reve¬
lation that a particular bank provided services to a major drug dealer might well lead to some
social condemnation, even during a time when there were no formal rules prohibiting such
dealings.

COMBATING PREDICATE CRIMES 121

Some of the price charged for money-laundering services may reflect skill
in the methods used for hiding the origins of money, but one can assume
that such skills are in ready supply and that enough of those with the skills
can be persuaded to commit this crime for an appropriately high fee. Con¬
sequently, incarcerating, say, a few hundred launderers will not reduce the
pool of competent labor substantially, though it may raise the price that has
to be paid to obtain that labor, redistributing criminal income and induc¬
ing entry by new participants.
The effective cost of money-laundering services is not fully reflected in
the price charged. In addition to search time, the effective cost includes reli¬
ability in a number of dimensions such as integrity (delivering the speci¬
fied amount of money), timeliness (delivering the money when promised),
and scope (having a variety of methods of delivery to different locations).
Quality among suppliers may be differentiated, reflecting an institution's
or an individual's capacity to deliver services reliably, a particularly impor¬
tant consideration in the criminal world. Thus, an observed reduction in
prices may reflect a shift from higher- to lower-quality money-laundering
services.

Multiple Markets

An important analytic complexity is that there may be more than one market
for money-laundering services, depending on the predicate offense and the
amount that needs to be protected. For example, as was illustrated in chap¬
ter 3, laundering $1 million in drug revenues from the United States to
Mexico may require a higher percentage payment than laundering the same
amount in a bankruptcy fraud, simply because the launderer of drug mon¬
eys incurs risks of more serious penalties from law enforcement as well as
greater potential physical risks (violent retaliation for failure to protect
assets). If the transaction involves actual cash, it may also be more expensive
(per dollar) to launder large sums than small, for the same reasons. For non¬
cash operations, the relationship between size and cost may go the other way.
Launderers may also "specialize," either in terms of the kinds of funds
they accept or the kinds of institutions with which they transact. Money
laundering is also differentiated by phase; some launderers may not
provide full-service operations. For example, a simple currency exchange
bureau may only move money out of the United States (placement), but not
provide for the layering of the money so that it can no longer be traced, or
bring it back into the United States where the funds can be freely used with
no questions asked (integration). Black market peso brokers, on the other
hand, often serve both the supply and the demand ends of the market. They
first export the narco-dollars to Colombia (or arrange to purchase them for
resale in the United States), then exchange them for pesos for the cartels'
domestic use, and lastly provide dollars to Colombian importers who wish

122 CHASING DIRTY MONEY

to avoid the costs and bureaucracy of obtaining dollars legally. Thus, the
proceeds of drug sales in the United States may resurface, cleaned, in the
accounts of US companies that sell their products to Colombian businesses.
One segment of the market for money-laundering services may consist
of launderers employing a variety of methods for servicing customers,
depending on the latter's specific needs. In this conventional market, the
providers have multiple and independent customers and recruit agents
(for example, bank officials and casino employees) to ensure that they can
provide a range of services. Customers, who know other customers with
whom they share information, shift among launderers depending on the
price and quality of service. For example, there is anecdotal evidence that
drug dealers, circumspect though they may be, do share information about
money launderers.
Another segment of the market consists of almost accidental customer-
agent relationships. Customers do not find professional money launderers
whose principal activity is providing those services, but rather seek out
corrupt employees of financial intermediaries who service only one or a
small number of clients as a by-product of their legitimate occupations.
Search costs are high for both sides, reflecting the risks of disclosing need
or availability and the lack of any organizing focal point for the search.
For example, in its 1997-1998 Report on Money Laundering Typologies, the
Financial Action Task Force (FATF 1998) referred to a financier who allowed
a single drug trafficker to use his company to establish a source of funds.
The trafficker gave cash to the financial officer to deposit in a company
account, and the funds were then transferred to Monaco for the ostensible

purpose of buying Goya paintings. The paintings were fakes and, more¬
over, were never shipped, and the drug trafficker was the beneficiary of the
payments for the fake Goyas, receiving the million-dollar transfers. There
was no evidence that the financier's business provided money-laundering
services for other clients.

Both types of operations may be components of the market. Some cus¬

tomers in high-risk occupations (for example, cigarette smugglers) with
continuing needs in a number of locations may seek out specialized laun¬
derers. Others who have a one-time need may be content to find an acquain¬
tance capable and willing to provide the service to just one customer. There
is also self-laundering, for example through acquisition of a small business
that is on the "exempt" list at a bank (i.e., for which it is not necessary to file
a currency transaction report for large cash transactions that are consistent
with the regular pattern of the business). No one at the bank or any pro¬
fessional except perhaps a forgiving accountant needs to be involved, at
least in the placement stage. The Egmont Group (2000), which provides
accounts of investigations by financial intelligence units worldwide, tells of
a Western European family of drug traffickers that ran a currency exchange
bureau that served as both a cash-intensive front company and a means of
laundering money.

COMBATING PREDICATE CRIMES 123

 As discussed in chapter 3, although professional money launderers cer¬
tainly exist, they are surprisingly rare in reported cases. A great deal of the
revenues from crimes is self-laundered. This is important for both policy
and research purposes. The rationale behind the current AML regime is
based in part on the implicit assumption that the regime provides tools to
apprehend and punish a set of actors who provided a critical service for the
commission of certain kinds of crime and who had previously been beyond
the reach of the law — an assumption that makes the market model a use¬
ful heuristic device for analyzing the effects of laws and programs.
However, if money laundering is mostly done by predicate offenders or
by nonspecialized confederates, then the regime accomplishes much less.
There is no new set of offenders, just a new set of charges against the same
offenders, and the potential gains from the additional tool represented by
the AML regime, while valuable in increasing the efficiency of law enforce¬
ment, are likely to be substantially more modest than posited (Cuellar 2003).
For research purposes, this implies that the market-model concept is a
strained analogy. Price may not be well defined to most participants be¬
cause the service is rarely purchased. Risk may also be hard to observe
because it is derivative from participation in other elements of the crime.
Assessing how interventions increase risks and prices for those transac¬
tions that do involve stand-alone launderers will have only modest value.
Finally, stand-alone service providers may be scarce.
Thus, the market model may work well for some kinds of predicate
offenses and offenders, but less well for others. How this element of hetero¬
geneity in the money-laundering underworld affects the research agenda
for improving AML regimes will be taken up in chapter 8.

Performance Measures

To what extent does the market framework help assess the effectiveness of
an AML regime in reducing predicate crime? While useful for analyzing
some of the basic questions, the available data do not permit application for
assessing effectiveness. One source of difficulty is that the price of money¬
laundering services itself is not an adequate indicator. Enforcement, as
noted above, aims at both the demand and supply sides. Demand-side
efforts such as stings against customers have the effect of lowering observed
prices. By raising the nonmoney cost of purchasing money laundering,
which includes some risk of arrest, incarceration, and financial penalties,
such stings reduce demand. Supply-side efforts directed at the launderers
should raise the price. Both efforts should reduce the quantity of launder¬
ing and the net returns from crime, the ultimate goal, but price can then only
be interpreted along with estimates of quantity.
Quantity estimates, however, are not available, as there are no system¬
atic estimates of amounts laundered, either through US institutions or

124 CHASING DIRTY MONEY

globally (chapter 2). The US Congress has pressed FinCEN to develop esti¬
mates with a documented analytical base, but no such estimates have been
produced as yet. Nor is it likely that such estimates can be developed in the
foreseeable future, given the lack of even the beginnings of a data collec¬
tion apparatus to support this activity.
An alternative performance measure for assessing the effectiveness of
the AML regime in controlling predicate crime is the volume of predicate
offenses. Apart from the problem of developing a counterfactual — How
much predicate crime would have occurred without the AML regime? —
there is a fundamental problem of measuring predicate offense levels.
Consider again illegal drugs, the best studied of the activities generating
a demand for money laundering. While there are a number of possible
measures — total revenues, prices, or quantities consumed— none is precise
enough to be useful for analytical purposes. For example, the error band
around existing drug revenue estimates for the United States is very large,
with an official estimate of $50 billion in 1998 (Office of National Drug
Control Policy 2000) that should be viewed as the center point of a uniform
distribution from $25 billion to $75 billion. A decline of 25 percent in a five-
year period would be hard to detect with confidence.
Alternatively, one might use drug prices, since the mechanism by which
money-laundering controls are expected to reduce drug use is by raising
the cost of distribution rather than by reducing demand. A recent National
Research Council report (Manski, Pepper, and Petrie 2001) expressed con¬
siderable skepticism that the current system of data collection for prices
could detect any but the very largest changes in prices. Moreover, there
have been large revisions in the estimated prices for cocaine and heroin for
a given year in successive estimation efforts, reinforcing the sense of frailty
in those estimates.
A third potential measure is the number of dependent drug users. The
AML regime is intended to support drug policy efforts to reduce this num¬
ber. Estimates of the number of dependent users are also imprecise, though
less so than estimates of quantities or prices. The official estimates (which
are subject to frequent and substantial revisions) do not provide error-
bands, but any realistic assessment of their precision would suggest that a
change of 5 percent would be difficult to detect.
Moreover, it is unlikely that the AML regime itself could have very large
effects on the extent of drug use. Low-level drug dealers earn too little to
require money-laundering services, yet they account for the bulk of total
earnings. Price markups along the distribution system (conservatively
estimated at 50 percent at each level) show that more than 60 percent of
revenues go to low-level wholesalers and retailers, who are predominantly
independent agents rather than employees of larger organizations. At the
peak of the crack cocaine market in 1988, average annual earnings of
retail drug dealers in Washington, DC, were estimated at $28,000 (Reuter,

COMBATING PREDICATE CRIMES 125

MacCoun, and Murphy 1990). More recent studies report much lower
earnings (e.g., Bourgois 1995, Levitt and Venkatesh 2000).
High-level dealers, the only ones who need money-laundering services,
account for no more than 25 percent of total drug revenues. Assume that
in the current regime money launderers charge customers approximately
10 percent of the amount laundered. Now assume that an improved system
raised the price for money-laundering services by half, to 15 percent. The
result would be an increase in the price of drugs of only 1.25 percent, far too
small to be picked up by existing monitoring systems. This is not an argu¬
ment that money-laundering controls are neither effective nor cost effective,
but only that their success cannot be empirically assessed this way.
The performance assessment situation is even less promising for crimi¬
nal offenses other than drug distribution. Systematic estimates of the vol¬
ume of these crimes or revenues are often not available or are effectively
made up from thin air. It is implausible, for example, that one could reli¬
ably detect a reduction of 10 percent in the volume of (or revenues from)
embezzlement or corporate fraud.
Performance measurement is an increasingly important component of
responsive and responsible public policy, so the difficulty in finding cred¬
ible measures of AML regime performance in reducing crime is a major
problem. Perhaps more sophisticated versions of market models will help
in this respect, but their utility has not been established.

Improving Criminal Justice System Performance

This chapter noted earlier that AML regimes might have two other bene¬
fits in addition to controlling crime: improving the efficiency of the system
or catching offenders who otherwise would escape. Mariano-Florentino
Cuellar (2003) concedes that such regimes might have improved efficiency
in drug control and in reducing a few related criminal activities, but argues
that they have failed in the second area. The US AML regime principally has
been used to increase the penalties with which prosecutors can threaten
predicate offenders. The regime has had little success in apprehending pro¬
fessional money launderers or high-level criminals.
The paucity of cases against stand-alone launderers and investigations
that have their origin in money-laundering information supports the criti¬
cism that the AML regime has brought in few new offenders. There are no
systematic data on the origins of cases against major criminals such as prin¬
cipal drug dealers, so it is impossible to tell whether more of them are being
captured through money-laundering laws and investigations.

How Risky Is Money Laundering?

However crude, an estimate of how risky money laundering in the United

States has become as a result of the AML regime is helpful in assessing

126 CHASING DIRTY MONEY

regime performance. About 2,000 people are convicted of money-laundering
offenses (primary or otherwise) each year in the United States. For the
moment, assume that all of those convicted are providers of, rather than
customers for, the service. This assumption imparts an upward bias to our
risk estimate, since we know that some of those convicted are not stand¬
alone providers of money-laundering services.
To estimate risk, a figure for the total number of persons who launder
money is also needed. No such estimate is available, so an indicative cal¬
culation is all that can be offered. Assume total US money laundered annu¬
ally is near the low end of conventional estimates, say $300 billion. Only
20 percent of those convicted are reportedly involved with laundering
more than $1 million, but that is the amount involved in the specific trans¬
actions detected, not an annual flow. If an average money launderer handles
$10 million per annum (which might generate a gross income of $500,000 to
$1 million), then there would be 30,000 money launderers, and the proba¬
bility of conviction would be about 6.7 percent (2,000/30,000). For com¬
parison, there are estimates available that the probability of incarceration
for selling cocaine in the late 1980s was approximately 25 to 30 percent
(Reuter, MacCoun, and Murphy 1990). Though dated, these are the only
such estimates for an illegal market.
This exercise is highly speculative; there are other assumptions that might
generate a higher estimate of risk without overly straining plausibility. For
example, in addition to those who were convicted of money laundering,
there may be substantially more individuals for whom those charges were
dropped in exchange for information about the predicate offense or for
pleas to some other involvement in the predicate offense, even though the
individual's principal role was money laundering. If half of those who
were caught laundering money were convicted only on other charges, then
the risk figure might rise to 13.3 percent. For money launderers with valu¬
able legitimate labor market skills, that risk might generate a very high
premium for their services. Compared with drug dealers with little educa¬
tion, such professionals require a higher incentive to risk the same amount
of prison time.
These assumptions generously favor a finding that supports the effec¬
tiveness of the AML regime. Most of those convicted of money launder¬
ing are also convicted of other offenses, and many are probably
customers rather than providers. The assumption that each money laun¬
derer handles an average of $10 million per annum imparts a similar
upward bias; actual cases point to launderers with much lower volume.
That is certainly the case if many of them work for only one client. It is
quite plausible that even now, with an elaborate regime in place, money
launderers face a less than one in twenty probability of incarceration in
the United States. The financial penalties collected by the federal gov¬
ernment represent the most modest of taxes, even assuming low-end esti¬
mates of money laundering.

COMBATING PREDICATE CRIMES 127

 However, it must be reemphasized that this is not a complete assessment
of the effectiveness of the AML regime. The figures employed cover only
those individuals who were themselves involved directly in the money¬
laundering transaction. It may well be that SARs and other elements of the
regime generate useful evidence against larger numbers of drug dealers,
but that the final indictments and convictions are for the predicate crime
alone. The lack of information on this possibility is a major omission in the
current system of data collection.

Conclusions

For both conceptual and empirical reasons, it is impossible to assess directly

how much the current AML regime has reduced the volume of white-collar
crime, drug dealing, and other illicit market activity. Little information is
available about the prices charged by money launderers, and that price itself
is a poor representation of the total client cost of money laundering. Even
without having found enough data in the public domain to judge whether
prices have risen for particular types of transactions, it would appear that
money laundering is not a particularly risky business, given the record of
federal convictions. Nor does it seem plausible that a more effective anti¬
money laundering regime would increase costs to criminal offenders, even
drug dealers, enough to be observable with current data series.
The most useful assessment would be an index of the difficulty of laun¬
dering money, with difficulty measured as a combination of cost, risk, and
inconvenience. Such an assessment would require the systematic collection
of data on prices using everything from undercover operations to debrief¬
ing those arrested for purchasing or providing services. Such a data col¬
lection effort is part of the research agenda put forth in chapter 8.

128 CHASING DIRTY MONEY

 6
Protecting Financial System Integrity

An oft-stated goal of anti-money laundering (AML) regimes is to protect

the integrity of the financial system, particularly banks, which are at the
core of that system. Other types of financial and nonfinancial institutions
such as money service providers and real estate agents, which also are gen¬
erally covered by the AML regime, are important to financial system
development but less so for the day-to-day functioning of the economy.
Banking services are central to the smooth functioning of a market economy .
The role of AML regimes in supporting efforts to protect and facilitate
functional financial systems is well established. For example, the 2003-08
Strategic Plan of the US Treasury's Department's Financial Crime Enforce¬
ment Network (FinCEN 2003a, 8) links its AML activities directly to achiev¬
ing one of Treasury's key objectives, which is to "preserve the integrity
of financial systems." In Title III (the International Money Laundering
Abatement and Anti-Terrorist Financing Act Sec. 302 (a) (3)) of the USA
PATRIOT Act of 2001, the US Congress reported: "Money launderers sub¬
vert legitimate financial mechanisms and banking relationships by using
them as protective covering for the movement of criminal proceeds and the
financing of crime and terrorism, and, by so doing, can threaten the safety
of United States citizens and undermine the integrity of United States
financial institutions and of the global financial and trading system upon

which prosperity and growth depend."

Nor are AML regimes important only to the US financial system. William
Gilmore (1999, 83) quotes the Australian president of the Financial Action
Task Force (FATF), Tom Sherman, as stating in 1992: "Combating money
laundering is not just a matter of fighting crime but of preserving the integrity
of financial institutions and ultimately the financial system as a whole." 129
 This chapter examines the nature of the AML goal to protect financial sys¬
tem integrity and presents a framework for examining how well the regime
is accomplishing that goal. Looking at the limited available evidence through
the lens of that framework leads to the tentative conclusion that the AML
regime appears to have made substantial progress in protecting the financial
system, but we offer several qualifications regarding that progress.

Financial System Integrity as an AML Goal

Much of the initial focus of the global AML regime was on the core finan¬
cial system and particularly banks, since the banking system plays a cen¬
tral role in the collection and movement of funds. While the principal
objectives were to make it more difficult (expensive) for criminal offenders
to launder the proceeds of their crimes, and to employ the financial system
in the investigation and prosecution of those crimes, an important sub¬
sidiary objective has been to protect the integrity of the financial system
itself. Thus, for example, in a report on customer due diligence for banks,
the Basel Committee on Banking Supervision (2001, 2) noted that "know-
your-customer" policies and procedures "are critical to protecting the
safety and soundness of banks and the integrity of banking systems."
Society today disapproves of turning drug revenues into legitimate
funds. Financial institutions that accept money from drug dealers, even if
they do not face criminal charges, are perceived to be less than law-abiding.
The role of mainstream financial institutions, particularly banks, is to pro¬
vide public goods such as banking services and liquidity to the financial sys¬
tem and the economy, and they are expected to share and abide by generally
accepted social and ethical codes of behavior. Further, since core financial
institutions to varying degrees are regarded as quasi-public utilities with
access to such government safety nets as deposit insurance, access to the dis¬
count window, and payment services, they have now been called upon to
assist in supplying another public good, which is the prevention of money
laundering. In the process, they help protect the integrity of the core finan¬
cial system as a whole.
In this context, once the social objective to combat money laundering has
become well established, a bank's reputation suffers if it becomes associ¬
ated in the public mind with that crime, though the seriousness of that
decline in reputation may vary from society to society. In Switzerland, for
example, it can be very serious, since asset management accounts for as
much as half of Swiss banks' output, and private clients generate as much
as 85 percent of this business (Pieth and Aiolfi 2003, 20). The Swiss bankers'
association contracts for polls on the subject, and a 2004 survey of public atti¬
tudes toward Swiss banks found that 80 percent of respondents believed
that their banking institutions enjoy a positive reputation abroad (M.I.S.
Trend 2004). At the same time, 57 percent felt that Switzerland is not doing

130 CHASING DIRTY MONEY

enough to fight money laundering, but 76 percent want to maintain bank
secrecy. This nicely illustrates the tensions and cross-currents around this
subject. In other countries, moreover, reputation may not be as valuable an
asset, so banks may be less sensitive to being perceived as committing a
violation related to money laundering. It can be said, however, that if banks
and other financial institutions at the center of the financial system are to
play their assigned role in the economy, their customers should at the very
least trust them. Such confidence helps to prevent runs on banks, which
can undermine the stability not only of the financial system but also of the
economy as a whole.
Finally, given the AML regime, a financial institution that fails to estab¬
lish appropriate AML-compliance procedures incurs legal and financial
liability that can impact its bottom line as well as its reputation. Money¬
laundering regulations can be viewed as a way to insulate banks and sim¬
ilar institutions from direct connections with illegal activities. To this end,
supervisors of banks and other core financial institutions have over time
implemented a structure to induce institutions to take their AML respon¬
sibilities seriously. Supervisors have linked compliance to sound risk
management, which is central to minimizing the costs of financial inter¬
mediation to the institution and the system as a whole. They have embraced
a proactive supervisory posture in support of compliance with the overall
regime.
Senator John Kerry (D-Mass.), drawing on his experience as chairman and
ranking member of the US Senate Subcommittee on Terrorism, Narcotics,
and Operations, has described the interaction of organized crime, corrup¬
tion, money laundering, and a weakened financial system in Russia (Kerry
1997, 164-65). Criminal groups intent on hiding the proceeds of their crimes
gain control of banks, which allows them to corrupt the business and finan¬
cial system more broadly, which in turn taints the government and other
institutions of society. Ultimately the phenomenon spreads and mixes with
money laundered from crimes in other countries.
At the extreme, as argued in a report by the UK Performance and

Innovation Unit (2000, chapter 3, 7), "the accumulation of criminal assets

in a coun try's financial system can influence decisions about national bank¬
ing policies or about co-operation in international investigations, trans¬
parency and accountability rules." The case of the Bank of Credit and
Commerce International (BCCI) illustrates this extreme. BCCI caused sub¬
stantial disruption to the international financial system when its nefarious
activities were finally uncovered in 1991. The institution was found guilty
of numerous violations of the banking laws in a number of countries. BCCI
was a thoroughly corrupt institution that operated outside the laws and
regulations of a large number of countries, and its collapse resulted in
losses for a range of creditors, mostly individuals but some governments
as well. BCCI had been viewed by many of its customers as a sound insti¬
tution, and its operations supposedly had been examined — incompletely

PROTECTING FINANCIAL SYSTEM INTEGRITY 131

as it turned out — by supervisory authorities in major nations. Thus, the
AML regime has to be judged on the basis of the extent to which it protects
the integrity of international as well as national financial systems.

Evaluating Progress

How can the effectiveness of the AML regime in protecting the integrity of
national or international financial systems be measured? Indirect indicators,
if available and carefully interpreted, should provide a reasonable picture.
Examining actual money-laundering prosecutions can provide evidence on
general use of the financial system for money-laundering purposes (partic¬
ularly in the placement phase) and the nature of that use. It should be pos¬
sible to distinguish among institutions that are corrupt and actively solicit
money-laundering business, those that have willing or rogue employees
who provide such services on an ad hoc and noninstitutionalized basis, and
those that are unwitting accomplices in money-laundering operations.
Within this third category of unwitting accomplices, it also is important to
distinguish institutions that have deficient internal AML controls that may
contribute to facilitating money laundering.
With respect to the integrity of major national financial systems and the
global financial system, as a first approximation the test of success of the
AML regime should be whether institutions have been linked to either of
the first two money-laundering categories in terms of laundering the pro¬
ceeds of crimes committed in their home countries. A distinction must be

made between a bank's internal systems and business activities that aid
and abet money laundering and therefore can reasonably be associated
with weakening financial system integrity (a relatively low hurdle), and on
those internal systems and business activities that fail to stop money laun¬
dering and deter the underlying crime (a much higher hurdle).
In addition, scrutiny of suspicious activity reports (SARs) submitted by
and about institutions should aid in identifying each institution's potential
vulnerability, though there are biases and defects in such measures that
might warrant controls for other factors such as size, location, and clientele.
If institutions that show up frequently in prosecutions file small numbers of
SARs, that information might indicate that they are failing to meet their
responsibilities. However, extreme threats to the integrity of a country's
financial system, which might create financial instability on the scale of
BCCI, are difficult to detect or measure systematically other than after the
fact. BCCI was an example of a fraud and conspiracy conducted from within
the institution itself. Had the crimes been detected at an earlier date, the
implications of the BCCI collapse would have been essentially the same.
These types of data are not generally or systematically available to the
public in the United States or any other country, which constrains the abil¬
ity of researchers to conduct either type of analysis. For this study, we

132 CHASING DIRTY MONEY

developed two databases of cases. We drew upon those databases to sug¬
gest how the analysis of money-laundering cases might be used to produce
a fuller assessment of the progress of the AML regime in protecting core
financial system integrity.
The first database included international cases reported in seven annual
FATF reports on money-laundering typologies and in occasional reports
by the Egmont Group, an international association of financial intelligence
units (FIUs). Of 223 cases entered in the database, 185 involved core finan¬
cial institutions, exclusively banks, and excluding such entities as casinos
and insurance companies. Because these exercises were designed to iden¬
tify and share information about new methods of laundering money, they
are not necessarily a representative cross-section of money-laundering
prosecutions that in principle would best serve this type of analysis. Given
that qualification, 3 percent of the cases involved active solicitation by the
bank, and 3 percent involved the activities of rogue employees. In the
remaining 94 percent of cases, the banks were unwitting accomplices,
although it was not possible to determine whether or the extent to which
their internal AML controls (or lack of them) could be blamed.1
The second database included 60 cases of money laundering assembled
from descriptions in various sources, such as books, newspaper articles,
and the National Money Laundering Strategies.2 Because cases reported by
the media, in particular, are likely to be sensational, the bias probably is in
the direction of direct involvement by banking institutions or their employ¬
ees that were judged to be culpable. We were able to identify the role of
banks in 55 of these cases.3 Six involved active solicitation, four involved
rogue employees, and six fell under the heading of negligence with respect
to AML controls. However, in the remaining cases, which constituted 71 per¬
cent of the total, the banks were unwitting accomplices.4
Fourteen of the cases involved large banks, and of these, one case in¬
volved active solicitation, two involved rogue employees, four involved
negligence on the part of the banks, and seven involved banks as unwitting
accomplices to money laundering. With respect to the predicate crime, six
of the 14 were drug cases, all in the first half of the 1990s, and three were

1. Interestingly, 59 percent of the cases involved more than one bank as part of the money¬
laundering operation.

2. This sample is neither exhaustive nor necessarily random. Had we deliberately set out to
find cases in which large banks had been involved in money laundering, there certainly
would have been a different distribution of the nature of bank involvement in cases and types
of cases.

3. In 75 percent of the cases, the underlying (predicate) crime was committed at least in part
in the United States, while in 62 percent of cases the initial placement of proceeds of the crime
was in the United States.

4. Fairly consistent with the findings using the first database, 62 percent of the cases in this
second database involved more than one bank as part of the money-laundering case.

PROTECTING FINANCIAL SYSTEM INTEGRITY 133

corruption or kleptocracy cases, while real estate fraud and gambling
accounted for one case each. The three remaining cases involved bank vio¬
lations of money-laundering (Bank Secrecy Act) rules and regulations, one
(Bank of New York) was associated with a rogue employee, and two were
the result of negligence.
The anecdotal information provided through the second database of
cases suggests that in recent years only a small number of US institutions
have been involved in active solicitation of money-laundering business.
Even in cases where a financial institution is found to have actively solicited
money-laundering business, however, it is not necessarily closed down.
Such was the case of the Broadway National Bank in New York City, as
was described in box 5.1. In these situations, the interests of law enforce¬
ment authorities sometimes differ from those of the supervisory authori¬
ties, in that the former may value the deterrent effect of closing down an
institution while the latter are concerned about financial stability and not
provoking a costly run on the bank. A compromise is often found by re¬
moving or charging officers of the institution and imposing a large fine,
while at the same time rehabilitating the institution with new owners and
management under the watchful eyes of the supervisors.
In a case like that of Great Eastern Bank of Florida, also described in box
5.1, the bank received a substantial fine for what might be called gross neg¬
ligence, but its offense was not quite in the category of active solicitation of
money-laundering business.
Similarly, the number of large US institutions with willing employees
who facilitate money laundering from within appears to have been small
in recent years. The most prominent recent case involved the Bank of New
York, through which $7 billion was laundered by way of an account be¬
longing to Benex, a company linked with a purported Russian mobster.
One of the bank's vice presidents pleaded guilty to money laundering, but
the bank itself was not accused of wrongdoing (New York T imes, September 5,
1999, 3-1). It was, however, subjected to extensive and expensive super¬
visory guidance to improve its compliance with AML laws and regula¬
tions. This type of case as well as some in the third category — unwittingly
assisting money laundering in a context in which the institution's deficient
AML compliance regime contributed to the problem — now routinely result
in financial penalties and costly and ongoing supervisory scrutiny until the
institution has demonstrated improvements in its compliance procedures.
It appears possible that JP Morgan Chase will be subjected to similar for¬
mal sanctions in the wake of the criminal conviction of Beacon Hill Service
Corporation as an unlicensed money transmitter that used JP Morgan
Chase to make $6.5 billion in wire transfers over a six-year period
(Morgenthau 2004). We assume the institution has already been subjected
to informal sanctions as part of its annual supervisory examinations.
Supervisors also review SARs submitted by institutions in order to iden¬
tify an institution's potential vulnerability to rogue employees or weak AML
134 CHASING DIRTY MONEY
compliance systems. Along with regular reviews of an appropriate sam¬
ple of money-laundering prosecutions, supervisors could support efforts to
measure regime effectiveness, as well as make the regime more transparent,
by issuing a periodic public scorecard of AML regime progress in protecting
the integrity of the financial system.5
An illustration of some of these points came in the spring of 2004, when
the attention of the local Washington, DC, community focused on various
money-laundering allegations involving Riggs National Bank, a financial
institution that dates back 165 years and has served a number of presidents.6
What is known is that Riggs, a relatively small regional bank with about
$4.2 billion in deposits, catered extensively to deposits from the diplomatic
community, which reportedly accounted for almost 25 percent of its total.
An estimated 95 percent of Washington embassies had Riggs accounts.
Among other transgressions, the bank allegedly failed to submit required
reports involving large cash transactions by Saudi Arabian accounts. A
senior vice president was fired and is under criminal investigation for
participation in a possible money-laundering operation involving corrup¬
tion and the president of Equatorial Guinea. The bank is also suspected of
collaborating with former Chilean President Augusto Pinochet to hide
some of his wealth that may have had questionable origins. A $25 million
fine was imposed on the bank, and it was put under close management
scrutiny by the Office of the Comptroller of the Currency and the Federal
Reserve System, the severest penalty short of closing the institution. In
effect, the bank's reputation was ruined, and its owners were forced by a
combination of market forces and supervisory encouragement to put it

up for sale.7
The Riggs case illustrates three aspects of money laundering as it relates
to efforts to protect the integrity of the core financial system. First, Riggs
demonstrated "willful and systematic" lack of compliance with mandated
internal controls involving money laundering, and it was this failure rather
than an actual conviction for money laundering that triggered the regula¬
tors' stern responses. Second, a bank can so tarnish its reputation via such
failure that it is no longer viable as a stand-alone institution even if it
retains substantial institutional value. Third and related, some of the prin¬
cipal money laundering-related failures involved a special category of pri¬
vate banking — providing services to embassies and their governments. The

5. Although US bank supervisors routinely use this type of information in connection with
their AML examinations of individual banks, we were not able to gain access to the data.

6. The Riggs case was not included in the anecdotal database because the full story was still
coming out as our study was completed.

7. The offer by the PNC Financial Services Group valued at about $24 per share is higher than
the Riggs stock, which reported is closely held, has traded since 1998; Riggs stock had traded
as low as $10 a share since then and only slightly below $15 a share in the year before its
announced sale.

PROTECTING FINANCIAL SYSTEM INTEGRITY 135

clear lesson is that this type of private banking, along with government offi¬
cials, should not receive special treatment under the AML regime.

Conclusions

Several qualified conclusions can be drawn from evaluating the progress

of the AML regime in protecting the integrity of the core financial system.
The regime, which has now been in place in major jurisdictions for more
than 15 years, has altered how banks and other core financial institutions
approach their responsibilities and conduct business. The AML regime has
induced banks to take seriously their obligation to avoid direct contact with
criminal money by putting in place reporting systems and developed mon¬
itoring techniques that make them less attractive for money laundering.
The emerging global AML regime makes it more difficult to use banks and
mainstream financial institutions to place funds, although there is little
evidence that these efforts have made money laundering substantially
more expensive. Nevertheless, the global regime appears to have largely
achieved one of its primary goals, which is to eliminate the threat from
money laundering to the integrity of banking systems, at least for large
institutions in the major jurisdictions.
This is not to say that the threat has been permanently eliminated —
witness the Beacon Hill case referred to earlier — or that more cannot or
should not be done to promote further global progress. Financial institu¬
tions that have been unwitting accomplices to money laundering could
have had tighter AML controls in place. Those controls might have been
helpful in detecting the underlying crime or, more likely, might have
prompted the criminals to take their business elsewhere.
The assessment in this chapter should be qualified in four important
respects. First, the principal concern in connection with money laundering
and the integrity of core national financial systems and the global system has
to do with the placement stage of the three-stage laundering process. It is at
that stage that the financial institution, and by extension its financial system,
is most vulnerable to corruption and the loss of reputation. Preventing the
involvement of core financial institutions in money laundering at the layer¬
ing or integration stages is a different and more difficult matter.
The global AML regime has recently been extended to correspondent
banking relationships, which often are an integral aspect of the layering
and integration stages of money laundering. The USA PATRIOT Act con¬
tained some tough and groundbreaking provisions to address the corre¬
spondent banking aspect of the AML regime. In addition, recommendation
7 of the 2003 FATF Forty Recommendations incorporates new provisions
with regard to cross-border correspondent banking and similar relation¬
ships. Detection of money laundering in the layering and integration stages
of the anti-money laundering process requires close cooperation between

136 CHASING DIRTY MONEY

the AML regime's prevention and enforcement pillars. Extensive use of a
financial institution in a major jurisdiction in these later stages could
undermine the institution's stability, integrity, and reputation, in turn
undermining customer confidence in the financial system.
Second, financial institutions could be used more extensively and effec¬
tively in connection with the investigative element of the AML regime's
enforcement pillar. One example is in sting operations, and another is in
more effective use of SARs and similar reports. As noted in chapter 4, the
US private sector has been quite critical of the government's tendency to
operate as a one-way street on money laundering with respect to the pro¬
vision of information. Similar complaints can be found in the KPMG (2003)
report on the reporting regime for SARs in the United Kingdom. One
sometimes hears complaints that banks and their supervisors resist coop¬
erating with law enforcement authorities.8
Third, the jury is still out with respect to money laundering in connection
with the "private banking" activities of major international banks. There
have been major problems in the recent past, such as the difficulties that
occurred in Mexico as a result of the United States' Operation Casablanca
in 1998 (chapter 4). Such activities often involve the proceeds of crimes —
such as corruption — committed outside the home countries of major finan¬
cial institutions, and therefore only indirectly affect the integrity of those
countries' financial systems. Nevertheless, in response to these and other
concerns related to public confidence in banking systems, 12 major private
international financial institutions adopted Global Anti-Money Laundering
Guidelines for Private Banking in October 2000 (Wolfsberg Group 2002). 9 In
addition, recommendation 6 of the 2003 FATF Forty Recommendations
calls for enhanced due diligence with respect to "politically exposed per¬
sons" whose source of wealth may not be legitimate.
Finally, the tentative assessment that the global AML regime is generally
effective in protecting the integrity of core financial systems of major coun¬
tries leaves to one side the issue of institutions at least notionally head¬
quartered outside major financial centers, as well as their capacity to abuse
and undermine the integrity of the global financial system. Again, these

8. Martin Mayer wrote in the New York Times (January 14, 2004) that the Federal Reserve
Board would not allow regional banks to reveal the identity of purchasers of large blocks of
US currency. He did a disservice to the public and the AML regime by putting forward his
unsupported accusation. The Federal Reserve routinely shares such information with law
enforcement agencies.

9. The Wolfsberg Group revised its guidelines for private banking in 2002 and also adopted
principles on the suppression of this financing of terrorism and AML principles for corre¬
spondent banking. In 2003, the group issued a statement on procedures for monitoring,
screening, and searching, which carries CDD beyond initial customer contacts and establish¬
ment of an account relationship. The procedures can be used for investigative purposes and
have the potential to help identify layering operations.

PROTECTING FINANCIAL SYSTEM INTEGRITY 137

institutions and jurisdictions pose less of a direct threat to the financial sys¬
tems of the major countries, but the risks are not negligible. This concern,
as well as those regarding the proceeds of corruption often associated with
private banking, is more relevant to the role of the AML regime in target¬
ing the global "public bads" of terrorism, kleptocracy, and failed states,
which will be discussed in chapter 7.
While the global AML regime appears to have successfully protected the
integrity of banks via core financial institutions headquartered in the major
financial centers, the regime may not have reduced the total volume of
criminal proceeds laundered globally; the activity has perhaps been
pushed into more peripheral institutions and jurisdictions. As a result, the
AML regime has been expanded to pursue laundering taking place
through those institutions and locations. The net effect may have been pos¬
itive with respect to protecting the integrity of the core financial system but
minimal with respect to achieving other AML goals.

138 CHASING DIRTY MONEY

 7
Combating Global “Public Bads”

Three types of objectives for national and global anti-money laundering

(AML) regimes have been identified: (1) the central goal of reducing a
range of predicate crimes (chapter 5); (2) the more narrowly focused goal
of protecting the integrity of the core financial system (chapter 6); and (3) a
group of national security or systemic goals under the heading of global
"public bads" — terrorism financing, corruption and kleptocracy, and failed
states. In examining the progress of the AML regime in achieving this final
group of goals, this chapter will consider each of the "bads." The global
AML regime can and does play a constructive role in limiting each of these
global problems, but the regime is just one tool, the development and use
of which poses many challenges given the international context in which
these goals must be pursued.

Terrorism

The first complication in assessing AML regime effectiveness in combating

terrorism financing is that resources for such activities are laundered
before, as opposed to after, the crime. The traditional AML focus is on the
placement stage, when funds enter the financial system, but taking action
when funds are already in the hands of terrorist organizations can be too
late to stop the intended terrorist act.
Thus, the role of the AML regime with respect to terrorism financing is
principally prevention in the full defensive sense of the word. Financial
trails may be helpful after the terrorist act has occurred in terms of identi¬
fying and apprehending supporters of terrorism, but the goal is to stop

139
those acts before they happen. The anti-money laundering enforcement
pillar serves at best on the margin as a deterrent to terrorism, perhaps by
making potential terrorists recognize the difficulty in acquiring funds.
However, enforcement is more relevant to the actual blocking, seizure, and
confiscation of resources financing terrorism. As detailed by the GAO
(2003b), and R. E. Bell (2003) in the UK context, terrorist organizations use
a range of alternative or nontraditional means to earn, move, and store
their assets. Consequently, the full panoply of tools of the AML regime
must be used to combat terrorism financing. Bell also emphasizes that pri¬
vacy and human rights considerations are more complex than with respect
to conventional anti-money laundering because some sources of financing
are legitimate even if the uses are not.
The Eight Special Recommendations on Terrorist Financing adopted by
the Financial Action Task Force (FATF) in October 2001 (FATF 2001a)
draw a parallel between other forms of money laundering and the financ¬
ing of terrorism. Hastily assembled following the September 11, 2001, ter¬
rorist attack on the United States and based on the model of the original
FATF Forty Recommendations, the Eight Special Recommendations em¬
phasize issues that are more relevant to combating the financing of
terrorism — such as the role of nonprofit entities and informal funds trans¬
fer mechanisms- — than conventional money laundering. In general, the
Eight Special Recommendations reflect the fact that the AML regime as it
applies to terrorism financing is analogous to its application to other
money-laundering activities, but with some important differences in con¬
cept and application.
The Eight Special Recommendations call for (1) ratifying and imple¬
menting the 1999 UN International Convention for the Suppression of the
Financing of Terrorism (parallel to recommendation 1 of the original FATF
Forty Recommendations to ratify and implement the 1988 Vienna Conven¬
tion against Illicit Traffic in Narcotic Drugs and Psychotropic Substances);
(2) criminalizing the financing of terrorism (parallel to FATF recommenda¬
tion 4 on criminalizing money laundering); (3) freezing and confiscating
funds and assets of terrorists (parallel to FATF recommendation 7); (4) report¬
ing suspicious transactions (parallel to FATF recommendations 15-18);
(5) international cooperation (parallel to FATF recommendations 33-34 and
36-40); (6) measures against nontraditional channels of value transmission
(analogous to FATF recommendations 8 and 9, extending the framework
to nonbank financial and nonfinancial institutions); (7) special attention to
wire transfers (analogous to FATF recommendation 9); and (8) reviewing
laws and regulations that relate to nonprofit entities (analogous to FATF
recommendation 25 with respect to the abuse of shell corporations).
Of course, combating the financing of terrorism predated the Sep¬
tember 11 tragedy. In the wake of the bombing of Pan American flight 103
on December 21, 1988, over Lockerbie, Scotland, the 1989 G-7 Summit issued
a declaration on terrorism that focused on deterring terrorist acts. President

140 CHASING DIRTY MONEY

Clinton invoked the International Emergency Economic Powers Act on
January 23, 1995, to prohibit financial transactions with terrorists and
groups financing terrorism. Following the bombing in Dhahran, Saudi
Arabia, on June 26, 1996, the G-7 Summit issued another declaration on ter¬
rorism that included a reference to the financing of terrorism. The United
Nations also reached agreement in 1999 on the International Convention
for the Suppression of the Financing of Terrorism. In June 2000, the Bremer
Commission report on Countering the Changing Threat of International
Terrorism was submitted to the US Congress. A tug of war then ensued
within the new Bush administration about implementing the report's rec¬
ommendations, including US ratification of the UN International Con¬
vention on the Suppression of the Financing of Terrorism. The inter- and
intra-agency infighting was resolved only in the wake of the September 11
attacks, which put combating terrorism financing at the top of the agenda
of the US AML regime. September 1 1 also shifted some of the attention
away from an exclusive focus on state sponsorship of terrorism.
The effort to enhance the AML regime to block and seize funds intended
to finance terrorism was impressive even scaled by the enormity of the
World Trade Center attack. Richard Spillenkothen, director of the division
of supervision and regulation at the Federal Reserve Board, testified before
the US Senate: "The US government's response to the terrorist attacks on
September 11 has necessitated unprecedented cooperation among federal
bank supervisors, the private sector, law enforcement agencies, and the
international financial community" (Spillenkothen 2002, 1).
From the standpoint of effectiveness, if the AML regime were to prevent
a single major terrorism incident each year on the scale of those in Bali,
Indonesia on October 12, 2002, or Madrid, Spain on March 11, 2004, then it
surely would be judged worthwhile. However, it is difficult if not impossi¬
ble to establish connections between terrorism averted and any specific ele¬
ment of the AML regime; only where a specific terrorist plot is foiled, fairly
late in its development, can a connection be made. In turn, inasmuch as the
regime is justified on the basis of preventing international terrorism inci¬
dents, it is less likely to be examined on a cost-effectiveness basis as much
as on its efficacy in measurably contributing to preventing an incident.1
It would be a stretch to consider statistics on the incidence of global ter¬
rorism as a useful indicator of the effectiveness of the AML regime per se
in reducing terrorism, in part because prevention involves much more than
interdicting financial flows, and in part because of the low-cost nature of
most terrorist operations. Paul Pillar (2003) argues that financial instru¬
ments play a secondary role in counterterrorism because terrorism is cheap

1. Bell (2003, 120-21) in a thoughtful article on combating the financing of terrorism suggests
only indirect measures of success: number of forfeiture orders, value of assets diverted,
reduced revenues to terrorist organizations, and international cooperation indexed by active
participation in blocking operations.

COMBATING GLOBAL “PUBLIC BADS” 141

and the modest flows of funds are difficult to track. Unlike some other
criminal activities that are financed from the accumulated stock of the pro¬
ceeds of crime, financing of actual terrorist incidents primarily involves
small flows of financial resources in the tens or hundreds of thousands,
rather than millions, of dollars.
One should distinguish, however, between the marginal cost of a par¬
ticular terrorist operation and the much larger total cost of creating and
maintaining a sponsoring organization. In this sense, combating terrorism
financing has a good deal in common with combating money laundering
in the context of conventional organized criminal organizations such as the
Mafia, drug cartels, and street gangs. In addition, some countries and
organizations that sponsor terrorism are also involved in conventional
criminal activities such as drugs and cigarette smuggling.
Similarly, one cannot conclude much about the effectiveness of the AML
regime from the total amount of funds frozen or seized since September 11,
2001. The global total as of March 24, 2004, was estimated at $203 million
as a result of 173 jurisdictions having issued freeze orders.2
A slightly better measure of AML regime effectiveness might be the
amounts that are blocked, seized, or confiscated each time a new channel is
added to the list of sources of terrorist financing, such as a new charity or a
name of a new individual linked to such an activity. By this metric, results
since September 11, 2001, are open to a variety of interpretations. As of the
end of 2001, $112 million had been blocked or seized, which is a quite sub¬
stantial amount relative to what is needed for individual terrorist opera¬
tions. The amount 27 months later had risen to just over $200 million, but
only $3.5 million had been added to the total in the final eight of those
27 months, indicating that the process had slowed considerably. Some of the
previously blocked funds ($27.7 million) were used to help finance the new
government in Afghanistan, and some went to pay legal fees, so the $200 mil¬
lion understates the total of all funds block or seized on a continuous basis.3
Moreover, freezing often blocks a pipeline for fund transfer. Anecdotal
reports suggest that such actions have a negative effect on fundraising, but
they do not produce quantitative measures.
To illustrate the difficulty of going from these statistics to measures of
effectiveness in preventing terrorism, assume that the unadjusted total of
funds frozen, seized, or impeded was $300 million, or about $10 million a

2. This is the sum of blocked or frozen funds plus seized or confiscated funds. Funds are first
blocked or frozen and later may be seized or confiscated via a separate legal proceeding.
There may be some double counting in the data because they draw on a number of non-US
sources. Some funds have been released to finance government activities. The total includes
$64 million seized with the balance frozen. The data are from the US Treasury based on
Zarate (2004).

3. Some hands may have been blocked and later released by the courts in the countries
involved for lack of sufficient evidence to justify the initial blocking or subsequent seizure.

142 CHASING DIRTY MONEY

month over the 27 months from December 2001 through March 2004. If the
AML regime had a 33 percent success rate, which would be remarkable,
then about $20 million a month slipped through the net. On the one hand,
if each major terrorist act costs on average $400,000 — the FBI estimates the
budget for the September 11 attacks at between $300,000 and $500,000 (US
Treasury 2003, footnote 50) — as many as 1,350 such acts might have been
financed over the 21h-year period. On the other hand, US intelligence offi¬
cials have been quoted as saying that al Qaeda operates on a budget in the
tens of millions a year (Washington Post , December 20, 2003, A16), so an esti¬
mate of $300 million blocked, seized, or impeded over this period could be
considered significant. This arithmetic is intended neither as an indictment
of the AML regime as it has operated to combat terrorism financing since
September 11, nor to suggest that the larger amounts frozen or seized im¬
ply that more has slipped through the net. Rather, the aim is simply to
illustrate the challenge that the regime faces based on this metric, and to
demonstrate why we prefer to measure effectiveness through a flow test
(how much is blocked in each new antiterrorist financing action) rather
than a stock test (total amounts frozen or seized).
Pillar (2003) offers a more positive interpretation of events, citing the
interdiction of terrorist finances as "one of the biggest post-9/ 11 improve¬
ments of any of the counterterrorist instruments." In addition to the legal
powers expanded as a result of 9/11, he argues that the higher priority
attached to terrorism by the US government has stimulated more cooper¬
ation from foreign governments.
Prior to September 11, US and non-US official policy toward international
terrorism focused to a substantial degree on state-sponsored terrorism and
the use of economic sanctions to discourage the harboring of terrorist groups.
Even now, the United States continues to designate state sponsors of ter¬
rorism, foreign terrorist organizations, and specially designated terrorists,
as it has been doing since the early 1970s. This type of sanction has been
used against Afghanistan, Cuba, Iran, Iraq, Libya, Sudan, and Syria, with
limited impact. Sanctions or the threat of sanctions also can be lifted to pro¬
vide positive incentives in certain situations, as was done in the wake of
September 11. Gary Hufbauer, Jeffrey Schott, and Barbara Oegg (2001, 13)
conclude "it would be illusory to expect that the arsenal of economic sanc¬
tions can play more than a modest role in the war against terrorism." They
cite the limited success against state sponsors, terrorist groups, and major
drug lords. Nevertheless, this is an area where modest contributions can
be important.
As with other uses of the anti-money laundering regime, AML tools are
only one important instrument in the fight against the financing of terror¬
ism. Blocking or seizing funds intended to finance terrorism are only part of
the overall effort to combat such activities and their financing. As stated by
the US Treasury on August 29, 2003, in a press release about a UN report on
sanctions to block the flow of funds to al Qaeda: "The point isn't grabbing

COMBATING GLOBAL “PUBLIC BADS” 143

dollars in bank accounts when freezing orders go into place, it is destroying
the financial infrastructure of terrorism. That means seizing money, but it
also involves dismantling the channels of funding, deterring those who
would give aid and support to terrorists, and following the leads to terror¬
ist cells." In a modest step along these lines, the US Treasury on May 12,
2004, for the first time invoked a provision of the 2001 USA PATRIOT Act
(section 311) to declare the Commercial Bank of Syria — the principal
government-owned bank — an institution of "primary money laundering
concern" in connection with the bank having been used by terrorists. The
action prohibited US financial institutions from having any direct or indi¬
rect dealings with the Syrian institution or any of its subsidiaries.
In stressing the complexities and challenges of counterterrorist policy.

Pillar (2003, 217) describes terrorism as "a problem managed, never solved."
Among his recommendations are to disrupt terrorist infrastructure world¬
wide, to use all available methods while not relying heavily on any one
of them, to legislate sparingly, and to keep terrorist lists honest. Each of
Pillar's points provides perspective on the role of the AML regime in com¬
bating terrorism financing. The complexities of the phenomena of terrorism,
on the one hand, and money laundering, on the other, militate against sim¬
plistic, overarching solutions such as the creation of a new international
organization dedicated solely to issues involving terrorist financing, as
suggested by the Council on Foreign Relations (2002, 27) report on the sub¬
ject, but dropped from its 2004 update.
The call for honesty in compiling lists of terrorists suggests the need for
some care in blocking or seizing assets. Blocking assets in error, even well-
intentioned error, only to have them later released via a judicial process, can
undermine support for some of the more draconian (but perhaps necessary)
methods applied to combating the financing of terrorism. Ron Suskind
(2004, 191-93) describes the frantic US efforts to open a financial front in the
war on terror on the basis of evidence that might not stand up in court. He
concludes with a quotation he attributes to US Treasury General Council
David Aufhauser, "It was almost comical. We listed out as many of the
usual suspects as we could and said, 'Let's go freeze some of their assets.' "
In this context, any metric for judging the success of the AML regime in
combating the financing of terrorism will not be entirely satisfactory.4
However, the amount of assets frozen or blocked and seized or confiscated
is relevant to an overall assessment of the AML regime's contribution.
Consequently, after the initial success and more recent modest accom¬
plishments in combating such financing, US and other authorities have
stressed better compliance with international obligations in this area and

4. Metrics for judging the success of overall counterterrorism efforts such as arrests, convic¬
tions, and attacks prevented (more problematic to measure) often have little to do with the
AML regime per se. Recall the discussion in chapter 5 of the temporary surge in terrorism-
related SARs in the wake of the September 11 attack.

144 CHASING DIRTY MONEY

an increasing focus on informal money transfer mechanisms. The FATF has
issued best-practice papers on the freezing of terrorist assets (FATF 2003d)
and on combating the abuse of nonprofit organizations (FATF 2002a). The
G-7, under US leadership, has also exhorted other countries to cooperate
more effectively in combating terrorist financing. For example, the G-7
finance ministers and central bank governors issued a Joint Statement on
Combating Terrorist Financing in April 2004 following a meeting with the
finance ministers and central bank governors of 12 other countries, the
FATF president, and the heads of the International Monetary Fund (IMF),
World Bank, and the European Commission. They pledged closer cooper¬
ation and emphasized the importance of dealing collectively with the prob¬
lem of terrorist financing.
As with other objectives of the AML regime, international cooperation is
crucial to the success of the regime in combating the financing of terrorism.
Notwithstanding various statements of strong support, however, cooper¬
ation in this complex area is only as effective as the political will to take
costly or politically unpopular stands that might face domestic resistance.
For example, the FATF's 2002-03 annual report said that the full compli¬
ance rate with its Eight Special Recommendations on Terrorist Financing
was only 75 percent. The average full compliance rate by FATF members
with the original FATF Forty Recommendations as of the same date was
more than 90 percent.5 Although there were shortfalls in compliance with
each of the recommendations on terrorist financing, failures to comply
fully were most frequent with those relating to wire transfers and to ratifi¬
cation and implementation of UN conventions and resolutions. For exam¬
ple, as of April 2004 only 112 countries had carried out the first of the
FATF's special recommendations, which was to ratify the 1999 UN Inter¬
national Convention for the Suppression of the Financing of Terrorism.
Another 41 countries, including six FATF countries, had signed the con¬
vention but not yet ratified it.6 To cover the entire 191 members of the
United Nations another 38 countries also would have to accede to the con¬
vention via ratification.

In general, efforts to rapidly implement measures to combat the financ¬

ing of terrorism have met with considerable indifference and resistance. A
UN monitoring group (United Nations 2003a) reported that 108 members
had failed to submit reports on terrorism financing six months after the

5. Reporting on its pilot project with the World Bank to assess compliance with standards to
combat money laundering and terrorism financing, the IMF (2004c) also found that compliance
with the original FATF Forty Recommendations was much higher than with the Eight Special
Recommendations on Terrorist Financing. The assessment noted that in many countries, the
legislation necessary for the latter group of recommendations had not yet been adopted.

6. The FATF countries were Argentina, Belgium, Brazil, Germany, Greece, and Ireland. The
United States, leading the war on terrorism, did not itself ratify the convention until June
26, 2002.

COMBATING GLOBAL “PUBLIC BADS” 145

agreed submission date. The report said that 25 of those countries were of
special interest because of other information suggesting that al Qaeda or its
associates might be active within their borders. Among the most prominent
were Egypt and Indonesia. Eight European countries also had not submit¬
ted reports, including Ireland and Luxembourg.7 On the other hand, Saudi
Arabian authorities touted their passage of the FATF review of Saudi com¬
pliance with the 40 + 8 recommendations on anti-money laundering and ter¬
rorism financing (Financial Times, March 8, 2004, 3), notwithstanding the fact
that the actual review, while saying that Saudi Arabia was "compliant or
largely compliant" with the recommendations, pointed to the country's lack
of a clear definition of terrorist financing, failure to issue implementation
rules for its 2003 AML legislation, and shortcomings in the area of customer
due diligence (FATF 2004a, annex C, 9). A report by the Council on Foreign
Relations (2004), while praising some Saudi actions over the previous two
years, is also very critical of the implementation of its regime to combat
money laundering and the financing of terrorism.
In addition, a large number of countries have not put in place legislation
permitting them to freeze terrorist assets without delay, as called for under
UN Security Council Resolution 1373. Following the March 11, 2004, attack
in Madrid, the European Council of Heads of Government cited Resolution
1373 and called on all EU members to implement previous decisions on
freezing property and evidence by December 2004 and on confiscation of
proceeds by June 2004. The council did not list the countries that had thus
far failed to comply, but it is reasonable to conclude that there was more
than one.

With respect to tightening the US AML regime as it applies to the financ¬

ing of terrorism, the American Civil Liberties Union has challenged some of
the provisions of the USA PATRIOT Act on privacy grounds.8 Opponents
have criticized the elimination of the distinction between criminal and intel¬
ligence classifications, and have filed legal briefs opposing post-9/11 poli¬
cies.9 The effects of such resistance may well be taking their toll, for better or
for worse. After the US government charged US Islamic leader Abdurahman

7. Another example of the limited urgency attached to international efforts to extend the
AML regime to cover the financing of terrorism is the weak response of the 184 members of
the IMF to a voluntary questionnaire on anti-money laundering and countering terrorism
financing that was authorized in November 2001. By September 2002, only 49 countries had
responded, and four G-7 countries (Canada, Germany, Japan, and the United Kingdom) were
among those that had not.

8. Steinberg, Graham, and Eggers (2003) make a plea for public guidelines on how these
important new powers should be implemented.

9. On the other hand, Richard Falkenrath, former deputy homeland security adviser to US
President George W. Bush, wrote in the Financial Times (July 7, 2004, 13) that the failure of
European governments to eliminate this same "wall" between law enforcement and intelli¬
gence has prevented them from "connecting the dots" with respect to terrorist attacks in
Europe such as that which occurred in Madrid.

146 CHASING DIRTY MONEY

Almoudi with money laundering and engaging in prohibited financial
transactions in October 2003, the New York Times (October 24, 2003, A19)
noted that he had not been charged with terrorist financing, perhaps because
prosecutors were gun-shy as a result of mixed results in earlier such cases.
Regarding the previously mentioned UN report on sanctions directed at
blocking the flow of funds to al Qaeda, the Washington Post (August 29,
2003) reported that the Schengen Information System, used by 13 European
Union and two other European countries to monitor border crossings, con¬
tained only 40 of the 219 names on the UN list of suspect individuals and
institutions. Several members of the Schengen group reportedly are pre¬
cluded by their national laws from placing the names of their own citizens
on national watch lists without an appropriate judicial basis. Following the
Madrid attack, the EU resolved once again to step up its efforts to imple¬
ment a 2001 decision to create a Europe-wide arrest warrant.
As with other aspects of the AML regime, differences in legal as well as
regulatory structure and philosophy sometimes get in the way of cooper¬
ation. For example, Germany has criticized US and UK authorities for
resisting the extension of financial supervision regulations to "under¬
ground banks" such as the hawala system of international money trans¬
mission (Financial Times, June 18, 2003). But for the moment, US and UK
authorities are satisfied with applying registration, customer due diligence,
and suspicious activity report (SAR) requirements to these institutions,
stopping short of a full supervisory regime. Similar differences in regula¬
tory philosophy underlie opposing positions regarding the regulation of
hedge funds in Germany and France as opposed to the United States and
United Kingdom.
Such differences interfere with establishing a seamless global AML regime
as it applies to the financing of terrorism. To cite another example, Jochen
Sanio, a German then president of the Financial Action Task Force (FATF),
said in 2003 that "no German bank had yet filed a spontaneous report say¬
ing that someone had transferred money or was suspected of being involved
in the financing of terrorism" (Reuters News Services, January 14, 2003) — the
apparent implication being that such reporting requirements were of limited
utility. The United States has been pressing other countries to crack down
on cash transfers, in part out of concern that they are increasingly used to
channel funds to terrorist groups. On the other hand, a frequent complaint
abroad about the US financial enforcement apparatus is that when the
Treasury's Office of Foreign Assets Control (OF AC) requests that the assets
of certain individuals be frozen, it fails to provide the necessary information
that foreign authorities need to defend their actions in court.
The United States has also been urging other jurisdictions, starting with
its G-7 colleagues, to introduce cash reporting requirements along the lines
of its currency transaction reports (CTRs) and reports of international trans¬
portation of currency or monetary instruments (CMIRs). To date, however,
the US effort has not been successful. In its report on conducting due

COMBATING GLOBAL “PUBLIC BADS” 147

diligence on current customers with whom a client relationship had been
established prior to April 1994, Pricewaterhouse Coopers (2003, 64) notes:
"Firms said that they would be in favor of the current customer review if
they thought that it would stop even one terrorist act, but they thought that
the likelihood of this was so low that any benefit was negligible."
Costs as well as benefits figure in to decisions regarding application of
the AML regime to combating terrorist financing. For some poorer coun¬
tries, the costs are large and the direct benefits are not valued highly. An
appropriate response is stepped-up technical assistance, as was suggested
for the United States in the report by the Council on Foreign Relations
(2002, 25) on terrorist financing. The G-8 Counter Terrorism Action Group
has responded by addressing capacity building, and the FATF uses its
Technical Assistance Needs Assessment exercise for the same purpose.
In conclusion, the global AML regime has contributed to combating the
particular global pubic "bad" of financing terrorism. However, the regime
is just one part of what is a complex effort for which measures of success
are imprecise and indirect. Just like the pursuit of other AML objectives,
efforts to strengthen application of the global AML regime to terrorism
financing face the usual range of challenges associated with aligning dif¬
ferent national regimes and structures. In addition, combating terrorism
financing involves unique difficulties because of the relatively small
amounts of money involved, and because the resources that need to be
tracked, rather than deriving from a crime, finance crimes that authorities
desperately need to prevent.

Corruption and Kleptocracy

Corruption is a major public policy issue that has been called "the most
important economic issue facing the world today" (Hills 2001, 1). "The
extent of worldwide corruption is staggering," writes Daniel Kaufmann
(2003, 1). "A conservative estimate of the value of corrupt worldwide
income would not merely be measured in terms of billions of dollars, but
instead in the low US$ trillion range . . . [and] the additional indirect and

long-term costs would further enlarge such rough estimates."10

Corruption is a manifestation of the economic activity of rent seeking.
While empirical studies clearly have demonstrated the substantial econo¬
mic costs of corruption and the weak governance often associated with it,
and while the phenomenon is widely recognized and extensively studied.

10. Kaufmann's estimate of the total annual amount of corruption, which he describes as
"rough," is based on surveys. While the precision of his specific estimate is debatable, no one
disputes that the amounts involved are large, certainly in the hundreds of billions of dollars
annually.

148 CHASING DIRTY MONEY

actually designing policy approaches to reduce substantially the incidence
and extent of corruption is a major challenge (see Kaufmann, Kraay, and
Zoido-Lobaton 1999; Knack and Keefer 1995; and Mauro 1995).
Paolo Mauro (2002) offers a pair of theoretical models to explain the
strategic complementarity between corruption and weak governance. In
one model, corrupt behavior by a group of officials contributes to the cor¬
ruption of all officials and helps to explain the resulting multiple equilib¬
ria in which rampant corruption mires a country in the "bad" equilibrium.
In another model, corruption contributes to political instability because
one government official's corruption induces his or her colleagues to fol¬
low the example, which shortens the effective time horizon of the govern¬
ment and reduces its chances of reelection. The policy conclusions from
such models point toward comprehensive rather than gradual reforms to
move decisively away from the bad equilibrium. The conclusions also sup¬
port improved disclosure and transparency as devices to motivate public
opinion and ultimately discipline office holders.
Corruption is a broad phenomenon often found in relationships between
the public and private sectors, but some would argue that it occurs in purely
private sector activities as well.11 The World Bank (1997, 8) defines corruption
as "the abuse of public office for private gain." Transparency International
(1996, 7), the most prominent nongovernmental organization that has focused
on this issue, employs a less concise definition: "Corruption involves behav¬
ior on the part of officials in the public sector, whether politicians or civil ser¬
vants, in which they improperly and unlawfully enrich themselves, or those
close to them, by the misuse of the public power entrusted to them."
Corruption of a public official is connected de facto with money laun¬
dering, regardless of whether corruption is a predicate crime for a money¬
laundering prosecution in the jurisdiction where the corruption occurs.
The public official will place his or her proceeds where they are relatively
safe. The safe location may be a business or piece of real estate in the offi¬
cial's own country, but often it is in a foreign country, which reinforces the
cross-border link between money laundering and corruption.
Kleptocracy is corruption by high-level public officials who use their
positions systematically to line their pockets directly or indirectly with
funds from the public purse.12 Kleptocracy is a subcategory of political

11. The new UN Convention Against Corruption (United Nations 2003b) does not cover
private-to-private sector corruption, to the disappointment of some private sector observers.

12. Moody-Stuart (1997, 63) provides a nice classification of official corruption: "Five percent
of $200,000 will be interesting to a senior official below the top rank; 5 percent of $2 million is

in the top official's area; 5 percent of $20 million is real money for a minister or his key staff;
5 percent of $200 million justifies the serious attention of the head of state." The kleptocrats
are the last and the next-to-last levels of officials in this classification.

COMBATING GLOBAL “PUBLIC BADS” 1 49

corruption, which is more commonly found in societies with less well-
developed systems of justice and limited transparency in the public and
private sectors. Of course, the line between political corruption and polit¬
ical contributions is not an easy one to draw. The relativistic view of pub¬
lic corruption of Robert Neild (2002, 202) defines it as the "breaking by
public persons, for the sake of private financial or political gain, of the
rules of conduct in public affairs prevailing in a society in the period under
consideration." Not surprisingly, he is not optimistic that the standards
for public conduct in much of the world will soon evolve to a level close
to those found today in northwestern Europe.
The prevention pillar of the AML regime can play an important role in
reducing the incidence and scale of kleptocracy through rigorous appli¬
cation of customer due diligence (CDD) regarding "politically exposed per¬
sons," defined by the FATF (2003c) as "individuals who are or have been
entrusted with prominent public functions in a foreign country, for exam¬
ple heads of state or of government, senior politicians, senior government,
judicial or military officials, senior executives of state-owned corpora¬
tions, important political party officials." Applying AML procedures such
as CDD can help to limit the success of kleptocrats, since they generally
have little interest in keeping their ill-gotten gains invested in their own
jurisdictions, where any change in government might threaten control of
those assets.
The enforcement pillar can play a role as well via both the punishment of
kleptocrats and confiscation of the proceeds of their crimes. Both elements
of the enforcement pillar are important not only because they punish the
individuals but also because they send messages to junior officials who may
be tempted into similar behavior.
Such condign punishment outside of the jurisdiction of the underlying
offense is particularly relevant to bribery and corruption. Corrupt officials
in poor nations who control the justice system there may be unreachable
except through money-laundering investigations in other countries. For
example, Pavel Nickolayevich Lazarenko, the former prime minister of the
Ukraine, was indicted, placed under house arrest, and convicted in June
2004 in the United States for money laundering. This case was the first
brought under US money-laundering law using as a legal foundation the
proceeds of extortion that was committed entirely overseas. The underlying
offenses of receiving bribes and extortion, which notionally occurred in the
Ukraine even though the money did not change hands in Kiev, could not be
prosecuted there. Prosecution of the prime minister in the United States,
based on use of US banks and brokerage accounts, served to improve jus¬
tice in a salient and important way.
Similarly, the prosecution of kleptocrats and confiscation of the proceeds
of their crimes provide a positive example for limiting garden-variety cor¬
ruption in the home countries. Thus the incidence of corruption is linked

150 CHASING DIRTY MONEY

to the incidence of kleptocracy. Widespread corruption, in turn, can under¬
mine the integrity of the financial system.13
Assessing the effectiveness of the AML regime in reducing kleptocracy
more narrowly, and corruption more broadly, relies to some extent on indi¬
rect measures such as surveys and other indicators of the prevalence of
these phenomena, for example those put out by Transparency International
and the World Bank. In general, the AML regime is not the only or even the
major element contributing to success in combating kleptocracy, but it has
a role to play and can reasonably take some credit for any improvement,
or blame for some deterioration, in the incidence of the phenomenon.
A reasonable proximate measure of the effectiveness of the AML regime
in dealing with kleptocracy would be the flow of convictions of kleptocrats
or the amount of funds frozen and returned to their countries of origin.
Unfortunately, data to construct such a measure are not currently avail¬
able, and anecdotal information is uneven. Various public and private
international organizations have made ongoing efforts to raise the profile
of work against kleptocracy and corruption, but none has yet developed a
comprehensive database that would permit measuring progress in com¬
bating these crimes.
Transparency International's Global Corruption Report 2004 illustrates both
the progress to date and the need for systematic database construction. The
publication includes a summary table of the activities of 10 major kleptocrat
presidents and prime ministers who served from 1965 to 2002, starting with
Joseph Mobutu in Zaire and ending with Arnoldo Aleman in Nicaragua
(Transparency International 2004, 13). 14 They embezzled amounts believed
to have ranged from $78 million to $35 billion while in office. A few of the
individuals on the list, such as Aleman as well as Joseph Estrada in the
Philippines, were convicted of corruption, and some of the stolen funds
located in foreign banks were frozen and returned to their countries. Some
of the funds stolen by former president Sani Abacha of Nigeria also were
returned to that country.

Transparency International's list demonstrates that, even though a com¬

plete database is not yet available, information on kleptocracy and indirectly

13. Note that without the AML regime, criminal funds could be passed through the financial
system without any payment of additional bribes. In this sense the regime itself creates a
threat to the integrity of the financial system in the form of corrupt practices. The AML
regime, in turn, seeks to cleanse banks and other financial institutions not by reducing the
problem of bribes paid to violate regulations but by maintaining a distance between the
underlying criminal acts and the financial sector.

14. The others in order of the estimated size of their embezzlements are Suharto of Indonesia,
Marcos of the Philippines, Abacha of Nigeria, Milosevic of Serbia /Yugoslavia, Duvalier of
Haiti, Fujimori of Peru, Lazarenko of Ukraine, and Estrada of the Philippines. Mobuto ranks
third on the list and Aleman ninth. The list, of course, is not exhaustive. Any number of other
names could have been added — e.g., Benazir Bhutto of Pakistan, Chiluba of Zambia, Moi of
Kenya, and Taylor of Liberia.

COMBATING GLOBAL “PUBLIC BADS” 1 51

on the AML regime's role in its prevention and prosecution is gradually
being assembled. One reason for the slow progress is the general lack of
attention in the policy community to developing systematic measures of the
effectiveness of the AML regime. However, there are other reasons as well.
The global AML regime has been slow to incorporate public corruption, and
by extension kleptocracy, into its set of objectives. In the United States, for¬
eign corruption became a predicate offense for a money-laundering prose¬
cution only with the passage of the USA PATRIOT Act in 2001, despite earlier
executive branch proposals and pressures from other countries to do so.
The Organization for Economic Cooperation and Development (OECD)
ratified its Convention on Combating Bribery of Foreign Public Officials in
International Business Transactions in 1997. But action followed these good
intentions rather slowly. The convention did not enter into force until 1999,
when 12 countries had deposited their instruments of acceptance, approval,
or ratification.15 Many countries were slow to pass legislation necessary to
implement the convention, in particular criminalizing foreign bribery,
which the United Kingdom did only as part of legislation passed in the
wake of the September 11 tragedy. Prosecutions for foreign bribery, in fact,
have been quite limited outside the United States,16 and the OECD has been
criticized for weak enforcement monitoring. Finally, the convention itself
has come under fire for loopholes, such as noncoverage of foreign sub¬
sidiaries, bribery of officials of political parties, and private-sector bribery
(Heimann 2004).
At the international multilateral level, negotiations began on the UN
Convention Against Corruption in December 2000, and 95 countries signed
the final text in December 2003. 17 The UN convention includes prevention
measures such as establishment of anticorruption bodies, requires the crim¬
inalization of a range of acts of corruption, mandates increased international
cooperation, and, importantly, has a chapter on asset recovery. Of course,
the convention does not satisfy everyone — Transparency International, for
example, has criticized it for failing to address monitoring mechanisms and
for not mandating the criminalization of bribery.
With respect to kleptocrats, in particular, only recommendation 6 of the
2003 FATF Forty Recommendations explicitly addressed "politically ex¬
posed persons." One breakthrough in this area was the passage in May

15. As of March 2004, 34 countries, including five nonmembers of the OECD (Argentina,
Brazil, Bulgaria, Chile, and Slovenia) had adhered to the convention.

16. Bribery of a foreign official has been a crime under the US Foreign Corrupt Practices Act
(FCPA) since the 1970s, but prosecutions are infrequent. In May 2003, James Giffen and

ExxonMobil were charged under the FCPA in connection with payment of a $51 million "ser¬
vice fee" in Kazakhstan related to an oil contract. Giffen allegedly received more than $78 mil¬
lion from Mobil and other companies, passing some of the funds on to senior Kazakh officials.

17. At the end of March 2004, Nigeria announced that it would also sign.

152 CHASING DIRTY MONEY

2003 of UN Security Council Resolution 1483, which called for freezing the
funds, assets, and economic resources of Saddam Hussein and other senior
officials of the former Iraqi regime and their immediate family members.
However, to date there is little evidence that this requirement has pro¬
duced financial results.
Pursuing known or suspected kleptocrats before they have been driven
from power is politically sensitive for governments that often need the
cooperation of other governments on other issues, even if they are corrupt.
In this connection, a noteworthy development was the establishment in
2003 of a US government task force to systematically target high-level pre¬
sent and former Latin American officials involved in corruption (Nezv York

Times, August 23, 2003, A9). Such proactive concern with the issue of klep¬
tocrats appears to have produced some results. When Aleman of Nicaragua
was convicted of corruption in December 2003, a US judge approved forfei¬
ture of a multimillion-dollar Florida condominium and seizure of $150,000
from the former revenue director in Aleman's finance ministry. Another
important case that emerged in early 2004 was the investigation of a senior
official with the Riggs National Bank in Washington, DC (see chapter 6) for
alleged involvement with the kleptocratic activities of the president of
Equatorial Guinea, Teodoro Obiang, and his son.
It is no easy matter to deal with kleptocrats after they have been driven
from office. In part, this is because allegations of corruption may involve
the settling of political scores,18 or because new leaders may not be entirely
serious about taking action, as evidenced, for example, by neglecting to
provide information that might form the basis for legal action by another
country. One reason for such bait-and-switch tactics is that the requesting
authorities sometimes are in the process of following the example of their
kleptocratic predecessors rather than practicing what they preach. Even
when home country authorities are serious, they need authorities in other
jurisdictions to share their concern, devote resources to the matter, and
cope with due process requirements in their jurisdictions. Such complica¬
tions often impede the type of quick cooperation necessary to identify,
block, and confiscate the proceeds of high-level official corruption. As
usual, political and legal standards differ across jurisdictions.

18. Suspected political motivations for prosecutions in the home countries of alleged klepto¬
crats may sometimes underlie findings that the evidence presented by those authorities is too
weak to justify extradition or the freezing and seizure of assets. In a case that comes close to
allegations of kleptocracy, a Greek court, following the example of courts in Spain and the
United Kingdom, declined in 2003 to extradite oligarch Vladimir Gusinsky to Russia on
money-laundering charges possibly associated with capital flight, which is covered by
money-laundering statutes in some countries but is not compatible with FATF recommenda¬
tions. The reason cited for denying extradition was that Russian authorities had failed to sub¬
stantiate the charges, and the independent Greek legal system turned them down. The
European Court of Human Rights subsequently rebuked the Russian government for its pur¬
suit of Gusinsky (Financial Times, October 15, 2003, 4; May 20, 2004, 2).

COMBATING GLOBAL “PUBLIC BADS” 1 53

 The Swiss authorities scored a public relations coup with their early pur¬
suit of the assets of Nigeria's Abacha, but that case involved the Swiss legal
regime, which has fewer or less cumbersome due process protections than
those in many other developed countries. Authorities also benefited from
the Abacha family's apparent misperception that their ill-gotten gains were
safe in Swiss institutions.19 Other jurisdictions, however, have been less
successful in pursuing the Abachas' assets. The UK's Financial Services
Authority announced in 2001 that it had the power to investigate allega¬
tions against the Abachas and sanction institutions involved for failing to
comply with AML controls but not to freeze assets that may be the proceeds
of crime. UK authorities eventually did return about $8 million to Nigeria's
Economic and Financial Crimes Commission on the basis that cash pre¬
viously seized at Heathrow belonged to the Nigerian central bank. By the
second half of 2003, however, UK authorities had largely abandoned
efforts to track the Abacha family's assets, despite evidence assembled by
the Financial Services Authority that as much as $1.3 billion may have been
processed through London between 1996 and 2000. No reason was given for
the decision, but one can speculate either that Nigerian authorities had not
provided the United Kingdom with sufficient information to pursue the
matter, that the assets could not be reached, or that UK authorities reached
the conclusion that it was not worth the effort to seize assets and return them
to Nigeria, where there is no assurance that they might not be stolen again.
In the United States, the White House or the State, Treasury, Justice, or
Homeland Security Departments may receive requests for assistance with
respect to corruption by former officials in other countries. The US gov¬
ernment bureaucracy generally is not structured to respond quickly to such
requests, resulting in delays that at the least can be politically embarrass¬
ing. For example, the US government in late 1999 received a publicized
request from Indonesia for assistance in finding the assets of former presi¬
dent Mohamed Suharto. The request went unanswered for 10 months, and
the eventual response was that the Indonesian government would have to
provide more detailed information if the US government was to be of any
assistance.
Such instances were the principal motivation behind an action item in

the US Treasury's 2000 National Money Laundering Strategy that estab¬

lished an interagency working group to streamline and improve coordi¬
nation in handling such requests and inquiries. The group succeeded in
establishing new procedures, but under US law the government is still con¬
strained from responding proactively to assertions by foreign governments
that various former officials were corrupt and that their assets in the United
States should be turned over to the new government. Requests for assis-

19. It was reported in January 2004 that Swiss authorities are cooperating with Argentine
authorities investigating money laundering by 200 former officials of President Menem's
administration.

154 CHASING DIRTY MONEY

tance must contain a factual or legal basis to allow the US government to
act, such as evidence that the funds were the proceeds of a US crime (in¬
cluding corruption). Despite such constraints, the US government has
sometimes taken actions, as seen in the previously mentioned Lazarenko
and Aleman cases.
Kleptocracy and international bribery cases are complex and expensive,
as acknowledged even by those writing on behalf of proactive organiza¬
tions such as Transparency International. Fritz Heimann (2004, 129) notes:
"Prosecutors may be reluctant to bring foreign bribery cases because they
lack the professional resources to pursue complex international cases.
Procedures for obtaining evidence from abroad are often cumbersome and
often unproductive." He prescribes technical assistance in connection with
cases in developing countries.
The inquiry by Kenyan President Mwai Kibaki into corruption by the
government of his predecessor, Daniel arap Moi, offers a glimpse into both
the financial and political constraints to such initiatives. The inquiry has
required bringing in a costly team of business investigation consultants
working with forensic investigators. Moreover, since many of the alleged
recipients of earlier corruption payments are in the new government, the
political pressures to scale back the inquiry have been substantial.
Nigeria established the Economic and Financial Crimes Commission pri¬
marily to deal with private-sector corruption such as money laundering,
fraud, and tax evasion. Given Nigeria's history of public-sector corruption
and interference in legal proceedings, however, public- and private-sector
corruption are difficult to separate, and resistance is entrenched. The
commission faces an uphill battle in getting Nigeria to conform to the
global AML standards established by the FATF. Executive Chairman Nuhu
Ribadu was granted only 6.25 percent of his budget request for fiscal
year 2004 — $2.2 million out of a request for $35.5 million (Financial Times,
February 24, 2004, 3).
In recent years, G-8 summit meetings have addressed corruption issues.
Building on an initiative at the Evian Summit in 2003, the Sea Island Summit
produced agreements the following year with four countries — Georgia,
Nicaragua, Nigeria, and Peru — to combat corruption and promote trans¬
parency, with the latter focused on the process for granting licenses in the
extractive sectors. Complementary actions by the G-8 are limited to techni¬
cal assistance, denial of safe havens to convicted kleptocrats, and dealing
more effectively with recovering the proceeds of corruption. The last action
restated a commitment made the Okinawa Summit in 2000.

In conclusion, the hope is that in the future the AML regime and asso¬
ciated political structures will be more effective in dealing with klepto¬
crats before and after they have left power. Doing so would improve the
anticorruption climate, but substantive progress in the future is likely to
be slow, as it has been in the past. As part of this progress, better measures
of success than surveys of corruption perceptions should also be developed.

COMBATING GLOBAL “PUBLIC BADS” 155

Failed States

Kleptocracy and corruption generally are associated with failed states in

which the entire political structure has imploded or been perverted in the
pursuit of personal economic or political gain. Examples include Nigeria
during the rule of Sani Abacha and Myanmar/ Burma under the current
military junta. In failed states, governments and public institutions nor¬
mally expected to enforce laws are actively engaged in undermining or
ignoring them, which in turn contributes to economic, financial, and polit¬
ical instability. A high degree of internal or cross-border violence is often
associated with such states.
Kleptocracy and corruption, as well as other predicate crimes (such as
theft, extortion, and drugs) normally associated with money laundering,
contribute to the failure of states. Once a state begins to fail, the process
feeds on itself, and those in nominal control of the instruments of the gov¬
ernmental process have an interest in systematically employing those
instruments for personal rather than public gain. The systemic concern
associated with the global public "bad" of failed states is not just the break¬
down of law and order, but the fact that such states become a breeding
ground for other public "bads" such as terrorism and other forms of vio¬
lence, which may then be exported to other countries. Examples in recent
years include Afghanistan and the Democratic Republic of the Congo.
The prevention and enforcement pillars of the AML regime have a role
to play in identifying, isolating, and ultimately rehabilitating failed states.
However, the AML regime is ancillary to broader efforts directed at bring¬
ing about profound structural changes in the government and society.
Successful implementation of AML regimes requires resources that are
generally lacking in failed states because most of their financial resources
have been stolen, or because the revenue-raising apparatus of the public
sector has broken down. In addition, an effective AML regime requires the
political and institutional capacity lacking in failed states.

Money Laundering and Failed States

A number of indicators can be used to assess the relationship between failed

states and money laundering, as well as the effectiveness of the AML regime
in helping to reduce the number of such states. The focus here will be on the
linkages between failed states (based on two political indices and one eco¬
nomic index), money laundering (based on two indicators in particular
countries and territories), and corruption (based on the Transparency Inter¬
national index). The information used dates primarily from 2002 or 2003,
with the exception of data from the FATF's Non-Cooperative Countries and
Territories initiative, which since 1999 has assessed the extent to which
countries cooperate with global anti-money laundering efforts. A time series

156 CHASING DIRTY MONEY

for these various indicators would help measure the progress of the global
AML regime in this area, but consistent time series are not available for most
indicators.
Table 7.1, which summarizes information on failed states for 2002, draws
on what is known as the Polity IV study (Gurr, Harft, and Marshall 2003) and
on a study by Robert Rotberg (2003) on indicators of political failure. The lat¬
ter study describes failed states as "tense, deeply conflicted, dangerous and
contested bitterly by warring factions . . . [with] the following features: they
have internal violence, they can no longer deliver positive political goods to
their inhabitants, they cannot control borders, they have flawed institutions,
they have deteriorated or destroyed infrastructures" (Rotberg 2003, 5). Polity
IV concentrates on four indicators of violence or political instability: revolu¬
tions, ethnic wars, adverse regime changes, and genocides or "politicides."20
Polity IV and Rotberg identify the same seven countries as "failed states":
Afghanistan, Angola, Burundi, the Democratic Republic of the Congo,
Sierra Leone, Somalia, and Sudan. Rotberg also includes Liberia and Polity IV
includes Myanmar/Burma in the "failed" category. There is much less
concordance between Rotberg's category of 21 "weak states" — defined as
countries with a shorter list of dimensions of failure — and Polity IV's cate¬
gory of 20 "complex states," defined as countries with two or more linked
wars or crises (box 7.1). Six countries are common to the two lists of "weak"
or "complex" states, which will be characterized together here as "failing"
states.21
With respect to states that have failed or are failing economically, we draw
on the most recent Country Policy and Institutional Assessment (CPIA) by
the World Bank Group's International Development Association (IDA 2003).
A World Bank (2002) task force used this type of analysis in a report on what
it called "low-income countries under stress," characterized by weak poli¬
cies, institutions, and governance. We identified as economically "failed" or
"failing" those states that had CPIA ratings or "public-sector manage¬
ment and institutions" ratings in the fourth and fifth quintiles in 2002. 22
The 11 countries in the fifth quintile on both criteria were classified as

20. Adverse regime changes include abrupt shifts in patterns of governance, periods of severe
elite or regime instability, and shifts away from democracy and toward authoritarian rule.
"Politicides" are defined as civil disturbances that result in the deaths of a substantial portion
of a political group.

21. The countries in common are Colombia, Georgia, Indonesia, Lebanon, Sri Lanka, and
Tajikistan.

22. Elements of the rating for public-sector management and institutions are property rights
and rule-based governance, quality of budgetary and financial management, efficiency of rev¬
enue mobilization, quality of public administration, and transparency, accountability, and
corruption in the public sector (World Bank 2003b, 11). This is one of four rating categories
that make up the overall CPIA rating, which also includes economic management, structural
policies, and policies for social inclusion and equity. It should be noted that the CPIA exer¬
cise covers only countries that are eligible to borrow from the IDA.

COMBATING GLOBAL “PUBLIC BADS” 157

Table 7.1 Failed and failing states in 2002
Not
Failed states included
Failing states
on CPIA
Country Rotberg3 Polity IVb CPIAC Rotberg ld Polity IVe

Afghanistan X X X
Algeria X X
Angola XXX
CP IA' lists
Azerbaijan X
Bolivia X
Burkina Faso X
Burundi X X X
Cambodia X X
Cameroon X
Chad X X
Central African
Republic X
Colombia X X X
Comoros X
Congo,
Democratic
Republic of X X X
Congo,
Republic of X

Djibouti X
East Timor X X
Ecuador X X
Egypt X X
Fiji X X

Gambia, The X
Georgia X X X
Ghana X
Guinea X X
Guinea-Bissau X X

Guyana X
Haiti X X
Indonesia X X
Iran X X
Iraq X X
Israel X X
Kyrgyz Republic X
Kiribati X
Laos X X
Lebanon X X X
Liberia X X
Moldova X
Myanmar/Burma X X
Niger X X
Nigeria X
Pakistan X
Papua
New Guinea
Paraguay X
Philippines X X
Rwanda X

(table continues next page)

158
Table 7.1 (continued)
Not
Failed states included
Failing states
on CPIA
Country
Rotberg3 Rotbergd
Polity IVb
Sao Tome and Polity IVe
Principe CPIAc X
Senegal X lists
Sierra Leone X X CXPIA1
Solomon Islands X X
Somalia X X X

Sri Lanka
X X
Sudan X X X
Tajikistan X X X
Thailand X X
Togo X

Tonga X
Turkey X X
Uganda X
Uzbekistan X
Vanuatu X
Yemen X
Zimbabwe X X
11
Total 8 8 21 20 23 18

CPIA = Country Policy and Institutional Assessment rating

a. Rotberg defines “failed” states as “tense, deeply conflicted, dangerous and contested bitterly by
warring factions.”
b. State failure includes four types of events according to Polity IV: revolutions, ethnic wars, adverse
regime changes, and genocides or politicides.
c. In the fifth quintile in both the “overall” category and the “public-sector management and institu¬
tions” category of the CPIA rating.
d. States with one or more failed state features.
e. Two or more temporarily linked wars or crises.
f. In the fifth or fourth quintile of the “overall” category or the “public-sector management and institu¬
tions” category of the CPIA rating and defined as having not failed.
g. Countries rated as politically failed or failing states that are not included in CPIA ratings.
Sources: CPIA ratings in IDA (2003); Rotberg (2003); Gurr, Harft, and Marshall (Polity IV study)
(2003).

economically "failed" states; the other 23 countries in either the fourth or

fifth quintiles on either criterion were classified as economically "failing."
Table 7.1 shows that only two countries (Angola and Sudan) of the 11
cited as failed economically are listed as failed politically by Rotberg or
Polity IV. Five others are listed as failing politically (Haiti, Laos, Solomon
Islands, Tajikistan, and Zimbabwe).23 Three of the nine countries that are
listed as failed politically are classified as failing economically (Burundi,

23. Zimbabwe today would probably be classified as "failed" as opposed to "failing" politi¬
cally. The remaining four countries characterized as failed economically, but which are neither
failed nor failing politically, are the Central African Republic, Nigeria, Togo, and Uzbekistan.

COMBATING GLOBAL “PUBLIC BADS” 159

 Box 7.1 “Failed” or “failing” states
The Polity IV study (Gurr, Harft, and Marshall 2003) provides a time series of “failed
states” — states so ridden by conflict as to be essentially ungoverned — and what it calls
“complex” or failing states, which are sufficiently conflicted as to be in danger of becom¬
ing failed states. Although continuity is considerable, a state may be classified as failed
or failing in one year but not similarly classified the next or the previous year. Of the eight
states scored by Polity IV as failed in 2002, five were scored as failed or failing a decade
earlier in 1 992: Afghanistan, Angola, Burundi, the Democratic Republic of the Congo, and
Somalia. Liberia also was scored as failed in 1992 by Polity IV, but not in 2002, when it
was scored as failed by Rotberg (2003), the other index used in this examination. In addi¬
tion, three states were scored by Polity IV as failed or failing in 1 992 that were scored as
failing in 2002: Algeria, Lebanon (scored as failed in 1992), and Tajikistan. The only coun¬
try that was scored as failed or failing in 1 992 and was not in either category in 2002 was
Peru.
Polity IV classified 87 countries as failed or failing one or more years between 1 992 and
2001 . More than 75 percent of those classifications were on the combined list for 2002
shown in table 7.1 . The exceptions were two failed states (Bosnia over nine years and
Lesotho for two years) and six failing states: Albania (one year), Armenia (two), Belarus
(two), Nepal (one), Peru (one), and Zambia (one). More than 80 percent of the countries
that Polity IV lists as failed in at least one year between 1992 and 2001 are on the list in
table 7.1 as failed or failing in 2002. Again, Bosnia and Lesotho are the exceptions. Those
countries most frequently listed as failed (five years or more) are Afghanistan, the
Democratic Republic of the Congo, Lebanon, Sierra Leone, and Somalia.

the Democratic Republic of the Congo, and Sierra Leone).24 Most of the
remaining countries classified as failing politically had estimated incomes
per capita in 2002 too high to be included as IDA borrowers, for which the
threshold in fiscal year 2004 was $865.
For assessments of the seriousness of money laundering in various na¬
tional jurisdictions, we used information from the money-laundering and
financial crimes section of the International Narcotics Control Strategy Report
(INCSR) of the US State Department (2003), as well as the results of the
Financial Action Task Force's Non-Cooperative Countries and Territories
(NCCT) initiative.
The INCSR classified 190 national jurisdictions as to whether they are of
"primary concern" or "concern" with respect to money laundering, with a
final category sued for jurisdictions that are just "monitored." Jurisdictions
of primary concern are major money-laundering countries defined by statute
as those "whose financial institutions engage in currency transactions

24. The other four countries classified as politically but not economically failed were explic¬
itly not rated by the CPIA process, even though in principle their per capita incomes are low
enough that they could borrow from the IDA. Myanmar /Burma, Liberia, and Somalia are
inactive IDA borrowers, and Afghanistan lacked the information necessary to receive a CPIA
rating as of 2002. East Timor, a new IDA borrower, was not included in the 2002 CPLA exer¬
cise because of a lack of information.

160 CHASING DIRTY MONEY

Table 7.2 Inclusion of countries in INCSR ratings
and FATF reviews (number of countries)
INCSR rating

FATF treatment 78 Total

Primary concern3
Monitored0 114
Not reviewed 12 ern15
Conc26
24 76
Reviewed 41 9
105 1 0 236
Failedd 137 0 17
Failed/passede
197 3
Passed1
FATF members 5 6 30
53
Total 50
190
87
INCSR = International Narcotics Control Strategy Report
FATF = Financial Action Task Force

a. Jurisdictions whose financial institutions engage in currency transactions involving signifi¬

cant amounts of proceeds of narcotics trafficking.
b. Jurisdictions where money laundering takes place but is not considered a critical problem.
c. Jurisdictions reviewed that do not pose an immediate concern.
d. Countries that have failed to satisfy the FATF Non-Cooperative Countries and Territories
(NCCT) criteria.
e. Countries that initially failed to satisfy the NCCT criteria, but subsequently satisfied them.
f. Countries that initially satisfied the NCCT criteria.
g. Countries and territories that are members of the FATF and have not been subject to NCCT
reviews, but that have been subject to FATF mutual evaluations.

involving significant amounts of proceeds of narcotics trafficking."

However, the INCSR recognizes the difficulty of distinguishing between
the proceeds of narcotics trafficking and the proceeds of other serious
crime and, therefore, explicitly applies a broad definition of the scope of
money laundering.
The FATF's NCCT initiative — dubbed the "name and shame" process
because it calls countries to task for failing to cooperate with global
anti-money laundering efforts — is somewhat different in that it focuses on
a jurisdiction's compliance with criteria regarding its legal and regulatory
framework, international cooperation on money laundering, and resources
allocated to anti-money laundering activities (see box 4.3 in chapter 4).
Table 7.2 summarizes the overlap between the INCSR and FATF rating
and review processes. While there are important differences in coverage,
the FATF review process includes a substantial number of countries and
territories identified in the INCSR as jurisdictions of "concern" with respect
to money laundering (about 50 percent) and a higher number of jurisdic¬
tions of "primary concern" (about 80 percent). In all, 80 percent of FATF
members subjected to multilateral review, including the United States, are

rated by the INCSR as jurisdictions of "primary concern" or "concern"

because of the substantial amount of money laundering that occurs there
despite having adequate AML regimes in place. The six FATF members
rated as needing only to be "monitored" by the INCSR are Denmark,
Finland, Iceland, New Zealand, Norway, and Sweden.

COMBATING GLOBAL “PUBLIC BADS” 161

Table 7.3 Failed and failing states and INCSR ratings
(number of countries)
INCSR rating
Primary
Total Not
States and territories rated rated Total
Concern6 Monitored0
concern3
Political
1 0 5 6 3 9
Failed6 10
4 17 4
Failinge
Economic (additional) 31 35
1 0 2 3 1 4
Failed'
Failings 0 2 9 11 3 14

Total failed or failing 12 6 33 11

35 128
Not failed or failing 52 124
51 4 62
47 43
37 175 190
T otal states 85 15
Dependent or autonomous
territories 6 7 2 15 150 15
190 205
Total 53 50
87
INCSR = International Narcotics Control Strategy Report

a. Jurisdictions whose financial institutions engage in currency transactions involving signifi¬

cant amounts of proceeds of narcotics trafficking.
b. Jurisdictions where money laundering takes place but is not considered a critical problem.
c. Jurisdictions reviewed that do not pose an immediate concern.
d. Rotberg (2003) combined with Polity IV study (Gurr, Harft, and Marshall 2003); see notes
a and bin table 7.1.
e. Rotberg combined with Polity IV study (Gurr, Harft, and Marshall 2003); see notes band e
in table 7.1.

f. In the fifth quintile in both the “overall” category and the “public-sector management and
institutions” category of the Country Policy and Institutional Assessment (CPIA) rating.
g. In the fifth or fourth quintile of the “overall” category or the “public-sector management and
institutions” category of the CPIA rating, and rated as not failed.
Sources: Country Policy and Institutional Assessment (CPIA) ratings in IDA (2003); Rotberg
(2003); Gurr, Harft, and Marshall (Polity IV study) (2003); US Department of State (2003).

Because of the broader coverage of the INCSR, its rating system is used
here to assess the link between money laundering and "failed" or "failing"
states. Table 7.3 shows that only 18 (35 percent) of the 51 politically or eco¬
nomically failed or failing states that the INCSR rated are classified as juris¬
dictions of "primary concern" or "concern."25 This percentage is less than
the 52 percent of the 175 countries covered by both classifications that are
of "primary concern" or "concern." Those 18 "failed" or "failing" states
account for 20 percent of the 90 countries of "primary concern" or of "con-

25. Of the limited number of relevant FATF reviews — nine of the 62 failed or failing states —
seven states did not pass their initial reviews. The two that passed were Turkey, which is a
FATF member and is classified by Polity IV as politically failing, and Vanuatu, classified as eco¬
nomically failing on the basis of the CPIA review. However, the seven failed or failing states
account for only about 30 percent of the 23 jurisdictions that initially failed their FATF reviews.

162 CHASING DIRTY MONEY

cern" in the INCSR, less than the 29 percent of the 175 countries covered by
both classifications that are "failed" or "failing."26
This lack of a close association between money laundering and failed states
should not be particularly surprising. As argued by Donato Masciandaro
and Allesandro Portolano (2002), money launderers or their clients attach
high importance to keeping their money safe and like to exploit legal pro¬
tections to do so, which is no easy task in politically or economically failed
or failing states. The only politically failed state that is rated by the INCSR
as of primary concern is Myanmar/ Burma, and the only economically
failed state is Nigeria. Both failed their NCCT reviews. Myanmar/ Burma
might be considered in a different category of rogue states where there is
order accompanied by violence inflicted by the authorities. Such states —
North Korea could be considered to be in the same category — might be con¬
tent to flout international norms by dealing with criminal gangs that operate

globally.27

Money Laundering and Corruption

Table 7.4 presents a cross-tabulation of the assessment of the seriousness

of money laundering on the basis of the INCSR, and the assessment of the
extent of corruption on the basis of the Corruption Perceptions Index (CPI)
published by Transparency International (2003). A total of 23 (43 percent)
of the 53 countries that are in the fourth or fifth quintiles on the CPI and are

rated in the INCSR are classified as jurisdictions of "primary concern" or

"concern" as part of that rating process, less than the 56 percent of the
130 countries covered by both classifications that are classified as countries
of "primary concern" or "concern." Those states account for 32 percent of
the 73 jurisdictions that are classified as a "primary concern" or "concern"
by the INCSR and also are ranked by the CPI, less than 40 percent of the
130 countries covered by both classifications that are in the fourth or fifth
CPI quintiles.28 The association between money laundering and corruption
is not quite as weak as that between money laundering and failed states,
but it is still less than would be expected statistically if there were a posi¬
tive association.

26. The calculation excludes dependent or autonomous territories covered by the INCSR that
presumptively cannot be full-fledged failed or failing states, such as the Cayman Islands and
the Isle of Man.

27. North Korea, however, did not make any of our lists of failed or failing states in 2002, and

along with South Korea is rated by the INCSR only as a country of "concern" with respect to
money laundering.

28. The association is essentially neutral using the FATF reviews for cross-tabulation with the
CPI, but these cover a much smaller sample of countries. Of the 20 countries with CPI ratings
that did not pass their first reviews under the FATF's NCCT Initiative, eight (40 percent) were
rated in the fourth or fifth CPI quintile.

COMBATING GLOBAL “PUBLIC BADS” 163

Table 7.4 Money laundering and corruption (number of countries)
INCSR rating
Primary Total Not
rated rated
CPI3 quintile Total
15
concern15 Concern0 Monitoredd 28
Firth 6 4 15 25 3 28
Fourth 6 117 28 0
Fourth and fifth 19 30 53 3 56
12
First to third 77 0 133
77
31
43 130
27
Total CPI rated 30 3
10 57
20 30 60 12 72
Not ratede
190
Total 53 50 15 205
87
INCSR = International Narcotics Control Strategy Report
a. Corruption Perceptions Index (CPI), Transparency International (2003).
b. Jurisdictions whose financial institutions engage in currency transactions involving signifi¬
cant amounts of proceeds of narcotics trafficking.
c. Jurisdictions where money laundering takes place but is not considered a critical problem.
d. Jurisdictions reviewed that do not pose an immediate concern.
e. Rated by the INCSR but not by Transparency International. Includes 14 dependent and
autonomous territories that are rated by the INCSR.
Sources: INCSR, US Department of State (2003); CPI, Transparency International (2003).

As with the link between money laundering and failed states, many
countries for which the perception of corruption is high do not show up as
jurisdictions regarding which there is great concern about money laun¬
dering. The explanation may be similar: where there is a lot of corruption,
with the possible exception of corruption principally by senior government
officials, the proceeds of other crimes, over and above "living expenses,"
may not be safe. However, the data reviewed here do identify 11 countries
that are of "primary concern" or "concern" with respect to global money
laundering, are politically and/ or economically "failed" or "failing" states,
and also score poorly when it comes to corruption: Bolivia, Myanmar/
Burma, Ecuador, Haiti, Indonesia, Lebanon, Nigeria, Pakistan, Paraguay,
the Philippines, and Yemen.
The lack of evidence of linkages between failed states, money launder¬
ing, and corruption does not imply that there is no role for the global AML
regime's prevention and enforcement efforts in dealing with failed states.
The role, however, may be ancillary to broader efforts directed at bringing
about profound structural changes in these countries' governments and
societies. A credible AML regime requires the political and institutional
capacity necessary to put it in place. Implementation of an effective AML
regime also requires resources that failed states generally lack, either be¬
cause they have been stolen or because the government has limited or no
revenue-raising capacity.
Consider the case of Afghanistan, which is still classified as a failing state,
though one would hope a postconflict, recovering one. The 2003 INCSR

164 CHASING DIRTY MONEY

rated it only as a country to be "monitored," and Transparency International
does not yet cover it in its ratings of corruption, presumably because there
is insufficient international business activity to carry out the necessary sur¬
veys for the ranking process. Its apparent progress notwithstanding,
Afghanistan remains a potential venue for money laundering because of its
location and the fact that the opium sector may account for about half of
overall GDP (IMF 2003a, 2). A summary report in 2003 of the views of IMF
executive directors reflects these concerns. While commending the creation
of a supervision department in the central bank, the executive directors
"emphasized the importance of further developing regulatory and super¬
visory capacities and restructuring and privatizing state banks; urged
tighter control of informal mechanisms of financing, the introduction of
anti-money laundering, and controlling the financing of terrorism legisla¬
tion; . . . [and] were concerned about the serious risks posed by the rise in
poppy cultivation and the production of opium in Afghanistan to the levels
of the late 1990s." In encouraging Afghan authorities to prepare the ground¬
work for privatization of state-owned companies, the executive directors
"saw several key areas of reform as preconditions, including putting in
place a functioning financial system, a market-oriented regulatory frame¬
work, and a functioning and fair legal system to firmly establish the rule of
law and the security of property rights" (IMF 2003a, 3M).
In this context, it is not difficult to understand why it may be a number
of years before Afghanistan graduates from the status as failed or failing
state and scores well on anti-money laundering and corruption indices.

The NCCT Process and Global AML Standards

Besides strengthening compliance with the global AML regime, the FATF's
Non-Cooperative Countries and Territories initiative has played an impor¬
tant role in distinguishing between jurisdictions that reject making any
effort to comply with international norms and standards, and those juris¬
dictions that accept or can be persuaded to accept such a responsibility. A
country that can be persuaded to conform to global norms may be less
likely to become a failed or failing state.29 Two critical questions are: How
effective has the NCCT approach been in terms of using the threat of ad¬
verse publicity and application of countermeasures to motivate reform?
How much improvement has there been as a result of the NCCT process
and what can be expected going forward?

29. The NCCT "name and shame" process is a soft type of targeted sanction. The NCCT
process has been reasonably successful because the sanctions focus primarily and narrowly
on the financial sector, using market forces as incentive devices, and have substantial multi¬
lateral support. See Hufbauer, Schott, and Elliott (2004) for a full analysis of economic sanc¬
tions and the reasons for their successes and failures.

COMBATING GLOBAL “PUBLIC BADS” 165

 For the first question, it is instructive to look at the case of Nauru, a
Pacific island nation with a population of about 12,500 and an income per
capita estimated at $5,000. Nauru is so small that it is not a member of
either the IMF or the World Bank, and therefore is not included in the CPI A

rating. The INCSR, however, listed Nauru as a country of "primary con¬

cern" regarding money laundering, and the country failed its first NCCT
review in June 2000. Nauru passed AML legislation in August 2001 at least
partly in response to the FATF review, but the provisions in that legislation
with respect to supervision and regulation did not cover the offshore bank¬
ing sector. As a result, the FATF in December 2001 recommended the appli¬
cation of countermeasures to Nauru, which prompted the government to
amend its AML legislation to cover offshore banks, but did not lead to their
actual licensing and supervision. The legislation was further amended in
March 2003, but the FATF still would like Nauru "to take additional steps
to ensure that previously licensed offshore banks are no longer conducting
banking activity and no longer are in existence" as a condition for consid¬
ering the removal of countermeasures (FATF 2003b, 12).
What has all this activity accomplished with respect to Nauru? After all,
no money actually flows to or through the island in any physical sense in
connection with its offshore banking sector; the flows are virtual, electro¬
nic, and controlled from elsewhere.30 Reportedly, there has been a decline
in the number of shell banking organizations, or at least in the rate of
increase — as many as 400 such banks along with numerous other shell-
type entities were said to have been registered in Nauru at one point (FATF
2003b, 11). If they so choose, countries can forbid their banks or nonbanks
from dealing with those organizations to the extent that they know their
names or whether they are entities known to be incorporated or registered
in Nauru. In effect, Nauru and its offshore banking entities can be largely,
if not completely, quarantined.
There is some hope that Nauru may further amend its legislation and
implementation of it in order to cooperate more fully with the FATF and
international law enforcement authorities. While it appears that Nauru has
to some extent been made an example for purposes of a "demonstration
effect," it has a way to go.31 On the other hand, it seems unlikely that a large
amount of proceeds from crimes in industrial countries has been laundered
there — but it is impossible to prepare even a rough estimate without more
cooperation from the Nauruan authorities. What is clear is that Nauru has
made some progress, and that its money-laundering role has been at least
somewhat reduced as a result of the application of countermeasures.

30. Wechsler (2001, 48) notes that Nauru banks were involved in some of the money flows
from Russia through the Bank of New York in the late 1990s.

31. We met with the director of Nauru's police force (Bernard Junior Dowiyogo) in January
2003. He informed us that when requests came to him from abroad seeking cooperation in
international investigations, he passed them on to the responsible government agency and
had nothing further to do with them!

166 CHASING DIRTY MONEY

 Masciandaro and Portolano (2002) worry that the NCCT process is not
uniformly applied and that countries may not effectively implement their
AML regimes once they have passed the NCCT review. They argue for
complementary measures better to integrate jurisdictions into the global
financial system, as well as for the development of stronger sanctions for
noncompliance. Guy Stessens (2000) notes two types of "noncooperation"
with global AML standards. The first may be an unwillingness to cooperate,
in which case the appropriate response is to quarantine the country's finan¬
cial system to encourage it to modify its behavior. The second may be insuf¬
ficient capacity due to governance problems or a lack of financial or
nonfinancial resources. If the world is serious about establishing a compre¬
hensive and global AML regime, this latter type of noncooperation would
seem best addressed with technical assistance.

Regarding the broader issue of moving ahead with promoting compli¬

ance with global AML standards, it should be noted that 17 jurisdictions
that initially failed FATF reviews subsequently passed.32 Most were off¬
shore financial centers such as the Cayman Islands, Lebanon, Liechtenstein,
and Panama, but the list includes a number of countries of larger signifi¬
cance such as Egypt, Hungary, Israel, Russia, and Ukraine.33 As of early
2004, three sizable countries remained on the FATF's noncooperation list:
Indonesia, Nigeria, and the Philippines. These countries, whose combined
population tops 350 million, have made sufficient progress since their
FATF reviews so as not to be subject to countermeasures, with which the
Philippines was threatened in 2001. It remains to be seen, however, how
these countries will perform or be dealt with in the future. If they fail to
meet international standards, it will be much more challenging politically
and technically to apply a Nauru-type quarantine to them.
In addition, the monitoring of observance of global AML standards was
largely turned over to the IMF and the World Bank following development
of a common methodology with the FATF (chapter 4). This transfer of
responsibility has the benefit that reviews are now conducted by organi¬
zations with near-universal membership, rather than by a self-selected
group of countries. Because of the broader coverage, the reviews are con¬
ducted at arm's length rather than as a peer-review process that sometimes
limits frank criticism. The growing number of FATF and FATF-style re¬
gional groups also participate in and sometimes conduct the reviews. At
the conclusion of the 12-month IMF/World Bank pilot project, a joint re¬
port concluded that the wealthier the country, the more developed its

32. The FATF has not excluded the possibility, however, of listing a country or territory as
uncooperative if there is backsliding.

33. Russia made it all the way from FATF's "name and shame" list to full membership in the
FATF within a very short period principally because it had met the "minimum requirements"
for membership with respect to its AML regime and because of its classification as a "strate¬
gically important country" (FATF 2003a, 10-18).

COMBATING GLOBAL “PUBLIC BADS” 167

AML regime, though jurisdictions regardless of financial resources had
shortfalls with respect to frameworks for combating terrorism financing
(IMF 2004b, 2). The most frequent implementation weaknesses included
intragovernmental coordination problems, ineffective law enforcement,
weak supervision, inadequate controls in financial firms, and shortfalls in
the area of international cooperation.
Countries must "volunteer" for the IMF/ World Bank reviews, and while
countries can choose not to have the results published, the reviews are
available to other member governments. Still, while the peer pressure,
transparency, and accountability of the process are enhanced by the in¬
volvement of international institutions such as the IMF and the World
Bank, there is today less capacity to apply meaningful pressure than when
the reviews were carried out exclusively by the FATF.
For example, China, Colombia, India, Pakistan, Thailand, and Venezuela
are six large and politically important countries that are rated as a "primary
concern" regarding money laundering by the INCSR but were not re¬
viewed by the FATF.34 Will they volunteer or be induced to volunteer for
an IMF /World Bank review of their compliance with international stan¬
dards on anti-money laundering and combating terrorism financing, and
will they volunteer or can they be shamed into allowing those reports to be
published?35 In fact, a number of IMF executive directors (IMF 2004a, 4)
commented critically on the fact that the FATF has reserved the right to
conduct further rounds of its NCCT process on its own. Their view was
that if the FATF were to take the initiative to place new countries on its
NCCT list, the IMF should reconsider its collaborative work in this area.

They argued that the IMF's work on compliance with global standards and
codes is based on the uniform, voluntary, and cooperative nature of its
activities. Moreover, countries may be reluctant to volunteer for reviews if
they risk being subjected to sanctions as a consequence. On the other side,
a report by the Council on Foreign Relations (2004) has called for reinvest¬
ing the FATF with the authority to "name and shame" jurisdictions for
falling short in implementing their regimes for combating money launder¬
ing and the financing of terrorism.
In India, an anti-money laundering law passed in early 2003 has yet to
be fully implemented, and questions have been raised about several ele-

34. China and India have been targeted for membership by the FATF and will be subject to
FATF-style mutual evaluations if they become members. China has stated (Zhou 2004, 5) that
the FATF should be more representative.

35. The United Kingdom set a good example by volunteering to be reviewed under the
IMF/ World Bank program and allowing publication of the report. The United States also vol¬
unteered for a review of its compliance with standards and codes on money laundering and
the financing of terrorism, but not for the broader financial-sector assessment program. The
review was not carried out, and the United States has since "volunteered" for a FATF mutual
assessment, which is much less rigorous.

168 CHASING DIRTY MONEY

merits of the Indian AML regime as it is being put in place, among them
whether suspicious activity reports will be required or voluntary. Inter¬
views in India for this study suggest that money laundering is not per¬
ceived as a major problem there, which may be why motivation is lacking
to adopt and implement full-scale global AML standards. For example,
drugs are said not to be a problem.36 The informal money transfer system,
though technically illegal for international transactions, is widely used in
India because of the country's tight exchange regime and other types of
financial controls. But as these controls are in the process of being relaxed,
the perception seems to be that whatever problems there might be with
money laundering will likely fade away.
Apart from the question of whether countries will willingly subject
themselves to AML regime compliance reviews is the question of whether
the IMF — known for its tough stands on countries' monetary and fiscal
policies — can be equally tough on money laundering. The IMF has lever¬
age only when a country comes to it to request financial assistance, but it
is unlikely that improving implementation of an AML regime can be justi¬
fied as a condition for restoring macroeconomic stability — much less gain
the support from IMF member countries necessary for adoption.37
Take the example of the Philippines, whose AML regime was cited by
the FATF as inadequate, and which was subsequently threatened with
countermeasures. Compliance with FATF standards, however, could not
realistically have been required of the Philippines for access to IMF re¬
sources during 2000-03, when that country had an IMF-supported macro-
economic adjustment program.38 The link between money laundering and
macroeconomic stability is simply not sufficiently well developed to sell
such a linkage.

Another example is Vietnam, rated by the INCSR as a country warrant¬

ing "concern" but not "primary concern" regarding money laundering. An
IMF review of the Vietnamese economy in 2003, when the country was
operating under an IMF-supported adjustment program, urged passage
of "an effective anti-money laundering decree by year-end, followed by
rigorous implementation," and also emphasized "the importance of an

36. The actual level of drug trafficking in India is extremely difficult to estimate. The UN fig¬
ures suggest that India has more heroin addicts than any other nation, the combination of a
moderate prevalence rate and a huge population, but the estimates are highly conjectural
because of the lack of a national survey of drug use.

37. One possible exception might be assistance to failed states, where a case can be made that
a substantially improved AML regime could be key to the rehabilitation process, as in the case
of Afghanistan outlined earlier in this chapter.
on a
38. Wechsler (2001, 52) does point out, however, that when the United States abstained
vote on an IMF program for the Philippines in 2000 because of concerns that the country had

not fulfilled its previous fiscal, monetary, and financial-se ctor commitment s, the vote was
interpreted as a sign of concern about the weak Philippine AML regime.

COMBATING GLOBAL “PUBLIC BADS” 169

independent external audit of the central bank in accordance with inter¬
nationally accepted standards" (IMF 2003b, 4). Vietnam's reluctance to
agree to such an audit was holding up an IMF disbursement at the time,
and the Fund's support for Vietnam's adjustment program eventually was
terminated in April 2004 because of this issue, which had been judged to
be central to the IMF's macroeconomic policy concerns.
In sum, the answer to the two questions posed earlier about the effec¬
tiveness and improvement associated with the NCCT process is that there
has been progress. The global AML regime is now well established, and
FATF standards are broadly accepted and have been endorsed by the IMF
and World Bank. However, the success rate is far from 100 percent, and
effective implementation is difficult both to achieve and to assess.

Conclusion

While the surveys and databases cited in this section have shown some
broad connections between global anti-money laundering initiatives and
failed states, the major instruments for dealing with such states lie outside
the AML regime. At the same time, unless the global AML regime is used
proactively as a prevention tool, failing or failed states may become huge
gaps in the global AML regime, particularly in the case of large countries
such as Nigeria. Moreover, the links between large-scale kleptocracy,
money laundering, and the failure of states appear to be stronger than
those between garden-variety corruption, money laundering, and state
failure. That finding confirms a role for the AML regime in efforts to aid
failing and failed states, but it also suggests that the deeper connections
between the various global "bads" need to be studied further.

170 CHASING DIRTY MONEY

 8
Improving the Global AML Regime

While the initial focus of the US anti-money laundering (AML) regime was
the intersection of organized crime and drugs, the focus subsequently
widened to include the proceeds from many crimes. Today the AML re¬
gime has multiple goals, including those identified throughout this study:
reducing the incidence of crime, which includes but is far from limited to
drug-related crimes; protecting the integrity of the core financial system;
and controlling a number of global "public bads," particularly terrorism,
corruption and kleptocracy, and failed states.
From its earliest days, money laundering has frequently involved cross-
border transactions, since moving dirty money across borders is an effec¬
tive way to disguise its trail. With the increased globalization of the financial
system, money laundering has evolved into an activity affecting societies
and financial systems everywhere in the world. Money laundering is a prin¬
cipal area of abuse of the global financial system — what has been called a
dark side of globalization.
Based on our analysis of money laundering and assessment of the global
AML regime, this chapter presents conclusions and recommendations for
improving the global regime in seven areas: the appropriate scope of the
AML regime, the regime as a means to various ends, challenges to US
implementation of the AML regime, challenges to global implementa¬
tion of the regime, opportunities for international cooperation, implica¬
tions for domestic law enforcement, and a research agenda for an ongoing
and comprehensive assessment of the regime as it adjusts to changing
conditions.

171
Scope of the AML Regime

The scope and detail of the US and global AML regimes built up over a
period of less than two decades is impressive. If diligently implemented on
a global basis, the 2003 Forty Recommendations issued by the Financial
Action Task Force will expand the global regime even further, particularly
the prevention pillar.
The evolution of the enforcement pillar, with the exception of expanding
the number of predicate offenses covered by the AML regime, has been less
dramatic. In many jurisdictions enforcement still is minimal. Moreover, as
is often true of structures resting on two pillars, there is tension between
them including which should receive greater emphasis in implementation.
It is unlikely, and may well be undesirable, that the global AML regime
will continue to expand at its recent pace. Even in high-profile areas such
as terrorism, there are practical limits to extending the prevention pillar
and resource constraints to aggressive use of the enforcement pillar.
Money launderers can be expected to change their tactics in response to
enhancements in the global AML regime, so the task of combating money
laundering is not likely to get any easier in the years ahead. However, it is
reasonable to ask just how much can be expected of governments and the
private sector. Constructing a zero-tolerance regime that was consistent
with the smooth flow of finance would be costly and politically unaccept¬
able. At the margin, the broadly defined costs of extending the regime are
not likely to be worth the modest reduction in money laundering and small
contributions to ultimate goals to which such an extension would con¬
tribute. The challenge, especially with respect to prevention, is to identify
the margin where costs are roughly equal to expected benefits.
The 2003 FATF Forty Recommendations will provide additional chal¬
lenges in terms of compliance with, and implementation of, the global
AML regime. The standard for national regimes has been raised substan¬
tially through the explicit extension of the global regime to nonfinancial
businesses and certain professions. Applying the basic elements of the pre¬
vention pillar (customer due diligence and reporting) to certain profes¬
sionals such as lawyers and accountants will be resisted, and monitoring
and enforcing compliance (supervision and sanctions) will be even more
difficult. Almost all jurisdictions will fall short of perfection with respect to
design and certainly with respect to implementation.
The issue will be the impact of such shortfalls on overall acceptance of
the global regime. For example, if the United States is unwilling to apply the
prevention pillar of the AML regime to lawyers and accountants, will this
adversely affect US capacity to stiffen controls on terrorist financing? The
answer surely is yes, but a sufficient case has not yet been made to overcome
US political resistance to applying the prevention pillar more fully to these
professions. The examples cited in FATF (2004b, 24-27) of lawyers and
accountants acting as "gatekeepers" relate almost exclusively to their direct
172 CHASING DIRTY MONEY
involvement in facilitating money laundering. As they are already liable for
such activities under criminal statutes, little would be gained by subjecting
them to due diligence and reporting requirements.
Moreover, the costs that would be imposed by comprehensively extend¬
ing the AML regime to lawyers, in particular, could be substantial. Law
offices would be required to have a compliance officer and to develop mech¬
anisms to implement a customer due diligence (CDD) program, which
would involve training and internal and external audits. In addition, many
lawyers are solo practitioners or work in small firms with limited financial
and technical resources for tracking and linking dispersed information on
clients and their activities. Finally, lawyers' insurance requirements no doubt
would increase. Since there is no study on the role of lawyers in money
laundering in the United States, it is impossible to assess whether the cost
of extending the US AML regime to the legal or other professions is worth
the benefits. At the very least, however, professional organizations should
enhance their educational activities and standards in this area.

Money laundering is likely to evolve over the years in response to tech¬

nological, economic, and social developments. The AML regime will cor¬
respondingly have to be adapted, and that task will be made difficult by
the lack of hard information on either the costs or the effectiveness of the
regime. Although the financial and nonfinancial costs of the current regime
are most likely bearable for advanced economies and large institutions, they
loom far larger for less developed economies and smaller institutions. The
financial costs for institutions are to a considerable extent fixed, as in the
case of having to adapt existing data management systems to meet new
reporting requirements. These fixed costs are more difficult for smaller insti¬
tutions to absorb. The nonfinancial costs are often associated with dead¬
weight losses that are more burdensome for poor countries. In effect,
combating money laundering is a luxury good, a reality that argues for
increased technical assistance financed by the countries that particularly
value the benefits of the AML regime. It also argues for direct financial assis¬
tance to countries to increase the probability of effective implementation.
The designers of the AML regime are aware of these circumstances and
constraints, at least as they affect decisions involving financial institutions in
the major centers. Consistent with the 2003 FATF Forty Recommendations,
US authorities have embraced a risk-based approach in several areas. US
financial institutions are expected to apply one level of CDD scrutiny to nor¬
mal customers — having presorted them using a risk-based approach to their
actual and potential use of the institution— and a higher level for "politically
exposed persons," based essentially on the judgment of the institution. The
level of scrutiny applied to potential sources of terrorist financing is higher
still, as the authorities require institutions to do name checks for individuals
and organizations against lists supplied by the government.
Thus, it is possible to adjust the calibration of the risk-based system.
However, the nature and magnitude of the challenge is such that errors

IMPROVING THE GLOBAL AML REGIME 173

will occur. On the one hand, money laundering or terrorist financing will
slip into and through the financial system. On the other, individuals will
be wrongfully denied access to the financial system or subject to harassing
investigations, undermining the self-policing of compliance on which the
regime primarily relies. If the AML regime erects excessive barriers for
legitimate customers, transaction costs rise, the reputations of businesses
are unnecessarily tarnished because of false links to money laundering,
and the regime itself comes under stress.
Mariano-Florentino Cuellar (2003, 453) offers a completely different take
on the scope of the US regime and, implicitly, the global one. He argues that
the regime's scope — particularly the long list of predicate crimes on which
money-laundering charges can be based — is too broad. As a result, law
enforcement officials use the AML regime to prosecute those who commit¬
ted the underlying crime, but are distracted from focusing on the disruption
of criminal finance, which he argues was the original intent of US money¬
laundering legislation. He proposes revising the US money-laundering
statutes to provide a much narrower focus.1
The Cuellar position is implicitly predicated on an unsophisticated appli¬
cation of the market model of money-laundering activity and an associated
view that there are large numbers of stand-alone money launderers offering
their services to criminal networks. The evidence reviewed in chapters 3
and 5 of this study suggests that stand-alone money launderers are the
exception rather than the rule. In general, money laundering is an integral
part of the underlying offense, or the money launderer is integrated in an
overall criminal operation.
Cuellar (2003, 456) advocates the use of information technology and arti¬
ficial intelligence to review patterns of wire transfers to create "money pro¬
files" and uncover the operations of criminal finance. But it is questionable
whether developing such unique profiles in the transactions of money
launderers is even possible. The US Congress Office of Technology Assess¬
ment (OTA 1995), whose report Cuellar cites as supporting his view, is also
skeptical. Robert Litan (2004) makes a similar proposal that is more promis¬
ing as a research project: the US Financial Crime Enforcement Network
(FinCEN) could examine a broad range of financial transactions by various
types of criminals to look for patterns that could be shared with private-
sector institutions and might enhance the detection of money laundering.
If other information were brought to bear in the examination of wire
transfers, for example, origins and destinations, then Cuellar's approach
might be more fruitful. But the OTA report notes that this would likely lead

1. The National Association of Criminal Defense Lawyers (2001, 1) made a similar proposal
to rewrite the US money-laundering statutes. The association is critical of the "alarming
expansion" of the scope of those statutes that, as interpreted and applied, subject unwary indi¬
viduals and businesses to "overreaching investigations and prosecutions unrelated to drug
trafficking or organized crime."

174 CHASING DIRTY MONEY

to a large number of false positives and raise a host of privacy concerns
both domestically and internationally. Even if the technology were avail¬
able at low cost to provide a passport and vaccination certificate for each
dollar that moves through the global economy and the financial system, the
suppliers and purchasers of goods and services and the financial system
would balk at participating in such a regime. A "Big Brother" approach to
money laundering, much less other issues, generally is not acceptable in
democratic societies. The US political debate about renewing the USA
PATRIOT Act in 2005 illustrates that such an approach is not universally
accepted even in the special case of terrorism.
Moreover, if the AML regime becomes overloaded with requirements,
pressures will mount to roll it back in all dimensions. In the mid-1990s, the
US AML regime was scaled back by Congress with respect to the submis¬
sion of suspicious activity reports (SARs) in response to criticism that some
requirements were excessively burdensome. Cuellar (2003), among others,
advocates trimming the AML regime, and that view has considerable sup¬
port in the US Congress.

Means to What Ends?

The global AML regime is a means to an end and a tool to be used to help
achieve certain goals, but not an end in itself. In the United States as well
as many other jurisdictions, the principal initial motivation in criminaliz¬
ing money laundering was to support the war against illegal drugs and
associated criminal gangs.
This rationale was based implicitly on the market model for money¬
laundering services and the hypothesis that there were large numbers of
money launderers, associated with financial institutions, offering their
third-party services. The policy implication of this view was that closing
down the money launderers would sharply curtail the underlying crime.
As economists, we embarked on this study from the same starting point.
However, the evidence presented in chapters 3 and 5 suggests that the
applicability of an unsophisticated market model to money laundering is
limited, and that the hypothesis that there are a substantial number of
stand-alone money launderers is not supported.
Nevertheless, the AML regime as it has evolved is employed to pursue
a wide range of law enforcement and other social objectives, which has
two consequences: first, the regime should be recognized as a policy tool,
though only one of many, to achieve those objectives; and second, the AML
regime is no longer all about drugs. Recognizing that the AML regime is a
means to multiple ends, and not just directed at combating illegal drugs or
organized crime, means that assessing its effectiveness depends in part on
the particular objective being considered. This study has examined the con¬
tribution of the AML regime to three broad objectives: reducing crime,

IMPROVING THE GLOBAL AML REGIME 175

maintaining the integrity of the core financial system, and controlling the
global "public bads" of terrorism, corruption and kleptocracy, and failed
states. But any assessment is limited at this point because it has to rely on
indirect indicators of regime effectiveness. Better indicators can and should
be developed, and we discuss some possibilities in the research agenda
presented in this chapter. Data constraints aside, however, it should also
be pointed out that multiple objectives require multiple indicators. In par¬
ticular, even if it were possible to measure accurately the aggregate annual
volume of money laundering globally, the findings would not adequately
represent the precise contribution of the AML regime to combating terror¬
ism, inhibiting the drug trade, discouraging white-collar crime, or protect¬
ing the integrity of the core financial system.
Following the money was an important law enforcement tactic long
before money laundering was criminalized. However, criminalizing money
laundering expands that tactic by enlisting the private sector in the process,
and by allowing for targeting stand-alone professional launderers, to the
extent that they exist. Criminalization also provides the basis for develop¬
ing the prevention pillar, which is essential to protecting the integrity of the
core financial system.
The prevention pillar has some deterrent effect on certain methods of
money laundering, although it also creates incentives for the mutation of
the phenomenon. The prevention pillar (most notably the reporting ele¬
ment) provides information for use in investigation, which is part of
enforcement. The investigative element operates through several channels:
as a primary tool to bring criminals to justice, leading to stand-alone pros¬
ecutions; as a secondary tool to expand cases and develop leads with res¬
pect to matters and persons already under investigation; and as a tertiary
tool where money laundering merely turns up in the course of an investi¬
gation and adds to possible charges.
Two policy implications can be drawn from these observations. First, the
AML regime can be proactively used to pursue certain objectives, as in sting
operations that attract those seeking money-laundering services.2 And sec¬
ond, to be effective as a law enforcement tool, the AML regime must be a
two-way street in which the private sector not only supplies information but
also is enlisted to participate in achieving regime objectives.
Both of these policy implications are controversial, particularly the latter.
The private sector does not always welcome being coopted by government,
yet cooperation by private-sector institutions is often essential to success.
For its part, law enforcement often is reluctant to take private businesses
into its confidence out of fear of damaging its case either prior to or during
the prosecutorial stage. Given the extent to which the AML regime already
relies on the quasi-voluntary cooperation of various private-sector institu-

2. Cuellar (2003) implicitly endorses a proactive AML strategy, and James (2002, 6) takes a
similar position.

176 CHASING DIRTY MONEY

tions, US law enforcement authorities would likely best serve anti-money
laundering efforts by cooperating more closely with the private sector, par¬
ticularly when it comes to providing more information about what they are
looking for and why.

Implementation Challenges for the United States

The United States has taken the lead in developing and establishing the
global AML regime, recognizing that money laundering is inherently not
only an issue without borders but, more importantly, one for which borders
can impede progress. US authorities reached this conclusion in part as a
result of efforts to level the international playing field for US financial insti¬
tutions whose assistance is critical to effectively combat money laundering.
The United States faces several challenges going forward, including
implementing the 2003 FATF Forty Recommendations, determining the
future of the National Money Laundering Strategy (NMLS), and main¬
taining its global leadership role.

FATF Forty Recommendations

The 2003 FATF Forty Recommendations appear at first not to require a

great deal from the United States, but issues lurk below the surface in at
least two areas: application of the prevention regime to certain profession¬
als (particularly lawyers and accountants), and the treatment of special
purpose vehicles.
Many nations and territories face challenges from certain professions
that have banded together globally to resist participation in the AML
regime at the level recommended by the FATF. Resistance in the United
States is likely to be particularly acute because of the interplay of privacy
issues and the regulatory structure of the US federal system. Regulation
and (limited) supervision of professionals occurs largely at the state level,
which in part explains why the accounting profession has until recently
been lightly controlled.3 Regardless of current US arrangements, the fact
that these professions will be included at least on a pro forma basis in
regimes in other jurisdictions poses a problem for the United States.
The second issue for the United States with respect to the FATF Forty
Recommendations involves the establishment and role of shell corpora-

an absolute
3. Lawyers also have generally been regulated by the states, but federalism is not
barrier to federal regulation in the financial area. Much federal financial law explicitly pre¬
where it does not, the federal banking supervisors and, to a lesser
empts state law, and even
In
extent, the Securities and Exchange Commission have effectively preempted the states.
addition, the recent Sarbanes-Oxley legislation imposed a regulatory regime on lawyers rep¬
resenting corporations as a matter of federal securities law.

IMPROVING THE GLOBAL AML REGIME 177

tions and special purpose vehicles at home (i.e., Nevada or Delaware) or
abroad (i.e., the Cayman Islands or Bermuda) as part of a US business or
financial structure, as well as ongoing pressure from some other jurisdic¬
tions for greater transparency as regards such arrangements. Authorities
in many European jurisdictions favor the imposition of CDD and report¬
ing requirements on such entities.4 This issue involves the philosophy and
the structure of regulation. In the United States, the structure, again, is one
where the presumption is that matters are left to the states, and increased
federal involvement requires jumping political and constitutional hurdles.
The philosophy of regulation in the United States is one of limited regula¬
tion at every level — an activity is legal unless it has been decided it is ille¬
gal. In contrast, in many other nations with fully developed financial
sectors, such as those in continental Europe, the philosophy is more along
the lines that nothing is legal unless government has approved it.
The global AML regime associated with the 2003 FATF Forty Recommend¬
ations has relatively weak supervision and sanction elements in its preven¬
tion pillar, but they are there. The International Monetary Fund (IMF) and
World Bank will evaluate performance under the new standards. The FATF
has reserved the possibility of resuming its "name and shame" role. If the
United States is found to have a low level of compliance, its capacity to exert
leadership in this area will be adversely affected. Mark Pieth and Gemma
Aiolfi (2003) have criticized both the United States and the United Kingdom
for uneven application of the previous FATF Forty Recommendations.

National Money Laundering Strategy

The US Congress in 1998 mandated the annual development and publica¬

tion of a National Money Laundering Strategy (NMLS) by the executive
branch via an interagency process for five years, ending in 2003. Should the
mandate be renewed?
On balance, the NMLS was a constructive instrument to focus executive
branch attention and, more importantly, promote better interagency coop¬
eration on money-laundering issues. On the other hand, it was far from
successful in molding an integrated and coherent approach to money laun¬
dering, which must ultimately be based on a firm foundation of research
and analysis. The preparation of annual national strategies on money laun¬
dering should be reauthorized, but with several important modifications

4. Some US observers express similar views. Wechsler (2001) flagged the issue of Delaware
corporations, and Jack Blum, a prominent voice on such matters, has called for national reg¬
istration of corporations that identifies where they can be served with civil and criminal
process. In addition, a Financial Stability Forum (FSF 2000) report on offshore financial cen¬
ters focused attention on the need to better identify the beneficial owners of corporate vehi¬
cles established in those jurisdictions as part of the effort to enhance financial stability and
fight financial fraud.

178 CHASING DIRTY MONEY

to encourage more analytical content, less wheel spinning, and greater con¬
tinuity.5 We offer the following four recommendations with respect to a
revised NMLS mandate:

b A strategy document should be developed and publicly presented every

third year, with annual updates on progress. Annual strategies tend to
be wheel-spinning exercises. No sooner is the production cycle for the
last document completed than the cycle for the next one must begin.
The achievements of the previous strategy will not yet be manifest,
however, so the next strategy will likely have a defensive air to it.
Furthermore, money laundering is not such a rapidly evolving activ¬
ity as to require annual adjustments in the AML regime. The annual
progress updates can include amendments as needed in connection
with any changes in emphasis, as in the case of the shift to focusing on
terrorist financing in the wake of the September 11 attack.

■ A review of progress during the period in achieving previously iden¬

tified goals, which appeared only once in the previous five NMLS,
should be institutionalized. The constructive precedent in 2002 was not
followed in the 2003 NMLS, except for an appendix on online terrorist
financing that had been identified as an objective in 2002.

■ The NMLS should include and report analytical work on money laun¬
dering, which is essential to developing a transparent and defensible
policy. The 2001 and 2002 NMLS contained more analytical material
than the previous two strategies or the one that followed. Rigorous
analysis enhances communication, and an important role of the NMLS
is as a communication device not just between the executive and leg¬
islative branches but also with the general public and the world at large.

a Oversight hearings should be employed systematically in connection

with the NMLS and its annual updates on progress. Hearings enhance
communication among the branches of government and promote inter¬
agency cooperation, which is a chronic problem in dealing with issues
as complex as the AML regime.

Global Leadership

Although the United States has been a driving force in shaping the global
AML regime, continuing in that leadership role will require addressing a
number of challenges in the years ahead. The first concerns the issues
raised above with respect to leading by example in complying with the
2003 FATF Forty Recommendations. The power of persuasion depends in

5. This position is similar to that found in General Accounting Office (2003a), although we
reached our conclusions via a somewhat different route.

IMPROVING THE GLOBAL AML REGIME 179

part on a demonstrated commitment to global norms, even those that may
not be at the top of the US list of priorities. To the extent that the US author¬
ities are unable or unwilling to implement FATF recommendations regard¬
ing the legal and accounting professions, shell corporations, and special
purpose vehicles, the United States will have to explain and justify its posi¬
tion. One useful step in this regard would be for US authorities to articu¬
late that they are open to expanding the US AML regime in these areas once
it is demonstrated that there are substantial net benefits.
A related aspect of this challenge is whether the United States is willing to
follow the United Kingdom and submit to a full assessment of its financial-
sector regulation by the IMF and World Bank, including its compliance
with international standards in the area of money laundering. The United
States did volunteer for a review of its compliance with the old FATF Forty
Recommendations and the Eight Special Recommendations on terrorist
financing as part of the IMF/World Bank pilot program. The United States
has since volunteered for the FATF assessment of its regime, a review that
would employ the revised methodology endorsed by the FATF, IMF, and
World Bank.6 However, this type of review would not result in a formal
assessment or rating of the US regime, but rather only a report on US com¬
pliance with the FATF 40 + 8 recommendations.
The United States has argued, with some merit, that scarce IMF/ World
Bank resources to conduct such full assessments should be devoted to other
countries where risks to the global AML regime are greater. On the other
hand, the United States is in a position to set an example for the rest of the
world by submitting to the review, yet has elected not to do so. Moreover,
a case can be made that money laundering is substantial in the United States
and should receive special attention as part of the global AML regime.
Because of the central role of the US financial system, as well as the role of
the US dollar as a medium of exchange and a standard of value in the global
financial system, many if not most criminals want the proceeds of their
crimes ultimately to be in dollars and potentially integrated into the US
economy, even if the crimes were committed in other jurisdictions. The
United States should reconsider its position and volunteer for a full IMF /
World Bank assessment of its financial sector.
A second challenge to US global AML leadership is the choice of priori¬
ties. Over the past three years, the United States, understandably, has
emphasized combating terrorism financing but has not garnered as much
global support as it might have hoped (chapter 7). Terrorism financing is
not the top priority of most countries' AML regimes, and many of these
regimes are supported by very limited resources. Some countries may pre¬
fer to concentrate on extending the global regime to cover accountants,
smuggling, or tax evasion. Those efforts, even if they do not match US pri-

6. In the meantime, the US Treasury Web site has posted previous self-assessments of US
compliance with FATF standards on combating money laundering and terrorism financing.

180 CHASING DIRTY MONEY

orities, warrant increased bilateral and multilateral technical assistance.
The United States, along with other major countries, should consider step¬
ping up such assistance to jurisdictions that are struggling to establish or
implement their regimes.
A related third challenge to US global AML leadership involves the fre¬
quent criticism that the US regime is overly selective in the list of foreign
predicate crimes that qualify as money-laundering offenses under US law.
For example, the 2003 FATF Forty Recommendations designates 20 cate¬
gories of high-priority offenses that should be covered by national money¬
laundering statutes. However, seven of those offenses related to money
laundering — including sexual exploitation, trafficking in human beings,
and counterfeiting — cannot be prosecuted in the United States if the under¬
lying crime was committed abroad.
Foreign or domestic tax evasion — other than failure to pay US taxes on
the proceeds of a crime — also does not qualify as an underlying crime for
US prosecution as a money-laundering offense. While US prosecutors can
work around this lacuna and do not regard it as an impediment to an effec¬
tive US AML regime, the absence of foreign tax evasion as a predicate offense
under US law is often cited as impeding international cooperation.7 Latin
American leaders, for example, often complain privately that while the US
insists on cooperation on issues it considers important, the US itself often
fails to cooperate on issues of importance to other countries, such as eva¬
sion of taxes on income from assets held abroad. In this context, the US pol¬
icy becomes a barrier to leadership. Thus, US law should be changed to
make tax evasion, whether in the United States or elsewhere, a predicate
offense for money-laundering prosecution.
The United States also needs to be more forthcoming in helping to
enforce the tax laws of other countries. The interaction of capital flight and
tax evasion is a particular problem where governments have difficulty rais¬
ing adequate revenues to finance their expenditures, resulting in a run-up
of unsustainable stocks of sovereign debt to foreign as well as domestic
holders. Though macroeconomic policy failures in Latin American coun¬
tries and elsewhere often provide substantial inducement for capital flight,
the United States also is regarded as one of the world's leading tax havens,
which contributes to fiscal problems in other countries. US critics counter

that many countries' tax laws are flawed, but what country's are not, to one
extent or another? A compromise approach might be to condition increased
US cooperation in the tax area on criteria that apply to the structure and
efficiency of the other country's tax system. For example, the criteria might
address the balance between direct and indirect taxes, or the maximum tax
rate applied to earned income. One should not be too sanguine about the

7. Such criticism is not directed only at the United States. An IMF (2001, 24) review of Cyprus
as an offshore financial center observed that international cooperation would be strengthened

if Cyprus were to clarify that tax evasion is an offense under its money-laundering laws and
regulations.

IMPROVING THE GLOBAL AML REGIME 181

chances of agreement on mutually acceptable criteria, but the effort might
produce some positive results over time and reinforce the willingness of
other countries to cooperate with the United States on money laundering
in general, and the financing of terrorism in particular.
Chapters 4 and 7 explained how the legal framework of the AML regime
aimed at kleptocrats is in its infancy, especially with regard to enforcement.
The United States should ratify the new UN Convention Against Corruption
(United Nations 2003b) to accelerate the development process, and imple¬
ment streamlined procedures (which might require changes in US laws) to
facilitate cooperation with other countries in freezing, confiscating, and
returning the assets of kleptocrats to the home countries from which they
stole them.

Global Implementation Challenges

Even if one accepts the premise that expansion of the global AML regime
will slow in the years ahead, the regime still will have to be regularly
adjusted, in part to address the continually changing mechanisms of money
laundering and the tactics of those needing to launder the proceeds of their
crimes. The FATF's mandate was renewed in May 2004 for another eight
years. While it is premature to speculate about when the Forty Recom¬
mendations should again be revised, it is reasonable to expect that adjust¬
ments and a new set of recommendations will be required at some point in
the future. What is significant and disturbing about the announcement of the
renewal of the FATF's mandate is that it made no mention either of the
importance of research on, or ongoing assessments of the costs and benefits
of, an effective anti-money laundering regime.
In the meantime, the principal global issue is implementation of the AML
regime on the basis of the 2003 FATF Forty Recommendations. Monitoring
compliance with global AML standards has been largely turned over to the
IMF and World Bank following a common methodology worked out with
the FATF (chapters 4 and 7). This transfer of responsibility has the benefit
that reviews are now conducted by organizations with near-universal mem¬
bership, rather than by a self-selected group of countries that may be reluc¬
tant to render frank criticism of fellow members.
On the other hand, while peer pressure, transparency, and accountabil¬
ity are enhanced by the involvement of the IMF and World Bank, the fact
that their mandates are far more diffuse has substantially reduced the
capacity to apply meaningful leverage, as compared with a process that
was exclusively run by the FATF. It is unrealistic to presume that money
laundering will rise to the level of importance of fiscal, monetary, exchange
rate, and banking policies in connection with IMF support for member
countries' economic programs (chapter 7).
A clear challenge ahead to the global AML regime will be to balance the
role of the FATF, as the standard-setting body, with the involvement of the

182 CHASING DIRTY MONEY

international financial institutions. Countries that some members of the IMF
and World Bank might see as candidates for peer pressure in terms of
improving their AML regime compliance with international standards may
not be seen as candidates by other member nations or by the managements
of those institutions.
A related international challenge lies in the scope and utility of sanctions
invoked against countries that do not comply with the global regime. As
noted in chapter 7, experience with the FATF's Non-Cooperative Countries
and Territories (NCCT) initiative has generally been positive, but not com¬
pletely successful or without controversy. In response to the initiative's
"name and shame" approach, about 70 percent of jurisdictions with initial
shortcomings have brought their AML regimes into better alignment with
international norms. Only Nauru and Myanmar/Burma have been quar¬
antined. Although the process has not yet been fully effective in convinc¬
ing Nauru to mend its ways or in shutting down Nauru as a host to money
launderers, some improvements have been made. Nauru's role in money
laundering appears to have been reduced if not eliminated, but it never
was large! Myanmar/Burma has made even less progress.
An issue for the global system going forward is to develop criteria upon
which similar sanctions can be imposed on other jurisdictions. Is it realistic
to expect that such sanctions can in fact be imposed on jurisdictions that are
more relevant to the international financial system than the two countries
sanctioned to date? How will such countries as Indonesia, Nigeria, and the
Philippines respond to pressures from the FATF, and what will the FATF
do if their response is inadequate? Will the FATF try to add other countries
to its list of uncooperative jurisdictions and incur the displeasure of at least
some members of the IMF and World Bank? The standard for noncoopera¬
tion will have to be raised substantially for this to happen.
This is an issue regarding which one must consider carrots as well as
sticks. The concept of mutual recognition of the AML regimes of other
countries — or at least institutions from other countries that have passed
their FATF examinations or IMF/World Bank reviews — deserves more
consideration than it received in the 2003 FATF Forty Recommendations
or under the USA PATRIOT Act. As noted repeatedly in this chapter
already, serious consideration should be given to increased technical and
financial assistance to countries trying to establish and implement national
AML regimes on the basis of international standards.

International Cooperation

The AML regime serves multiple though generally concordant goals, de¬
spite different emphases in different jurisdictions. Conflicts do sometimes
arise as a result of different regulatory philosophies and issues regarding
the allocation of limited resources.

IMPROVING THE GLOBAL AML REGIME 183

 International cooperation is often a challenge not only because of differ¬
ences in AML regimes but because of differences across countries with
respect to the structure and development of their financial systems. For
example, the limited role of retail banking for both households and small
businesses in Switzerland leads to less emphasis on the formal ex ante
CDD aspects of the prevention pillar for Swiss banks and greater empha¬
sis on identifying actual money-laundering operations through a coopera¬
tive relationship among financial institutions and the authorities.
The issue of limited resources is relevant both to choosing objectives for
the global AML regime and determining to which of those selected objec¬
tives to allocate scarce governmental or private resources. Limiting money
laundering was a high-profile public good in the United States even before
September 11, 2001, but today it is still regarded as an unaffordable luxury
good in some other jurisdictions.
Developing countries, in particular, are sometimes ambivalent about
creating or giving priority to AML efforts. Brent Bartlett (2002) points out
three common reasons why: (1) money laundering brings funds from
developed economies to developing economies; (2) the crimes that gener¬
ate the funds occur in developed rather than developing countries; and
(3) money laundering adds to the demand for banking services in devel¬
oping countries. Bartlett argues that that the data do not support any of
these statements, that money laundering undermines confidence in the
financial system, and that developing countries in general share a broad
interest in an effective AML regime. Nevertheless, it is safe to say that
reducing global "public bads" is not a high priority for many of these coun¬
tries, which often are unattractive to begin with as locations for financial
transactions because of a lack of integrity of domestic financial institutions.
For countries with weak financial systems — as evidenced by extensive use
of informal value transfer mechanisms — money laundering may seem a
minor problem compared to inadequate supervision of lending practices
or tracking of nonperforming loans. The sophisticated human and organi¬
zational resources required to create an effective AML regime are particu¬
larly scarce in such countries, which is one reason why the IMF and World
Bank have been encouraged to develop technical assistance programs to
help them develop those resources.8
It is an article of faith to the authorities in industrial countries that all
nations need to have effective AML regimes, but resources are scarce. The
global threat posed by weaknesses in poor countries may be quite minor,
and complete convergence of national AML regimes is not necessary to
achieve an effective global regime. The trick is to identify weaknesses that
need to be addressed and regimes that need to be upgraded before they
become major problems for the system as a whole.

8. The World Bank (2003a) and IMF have developed a reference guide for that purpose.

184 CHASING DIRTY MONEY

 Having effective AML regimes in poor or small jurisdictions identified as
offshore financial centers (OFCs) is particularly important to the global sys¬
tem. The IMF has identified 44 such countries and territories, most of which
are classified as developing. These jurisdictions need unimpeded links to
mainstream financial centers — even if it reduces their attractiveness to some
investors. Thus, an AML regime that is certified by the FATF and inter¬
national financial institutions is a critical asset. The rapid response of the
Cayman Islands and most other OFC jurisdictions to the FATF's designa¬
tion of them as uncooperative suggests that they understand the importance
of cooperation.
Achieving a better alignment of national regimes with the global regime
is complex. Russia struggled to align its AML legislation with global stan¬
dards because of a desire to include broad and vaguely defined economic
crimes within the ambit of its law, in part out of concern about capital
flight. Chinese authorities have articulated a similar position in the context
of the release of money-laundering rules by the People's Bank of China
(Financial Times, January 14, 2003). India sees money laundering principally
as an activity induced by capital and exchange controls, which it is slowly
dismantling.
Continued progress in developing the global AML regime also will re¬
quire better alignment of predicate offenses internationally. The 2003 FATF
Forty Recommendations include a roadmap for cooperation with a list of
core offenses, but there are important omissions.
Greater and ongoing international cooperation also is needed to forge
consensus on global strategies that can adequately serve the disparate
needs of the global community. Intensified efforts since September 2001 to
combat the financing of terrorism illustrate both the potential for focusing
the tools of the AML regime on particular objectives and the pitfalls asso¬
ciated with a failure to recognize that some countries consider other objec¬
tives to be of equal or greater importance.
One relatively weak test of international cooperation on money laun¬
dering will be the UN Convention Against Corruption (United Nations
2003b). To be effective, the convention must be ratified by the major
nations, many of which will have to change their domestic legislation and
then actually implement the convention's provisions, in particular its asset
recovery provisions.
The enforcement pillar of the global AML regime, often constrained by
tensions and differences between national criminal justice systems, will also
have to evolve in the years ahead. For example, more fully streamlined
modalities of information sharing in money-laundering cases should be
established between law enforcement authorities to replace the cumbersome
multistep processes of existing multilateral assistance arrangements that
require the use of diplomatic channels. Global standards with respect to asset
freezes and forfeitures are also needed so that such actions are not only inter¬
nationally coordinated but also comprehensive in their effect. A third area

IMPROVING THE GLOBAL AML REGIME 185

where progress is needed is extradition. International agreement is needed
to streamline procedures in connection with certain money-laundering
offenses, perhaps those 20 designated by the FATF to be particularly impor¬
tant. Members of the European Union are considering implementing such
procedures in connection with a specified list of crimes.
Consideration should also be given to producing a global equivalent of
the US State Department's International Narcotics Control Strategy Report
(INCSR) on money laundering. Although the INCSR identifies nations that
the US believes have major money-laundering problems, the principal con¬
cern is with money laundering that facilitates drug trafficking. The report
correctly states that this cannot be separated from money laundering prob¬
lems more generally. Nonetheless, the INCSR presents a US perspective,
whereas a periodic report from an international agency — such as the United
Nations Office on Drugs and Crime, which has itself no actual regulatory
responsibility — that rated nations in terms of the extent and nature of their
money-laundering problems might serve a useful global educative func¬
tion and support better analysis of money laundering and its control.9 The
ratings could draw on but be separate from IMF/World Bank reviews and
FATF mutual assessments.

Domestic Law Enforcement

Regulation and supervision alone can do much to keep financial institu¬

tions clean of money laundering. Chapter 6 showed that the global AML
regime has been quite effective in meeting this goal with respect to the core
financial systems of the major financial centers. However, this success will
tend to push money laundering to other less regulated and supervised
channels, which in turn points to the importance of effective law enforce¬
ment cooperation across the global AML regime.
Unfortunately, it is difficult to judge the effectiveness of current money¬
laundering investigations because assessment of the enforcement pillar lags
that of the prevention pillar, perhaps inevitably, since the former requires
much more detailed knowledge of the underlying reality. Still, enforcement
assessments have not even been attempted, so definitively determining
their feasibility is difficult. Even the greater focus on law enforcement envis-

9. Winer (2002) proposes that the United Nations establish a "white list" of financial institu¬
tions committed to the global AML regime that have passed rigorous reviews of their compli¬
ance with and implementation of AML standards. Once on the white list, they would become
exclusively eligible to receive and manage funds from the United Nations and other interna¬
tional organizations. This proposal apparently has received little attention since it was put for¬
ward two years ago. The cost of the administrative apparatus it proposes might exceed the
benefits to the AML regime as a whole, but such a proposal nevertheless represents the type
of imaginative thinking on international cooperation that warrants consideration.

186 CHASING DIRTY MONEY

aged in the 2003 FATF Forty Recommendations centers around organiza¬
tion and resources, which are important but do not emphasize results.
Money-laundering convictions in the United States average no more
than 2,000 annually, even on a generous measure. Given the suspected
scope of the activity, this suggests that money laundering is not a very risky
activity, particularly when one considers that most convicted launderers
are associated with sums of less than $1 million. A very speculative esti¬
mate of the risk of conviction faced by money launderers is about 5 percent
annually. However, some who provide money-laundering services may
only be convicted on other charges, probably related to the predicate
offenses that can generate longer sentences.10 This may be the result of plea
bargaining or of a decision to bring only the charges with the longest sen¬
tences. Thus, there is no way to measure the actual risk a money launderer
faces of going to prison. US seizures and forfeitures of $700 million annu¬
ally suggest that either a trivial fraction of laundered money is seized, or
that much less is laundered than is indicated by official statements about
the scale of the activity.
Interviews with both former and current law enforcement officials sug¬
gest that the existing regulatory system and the information it generates is
not well used in prosecutions. The US Customs and Internal Revenue
Services use suspicious activity reports skillfully for specialized purposes,
but the reports are rarely used to initiate investigations; instead they are
used as additional information for making a case that has originated with
another type of lead. Indeed, apart from sting operations, money launder¬
ing is rarely the initial offense for an investigation. Some knowledgeable
observers have described the sting operations of the early 1990s, when a
number of drug traffickers were brought down employing such techniques,
as the heyday of money-laundering investigations. But they add that the
culture of the major federal law enforcement organizations has since be¬
come less hospitable to making money laundering a central investigative
focus. The deemphasis of drug investigations has apparently led to less
commitment to money-laundering expertise in some key law enforcement
organizations. The shift in focus by federal enforcement to countering ter¬
rorism financing also has likely been at the expense of more general money¬
laundering investigations.
However, the reduced emphasis on large-volume money-laundering
operations also may reflect the training and orientation of the federal law
enforcement community. Money laundering is complex and often difficult
to follow for anyone without highly specialized knowledge of finance, and
relatively few agents and prosecutors are in a position to acquire the skills
necessary to pursue such cases. This is not a problem that can be solved

10. Cuellar (2003) provides some evidence that prosecutors tend to pursue the easier money¬
laundering charges for low-level perpetrators rather than the more-difficult-to-establish
charges for the underlying crime or major financing networks.

IMPROVING THE GLOBAL AML REGIME 187

with a simple recommendation, but we do believe that finding a way to
give more priority to money laundering as the initiating offense, and mak¬
ing better use of the existing SAR database, could improve the effective¬
ness of the AML regime in fighting crime.
The federal government, of course, has often run undercover money¬
laundering operations, sometimes to considerable investigative effect.
However, these operations pose two interesting problems that have re¬
ceived little attention. First, in order to generate business, agents offer low-
priced money-laundering services, which may drive down prices and thus
actually facilitate laundering. This is a theoretical possibility and a difficult
one to investigate, since it depends on how well the market actually oper¬
ates. The market effect may well be insignificant, but it is a concern that
should nevertheless be explicitly addressed when setting up these opera¬
tions. Second, the operations do facilitate, temporarily at least, the work¬
ings of criminal enterprises. To catch senior drug dealers in Operation
Polar Cap, a sting operation in the mid-1980s, federal agents allowed large
amounts of money to be laundered successfully back to Panama. The trade¬
offs in terms of the targets apprehended may well be reasonable, but again
these are issues that warrant explicit consideration and assessment.

Research Agenda

One clear finding of this study is that there has been little research on either
money laundering or the anti-money laundering regime. The law enforce¬
ment community simply does not use the types of systematic measurements
of inputs and outputs necessary to allocate scarce enforcement resources
effectively.
The multiple goals and complex cost considerations of the AML regime
point to the potential of using the cost-effectiveness of AML enforcement as
a framework to assess the regime. The implied outcome measure is the cost
of law enforcement with an AML regime in place as compared to the cost of
enforcement without it. While it would be difficult to assemble the infor¬

mation necessary to estimate this difference in any formal sense — "experi¬

menting" per se is impossible, and the effects of the regime on costs are too
subtle to be measured — such an approach suggests at least at the concep¬
tual level the value of obtaining data on the number of cases in which
money-laundering controls generated valuable information, or provided
the legal basis for conviction and incarceration, and on the costs of those
controls.
Empirical research on money laundering is even more challenging than
for most other criminal activities. There are no victimization surveys of the
kind that allow measurement of the volume of other white-collar crimes.
Population surveys that provide so much insight into drug distribution
and other consensual crimes are unlikely to provide much information

188 CHASING DIRTY MONEY

about the narrow set of participants involved in money laundering, most
of whom are often closely linked to, if not part of, the crimes.
Nevertheless, the AML regime ought to be susceptible to research. Formi¬
dable bureaucratic obstacles arise from the complex structure of law enforce¬
ment, which involves multiple agencies that are parts of different cabinet
departments, each with its own mission, expertise, and data systems. Data
do not travel easily across such bureaucratic landscapes.
An active AML research program will not resolve every issue at hand,
but six broad recommendations for a research agenda are offered below.

1 . Create a Database of Detected Money-Laundering Transactions

A great deal of information can be gathered about money laundering
through compilation of investigations in various countries. However, this
information has yet to be systematically assembled into a searchable and
researchable database. A starting point for money-laundering studies
should be the creation of just such a database that contains all the inves¬
tigative information, as has been done over the past 30 years in the study of
terrorism by organizations such as the RAND Corporation (e.g., Fowler
1981). For each transaction, data would be available inter alia on the predi¬
cate offense, the price paid to the launderer and the total cost of laundering
to the customer, the stages of money laundering covered by the transaction,
and the characteristics of the customer and the provider.
The database would not constitute a representative sample of money¬
laundering operations, but rather the results of investigative decisions.
Having such data available, however, would facilitate analyses of how
prices vary across transactions, how they have changed over time for par¬
ticular kinds of transactions, and how the patterns of utilization of differ¬
ent institutions vary across countries, offenses, and time. The results would
only be suggestive, but nevertheless would provide insights that are cur¬
rently not available.

2. Track Usage of Suspicious Activity Reports

The suspicious activity reporting system in the United States has expanded
over time, particularly since the mid-1990s, and now represents a substan¬
tial effort by both the government and the private sector. The SARs data¬
base contains a vast amount of information assembled at considerable
expense. The risk of information overload is real and substantial, and some
argue that the system is essentially dysfunctional— valuable warnings and
leads are lost, mislaid, or overlooked. No US agency has yet assessed how
effective SARs are in contributing to the arrest and conviction of money
launderers, or to achieving the specific goals of the AML regime such as
combating terrorism financing. The GAO has tried to arrive at an assess¬
ment but has been unable to do so.

IMPROVING THE GLOBAL AML REGIME 189

 To be useful, the SARs database needs to be made available to many
agencies that have neither the incentive nor the means to provide FinCEN
with feedback about the outcome of their use of the information. For exam¬
ple, a particular SAR may be just one of many pieces of evidence that lead
to a conviction many months if not years after the initial SARs database
query. The inquiry may be initiated by one agency, but the information is
passed on to another that then uses it to obtain a conviction.
Creating a system to report all successful uses of SARs is a daunting and
expensive task. Universal reporting on outcomes may not be necessary,
however; periodic targeted studies might be sufficient. A sample of queries
could be followed up and traced to the end of the initial investigation, or a
sample of cases involving crimes likely to be related to money laundering
could be examined to determine whether and how SARs played a role in
the convictions.11
Given the continued pressure to expand the range of institutions re¬
quired to report suspicious activities, an effort must be made to assess
whether the system does in fact produce enforcement information of suf¬
ficient value to justify that expansion. Without getting into specific details
of the appropriate investigative approach for such a study, it might lead to
improving the design of the SAR system in terms of the identification,
selection, and coverage of information that is worth gathering.

3. Sponsor Research
More sponsored research on money laundering and how to improve the
AML regime is needed at the national and international levels. In the United
States, research hands should be earmarked in the budgets of the principal
agencies responsible for the AML regime (the Homeland Security, Justice,
and Treasury Departments). Research program design and results should
be subject to outside evaluation, with a strong presumption that the results
will be published. Other agencies responsible for supervision of the finan¬
cial system, such as the Federal Reserve and the Securities and Exchange
Commission, should be encouraged to have parallel programs.
At the international level, the FATF, IMF, and World Bank should sup¬
port cooperative research programs.

4. Maintain a Scorecard on Core Financial System Integrity

Short of developing a more complete database of money-laundering cases,
supervisory authorities should at least regularly assess (perhaps every two
years) the extent to which money laundering threatens the integrity of the
core financial system. If necessary, law enforcement authorities in the
United States and other countries could assist in carrying out such reviews.

11. Gold and Levi (1994) conducted such analyses using data from the United Kingdom,
which has made more progress with assessing SARs than has the United States.

190 CHASING DIRTY MONEY

 Chapter 6 provided a prototype for this type of research project. Cases
that involve core financial institutions should be classified into categories
based on whether there was institutional solicitation, solicitation by a rogue
officer, or unwitting participation or negligence. Cases should then be cross-
classified by the size of the financial institutions involved. This exercise
would provide an initial baseline against which subsequent reviews could
assess progress of the AML regime in protecting the integrity of the core
financial system. The results of these assessments should be published.

5. Measure AML Regime Costs

A major shortcoming of the AML regime is the lack of hard data on what
it costs. Chapter 4 drew on the limited literature available and attempted
to come up with a rough estimate of the gross financial costs of the US
regime, including costs to taxpayers incurred by government, costs imposed
on financial and nonfinancial businesses and institutions, and costs for cus¬
tomers. Former US Treasury Secretary Paul O'Neill was right to push for
work on this topic, and it is unfortunate that he was not more successful.
Developing more and better data on AML regime costs should be part of
an overall research program.

6. Use Economic Modeling

While the first five suggestions for the research agenda have emphasized
empirical research, establishing a firm empirical basis for decisions about
the AML regime also requires developing models of money laundering.
Theoretical economists have begun to work in this area, but at a very aggre¬
gate level that reflects only the most schematic knowledge of money laun¬
dering (Masciandaro 1999; Masciandaro and Portolano 2002; and Mauro
2002).
Also needed is a microeconomic research agenda that examines such ele¬
ments of money-laundering analysis as the market model for laundering
services, as well as how the activity might respond to different kinds of reg¬
ulations and other interventions. The theoretical research could be strength¬
ened by the empirical analyses recommended above.

Final Comments

The global regime that has been constructed to combat money laundering

is as complex as the phenomenon itself, in part because of the regime's

multiple goals, but also because of institutional and attitudinal differences
across countries. These circumstances will continue to shape the interna¬
tional regime and strategies to control money laundering.
With the 2003 revision of the FATF Forty Recommendations on money
laundering and the relatively recent promulgation of the Eight Special

IMPROVING THE GLOBAL AML REGIME 191

Recommendations on Terrorist Financing, the global AML system con¬
fronts new challenges with respect to compliance, implementation, and
sanctions. Even the United States, one of the principal architects of the
AML regime, is unlikely to comply fully with the letter of the new FATF
framework. To ensure an effective strategy going forward, the United
States should submit to a full IMF review of its financial sector, including
review of its anti-money laundering and terrorism financing policies. In
addition, the United States should expand the list of foreign offenses that
can lead to US money-laundering prosecutions to include tax evasion as
well as other serious crimes not now covered and step up cooperation with
other jurisdictions on actions against corruption and kleptocracy.
Different countries have different AML priorities, but cooperation is
essential to achieve an effective overall global strategy. Ratification and
effective implementation of the UN Convention Against Corruption will
be a test for the AML system as a whole. The global AML regime clearly
needs further development and promulgation of anti-money laundering
strategies at the international as well as national levels. Cooperation with
the private sector also should be enhanced, and more technical and finan¬
cial assistance should be made available to poorer jurisdictions.
At present, there is no empirical base to assess the effectiveness of the
current AML regime in terms of suppressing money laundering and the
predicate crimes that generate it. Sifting of the limited available informa¬
tion suggests that the global AML regime has made progress in the general
area of prevention, but without much effect on the incidence of underlying
crimes. Critics argue that the regime has done little more than force money
launderers to change their methods. Felons' lives are a bit more difficult
and a few more are caught, but there is little change in the extent and char¬
acter of either laundering or crime. Critics may well be right. To rebut such
charges and continue to expand, the AML regime needs to demonstrate
progress toward its announced goals. The research agenda set out in this
study, if implemented conscientiously along with the other recommenda¬
tions made here, could contribute substantially to further progress in com¬
bating money laundering for many years to come.

192 CHASING DIRTY MONEY

References

ABA (American Bar Association) Task Force on Gatekeeper Regulation and the Profession.
2003a. Report to the House of Delegates. Photocopy (February).
ABA (American Bar Association) Task Force on Gatekeeper Regulation and the Profession.
2003b. Response to the Advance Notice of Proposed Rulemaking and Request for Comments.
Photocopy (June 9).
Abt, Vicki, James F. Smith, and Eugene Martin Christiansen. 1985. The Business of Risk:
Commercial Gambling in Mainstream America. Lawrence, KS: University of Kansas Press.
Association of Certified Fraud Examiners. 2002. 2002 Report to the Nation on Occupational Fraud
and Abuse. Austin, TX. Photocopy.
Bartlett, Brent. 2002. The Negative Effects of Money Laundering on Economic Development.
Platypus Magazine 77: 18-23.
Basel Committee on Banking Supervision. 1988. Prevention of Criminal Use of the Banking
System for the Purpose of Money-Laundering. Basel: Basel Committee on Banking
Supervision.
Basel Committee on Banking Supervision. 2001. Customer Due Diligence for Banks. Basel
Committee Publications 85. Basel: Basel Committee on Banking Supervision.
Bell, R. E. 2001. Discretion and Decision Making in Money Laundering Prosecutions. Journal
of Money Laundering Control 5, no. 1: 42-51.
Bell, R.E. 2003. The Confiscation, Forfeiture, and Disruption of Terrorist Finances. Journal of
Money Laundering Control 7, no. 2: 105-25.
BIS (Bank for International Settlements). 2004. International Banking and Financial Market
Developments. Quarterly Review (March).
Blades, Derek, and David Roberts. 2002. Measuring the Non-observed Economy Statistics. OECD
Brief 5. Paris: Organization for Economic Cooperation and Development.
Board of Governors of the Federal Reserve System. 2002. Report to Congress in Accordance with
Section 356(c) of the USA PATRIOT Act. Washington: Board of Governors of the Federal
Reserve System.
Bourgois, Philippe. 1995. In Search of Respect: Selling Crack in El Barrio. New York: Cambridge
University Press.
Buencamino, Leonides, and Sergei Gorbunov. 2002. Informal Money Transfer Systems: Op¬
portunities and Challenges for Development Finance. United Nations Department of
Economic and Social Affairs (DESA) Discussion Paper 26. November.
193
Bureau of Justice Statistics. 2003. Money Laundering Offenders, 1994-2001. Washington: Bureau
of Justice Statistics.
Byrne, Jotrn J. 2004. Improving Financial Oversight: A Private Sector View of Anti-Money
Laundering Efforts. Testimony on behalf of the American Bankers Association before the
House Financial Services Subcommittee on Oversight and Investigations. Photocopy
(May 18).
Calzaroni, Manlio. 2001. The Exhaustiveness of Production Estimates: New Concepts and
Methodologies. Organization for Economic Cooperation and Development, Paris.
Photocopy.
Camdessus, Michel. 1998. Speech to the Financial Action Task Force. Paris.
Caprio, Gerald, and Patrick Honohan. 2001. Finance for Growth: Policy Choices in a Volatile
World. New York: Oxford University Press for the World Bank.
Carlson, Kenneth, Herbert Weisberg, and Naomi Goldstein. 1984. Unreported Taxable Income
from Selected Illegal Activities. Cambridge. MA: Abt Associates, Inc.
CCBE (Council of the Bars and Law Societies of the European Union). 2003. Joint Statement
by the International Legal Profession on the Fight Against Money Laundering, www.
ccbe.org/doc/En/ signed_statement_030403_en.pdf.
Center for Global Development. On the Brink: Weak States and US National Security. A Report
of the Commission on Weak States and US National Security. Washington: Center for
Global Development.
Council on Foreign Relations. 2002. Terrorist Financing. New York: Council on Foreign Relations.
Council on Foreign Relations. 2004. Update on the Global Campaign Against Terrorist Financing.
New York: Council on Foreign Relations.
Cuellar, Mariano-Florentino. 2003. The Tenuous Relationship Between the Fight against
Money Laundering and the Disruption of Criminal Finance. The Journal of Criminal Law
and Criminology 93: 312M66.
Da Vanzo, J., J. Hawes-Dawson, R. B. Valdez, and G. Vemez. 1994. Surveying Immigrant Com¬
munities: Policy Imperatives and Technical Challenges. Santa Monica, CA: RAND Corporation.
Dixon, H. 1999. Controversy: On the Use of the "Hidden Economy" Estimates. Economic
Journal 109: F335-F337.
Egmont Group. 2000. FIU's in Action: 100 Cases from the Egmont Group. London: Egmont Group
of Financial Intelligence Units.
Eizenstat, Stuart. 2000. Efforts to Combat Money Laundering. Congressional Testimony
before the Committee on Banking and Financial Services. Federal News Service (March
9).
Eldridge, James E. 1986. The Bank Secrecy Act: Privacy, Comity, and the Politics of
Contraband. North Carolina Journal of International Law and Commercial Regulation (sum¬
mer): 677-96.
Elliehausen, Gregory. 1998. The Cost of Bank Regulation: A Review of the Evidence. Federal
Reserve Staff Study 171. Washington: Board of Governors of the Federal Reserve System.
El Qorchi, Mohammed, Samuel Munzele Maimbo, and John F. Wilson. 2003. Informal Funds
Transfer Systems: An Analysis of the Informal Haiuala System. International Monetary Fund
Occasional Paper 222. Washington: International Monetary Fund.
European Commission. 2003. Eurobarometer: Public Opinion in the European Union. Brussels:
European Commission.
Eurostat. 1995. European System of Accounts. Brussels: Eurostat.
FATF (Financial Action Task Force). 1998. 1997-1998 Report on Money Laundering Typologies.
Paris: Financial Action Task Force.

FATF (Financial Action Task Force). 1999. 1998-1999 Report on Money Laundering Typologies.
Paris: Financial Action Task Force.

FATF (Financial Action Task Force). 2000. Report on Money Laundering Typologies 1999-2000.
Paris: Financial Action Task Force.
FATF (Financial Action Task Force). 2001a. Special Recommendations on Terrorist Financing.
Paris: Financial Action Task Force.

194 CHASING DIRTY MONEY

FATF (Financial Action Task Force). 2001b. Report on Money Laundering Typologies 2000-2001.
Paris: Financial Action Task Force.

FATF (Financial Action Task Force). 2002a. Combating the Abuse of Non-Profit Organizations:
International Best Practices. Paris: Financial Action Task Force.

FATF (Financial Action Task Force). 2002b. Report on Money Laundering Typologies 2001-2002.
Paris: Financial Action Task Force.

FATF (Financial Action Task Force). 2003a. Annual Report 2002-2003. Paris: Financial Action
Task Force.

FATF (Financial Action Task Force). 2003b. Annual Revieiv of Non-Cooperative Countries and
Territories. Paris: Financial Action Task Force.
FATF (Financial Action Task Force). 2003c. The Forty Recommendations. Paris: Financial Action
Task Force.

FATF (Financial Action Task Force). 2003d. Freezing of Terrorist Assets: International Best
Practices. Paris: Financial Action Task Force.

FATF (Financial Action Task Force). 2003e. Report on Money Laundering Typologies 2 002-2003.
Paris: Financial Action Task Force.

FATF (Financial Action Task Force). 2004a. Annual Report 2003-2004. Paris: Financial Action
Task Force.
FATF (Financial Action Task Force). 2004b. Report on Money Laundering and Terrorist Financing
Typologies 2003-2004. Paris: Financial Action Task Force.
FEE (Federation des Experts Comptables Europeens). 1999. Charter of the European Pro¬
fessional Associations in Support of the Fight against Organized Crime, www.fee.be/
european/PDFs/Charter%20organisedCrime.pdf.
FEE (Federation des Experts Comptables Europeens). 2003. Fact Sheet FST 004. www.fee.
be/secretariat/PDFs/Fact%20Sheet/FST004. Money. Laundering. pdf.
Feige, Edgar L. 1979. How Big is the Irregular Economy? Challenge 22(1): 5-13.
Feige, Edgar L. 1997. Revised Estimates of the Underground Economy: Implications of the US
Currency Held Abroad. In The Underground Economy: Global Evidence of its Size and Impact,
eds. Owen Lippert and Michael Walker. Vancouver: Fraser Institute.
Fields, Greg, and Mimi Whitfield. 2001 . Prosecutions Helping Lift the Veil of Secrecy from
Offshore Banks. www.clevelandtour.com/reyesboard/messages/7.html.
FinCEN (Financial Crimes Enforcement Network). 2002. Use of Currency Transaction Reports.
Report to Congress. Washington: US Department of the Treasury.
FinCEN (Financial Crimes Enforcement Network). 2003a. 2 003-2008 Strategic Plan. Wash¬
ington: US Department of the Treasury.
FinCEN (Financial Crimes Enforcement Network). 2003b. The SAR Activity Review Report: Tips
and Issues. US Department of the Treasury Bulletin 5. Washington: US Department of the
Treasury.
FinCEN (Financial Crimes Enforcement Network). 2004. The SAR Activity Reviezv: By the Num¬
bers. Washington: US Department of the Treasury.
Fowler, William Warner. 1981. Terrorism Data Bases: A Comparison of Methods, Missions and
Systems. Santa Monica, CA: RAND Corporation.

Freiberg, Arie, and Richard Fox. 2000. Evaluating the Effectiveness of Australia's Confiscation
Laws. Australian and New Zealand Journal of Criminology 33, no. 3: 239-65.
FSF (Financial Stability Forum). 2000. Report of Working Group on Offshore Financial
Centres. Financial Stability Forum, Basel, Switzerland (April).
FSF (Financial Stability Forum). 2004. Review of Offshore Financing Centres Initiative.
Financial Stability Forum press release. Basel, Switzerland (April 5).
GAO (General Accounting Office). 1996. Money Laundering: A Framework for Understanding US
Efforts Overseas. GAO/GGD 96-105. Washington: General Accounting Office.
GAO (General Accounting Office). 1998. FinCEN's Law Enforcement Support Role is Evolving.
GAO/GGD 98-117. Washington: General Accounting Office.
GAO (General Accounting Office). 2002. Money Laundering: Extent of Money Laundering through
Credit Cards is Unknown. GAO/GGD 02-670. Washington: General Accounting Office.

REFERENCES 195
GAO (General Accounting Office). 2003a. Combating Money Laundering: Opportunities Exist to
Improve the National Strategy. GAO/GGD 03-813. Washington: General Accounting
Office.
GAO (General Accounting Office). 2003b. Terrorist Financing: US Agencies Should Systematically
Assess Terrorists' Use of Alternative Financing Mechanisms. GAO 04-163. Washington:
General Accounting Office.
Gilmore, William C. 1999. Dirty Money: The Evolution of Money Laundering Countermeasures.
Strasbourg, France: Council of Europe Publishing.
Godshaw, G. P., R. Koppel, and R. Pancoast. 1987. Anti-Drug Law Enforcement Efforts and Their
Impact. Bala Cynwyd, PA: Wharton Econometric Forecasting Associates.
Gold, Michael, and Michael Levi. 1994. Money-Laundering in the UK: An Appraisal of Suspicion-
Based Reporting. London: Police Foundation.
Grant Thornton. 1992. Regulatory Burden Phase II - National Opinion Survey of Community
Banks. Study prepared for the Independent Bankers Association of America. Photocopy
(June).
Grant Thornton. 1993. Regulatory Burden: The Cost to Community Banks. Study prepared
for the Independent Bankers Association of America. Photocopy (January).
Gurr, Ted, Barbara Harft, and Monty G. Marshall. 2003. State Failure Task Force Report: Phase
III (Polity IV). McLean, VA: Science Applications International Corporation (SAIC).
Gutmann, P. M. 1977. The Subterranean Economy. Financial Analysts Journal 26, no. 34.
Hermann, Fritz. 2004. Will the OECD Convention Stop Foreign Bribery? In Global Corruption
Report 2004, Transparency International. London: Pluto Press.
Hills, Roderick M. 2001. The Economics of Corruption. Looking Ahead 23, no. 2: 3-6.
HM Treasury (United Kingdom). 2002. Consultation Document on Proposed Revision of the
Money Laundering Regulations 1993 and 2001. Photocopy (November).
Hufbauer, Gary Clyde, Jeffrey J. Schott, and Barbara Oegg. 2001. Using Sanctions to Fight
Terrorism. Policy Brief 01-11. Washington: Institute for International Economics.
Hufbauer, Gary Clyde, Jeffrey J. Schott, and Kimberly Ann Elliott. 2004. Economic Sanctions
Reconsidered. 3rd edition (forthcoming). Washington: Institute for International Economics.
IDA (International Development Association). 2003. Allocating IDA Funds Based on Per¬
formance. Washington: World Bank.
IF AC (International Federation of Accountants). 2002. Anti-Money Laundering. Photocopy.
www.ifac.org.
IMF (International Monetary Fund). 2001. Cyprus: Assessment of the Offshore Financial Sector.
Washington: International Monetary Fund.
IMF (International Monetary Fund). 2003a. IMF Concludes 2003 Article IV Consultation with
the Islamic State of Afghanistan. IMF Public Information Notice 03/147, Washington.
IMF (International Monetary Fund). 2003b. IMF Concludes 2003 Article IV Consultation with
Vietnam. IMF Public Information Notice 03/140, Washington.
IMF (International Monetary Fund). 2003c. Offshore Financial Centers — The Assessment
Program — A Progress Report and the Future of the Program. Washington: International
Monetary Fund.
IMF (International Monetary Fund). 2003d. United Kingdom: Financial System Stability
Assessment. Washington: International Monetary Fund.
IMF (International Monetary Fund). 2004a. IMF Executive Board Establishes Standard for
Anti-Money Laundering and Combating the Financing of Terrorism Assessments. IMF
Public Information Notice 04/33, Washington: International Monetary Fund.
IMF (International Monetary Fund). 2004b. Japan: Report on the Observance of Standards and
Codes — FATF Recommendations for Anti-Money Laundering and Combating the
Financing of Terrorism. Washington: International Monetary Fund.
IMF (International Monetary Fund). 2004c. Twelve-Month Pilot Program of Anti-Money
Laundering and Combating the Financing of Terrorism Assessments: Joint Report Prepared
by IMF and World Bank. International Monetary Fund, Washington.

196 CHASING DIRTY MONEY

Inter-American Dialogue. 2004. All in the Family: Latin America's Most Important International
Financial Flow. Washington: Inter- American Dialogue.
Interpol. 2003. Funds Derived from Criminal Activities (FOP AC). www.interpol.int/Public/
FinancialCrime / FOP AC / default .asp .
Inter-University Consortium for Political and Social Research. 2003. National Archive of
Criminal Justice Data 1995 and 2000. www.icpsr.umich.edu/NACJD/index.html.
Isachsen, Arne J., and Steinar Strom. 1980. The Hidden Economy, the Labor Market and Tax
Evasion. Scandinavian Journal of Economics 82, no. 2: 304-11.
Isachsen, Arne J., and Steinar Strom. 1985. The Size and Growth of the Hidden Economy in
Norway. Review of Income and Wealth 31, no. 1: 21-38.
Jackson, Dudley. 2000. The New National Accounts: An Introduction to the System of National
Accounts 1993 and the European System of Accounts 1995. Cheltenham: Edward Elgar.
James, Alvin, Jr. 2002. Statement before the US Senate Committee on Banking Housing and
Urban Affairs. Photocopy (October 3).
Joseph, Lester M. 2001. Money Laundering Enforcement: Following the Money. Economic
Perspectives: An Electronic Journal of the U.S. Department of State, http:/ /usinfo. state,
gov /journals /ites/ 0501 / ijee/justice.htm.
Kaplan, Lawrence J., and Salvatore Matteis. 1967. The Economics of Loansharking. American
Journal of Law and Sociology 27:244
Kaufmann, Daniel. 2003. Anti-Corruption within a Broader Developmental and Governance
Perspective: Some Lessons from Empirics and Experience. Statement by the Head of the
World Bank Delegation to the High Level Political Signing Conference for the UN
Convention Against Corruption, Merida, Mexico. Photocopy (December 9-10).
Kaufmann, Daniel, Aart Kraay, and Pablo Zoido-Lobaton. 1999. Governance Matters. World
Bank Policy Research Working Paper 2196. Washington: World Bank.

Kerry, John. 1997. The New War: The Web of Crime that Threatens America's Security. New York:
Simon and Schuster.
Key, Jack. 1979. Testimony before the Senate Permanent Subcommittee on Investigations of the
Committee on Governmental Affairs, 96th Congress, 1st Session, December 7, 11, 12, 13, and
14, 1979 (Illegal Narcotics Profits). Washington: US Government Printing Office.
Knack, Stephen, and Philip Keefer. 1995. Institutions and Economic Performance: Cross-Country
Tests Using Alternative Institutional Measures. Economics and Politics 7, no. 3: 207-27.
KPMG. 2003. Money Laundering: Review of the Regime for Handling Suspicious Activity Reports.
London: National Criminal Intelligence Service.
Levi, Michael. 2002. Money Laundering and Its Regulation. Annals of the American Academy of
Political and Social Sciences, no. 582: 181-94.
Levitt, Steven, and S. Venkatesh. 2000. An Economic Analysis of a Drug-Selling Gang's
Finances. Quarterly Journal of Economics 115, no. 3: 755-89.
Litan, Robert. 2004. Drafting Finance in the War on Drugs. Council on Foreign Relations, New
York. Photocopy (January).

Manski, Charles, John Pepper, and Carol Petrie, eds. 2001. Informing America's Policy on Illegal
Drugs: What We Don't Know Keeps Hurting Us. Washington: National Academies Press.
Masciandaro, Donato. 1999. Money Laundering: The Economics of Regulation. European
Journal of Law and Economics 7, no. 3: 225-40.
Masciandaro, Donato, and Allesandro Portolano. 2002. Inside the Black (List) Box: Money
Laundering, Lax Financial Regulation, Non-Cooperative Countries. A Law and Economics
Approach. Paulo Baffi Center, Bocconi University and Bank of Italy. Photocopy (May).
Mauro, Paolo. 1995. Corruption and Growth. Quarterly Journal of Economics 110, no. 3: 291-303.
Mauro, Paolo. 2002. The Persistence of Corruption and Slow Economic Growth. IMF Working
Paper WP/02/213. Washington: International Monetary Fund.
Mirus, Rolf, and Roger S. Smith. 1997. Canada's Underground Economy: Measurement and
Implications. In The Underground Economy: Global Evidence of its Size and Impact, eds. Owen
Lippert and Michael Walker. Vancouver: Fraser Institute.

REFERENCES 197
M.I.S. Trend. 2004. Les Questions Bancaires Actuelles 2004: Opinions et Representations

des Citoyens Suisses. Institut pour l'Etude des Marches et les Sondages d'Opinion,
Lausanne, Switzerland. Photocopy.
Moody-Stuart, George. 1997. The Grand Corruption: How Business Bribes Damage Developing
Countries. Oxford: World View Publishing.
Morgenthau, Robert M. 2004. Testimony before the US Senate Finance Committee. July 21, 2004.
Muscato, Frank. 2003. Testimony before Field Hearing on Organized Retail Theft: Conduit of
Money Laundering, House Committee on Government Reform, Subcommittee on
Criminal Justice, Drug Policy, and Human Resources, Washington. Congressional Record
(November 10).
National Association of Criminal Defense Lawyers (NACDL). 2001. Proposals to Reform Federal
Money Laundering Statutes. NACDL Money Laundering Task Force. Photocopy.
National Crime Authority. 1991. Taken to the Cleaners: Money Laundering in Australia, vol. 1.
Canberra, Australia: National Crime Authority.
National Criminal Intelligence Service (NCIS). 2003. United Kingdom Threat Assessment of
Serious and Organized Crime. London: National Criminal Intelligence Service.
Neild, Robert. 2002. Public Corruption: The Dark Side of Social Evolution. London: Anthem Press.
OECD (Organization for Economic Cooperation and Development). 1998. Harmful Tax
Competition: An Emerging Global Issue. Paris: OECD.
OECD (Organization for Economic Cooperation and Development). 2000. Towards Global Tax
Cooperation: Progress in Identifying and Eliminating Harmful Tax Practices. Paris: OECD.
OECD (Organization for Economic Cooperation and Development). 2002. Measuring the Non-
Observed Economy — A Handbook. Paris: OECD.
Office of National Drug Control Policy. 2000. What America's Users Spend on Illicit Drugs,
1988-1998. Washington: US Government Printing Office.
Office of National Drug Control Policy. 2001. What America's Users Spend on Illicit Drugs, 1988-
2000. Washington: US Government Printing Office.
OTA (Office of Technology Assessment). 1995. Information Technologies for the Control of Money
Laundering. Washington: US Government Printing Office.
Oxfam. 2000. Tax Havens: Releasing the Hidden Billions for Poverty Eradication. Policy
Research Department of Oxfam (Great Britain). Photocopy.
Passas, Nikos. 2000. Informal Value Transfer Systems and Criminal Organizations: A Study
into So-called Underground Banking Networks. Ministry of Justice, The Hague, Neth¬
erlands. Photocopy.
Performance and Innovation Unit of the UK Cabinet Office. 2000. Recovering the Proceeds of
Crime, www.number-0.gov.uk/ su./criminal/ recovering/ contents.htm.
Pieth, Mark, and Gemma Aiolfi. 2003. Anti-Money Laundering: Leveling the Playing Field.
Basel Institute of Governance, Basel, Switzerland. Photocopy (June).
Pillar, Paul R. 2003. Terrorism and US Foreign Policy. Washington: Brookings Institution.

President's Commission on Organized Crime. 1987. The Impact: Organized Crime Today.
Washington: US Government Printing Office.
Pricewaterhouse Coopers. 2003. Anti-Money Laundering Current Customer Revieiu Cost Benefit
Analysis. London: Financial Services Authority.

Public Broadcasting Service (PBS). 2000. Frontline: Drug Wars Interview with "David."
www.pbs.org/wgbh/pages/frontline/ shows/ drugs/interviews/ david.html.
Rahn, Richard, and Veronique de Rugy. 2003. Threats to Financial Privacy and Tax Competition.
Cato Institute Policy Analysis 491. Washington: Cato Institute.
Reinicke, Wolfgang H. 1998. Global Public Policy: Governing without Governments? Washington:
Brookings Institution.
Reuter, Peter. 1986. The Social Costs of the Demand for Quantification. Journal of Policy
Analysis and Management 5(4): 807-12.
Reuter, Peter, Robert MacCoun, and Patrick Murphy. 1990. Money From Crime. Santa Monica,
CA: RAND Corporation.
Robinson, Jeffrey. 1996. The Laundrymen: Inside Money Laundering. New York: Arcade Publishing.

198 CHASING DIRTY MONEY

Rotberg, Robert 1. 2003. State Failure and State Weakness in a Time of Terror. Washington: World
Peace Foundation.

Schneider, Friedrich. 2002. The Size and Development of the Shadow Economies and Shadow
Economy Labor Force of 21 OECD Countries: What Do We Really Know? Photocopy.
Schneider, Friedrich, and Dominik Enste. 2000. Shadow Economies: Size, Causes, and
Consequences. The Journal of Economic Literature 38, no. 1: 77-114.
Simon, Carl, and Ann Witte. 1982. Beating the System: The Underground Economy. Boston:
Auburn House.
Spillenkothen, Richard. 2002. Testimony before the Committee on Banking, Housing, and
Urban Affairs, US Senate. Board of Governors of the Federal Reserve System, Wash¬
ington. Photocopy (January 29).
Stessens, Guy. 2000. Money Laundering: A New International Law Enforcement Model. Cam¬
bridge, UK: Cambridge University Press.
Steinberg, James B., Mary Graham, and Andrew Eggers. 2003. Building Intelligence to Fight
Terrorism. Brookings Institution Policy Brief 125 (September). Washington: Brookings
Institution.

Suskind, Ron. 2004. The Price of Loyalty: George W. Bush, the White House, and the Education of
Paul O'Neill. New York: Simon and Schuster.
Suss, Esther C., Oral H. Williams, and Chandima Mendis. 2002. Caribbean Offshore Financial
Centers: Past, Present and Possibilities for the Future. IMF Working Paper WP/02/88.
Washington: International Monetary Fund.
System of National Accounts. 1993. Brussels, New York, Paris, Luxembourg: Eurostat, IMF,
OECD, UN, World Bank.
Tanzi, Vito. 1980. The Underground Economy and Tax Evasion in the United States: Estimates
and Implications. Banca Nazionale del Lavoro Quarterly Review 135.
Tanzi, Vito. 1999. Uses and Abuses of the Estimates of the Underground Economy. Economic
Journal 109: 338M7.
Transparency International. 1996. The TI Sourcebook. Berlin: Transparency International.
Transparency International. 2003. Transparency International Corruption Perceptions Index.
Berlin: Transparency International.
Transparency International. 2004. Global Corruption Report 2004. London: Pluto Press.
Truman, David R. 1995. The Jets and the Sharks are Dead: State Statutory Responses to
Criminal Street Gangs. Washington University Law Quarterly 73(2): 683-735.
UNECE (UN Economic Commission for Europe). 2002. Inventory of National Practices in
Estimating Hidden and Informal Economic Activities for National Accounts. Geneva:
UN Economic Commission for Europe.
United Nations. 2003a. Second Report of the Monitoring Group Established Pursuant to
Security Council Resolution 1363 (2001). United Nations Security Council S/2003/1070,
New York.
United Nations. 2003b. UN Convention Against Corruption. United Nations, New York.
US Department of State. 2003. International Narcotics Control Strategy Report. Washington:
United States Department of State.
US Department of the Treasury. 1999. The National Money Laundering Strategy for 1999.
Washington: US Department of the Treasury.
US Department of the Treasury. 2000. The National Money Laundering Strategy for 2000.
Washington: US Department of the Treasury.
US Department of the Treasury. 2001. The National Money Laundering Strategy for 2001.
Washington: US Department of the Treasury.
US Department of the Treasury. 2002. The National Money Laundering Strategy for 2002.
Washington: US Department of the Treasury.
US Department of the Treasury. 2003. The National Money Laundering Strategy for 2003.
Washington: US Department of the Treasury.
US Senate Minority Staff of the Permanent Subcommittee on Investigations. 2001a. Cor¬
respondent Banking: A Gateway for Money Laundering (February 5). http:/ / freedom.
orlingrabbe.com/money_laundering/correspondent_banking.pdf.

REFERENCES 199
US Senate Minority Staff of the Permanent Subcommittee on Investigations. 2001b.
Your Banker Is a Snitch, http:/ / freedom.orlingrabbe.com/lftimes/banker_snitch.htm.
US Sentencing Commission. 2002. 2001 Sourcebook of Federal Sentencing Statistics. Washington:
United States Sentencing Commission.
US Sentencing Commission. 2003. Monitoring of Federal Criminal Sentences, 1995 and 2000.
Dataset. www.ICPSR.umich.edu/NACJD/archive.html.
van Duyne, Petrus C. 2003. Money Laundering Policy: Fears and Facts. In Criminal Finances
and Organising Crime in Europe, eds. Petrus C. van Duyne, Klaus von Lampe, and James
L. Newell. Nijmgen, Netherlands: Wolf Legal Publishers.
Wechsler, William F. 2001. Follow the Money. Foreign Affairs 80(4) (July/ August): 40-57.
Winer, Jonathan M. 2002. Illicit Finance and Global Conflict. Paper prepared for the Fafo
Program for International Cooperation and Conflict Resolution. Photocopy (March 25).
Wolfsberg Group. 2002. Global Anti-Money Laundering Guidelines for Private Banking.
www.wolfsberg-principles.com/wolfsberg_principles_lst_revision.html.
Woolner, Ann. 1994. Washed in Gold: The Story behind the Biggest Money Laundering Investigation
in US History. New York: Simon and Schuster.
World Bank. 1997. The World Development Report 1997. Washington: World Bank.
World Bank. 2002. World Bank Group Work in Low-Income Countries under Stress: A Task Force
Report. Washington: World Bank.
World Bank. 2003a. Reference Guide to Anti-Money Laundering and Combating the Financing of
Terrorism. Washington: World Bank.
World Bank 2003b. World Development Indicators. Washington: World Bank.
World Bank. 2004. Doing Business in 2004: Understanding Regulation. Washington: World Bank.
Zarate, Juan C. 2004. Testimony before the House International Relations Subcommittee on
the Middle East and Central Asia. Photocopy (March 24).
Zhou, Xiaochuan. 2004. Statement to the International Monetary and Financial Committee.
International Monetary Fund, Washington. Photocopy (April 24).

200 CHASING DIRTY MONEY

Glossary and Acronyms

AML. Anti-money laundering.

Annunzio-Wylie Money Laundering Act (1992). US statute that inter

alia introduced suspicious activity reporting by financial institutions and
made it a crime to operate an illegal money transmitting business.

Anti-Drug Abuse Act (1988). US statute that inter alia gave the IRS the
power to seize property involved in the breach of money laundering laws.
AOC. US Administrative Office of the Courts.

APG. Asia-Pacific Group on Money Laundering. Established in 1998. The

group's 26 current members are Australia, Bangladesh, Brunei Darussalam,
Chinese Taipei, Cook Islands, Fiji, Hong Kong, India, Indonesia, Japan,
Macau, Malaysia, Marshall Islands, Nepal, New Zealand, Niue, Pakistan,
Republic of Korea, Palau, the Philippines, Samoa, Singapore, Sri Lanka,
Thailand, the United States, and Vanuatu.

AUSTRAC. Australian Transaction Report and Analysis Center. Australia's

anti-money laundering regulator and financial intelligence unit.
Basel Committee. Basel Committee on Banking Supervision, sponsored
by the G-10 central banks; normally meets at the BIS.
BCCI. Bank of Credit and Commerce International.

BIS. Bank for International Settlements.

BSA. Bank Secrecy Act (1970). First US legislation on reporting and mon¬
itoring of currency transactions.

201
Caricom. Caribbean Community. Established in 1990. The community's
26 current members are Anguilla, Antigua and Barbuda, Aruba, Bahamas,
Barbados, Belize, Bermuda, British Virgin Islands, Cayman Islands, Costa
Rica, Dominica, Dominican Republic, Grenada, Haiti, Jamaica, Mont¬
serrat, Netherland Antilles, Nicaragua, Panama, St. Kitts and Nevis,
St. Lucia, St. Vincent and the Grenadines, Suriname, Trinidad and Tobago,
Turks and Caicos Islands, and Venezuela.

CCE. Continuing Criminal Enterprise Act. US legislation directed at com¬

bating organized crime, passed in 1970 along with the Racketeer Influenced
and Corrupt Organizations (RICO) Act.

CDD. Customer due diligence. Procedure mandated in some countries

for banks, financial institutions, some nonfinancial businesses, and certain
professions to obtain information on customers before doing business with
or for them.

CFATF. Caribbean Financial Action Task Force. Established in 1990. Its

26 current members are Anguilla, Antigua and Barbuda, Aruba, Bahamas,
Barbados, Belize, Bermuda, British Virgin Islands, Cayman Islands, Costa
Rica, Dominica, Dominican Republic, Grenada, Haiti, Jamaica, Montserrat,
Netherlands Antilles, Nicaragua, Panama, St. Kitts and Nevis, Saint Lucia,
Saint Vincent and Grenadine, Suriname, Trinidad and Tobago, Turks and
Caicos Islands, and Venezuela.

CFT. Combating the Financing of Terrorism.

CICAD. Inter- American Drug Abuse Control Commission. An agency of

the Organization of American States (OAS) established in 1986, with 34
current members.

CIP. Customer identification program. Term used in US regulations for

CDD.

CIS. Commonwealth of Independent States. A group of countries that

were part of the former Soviet Union with the exception of the Baltic
countries.

CMIR. Report of International Transportation of Currency or Other

Monetary Instruments. Report that must be filed to export more than
$10,000 in currency from the United States.

CoE. Council of Europe. Established in 1949, with 45 current members.

CPI. Corruption Perceptions Index. An index published since 1995 by
Transparency International that ranks nations according to their perceived
level of corruption.

CPIA. Country Policy and Institutional Assessment. Annual report by

the World Bank Group's International Development Association (IDA) to
rate the performance of borrowers.

202 CHASING DIRTY MONEY

Crime Control Act (1990). US statute that inter alia established the US finan¬
cial intelligence unit, the Financial Crime Enforcement Network (FinCEN).
CTAG. Counter Terrorism Action Group of the G-8.

CTIF-CF1. Belgian Financial Intelligence Processing Unit.

CTR. Currency transaction report. Form that US financial institutions
are required to use to report currency transactions, generally exceeding
$10,000, to the Internal Revenue Service.

DEA. US Drug Enforcement Administration.

Drug Trafficking Offenses Act (1986). UK statute categorizing money

laundering as criminal offense.

EDU. Europol Drugs Unit.

Egmont Group of Financial Intelligence Group. International group of

FIUs established in 1995 to share experiences, promote the creation of
FIUs in nonmember states, and exchange data. The group has 84 current
members.

Eight Special Recommendations on Terrorist Financing. Recommen¬

dations issued in October 2001 by the FATF to combat the funding of ter¬
rorist acts and terrorist organizations.

ESAAMLG. Eastern and Southern Africa Anti-Money Laundering Group.

Established in 1999. The group's 11 current members are Botswana, Kenya,
Malawi, Mozambique, Mauritius, Namibia, South Africa, Seychelles,
Swaziland, Uganda, and Tanzania.

EU. European Union.

European System of Accounts. National accounting standards and def¬

initions established in 1995 for countries in the European Union. Known as
ESA95, the system is compatible with but somewhat more specific than the
System of National Accounts (SNA93) used by most coun tries in the world.
Europol. European Law Enforcement Organization.

FATF. Financial Action Task Force. Established by the G-7 Summit in

Paris in 1989 to examine measures to combat money laundering. Its 33
current members are Argentina, Australia, Austria, Belgium, Brazil,
Canada, Denmark, the European Commission, Finland, France, Germany,
Greece, the Gulf Cooperation Council, Hong Kong, Iceland, Ireland,
Italy, Japan, Luxembourg, Mexico, the Netherlands, New Zealand,
Norway, Portugal, the Russian Federation, Singapore, South Africa,
Spain, Sweden, Switzerland, Turkey, the United Kingdom, and the
United States.

FCPA. Foreign Corrupt Practices Act (1977). US statute that inter alia
prohibited the bribery of foreign government officials by US persons and
prescribes accounting and record-keeping practices.

GLOSSARY AND ACRONYMS 203

FDIC. Federal Deposit Insurance Corporation.

FEE. European Federation of Accountants.

FHT. Interpol Financial and High Tech Crimes Sub-Directorate.

FinCEN. Financial Crimes Enforcement Network. US financial intelli¬

gence unit established in 1990 in the US Treasury Department.

FIU. Financial intelligence unit. National center for the reception, analy¬
sis, and dissemination of suspicious transaction reports and other infor¬
mation regarding potential money laundering or terrorism financing.

FOPAC. Fonds Provenant d'Activites Criminelles. Criminal Funds

Investigation Group (Interpol).

Forty Recommendations. Recommendations issued by the FATF to set

international anti-money laundering standards. The recommendations were
first issued in 1990, slightly revised in 1996, and thoroughly revised in 2003.
FSA. UK Financial Services Authority.

FSAP. Financial Sector Assessment Program. Joint reviews of financial

systems in member countries by the International Monetary Fund and the
World Bank.

FSF. Financial Stability Forum. A multinational group established in 1999

by the G-7 to promote cooperation among financial authorities, super¬
visory groups, and international institutions.

FSS A. Financial System Stability Assessment by the International Monetary

Fund.

G AFISUD. Regional Financial Action Task Force on Money Laundering in

Latin America. Established in 2000. Its nine current members are Argentina,
Bolivia, Brazil, Chile, Colombia, Ecuador, Paraguay, Peru, and Uruguay.

GAO. US General Accounting Office (renamed the US Government

Accountability office in June 2004).

G-7. Group of Seven, established in 1975. Members are Canada, France,

Germany, Italy, Japan, the United Kingdom, and the United States.

G-8. Group of Eight: G-7 plus Russia.

G-10. Group of Ten, established in 1962. Its 11 current members are

Belgium, Canada, France, Germany, Italy, Japan, the Netherlands, Sweden,
Switzerland, the United Kingdom, and the United States.

Hawala. Arabic word for bill of exchange or promissory note. An infor¬

mal money transfer system alternative or parallel to traditional banking or
financial channels.

204 CHASING DIRTY MONEY

Hawaladar. Hawala operator.

IAIS. International Association of Insurance Supervisors.

IFAC. International Federation of Accountants.

IFIs. International financial institutions, primarily the International Mon¬

etary Fund and the World Bank.

IFT. Informal funds transfer system.

ILO. International Labor Organization.

IMF. International Monetary Fund.

IMFC. International Monetary and Financial Committee of the Inter¬

national Monetary Fund.

IMoLIN. International Money Laundering Information Network, estab¬

lished in 1996.

INCSR. US State Department's International Narcotics Control Strategy

Report.

International Money Laundering Abatement and Anti-Terrorist Financing

Act (2001). Title III of the USA PATRIOT Act, incorporating proposals to
strengthen and expand the US AML regime.

IOSCO. International Organization of Securities Commissions. 105 cur¬

rent members.

IRS. US Internal Revenue Service.

JAFIO. Japan Financial Intelligence Office.

KYC. Know Your Customer. Another name for customer due diligence
(CDD).

MLCA. Money Laundering Control Act (1986). US statute that inter alia
made money laundering a crime.

MLSA. Money Laundering Suppression Act (1994). US statute that inter

alia created the BSA Advisory Group, merged FinCEN and the Office of
Financial Enforcement, and authorized the AML regulation of casinos.

Money Laundering and Financial Crimes Strategy Act (1998). US statute

that inter alia mandated the National Money Laundering Strategy (NMLS)
reports for five years (1999-2003).
Money Laundering Prosecution Improvements Act (1988). US statute
that inter alia introduced liabilities and fines for facilitators of money¬
laundering activities, expanded the definition of financial institutions,
and provided the legal basis for sting operations.

GLOSSARY AND ACRONYMS 205

MONEYVAL. European Regional Financial Action Task Force, which
consists of the Council of Europe Selected Committee of Experts on the
Evaluation of Anti-Money Laundering Measures (2002). Formerly the
PC-R-EV.

MOT. Meldpunt Ongebruikelijke Transacties (Office for the Disclosure

of Unusual Transactions). Dutch financial intelligence unit.

MROS. Swiss Money Laundering Reporting Office.

MSB. Money service businesses involved in issuing money orders and
traveler's checks, money transmission services, check cashing, currency
exchange, etc.

NCCT. Non-Cooperative Countries and Territories identified by the

FATF as not meeting global AML standards.

NCIS. UK National Criminal Intelligence Service.

NMLS. National Money Laundering Strategy. Annual US report to the
US Congress from 1999-2003 on anti-money laundering efforts.

OAS. Organization of American States. Established in 1948. With 35 cur¬

rent members.

OECD. Organization for Economic Cooperation and Development.

Established in 1961. With 30 current members.

OFAC. US Treasury Department's Office of Foreign Assets Control.

OFC. Offshore financial center.

OGBS. Offshore Group of Banking Supervision. Its 19 current members

are Aruba, Bahamas, Bahrain, Barbados, Bermuda, Cayman Islands,
Cyprus, Gibraltar, Guernsey, Hong Kong, Isle of Man, Jersey, Labuan,
Macau, Mauritius, Netherlands Antilles, Panama, Singapore, and Vanuatu.

ONDCP. US Office of National Drug Control Policy.

PC-R-EV. Council of Europe Selected Committee of Experts on the

Evaluation of Anti-Money Laundering Measures (1997). The committee
conducted self-assessment and mutual assessment exercises of anti¬
money laundering measures in Council of Europe countries. Replaced by
MONEYVAL in 2002.

PEP. Politically exposed person. Individuals in prominent public posi¬

tions in their countries.

Polity IV. A 2003 study by Gurr, Harft, and Marshall of political regime
characteristics and transitions.

RICO. Racketeer Influenced and Corrupt Organizations Act (1986). US

statute that inter alia targeted the profits of criminal activities committed
by organized crime.

206 CHASING DIRTY MONEY

ROSC. Report on Observance of Standards and Codes. Report by the
International Monetary Fund on the extent to which members observe cer¬
tain international standards and codes.

SAR. Suspicious activity report. Form that US financial institutions are

required to use to report suspicious transactions, generally exceeding
$5,000.

Smurf. (1) To divide large illicit bank deposits into several deposits, each
less than $10,000, so they will not be subject to a CRT; also known as "struc¬
turing" of deposits. (2) Slang for a courier who makes such structured
deposits.

STR. Suspicious transaction report. Report that financial institutions

must file to a country's financial intelligence unit if a transaction is sus¬
pected to be linked to criminal activity or to terrorism. Used in certain
countries other than the United States.

Strasbourg Convention (1990). Convention that required each Council of

Europe member to adopt legislation that criminalizes money laundering.

System of National Accounts. National accounting standards and defi¬

nitions established in 1993 and used by most countries in the world today.
Often referred to as SNA93.

Technical Assistance Needs Assessment. A FATF exercise.

TEOAF. US Treasury Executive Office for Asset Forfeiture.

TI. Transparency International (see CPI).

TRACFin. French Unit for Processing Information and Action against

Illegal Financial Flows.

UNECE. UN Economic Commission for Europe.

UNDCCP. United Nations Office for Drug Control and Crime Prevention.

UNODC. United Nations Office on Drugs and Crime.

USA PATRIOT Act. Uniting and Strengthening of America by Provid¬

ing Appropriate Tools Required to Intercept and Obstruct Terrorism Act
(2001). US statute on money laundering and terrorism financing enacted
after the September 11, 2001, terrorist attack.

Vienna Convention (1988). UN convention against illicit traffic in nar¬

cotics drugs and psychotropic substances.

Wolfsberg Principles. Joint anti-money laundering initiative by 12 inter¬

national banks setting common standards on private banking and other
areas. The 12 banks are ABN AMRO, Santander Central Flispano, Bank of
Tokyo Mitsubishi, Barclays, Citigroup, Credit Suisse Group, Deutsche
Bank, Goldman Sachs, HSBC, JP Morgan Chase, Societe Generale, and
UBS.

GLOSSARY AND ACRONYMS 207

Index

ABA. See American Bar Association

costs, 93-94 of, 98
concepts
Abacha, Sani, 151n, 154
accountants, and AML responsibilities, 62b-63b to customers, 95
ACLU. See American Civil Liberties Union
to general public, 6, 101-02
Afghanistan, 143, 156, 160b, 164—65 to government,
al Qaeda, 146, 147 measuring, 191 95-96
blocking funds to, 143-44 to private sector, 98
budget of, 143 and crime reduction, 119-20, 120
Aleman, Amoldo, 151, 153
developing countries' efforts, 184, 185
Almoudi, Abdurahman, 146-47 differences among national practices, 91-92
American Bar Association (ABA) effect on accountants, auditors, 62b-63b
on lawyers' reporting requirements, 60-61 effect on core financial institutions, 49, 54—56
on real estate regulation, 58 effect on non-core financial institutions, 56-57
American Civil Liberties Union (ACLU), 146 effect on nonfinancial businesses, 57-59
AML. See anti-money laundering effectiveness of, 6, 126
Andorra, 67 enforcement pillar, 4—5, 46-49, 47/
Angola, 159 expanded regulations of, 7
Annan, Kofi, 91 and failing states, 164
Annunzio-Wylie Money Laundering Act, 50f, 55n, global regime, 1, 46, 78-79, 191, 192
61
challenges for, 182-83
description of, 193 effectiveness of, 5
Impact on banks, 65, 65n and electronic financing, 73b
Anti-Drug Abuse Act, 501, 66, 68 enforcement efforts, 89-93, 172
description of, 193 evolution of, 50f-53f
Antigua, 89 international cooperation efforts, 183-86
anti-money laundering (AML). See also specific international institutions, roles of, 182-83
countries and NCCT, 165-70
assistance to failed states, 156 "noncooperation" with, 167
and banking, 6, 136, 137-38 prevention efforts, 59-60, 79-89, 172
and blocked, frozen funds, 142, 142m, 143, 144 recommendations for, 171, 179-86
and cooperation with private sector, 176-77 relevant regional, UN conventions, 89-90
and corruption, 164 SARs, filed by country, 116 1
cost-effectiveness of, 173-74
scope of, 45, 172-75

209
anti-money laundering, global regime — continued Great Eastern Bank of Florida, 108b
US role in, 49, 179-82 involvement in money laundering, 34
"white list," 186m
reputation of, 130-31
goals of, 3-4, 6-7, 139, 171 response to USA PATRIOT Act, 61-62
IMF/World Bank reviews of, 167-68 Switzerland, 80, 81
Barbuda, 89
and integrity of financial systems, 129-32
Barr, Bob, 70
and kleptocracy, 150-53, 151n
measures of effectiveness, 142-43 Basel Committee on Banking Supervision, 80, 82,
money profiles, 174 130, 193
performance measures for, 124-26 BCCI. See Bank of Credit and Commerce
International
prevention pillar, 4-5, 46-49, 47 f, 176
reporting requirements, 54-55, 54f, 59 n Beacon Hill Service Corporation, 29 n, 134, 136
for lawyers, 59 n, 60-61, 172-73, 182 bearer securities, 31-32
and real estate transactions, 58-59 Belgium, 13
for professionals, 59-61, 64-65 SAR filings, 116t
research underground economy, 14f, 18
international cooperative support for, 190 Benevolence International Foundation, 2b
Benex, 134
lack of, 188-89
recommendations for, 189-91 BIS. See Bank of International Settlements
societal contributions of, 175-76 "blue-collar" crime, and money laundering, 41,
and technology use, 174 41f

and terrorist financing, combating, 139-48, 146m Blum, Jack, 178m

tools for, 2b Bosnia, 160b
AOC. See United States, Administrative Office of Bremer Commission report, 141
the Courts bribery. See also corruption
APG. See Asia /Pacific Group on Money international, 155
Laundering and money laundering, 41 f, 42
Argentina, SAR filings, 117 Broadway National Bank, 108b, 134
arson (fraud), proceeds from, 22f BSA. See Bank Secrecy Act
Arthur Young & Co., 63 b Bullock, Barbara A., 27b
Asia/Pacific Group on Money Laundering (APG), Burma. See Myanmar/Burma
51f, 84, 193 Burundi, 159
Association of Certified Fraud Examiners, 20 Bush, George W., 76-77
Bush administration
auditors, and AML responsibilities, 62b-63b
Australia and NMLS, 74, 75-76, 77
AML regime, 79 taxation issues, 77

compliance with Forty Recommendations, 82 businesses, nonfinancial

National Crime Authority, 39 reporting requirements, 54f
SAR filings, 116f, 118 and US AML regime, 57-59, 57m
underground economy, 14f
Australian Transaction Report and Analysis Cadavid, German, 38f
Center (AUSTRAC), 118, 193 Camdessus, Michael, 13
Austria, 14t Canada
regulation of lawyers, 82
Bahamas, the, 88 underground economy, 14f, 18
Bahrain, 88 Caribbean Community (Caricom), 194
Bank of Credit and Commerce International Caribbean Financial Action Task Force (CFATF),

(BCCI), 131-32 84,194

Bank of International Settlements (BIS), 79 cash smuggling, stockpiling, 28
Bank of New York, 134, 166 n and predicate crimes, 33f
Bank Secrecy Act (BSA), 1, 50 1, 55, 65, 66, 193 casinos, 28-29
Advisory Committee, 61 and link to predicate crimes, 33f
cost of compliance, 98 numerical description for, 64
banking industry reporting requirements, 57
Bank of New York, 134 Cayman Islands, 88, 167, 185
Broadway National Bank, 108b, 134 CCE. See Continuing Criminal Enterprise Acts
Chiasso scandal, 80 CDD. See customer due diligence
and financial system integrity, 129, 130, 131, Celent Communications, AML cost estimates,
99-100, lOOn
133-35, 133m, 135m

210 CHASING DIRTY MONEY

CFATF. See Caribbean Financial Action Task Cuba, 143
Force
Cuellar, Mariano-Florentino
check-cashing agencies, 55 n critique of US AML regime, 39M0, 174
Chiasso banking scandal, 80
currency exchange, 31 b, 122-23
Chiluba, Frederick, 151n and predicate crimes, 331
China currency transaction reports (CTR), 34, 511, 55, 55 n
AML review of, 168 description of, 195
FATF membership, 168n and excess reporting, 94

CICAD. See Inter- American Drug Abuse Control currency-demand approach, 11-12
Commission
critique of, 15-16
CIP. See customer identification program currency-ratio approach, 12n
CIS. See Commonwealth of Independent States customer due diligence (CDD), 54 n, 541, 56, 56n
Clinton, Bill, 141 costs to general public, 101-02
Clinton administration description of, 194
and NMLS, 74, 75-76, 77 FATF recommendations, 59 n
taxation issues, 77 and prevention of money laundering, 46
CMIR. See Report of International Transportation and real estate, 58
of Currency or Other Monetary UK implementation of, 82
Instruments
US promotion of, 147-48
Colombia, 168 customer identification program (CIP), 54 n
Commercial Bank of Syria, 144 Cyprus, and tax evasion, 181«
Commonwealth of Independent States (CIS), 194 Czech Republic, 90
complex states, 157, 160b
Dauber, Alfred, 108b
confiscation of property, 69-70
Denmark, 141
Congo, Democratic Republic of the, 156, 160, 160b
Continuing Criminal Enterprise (CCE) Acts, 69, developing countries
194 AML efforts by, 184, 185
Cook Islands, 86 b impact of money laundering on, 93
corruption, 148-49. See also kleptocracy domestic money laundering, 26
definitions of, 149
drug addicts, estimates of sales to, 21-22
monetary classification of, 149 drug trafficking
in India, 169, 169n
and money laundering, 41f, 42, 156, 163-65, 1641
public, 150 and money laundering, 26b, 331, 34, 41, 411
and weak governance, 149 and money-laundering convictions, 110, 1111,
corruption perceptions index (CPI), 163 113
description of, 194 in US, 20, 221
and INCSR rating, 1641 Duvalier, Jean-Claude, 151n
Council of Europe Selected Committee of Experts
on the Evaluation of Anti-Money Eastern and Southern Africa Anti-Money
Laundering Measures (PC-R-EV), 198 Laundering Group (ESAAMLG), 84, 195
Council on Foreign Relations, 64 EDU. See Europol Drugs Unit
counterfeiting Egmont Group of Financial Intelligence Units, 2b,
123
and money-laundering methods, 331
proceeds from, in US, 221 description of, 92, 195
counterterrorism, 144, 144n
Egypt, 85, Stuart,
Eizenstat, 146, 16762 b
Country Policy and Institutional Assessment
(CPIA), 157, 194 electronic finance, 73b
embezzlement, 27b, 29 b
ratings, 157n, 1581-591
CPI. See corruption perceptions index cases, 151, 151n
credit cards
and money-laundering convictions, 110, 1111
and advance payments, 31b enforcement pillar, T-5, 4 7f. See also under
as a money-laundering method, 34 anti-money laundering
and predicate crimes, 331 and corruption, kleptocracy, 150
Credit Suisse, 80 and electronic finance, 73b
Crime Control Act, 501, 195 hypothetical example, 48
criminal income
key elements of, 46
and GDP, 20, 201, 21/ and NMLS action items, 751
Enron, 42
reliability of estimates, 20-23
US estimates, 221 Equatorial Guinea, 135, 153
CTR. See currency transaction reports ESA95. See European System of Accounts

INDEX 211
ESAAMLG. See Eastern and Southern Africa recommendation 21, 85m, 86b
Anti-Money Laundering Group recommendation 24b, 61, 83m
Estrada, Joseph, 151n recommendation 27, 91
euro notes, 93 n recommendation 29, 83
European Council Heads of Government recommendations
and terrorist financing, combating, 146 on CDD, 59m
European Federation of Accountants (FEE), 63b for nonfinancial businesses, 59, 83m
European System of Accounts (ESA95), 17, 18m, for professionals, 59-61, 59m
195
for reporting requirements, 60-61
European Union
regional groups, shortfalls of, 167-68
evolution of AML regime, 501-531 review, evaluation of AML regimes, 86-87
Commission Decision of 1994, 18n and Seychelles government, 85m
euro notes, 93n tax evasion, 91

money-laundering directive, 501-531, 81m, 90, terrorist financing, combating, 140

and threat of terrorism, 118-19 Financial Action Task Force on Money
underground economy, 14f Laundering in Latin America (GAFISUD),
Europol Drugs Unit (EDU), 51 f, 92
Evian Summit, 155 Financial84 Crime Enforcement Network (FinCEN),
extradition, 186 56, 56m, 70, 106, 174
ExxonMobil, 152n description of, 196
failure to file SARs, 108b

failed, failing states, 157, 158f-59f, 159 and real estate regulation, 58-59
definition of, 157 financial institutions. See also banking industry;
and INCSR ratings, 1621 insurance industry; securities transactions
and money laundering, 156-63 defined by FATF, 56m
noncompliance with FATF, 162m financial institutions, core

politically versus economically, 159-60, 162-63, affected by AML efforts, 49, 54-56
1621 numerical description for, 64, 65
Falkenrath, Richard, 146n and prevention pillar, 56
FATF. See Financial Action Task Force
reporting requirements, 541
FCPA. See Foreign Corrupt Practices Act supervision of, 55
Federal Reserve Board
tracking, assessing system integrity, 190-91
and cooperation with law enforcement, 137m financial institutions, non-core
Federal Reserve System affected by AML efforts, 56-57
Board of Governors, 25 numerical description for, 64
FEE. See European Federation of Accountants reporting requirements, 541
Fernandez, Nilo, 381 financial intelligence units (FIU), 92, 196
Financial Action Task Force (FATF), 2b, 531 Financial
196 Sector Assessment Program (FSAP), 87m,
assessment of AML regimes, 86b
description of, 195 Financial Stability Forum (FSF), 196
Eight Recommendations, 140, 145, 195 financial system integrity
enforcement power, 85 and AML activities, 129-32
establishment of, 81
evaluating progress in, 132-36
and financial system integrity, 129 Financial System Stability Assessment (FSSA), 87m
Forty Recommendations, 7, 501, 511, 531, 81-83 FinCEN. See US Treasury Department, Financial
Crime Enforcement Network
compliance challenges, 172-73
fine art, 31b
description of, 196
enforcement issues, 92 Finland, 141

implementation of, 82 FIU. See financial intelligence units

and kleptocracy, 152-53 Foreign Corrupt Practices Act (FCPA), 501, 152m,
196
and predicate crimes, 90-91
IMF compliance with, 145m Foreign Narcotics Kingpin Designation Act, 521
and INCSR ratings, 161-63, 1611 forfeiture of property, 69-70
member countries, 84, 84m, 145m, 195 France, 80, 90

Middle East-North Africa, 84 SAR filings, 1161

and NCCT, 86b, 89, 156-57, 160, 161 underground economy, 141
fraud
noncooperation list, 167, 167m
recommendation 3, 92 and money-laundering convictions, 1111
recommendation 7, 136 and money-laundering methods, 331
recommendation 19, 83m in US, 20, 221

212 CHASING DIRTY MONEY

FSAP, See Financial Sector Assessment Program IMoLIN. See International Money Laundering
FSF. See Financial Stability Forum Information Network
FSSA. See Financial System Stability Assessment India, 84, 185
Fujimori, Alberto, 151n
AML regime of, 168-69
drug trafficking in, 169, 169n
GAFISUD. See Financial Action Task Force on FATF membership, 168n
Money Laundering in Latin America Indonesia, 85, 86 b, 141, 146, 151k, 154
gambling informal funds transfer, 30b 83, 84
casinos, 28-29, 33 f, 57, 64 inmate survey, on money laundering, 112-13
horse racing, 29 INCSR. See International Narcotics Control
lotteries, 29 Strategy Report
in US, 20, 22f insurance industry
GAO. See US General Accounting Office FATF Recommendations, 83k
Gatekeepers Initiative, 60 and predicate crimes, 331
Georgia, 155 UK regulation of, 82
Germany, 80 insurance policies, 29-30
and banking requirements, 147 Inter-American Drug Abuse Control Commission
money-laundering convictions, 118 (CICAD), 194
underground economy, 13, 14f Internal Revenue Service (IRS), 55
Gilmore, William C., 79 International Association of Insurance

gold, 31b Supervisors (IAIS), 80

Grant Thornton, AML cost estimates by, 98 International Emergency Economic Powers Act,
Great Eastern Bank of Florida, 108b, 134 141
Greece, 13 International Federation of Accountants, 63b
AML efforts, 90 International Monetary Fund (IMF)
underground economy, 14f assessment of offshore financial centers, 88
Griffen, James, 152n compliance with FATF recommendations, 145k,
168
Group of Eight (G-8)
members of, 196 position on Afghanistan regulations, AML
Moscow Ministerial Conference, 60 efforts, 165
Okinawa Summit, 155 review of AML regime compliance, 86-87, 87k,
summit meetings, 155 168-70
review of US AML regime, 180
Group of Seven (G-7)
members of, 196 International Money Laundering Information
Paris Economic Summit, 81 Network (IMoLIN), 511
position on global AML, 78 International Narcotics Control Strategy Report
terrorism declarations, 140, 141 (INCSR) ratings

Group of Ten (G-10), 79 and corruption perceptions index, 1641

members of, 196 of failed, failing states, 1621
response to US advocacy of reporting require¬ and relative FATF treatment, 161-63, 1611, 1621
ments, 80 International Organization of Securities
Gusinsky, Vladimir, 153n Commissions (IOSCO), 80
Internet, and money laundering, 73b
Haiti, 151k, 159
Interpol, High Tech Crimes Sub-Directorate, 92-93
hawalas, 2b, 83, 84, 147, 196, 197 IOSCO. See International Organization of
HealthSouth, 25 Securities Commissions
Hirsch, Robert, 38f Iran, 143
Holmes, Leroy, 2 7b Iraq, 143
Hong Kong, 88 Ireland, 141, 146
Hood, Clyde, fraud suit against, 26 n IRS. See Internal Revenue Service
horse racing, 29 Israel, 167
human trafficking Italy, 13, 18, 18k
and underground banking, 28 b
in US, 22, 221
Japan, 88
Hungary, 167 SAR filings, 1161, 117
underground economy, 13, 14 1
LAIS. See International Association of Insurance JP Morgan Chase, 134
Supervisors
IMF. See International Monetary Fund Kenya, 151m
immigrants, illegal. See human trafficking Kerry, John, 70, 131

INDEX 213
Kibaki, Mwai, 155 data compilation for, 189
Klein, Adam, 66 n definition of, 9, 65m
domestic, 26
kleptocracy, 43, 71, 134, 149-55, 151m. See also
bribery; corruption and drug trafficking, 34
embezzlement cases, 151, 151m economic modeling of, 191
monetary classification of, 149m and electronic finance, 73b
and money laundering, 156 elements of, 25
prosecution of, 150, 152n, 153n and failed states, 156-63
US pursuit of foreign offenders, 153-55 financial penalties for, 113-14, 114f
"know your customer" (KYC), 130, 197 and gambling, 28-29
KPMG, AML cost estimates, 99, 100, 100m impact of globalization, deregulation, 78
impact on developing countries, 93
Lansky, Meyer, 121 informal value transfer systems, 30b
Laos, 159
inmate survey regarding, 112-13
lawyers. See also professionals and insurance policies, 29-30
as an agent for money laundering, 34 and kleptocracy, 156
Canadian regulation of, 82 and legitimate business ownership, 30 b
and CDD, reporting requirements, 59m, 60-61 phases of, 3
expansion of AML responsibilities, 172-73 policy guidance, consequences, 23-24
and FATF Recommendations, 60-61 and predicate crimes, 22-23, 69, 90, 111b
Lazarenko, Pavel Nickolayevich, 150, 151m prevention efforts, 46. See also prevention pillar
Lebanon, 167
prosecution, conviction data, 108-09, 109-12,
Lesotho, 160b 109b, 1101, Hit, 187
Liberia, 67, 151m prosecution of, 69
Libya, 143 and purchase of goods, 31b
licensing, 57 and real estate transactions, 31 b
Liechtenstein, 67, 85, 86, 167 and relationship to national accounts, 18
lotteries, 29 research, lack of, 8, 188-89
Luxembourg, 88, 90, 146 risk of, in the US, 127-28
and securities transactions, 30-32
macroeconomic estimates "shell" corporations, 30b-31b
currency-demand approach, 11-12 "smurfing," 26 b, 30 b, 199
of money laundering, 10, 11-18 social consequences of, 42-43
and national income accounts, 16 US law enforcement of, 186-88
mafia, 118 wire, electronic funds transfers, 30 b
March 11, 2004, EU responses to, 146, 147 Money Laundering and Financial Crimes Strategy
Marcos, Ferdinand, 151n Act, 51b, 70, 197
Marshall Islands, 67 Money Laundering Control Act (MLCA), 50t, 65,
Mathewson, John, 36 197
Maxwell, Robert, 29, 29 b, 124 Money Laundering Prosecution Improvements
Maxwell Group Newspaper PLC, 29 b Act, 197
Mayer, Martin, 137w Money Laundering Suppression Act, 51b, 197
Mexico
money profiles, 174
SAR filings, 117 money service bureaus, 56, 198
US sting operation in, 91 money-laundering estimates, 4, 5, 23
microeconomic estimates approaches to, 10, 16. See also macroeconomic
of money laundering, 10, 19-23 estimates; microeconomic estimates
Milosevic, Slobodan, 151m currency-demand approach to, 15-16
MLCA. See Money Laundering Control Act macroeconomic approach to, 11-18
Mobutu Sese Seko, 151m microeconomic approach to, 19-23
Moi, Daniel arap, 151m, 155 of OECD countries, 13, 14b
Monaco, 67
UNECE
for US, 9 report, 17-18
money laundering. See also self-laundering
bank involvement in, 34, 133-35, 133m, 135m money-laundering methods, 27-32
blue-collar crime, 41, 41f and corresponding crimes, 32, 33f, 34
and cash smuggling, 28 money-laundering offenses
and corruption, 149, 156, 163-65, 164f classification of, 40M3, 41b
and credit card advance payments, 31b dimensions of, 40-41
criminalization of, 176 money-laundering services
and currency exchange, 31b, 122-23 agents of, 34-35

214 CHASING DIRTY MONEY

 markets for, 5-6, 8, 35-36, 39M0, 120-24 Okinawa Summit, 155

policy-relevant price, 35 O'Neill, Paul, 67, 9 7n

prices for, 35-36, 37f— 38f, 121-22 Operation Casablanca, 91, 137
professional agents for, 36 Operation Dinero, 68
MONEYVAL, 198 Operation Highbind, 68
Myanmar/ Burma, 85, 86b, 163, 183 Operation Juno, 68
Operation Polar Cap, 36, 41, 188
"name and shame" process, 85, 86b Organization for Economic Cooperation and
National Drug Control Strategy, 71 Development (OECD), Convention on
national income accounts and money-laundering Combating Bribery of Foreign Public
estimates, 16 Officials in International Business
Transactions, 152
National Money Laundering Strategy (NMLS), 7-8,
9-10, 46, 52 1, 70-72, 74-77, 198 description of, 198
action items, 71, 72, 74f, 75, 75f and tax havens, 67, 6 In
and AML efforts, 75f OTA. See US Office of Technology Assessment
Bush administration, 74, 75-76, 77
Clinton administration, 77 Pakistan, 84, 151)1, 168
content of, 70-72 Pan American flight 103, 140
Panama, 167
continuity, differences among, 74-77, 74f, 75f
critique of, 178-79 PC-R-EV. See Council of Europe Selected
professionals, expanding the role of, 76-77 Committee of Experts on the Evaluation of
recommendations for revisions to, 179 Anti-Money Laundering Measures
shortcomings of, 71-72, 76, 77 PEP. See politically exposed person
taxation issues, 77 Peru, 151 n, 155, 160b
Nauru, 85, 86b, 165-66, 166n, 183 Philippines, 85, 86 b, 151)7, 167, 169, 169)7
NCCT. See Non-Cooperative Countries and Pillar, Paul R., 144
Territories initiative Pinochet, Augusto, 135
Netherlands PNC Financial Services Group, 135n
money-laundering convictions, 118 politically exposed person (PEP), 198
SAR filings, 1161, 117 "politicides," 157n
underground economy, 141 Polity IV study, 157, 159, 198
Netherlands Antilles, 88 failed, failing states, 158t-59f, 160b
New Zealand, 141 Portugal, 13, 14f, 90
Nicaragua, 151n, 155 precious metals, stones, 57-59, 64
Nigeria, 85, 86 b, 155 numerical description for, 64
NMLS. See National Money Laundering Strategy reporting requirements, 57
Non-Cooperative Countries and Territories initia¬ predicate crimes
tive (NCCT), 89 categories of, 4
description of, 165, 165n, 198 confiscation of proceeds from, 69-70
effectiveness of, 165-69 and Forty Recommendations, 90-91
and FATF recommendations, 86b, 89, 156-57, and money laundering, 22-23, 32, 33f, 34, 69, 90,
160,161 lilt

Nauru example, 165-66, 166n prosecution of, 69

nonfinancial businesses, FATF recommendations prevention pillar, 4—5, 47/ 176. See also under
for, 59, 83n anti-money laundering
North Korea, 163, 163)7 and corruption, kleptocracy, 150
Norway, 141 and electronic finance, 73b
hypothetical example, 47M8
Obiang, Teodoro, 153 key elements of, 46
OECD. See Organization for Economic and NMLS action items, 75f
Cooperation and Development numerical descriptions for, 64-65
in US, 54f
offshore financial centers (OFC), 87-89, 185
assessment of, 88 Pricewaterhouse Coopers, 99, 99n, 100n, 101
and business loss, 88-89 private sector, cooperation with AML efforts,
cross-border assets, 88-89, 89/, 891 176-77
IMF assessment of, 88 production boundary, 17
Nauru example, 166 professional services, for money laundering, 34
regulation of, 178n professionals. See also accountants; auditors;
Offshore Group of Banking Supervision (OGBS), lawyers
501, 198 as agents for money laundering, 34-35

INDEX 215
professionals — continued Sherman, Tom, and financial system integrity, 129
and CDD requirements for, 60 Sierra Leone, 160

expansion of AML responsibilities, 61, 76-77, Singapore, 88

smuggling
172-73
FATF recommendations for, 59-61, 59m and money-laundering convictions, lilt
involvement in predicate crimes, 33f and predicate crimes, 33 f
numerical description for, 64 "smurfing," 26 b, 30 b, 199
and reporting requirements, 54f, 58-59, 59n, 60 SNA. See System of National Accounts
property, forfeiture of, 69-70 Solomon Islands, 159
prostitution, in US, 19, 22f Somalia, 160b

Spain, 14f, 90, 141

Racketeer Influenced and Corrupt Organization Spence, Richard, 38f
Act (RICO), 50f, 69, 198 Spillenkothen, Richard, 141
real estate transactions, 31b
sting operations, 68, 91, 187, 188
and AML efforts, 57-58 stolen goods, revenue estimates of, 21
numerical description for, 64 STR. See suspicious transaction reports
and predicate crimes, 33( Strasbourg Convention, 199
Regional Financial Action Task Force on Money Sudan, 143, 159
Laundering in Latin America (GAFISUD), Suharto, Mohamed, 151m, 154
196 Summers, Lawrence, position on global AML, 78
Regulatory DataCorp, 64 suspicious activity reports (SAR), 34, 48, 51f, 65
Report of International Transportation of description of, 199
Currency or Other Monetary Instruments failure to file reports, 108b
(CMIR), 55, 194 filings, by country, 116f
reporting requirements filings in US, 106-08, 107/
for lawyers, 59 n, 60-61, 172-73, 182 and financial system integrity, 132
as a means to reduce crime, 68, 119
for professionals, 54t, 58-59, 59m, 60, 61, 76-77
and real estate transactions, 58-59 quality of, 94
Reports on Observance of Standards and Codes tracking the use of, 189-90
(ROSC), 87m, 199 versus STR, 82-83
Ribadu, Nuhu, 155 suspicious transaction reports (STR)
RICO. See Racketeer Influenced and Corrupt description of, 199
Organization Act versus SAR, 82-83
Sweden, 14f
Riggs National Bank, 135, 135m, 153
robbery. See theft Swiss Bankers Association, 80
ROSC. See Reports on Observance of Standards Switzerland
and Codes AML regime, 80
Rotberg study, 157, 159 banking system, 80, 81, 130
failed, failing states, 158t-59f SAR filings, 116f, 117
Rubin, Robert, 119m underground economy, 14b
Russia, 85, 86b, 167, 167m, 185
Syria, 143, 144
kleptocracy, 153m System of National Accounts (SNA), 17, 199

Saccoccia, Stephen, 38f Tajikistan, 159

Sanio, Jochen, 147 Tanzi, Vito, 11, 15
SAR. See suspicious activity reports tax competition, 91
Saudi Arabia, 135, 146
US position, 67m
Scandinavia, 13 tax evasion, 181, 181m
Schengen Information System, 147 crimes, 20m
and FATF, 91
Scrushy, Richard, 25-26
Sea Island Summit, 155 impact on national accounts, 12
securities transactions, 30-32 and money laundering, 33f, 66-67, lilt
"put" and "call" transactions, 32 proceeds from, in US, 22f
tax havens, 67
and predicate crimes, 33f
taxation issues
self-laundering, 23m, 27b, 29
September 11, 2001, 141 Bush versus Clinton approaches, 77
Serbia/Yugoslavia, 151m in NMLS reports, 77
Seychelles, 85, 85m Taylor, Charles, 151m
shell corporations, 30b-31b terrorism
and predicate crimes, 33f effect on AML regime efforts, 140-41

216 CHASING DIRTY MONEY

 and money laundering, 33 1, 41 1, 42-43 Administrative Office of the Courts (AOC), 109,
sanctions against, 143 109f, UOf
AML regime
threat to Europeans, 118-19
terrorism financing, combating, 1, 2b-3b, 139-42, advocacy of banking supervision, 79-80
141m, 143, 144, 145 and CDD requirements, 60, 173
costs of, 141-42, 143 compared with other regimes, 79
FATF recommendations on, 140, 145 costs of, 93-95, 94n, 103
measures of success, 144, 144m
to general public, 101-02
priority among nations, 180-81 to government, 95-98, 96/, 97 n
UN report on, 145M6, 147 gross financial, 93-94
Thailand, 168
to private sector, 98-101
theft, in US, 22f Cuellar's critique of, 174
transaction measures versus value-added mea¬ directives, evolution of, 50f-53f
sures, 13m enforcement efforts, 65-68, 96/

Transparency International, 163 FATF challenges for, 177-78

Turkey, 85m and foreign predicate crimes, 181-82
global leadership by, 49, 179-82
Ukraine, 86 b, 150, 151n, 167 goals for, 741, 77
IMF/World Bank review of, 168m, 180, 192
underground banking and human trafficking, 28b
underground economies, 11 investigation, 68-70
and kleptocracy, 182
in Belgium, 141, 18
in Canada, 14t, 18 law enforcement of, 186-88
estimates for OECD countries, 13, 14t legislation compromise, 74n
NMLS action items, 74f, 75f
estimating the activity of, 16, 17
and GDP, 13, 14f and nonfinancial businesses, 57-59, 57m
outlays, 96/, 9 7n
in Italy, 13, 14f, 18, 18«
in Japan, 13, 14f prevention efforts, 46, 49, 54-65, 54f
applied to core financial institutions, 56
legal versus illegal, 12-13, 12t
numerical
outlays, 96/descriptions, 64-65
monetary versus nonmonetary transactions, 12,
12t
prevention pillar, 54t
"reasonable consensus definition/' 12
professionals, increased regulation of, 56m, 60,
and relationship to national accounts, 18 61
in Spain, 13, 14f
Tanzi definition of, 12 public-private cooperation, 61

taxonomy of, 12f reporting requirements, 60, 68, 147-48, 173-74

SARs, role of, 68, 106-08, 107/ 1161
transaction versus value-added measures, 13m
in US, 13, 18 scope of, 118-19
UNECE. See United Nations, Economic strategies, 70-78
and tax evasion, 66-67, 181
Commission for Europe
terrorist financing, combating, 180-81
United Kingdom, 88
arson (fraud), proceeds from, 22f
AML regulations for professionals, 63b Bureau of Justice Statistics
Drug Trafficking Offenses Act, 50f, 51f, 195 inmate survey regarding money laundering,
implementation of Forty Recommendations, 82
112-13
Performance and Innovation Unit, 131
money-laundering prosecution, conviction
Proceeds of Crime Act, 52f, 53f, 63 b
data, 109f, 1101
pursuit of Abacha, 154 criminal income, 20f
SAR filings, 116f estimates, 20-23, 22f
underground economy, 14 1 and GDP, 20, 201, 21/
United Nations
gambling, 20, 221. See also under gambling
Convention Against Corruption, 149n, 152, 182, implementation of Forty Recommendations, 82
185
organized crime in, 19, 201
Drug Convention, 70 President's Commission on Organized Crime,
Economic Commission for Europe (UNECE), 17
report on terrorist financing, 145—46, 147 19
prosecution, conviction data, 109, 1091, 1101, 187
Resolution 1373, 146
prostitution, 19, 221
Security Council Resolution 1483, 153 theft in, 221

"white list," 186n pursuit of foreign kleptocrats, 153-55

United States. See also specific agencies and risk of money laundering in, 127-28
departments underground economy, 13, 141, 18

INDEX 217
US Customs Service, entry, exit inspections, 28, Venezuela, 168
28 n Vershish, Joseph, 108b
US Drug Enforcement Agency Vienna Convention, 198
sting operations, 68 Vietnam, 93, 169-70
US General Accounting Office (GAO), 9 Volcker, Paul, 79
excessive CTRs, 94
weak states, 157
and filing of SARs, 107-08
US Office of Technology Assessment (OTA) and Weinig, Harvey, 38f
reporting requirements, 174-75 Wharton Econometrics Forecasting Associates, 19
US Sentencing Commission "white list," 186m
money-laundering prosecution, conviction data, white-collar crime and money laundering, 41b, 42
109-12, 109t, Hit wire, electronic funds transfers, 30b, 33f
US State Department Wolfsberg Principles, 198
INCSR ratings, 186 World Bank
US Treasury Department CPIA ratings, 157, 157n, 158f-59f, 160 n
Financial Crime Enforcement Network IDA borrowers, 160n
(FinCEN), 54-55, 56, 56 n, 58-59, 70, 106, 108b, report on weak, failing states, 157, 159
174, 196
review, evaluation of AML compliance, 86-87,
Office of Foreign Assets Control, 147 168-70, 180, 182, 183
USA PATRIOT Act, 49, 53f, 56, 57, 59, 61, 198, 199
Zambia, 151n
banking industry response to, 61-62
criticism of, 146-47 Zimbabwe, 159, 159m
and regulation of professionals, 56n, 58, 62 b
and reporting requirements, 56 n, 65
and Title III, 74n, 129

218 CHASING DIRTY MONEY

 Other Publications from the 15 Trade Policy for Troubled Industries*
Institute for Gary Clyde Hufbauer and Howard R. Rosen
March 1986 ISBN 0-88132-020-X
International Economics 16 The United States and Canada: The Quest for
Free Trade* Paul Wonnacott, with an
* = out of print
appendix by John Williamson
March 1987 ISBN 0-88132-056-0
POLICY ANALYSES IN
17 Adjusting to Success: Balance of Payments
INTERNATIONAL ECONOMICS Series
Policy in the East Asian NICs*
Bela Balassa and John Williamson
1 The Lending Policies of the International
June 1987, rev. April 1990 ISBN 0-88132-101-X
Monetary Fund* John Williamson 18 Mobilizing Bank Lending to Debtor
August 1982 ISBN 0-88132-000-5 Countries* William R. Cline
2 "Reciprocity": A New Approach to World June 1987 ISBN 0-88132-062-5
Trade Policy?* William R. Cline 19 Auction Quotas and United States Trade
September 1982 ISBN 0-88132-001-3
Policy* C. Fred Bergsten, Kimberly Ann
3 Trade Policy in the 1980s* Elliott, Jeffrey J. Schott, and Wendy E. Takacs
C. Fred Bergsten and William R. Cline
September 1987 ISBN 0-88132-050-1
November 1982 ISBN 0-88132-002-1
20 Agriculture and the GATT: Rewriting the
4 International Debt and the Stability of the Rules* Dale E. Hathaway
World Economy* William R. Cline
September 1987 ISBN 0-88132-052-8
September 1983 ISBN 0-88132-010-2 21 Anti-Protection: Changing Forces in United
5 The Exchange Rate System,* Second Edition States Trade Politics*
John Williamson I. M. Destler and John S. Odell
Sept. 1983, rev. June 1985 ISBN 0-88132-034-X
September 1987 ISBN 0-88132-043-9
6 Economic Sanctions in Support of Foreign 22 Targets
Policy and Indicators: A Blueprint for the
Policy Goals* International Coordination of Economic
Gary Clyde Hufbauer and Jeffrey J. Schott
October 1983 ISBN 0-88132-014-5 John Williamson and Marcus H. Miller
7 A New SDR Allocation?* John Williamson
September 1987 ISBN 0-88132-051-X
March 1984 ISBN 0-88132-028-5
23 Capital Flight: The Problem and Policy
8 An International Standard for Monetary
Responses* Donald R. Lessard and
Stabilization* Ronald L. McKinnon
John Williamson
March 1984 ISBN 0-88132-018-8
December 1987 ISBN 0-88132-059-5
9 The Yen/Dollar Agreement: Liberalizing 24 United States-Canada Free Trade: An
Japanese Capital Markets* Jeffrey A. Frankel Evaluation of the Agreement*
December 1984 ISBN 0-88132-035-8
Jeffrey J. Schott
! 10 Bank Lending to Developing Countries: The
April 1988 ISBN 0-88132-072-2
Policy Alternatives* C. Fred Bergsten,
William R. Cline, and John Williamson 25 Voluntary Approaches to Debt Relief*
John Williamson
April 1985 ISBN 0-88132-032-3
Sept. 1988, rev. May 1989 ISBN 0-88132-098-6
1 11 Trading for Growth: The Next Round of 26 American Trade Adjustment: The Global
Trade Negotiations* Impact* William R. Cline
Gary Clyde Hufbauer and Jeffrey J. Schott March 1989 ISBN 0-88132-095-1
September 1985 ISBN 0-88132-033-1 27 More Free Trade Areas?*
! 12 Financial Intermediation Beyond the Debt
Jeffrey J. Schott
Crisis* Donald R. Lessard, John Williamson
May 1989 ISBN 0-88132-085-4
September 1985 ISBN 0-88132-021-8 28 The Progress of Policy Reform in Latin
! 13 The United States-Japan Economic Problem* America* John Williamson
C. Fred Bergsten and William R. Cline
January 1990 ISBN 0-88132-100-1
October 1985, 2d ed. January 1987 29 The Global Trade Negotiations: What Can Be
ISBN 0-88132-060-9 Achieved?* Jeffrey J. Schott
I 14 Deficits and the Dollar: The World Economy
September 1990 ISBN 0-88132-137-0
at Risk* Stephen Marris 30 Economic Policy Coordination: Requiem or
December 1985, 2d ed. November 1987
Prologue?* Wendy Dobson
ISBN 0-88132-067-6
April 1991 ISBN 0-88132-102-8
 49

31 The Economic Opening of Eastern Europe* Cooperating with Europe's Monetary Union
John Williamson C. Randall Henning
May 1991 ISBN 0-88132-186-9 ISBN 0-88132-245-8
50 May 1997
32 Eastern Europe and the Soviet Union in the
Renewing Fast Track Legislation* I. M. Destler
World Economy* September 1997 ISBN 0-88132-252-0
Susan M. Collins and Dani Rodrik 51 Competition Policies for the Global Economy
May 1991 ISBN 0-88132-157-5 Edward M. Graham and J. David Richardson
33 African Economic Reform: The External November 1997 ISBN 0-88132 -249-0
Dimension* Carol Lancaster 52
Improving Trade Policy Reviews in the World
June 1991 ISBN 0-88132-096-X Trade Organization Donald Keesing
34 Has the Adjustment Process Worked?* April 1998 ISBN 0-88132-251-2
Paul R. Krugman 53 Agricultural Trade Policy: Completing the
October 1 991 ISBN 0-88132-116-8 Reform Timothy Josling
35 From Soviet disunion to Eastern Economic
April 1998 ISBN 0-88132-256-3
54
Community?* Real Exchange Rates for the Year 2000
Oleh Havrylyshyn and John Williamson Simon Wren Lewis and Rebecca Driver
October 1991 ISBN 0-88132-192-3
April 1998 ISBN 0-88132-253-9
55 The Asian Financial Crisis: Causes, Cures,
36 Global Warming The Economic Stakes*
William R. Cline
and Systemic Implications Morris Goldstein
May 1992 ISBN 0-88132-172-9 June 1998 ISBN 0-88132-261-X
37 Trade and Payments After Soviet 56 Global Economic Effects of the Asian

Disintegration* John Williamson Currency Devaluations

June 1992 ISBN 0-88132-173-7 Marcus Noland, LiGang Liu, Sherman
38 Trade and Migration: NAFTA and Robinson, and Zhi Wang
Agriculture* Philip L. Martin July 1 998 ISBN 0-88132-260-1
October 1993 ISBN 0-88132-201-6 57 The Exchange Stabilization Fund: Slush
39 The Exchange Rate System and the IMF: A Money or War Chest? C. Randall Henning
Modest Agenda Morris Goldstein
May 1 999 ISBN 0-88132-271-7
June 1995 ISBN 0-88132-219-9 58 The New Politics of American Trade: Trade,
40 What Role for Currency Boards? Labor, and the Environment
John Williamson I. M. Destler and Peter J. Balint
September 1995 ISBN 0-88132-222-9 October 1 999 ISBN 0-88132-269-5
41 Predicting External Imbalances for the 59 Congressional Trade Votes: From NAFTA
United States and Japan* William R. Cline Approval to Fast Track Defeat
September 1 995 ISBN 0-88132-220-2 Robert E. Baldwin and Christopher S. Magee
42 Standards and APEC: An Action Agenda* February 2000 ISBN 0-88132-267-9
John S. Wilson 60 Exchange Rate Regimes for Emerging
October 1995 ISBN 0-88132-223-7 Markets: Reviving the Intermediate Option
43 Fundamental Tax Reform and Border Tax
John Williamson
Adjustments* Gary Clyde Hufbauer September 2000 ISBN 0-88132-293-8
61 NAFTA and the Environment: Seven Years
January 1996 ISBN 0-88132-225-3
44 Global Telecom Talks: A Trillion Dollar Later Gary Clyde Hufbauer, Daniel
Esty, Diana Orejas, Luis Rubio, and
Deal*
Ben A. Petrazzini
Jeffrey J. Schott
June 1996 ISBN 0-88132-230-X October 2000 ISBN 0-88132-299-7
45 WTO 2000: Setting the Course for World 62 Free Trade between Korea and the United
Trade Jeffrey J. Schott States? Inborn Choi and Jeffrey J. Schott
April 2001 ISBN 0-88132-311-X
46 September 1996Economic
The National ' ISBN 0-88132-234-2
Council: A Work in 63 New Regional Trading Arrangements in the
Asia Pacific?
Progress * I. M. Destler
November 1996 ISBN 0-88132-239-3 Robert Scollay and John P. Gilbert
47 The Case for an International Banking
May 2001 ISBN 0-88132-302-0
Standard Morris Goldstein 64 Parental Supervision: The New Paradigm
April 1997 ISBN 0-88132-244-X for Foreign Direct Investment and
48 Transatlantic Trade: A Strategic Agenda* Development
Ellen L. Frost Theodore H. Moran

May 1997 ISBN 0-88132-228-8 August 2001 ISBN 0-88132-313-6

 65 The Benefits of Price Convergence:
Japan in the World Economy*
Speculative Calculations Bela Balassa and Marcus Noland
Gary Clyde Hufbauer, Erika Wada, 1988 ISBN 0-88132-041-2
and Tony Warren America in the World Economy: A Strategy for
December 2001 ISBN 0-88132-333-0 the 1990s* C. Fred Bergsten
66 Managed Floating Plus 1988 ISBN 0-88132-089-7
Morris Goldstein
Managing the Dollar: From the Plaza to the
March 2002 ISBN 0-88132-336-5 Louvre* Yoichi Funabashi
! 67 Argentina and the Fund: From Triumph 1988 , 2d. ed. 1989 ISBN 0-88132-097-8
to Tragedy United States External Adjustment and the World
Michael Mussa
Economy* William R. Cline
July 2002 ISBN 0-88132-339-X May 1989 ISBN 0-88132-048-X
I 68 East Asian Financial Cooperation Free Trade Areas and U.S. Trade Policy*
C. Randall Henning Jeffrey J. Schott, editor
September 2002 ISBN 0-88132-338-1 May 1989 ISBN 0-88132-094-3
* 69 Reforming OPIC for the 21st Century Dollar Politics: Exchange Rate Policymaking in
Theodore H. Moran
the United States*
May 2003 ISBN 0-88132-342-X I.M. Destler and C. Randall Henning
I 70 Awakening Monster: The Alien Tort September 1989 ISBN 0-88132-079-X
Statute of 1789 Latin American Adjustment: How Much Has
Gary C. Hufbauer and Nicholas Mitrokostas Happened?* John Williamson, editor
July 2003 ISBN 0-88132-366-7 April 1990 ISBN 0-88132-125-7
! 71 Korea after Kim Jong-il The Future of World Trade in Textiles and
Marcus Noland
Apparel* William R. Cline
January 2004 ISBN 0-88132-373-X 1987, 2d ed. June 1999 ISBN 0-88132-110-9
! 72 Roots of Competitiveness: China's Evolving Completing the Uruguay Round: A Results-
Agriculture Interests Daniel H. Rosen, Oriented Approach to the GATT Trade
Scott Rozelle, and Jikun Huang Negotiations* Jeffrey J. Schott, editor
July 2004 ISBN 0-88132-376-4 September 1990 ISBN 0-88132-130-3
Economic Sanctions Reconsidered (2 volumes)
Economic Sanctions Reconsidered:
; BOOKS
Supplemental Case Histories
Gary Clyde Hufbauer, Jeffrey J. Schott, and
j IMF Conditionality* John Williamson, editor Kimberly Ann Elliott
! 1983 ISBN 0-88132-006-4 1985 , 2d ed. Dec. 1990 ISBN cloth 0-88132-115-X
Trade Policy in the 1980s* William R. Cline, editor
ISBN paper History
Economic Sanctions Reconsidered: 0-88132-105-2
and
j 1983 ISBN 0-88132-031-5 Current Policy
Subsidies in International Trade*
j Gary Clyde Hufbauer and Joanna Shelton Erb Gary Clyde Hufbauer, Jeffrey J. Schott, and
Kimberly Ann Elliott
ji International
1984 ' Debt: ISBN Risk
Systemic 0-88132-004-8
and Policy December 1 990 ISBN doth 0-88132-140-0
I Response* William R. Cline ISBN paper 0-88132-136-2
1984 ISBN 0-88132-015-3 Pacific Basin Developing Countries: Prospects for
' Trade Protection in the United States: 31 Case the Future* Marcus Noland
i Studies* Gary Clyde Hufbauer, Diane E. Berliner, January 1991 ISBN cloth 0-88132-141-9
| and Kimberly Ann Elliott
ISBN paper 0-88132-081-1
! 1986 ' ISBN 0-88132-040-4 Currency Convertibility in Eastern Europe*
’ Toward Renewed Economic Growth in Latin John Williamson, editor
America* Bela Balassa, Gerardo M. Bueno, Pedro- October 1 991 ISBN 0-88132-128-1
Pablo Kuczynski, and Mario Henrique Simonsen International Adjustment and Financing: The

1986 ’ ISBN 0-88132-045-5 Lessons of 1985-1991* C. Fred Bergsten, editor

Capital Flight and Third World Debt* January 1992 ISBN 0-88132-112-5
j Donald R. Lessard and John Williamson, editors North American Free Trade: Issues and
1987 ISBN 0-88132-053-6 Recommendations*
The Canada-United States Free Trade Agreement: Gary Clyde Hufbauer and Jeffrey J. Schott
The Global Impact* April 1992 ISBN 0-88132-120-6
Jeffrey J. Schott and Murray G. Smith, editors Narrowing the U.S. Current Account Deficit*
1988 ISBN 0-88132-073-0 Allen J. Lenz
World Agricultural Trade: Building a Consensus* June 1992 ISBN 0-88132-103-6
William M. Miner and Dale E. Hathaway, editors The Economics of Global Warming
1988 ISBN 0-88132-071-3 William R. Clin e/June 1992 ISBN 0-88132-132-X
US Taxation of Internaitonal Income: Blueprint Estimating Equilibrium Exchange Rates
for Reform* Gary Clyde Hufbauer, assisted John Williamson, editor
by Joanna M. van Rooij
September 1994 ISBN 0-88132-076-5 i
October 1 992 ISBN 0-881 32-134-6 Managing the World Economy: Fifty Years After
Bretton Woods Peter B. Kenen, editor
Who's Bashing Whom? Trade Conflict in High-
Technology Industries Laura D' Andrea Tyson September 1994 ISBN 0-88132-212-1
November 1992 ISBN 0-88132-106-0 Reciprocity and Retaliation in U.S. Trade Policy
Korea in the World Economy* II SaKong Thomas O. Bayard and Kimberly Ann Elliott
January 1993 ISBN 0-88132-183-4 September 1994 ISBN 0-88132-084-6
Pacific Dynamism and the International
The Uruguay Round: An Assessment*
Economic System* Jeffrey J. Schott, assisted by Johanna W. Buurman
C. Fred Bergsten and Marcus Noland, editors November 1994 ISBN 0-88132-206-7
May 1993 ISBN 0-88132-196-6 Measuring the Costs of Protection in Japan*
Economic Consequences of Soviet Disintegration* Yoko Sazanami, Shujiro Urata, and Hiroki Kawai
John Williamson, editor
January 1995 ISBN 0-88132-211-3
May 1993 ISBN 0-88132-190-7 Foreign Direct Investment in the United States,
Reconcilable Differences? United States-Japan 3d ed., Edward M. Graham and Paul R. Krugman
Economic Conflict* ISBN 0-88132-204-0
January 1995
C. Fred Bergsten and Marcus Noland The Political Economy of Korea-United States
June 1993 ISBN 0-88132-129-X
Does Foreign Exchange Intervention Work? C. Fred Bergsten and II SaKong, editors
Cooperation*
Kathryn M. Dominguez and Jeffrey A. Frankel February 1995 ISBN 0-88132-213-X*
September 1993 ISBN 0-88132-104-4 International Debt Reexamined* William R. Cline
Sizing Up U.S. Export Disincentives* February 1995 ISBN 0-88132-083-8
J. David Richardson American Trade Politics, 3d ed., I.M. Destler
September 1993 ISBN 0-88132-107-9 April 1995 ISBN 0-88132-215-6
NAFTA: An Assessment Managing Official Export Credits: The Quest for
Gary Clyde Hufbauer and Jeffrey J. Schott/ rev. ed. a Global Regime* John E. Ray
October 1 993 ISBN 0-88132-199-0
July 1995 ISBN 0-88132-207-5
Adjusting to Volatile Energy Prices
Asia Pacific Fusion: Japan's Role in APEC*
Philip K. Verleger, Jr. Yoichi Funabashi
November 1993 ISBN 0-88132-069-2 October 1995 ISBN 0-88132-224-5
The Political Economy of Policy Reform Korea-United States Cooperation in the New
John Williamson, editor
World Order*
January 1994 ISBN 0-88132-195-8 C. Fred Bergsten and II SaKong, editors
Measuring the Costs of Protection February 1 996 ISBN 0-88132-226-1
in the United States
Why Exports Really Matter!* ISBN 0-88132-221-0
Gary Clyde Hufbauer and Kimberly Ann Elliott Why Exports Matter More!* ISBN 0-88132-229-6
January 1994 ISBN 0-88132-108-7 J. David Richardson and Karin Rindal

The Dynamics of Korean Economic Development* July 1995; February 1996

Cho Soon Global Corporations and National Governments
March 1994 ISBN 0-88132-162-1 Edward M. Graham

Reviving the European Union* May 1996 ISBN 0-88132-111-7

C. Randall Helming, Eduard Hochreiter, and Gary Global Economic Leadership and the Group of
Clyde Hufbauer, editors Seven C. Fred Bergsten and C. Randall Hennim
April 1994 ISBN 0-88132-208-3 May 1996 ISBN 0-88132-218-0
China in the World Economy Nicholas R. Lardy The Trading System After the Uruguay Round*
April 1994 ISBN 0-88132-200-8 John Whalley and Colleen Hamilton
Greening the GATT: Trade, Environment, and July 1996 ISBN 0-88132-131-1
the Future Daniel C. Esty Private Capital Flows to Emerging Markets After

July 1994 ISBN 0-88132-205-9 the Mexican Crisis* Guillermo A. Calvo

Morris Goldstein, arid Eduard Hochreiter
Western Hemisphere Economic Integration*
Gary Clyde Hufbauer and Jeffrey J. Schott September 1 996 ISBN 0-88132-232-6
July 1994 ISBN 0-88132-159-1 The Crawling Band as an Exchange Rate Regime
Currencies and Politics in the United States, Lessons from Chile, Colombia, and Israel
Germany, and Japan John Williamson
C. Randall Henning September 1996 ISBN 0-88132-231-8
September 1994 ISBN 0-88132-127-3
Flying High: Liberalizing Civil Aviation in the Toward A New International Financial
Asia Pacific* Architecture: A Practical Post-Asia Agenda
Gary Clyde Hufbauer and Christopher Findlay Barry Eichengreen
November 1996 ISBN 0-88132-227-X
February 1 999 ISBN 0-881 32-270-9
Measuring the Costs of Visible Protection Is the U.S. Trade Deficit Sustainable?
in Korea* Namdoo Kim Catherine L. Mann
November 1996 ISBN 0-88132-236-9
September 1 999 ISBN 0-88132-265-2
The World Trading System: Challenges Ahead Safeguarding Prosperity in a Global Financial
Jeffrey J. Schott System: The Future International Financial
December 1996 ISBN 0-88132-235-0 Architecture, Independent Task Force Report
Has Globalization Gone Too Far? Dani Rodrik Sponsored by the Council on Foreign Relations
March 1997 ISBN cloth 0-88132-243-1 Morris Goldstein, Project Director
Korea-United States Economic Relationship* October 1999 ISBN 0-88132-287-3
C. Fred Bergsten and II SaKong, editors Avoiding the Apocalypse: The Future of the
March 1997 ISBN 0-88132-240-7 Two Koreas Marcus Noland
Summitry in the Americas: A Progress Report June 2000 ISBN 0-88132-278-4
Richard E. Feinberg Assessing Financial Vulnerability: An Early
April 1997 ISBN 0-88132-242-3 Warning System for Emerging Markets
Corruption and the Global Economy Morris Goldstein, Graciela Kaminsky, and
Kimberly Ann Elliott Carmen Reinhart
June 1997 ISBN 0-88132-233-4 June 2000 ISBN 0-88132-237-7
Regional Trading Blocs in the World Economic Global Electronic Commerce: A Policy Primer
System Jeffrey A. Frankel Catherine L. Mann, Sue E. Eckert, and Sarah
October 1997 ISBN 0-88132-202-4 Cleeland Knight
Sustaining the Asia Pacific Miracle:
July 2000 ISBN 0-88132-274-1
Environmental Protection and Economic The WTO after Seattle Jeffrey J. Schott, editor
Integration Andre Dua and Daniel C. Esty
July 2000 ISBN 0-88132-290-3
October 1997 ISBN 0-88132-250-4 Intellectual Property Rights in the Global
Trade and Income Distribution William R. Cline Economy Keith E. Maskus
November 1997 ISBN 0-88132-216-4
A ugust 2000 ISBN 0-88132-282-2
Global Competition Policy The Political Economy of the Asian Financial
Edward M. Graham and J. David Richardson Crisis Stephan Haggard
December 1997 ISBN 0-88132-166-4
A ugust 2000 ISBN 0-88132-283-0
Unfinished Business: Telecommunications after Transforming Foreign Aid: United States
the Uruguay Round Assistance in the 21st Century Carol Lancaster
Gary Clyde Hufbauer and Erika Wada August 2000 ISBN 0-88132-291-1
December 1997 ISBN 0-88132-257-1 Fighting the Wrong Enemy: Antiglobal Activists
Financial Services Liberalization in the WTO and Multinational Enterprises Edward M. Graham
Wendy Dobson and Pierre Jacquet September 2000 ISBN 0-88132-272-5
June 1998 ISBN 0-88132-254-7 Globalization and the Perceptions of American
Workers
Restoring Japan's Economic Growth
Adam S. Posen Kenneth F. Scheve and Matthew J. Slaughter
September 1998 ISBN 0-88132-262-8 March 2001 ISBN 0-88132-295-4
Measuring the Costs of Protection in China World Capital Markets: Challenge to the G-10
Zhang Shuguang, Zhang Yansheng, and Wan Wendy Dobson and Gary Clyde Hufbauer,
Zhongxin assisted by Hyun Koo Cho
November 1998 ISBN 0-88132-247-4
Foreign Direct Investment and Development: May 2001for 'Free ISBN
Prospects Trade0-88132-301-2
in the Americas
The New Policy Agenda for Developing
Jeffrey J. Schott
Countries and Economies in Transition
August 2001 ISBN 0-88132-275-X
Theodore H. Moran Toward a North American Community:
December 1998 ISBN 0-88132-258-X Lessons from the Old World for the New
Behind the Open Door: Foreign Enterprises Robert A. Pastor
in the Chinese Marketplace August 2001 ISBN 0-88132-328-4
Daniel H. Rosen Measuring the Costs of Protection in Europe:
January 1999 ISBN 0-88132-263-6 European Commercial Policy in the 2000s
Patrick A. Messerlin

September 2001 ISBN 0-88132-273-3

Job Loss from Imports: Measuring the Costs Has Globalization Gone Far Enough? The Costs
Lori G. Kletzer of Fragmented Markets
ISBN 0-88132-296-2 Scott Bradford and Robert Z. Lawrence
September 2001
No More Bashing: Building a New Japan-United February 2004 ISBN 0-88132-349-7
States Economic Relationship C. Fred Bergsten, Food Regulation and Trade: Toward a Safe
Takatoshi I to, and Marcus Noland and Open Global System
October 2001 ISBN 0-88132-286-5 Tim Josling, Donna Roberts, and David Orden
Why Global Commitment Really Matters! March 2004 ISBN 0-88132-346-2
Howard Lewis III and J. David Richardson Controlling Currency Mismatches in
October 2001 ISBN 0-88132-298-9 Emerging Markets
Leadership Selection in the Major Multilaterals Morris Goldstein and Philip Turner
Miles Kahler
April 2004 ISBN 0-88132-360-8
November 2001 ISBN 0-88132-335-7 Free Trade Agreements: US Strategies
The International Financial Architecture: and Priorities

What's New? What's Missing? Peter Kenen Jeffrey J. Schott, editor

November 2001 ISBN 0-88132-297-0 April 2004 ISBN 0-88132-361-6
Delivering on Debt Relief: From IMF Gold to Trade Policy and Global Poverty
a New Aid Architecture William R. Cline

John Williamson and Nancy Birdsall, June 2004 ISBN 0-88132-365-9

with Brian Deese Transforming the European Economy
Martin Neil Baily and Jacob Kirkegaard
April 2002 ISBN 0-88132-331-4
Imagine There's No Country: Poverty, Inequality, September 2004 ISBN 0-88132-343-8
and Growth in the Era of Globalization Bailouts or Bail-ins? Responding to Financial
Surjit S. Bhalla Crises in Emerging Economies
Nouriel Roubini and Brad Setser
September 2002 ISBN 0-88132-348-9
Reforming Korea's Industrial Conglomerates September 2004 ISBN 0-88132-371-3
Edward M. Graham Chasing Dirty Money: The Fight Against
Money Laundering
January 2 003 ISBN 0-88132-337-3
Peter Reuter and Edwin M. Truman
Industrial Policy in an Era of Globalization:
Lessons from Asia November 2004 ISBN 0-88132-370-5
Marcus Noland and Howard Pack
March 2003 ISBN 0-88132-350-0
SPECIAL REPORTS
Reintegrating India with the World Economy
T. N. Srinivasan and Suresh D. Tendulkar
March 2003 ISBN 0-88132-280-6 1 Promoting World Recovery: A Statement on
After the Washington Consensus: Global Economic Strategy*
Restarting Growth and Reform in by Twenty-six Economists from Fourteen Countries-
Latin America Pedro-Pablo Kuczynski December 1982 ISBN 0-88132-013-7
and John Williamson, editors 2 Prospects for Adjustment in Argentina,
March 2003 ISBN 0-88132-347-0 Brazil, and Mexico: Responding to the Debt Crisis
The Decline of US Labor Unions and John Williamson, editor
the Role of Trade Robert E. Baldwin 3 Inflation and Indexation: Argentina, Brazil,

June 2003 ISBN 0-88132-341-1 and Israel* John Williamson, editor

Can Labor Standards Improve under March 1985 ISBN 0-88132-037-4
Globalization? 4 Global Economic Imbalances*
Kimberly Ann Elliott and Richard B. Freeman C. Fred Bergsten, editor
ISBN 0-88132-332-2 March 1986 ISBN 0-88132-042-0
June 2003
Crimes and Punishments? Retaliation 5 African Debt and Financing*
under the WTO Carol Lancaster and John Williamson, editors
Robert Z. Lawrence
May 1986 ISBN 0-88132-044-7
October 2003 ISBN 0-88132-359-4 6 Resolving the Global Economic Crisis: After
Inflation Targeting in the World Economy
Edwin M. Truman Wall Street*
by Thirty-three Economists from Thirteen
Countries
October 2003 ISBN 0-88132-345-4
Foreign Direct Investment and Tax December 1987 ISBN 0-88132-070-6
Competition John H. Mutti
November 2003 ISBN 0-88132-352-7
7 World Economic Problems* The Impact of Foreign Direct Investment
Kimberly Ann Elliott and John Williamson, on Development: New Measures, New
editors Outcomes Magnus Blomstrom, Edward
April 1988 ISBN 0-88132-055-2 Graham, and Theodore Moran, editors
Reforming World Agricultural Trade* China's Entry into the World Economy
Richard N. Cooper
by Twenty-nine Professionals from Seventeen
Countries American Trade Politics, 4th ed.
1988 ISBN 0-88132-088-9 I. M. Destler
8 Economic Relations Between the United The ILO in the World Economy
States and Korea: Conflict or Cooperation?* Kimberly Ann Elliott
Thomas O. Bayard and Soogil Young, editors Reforming Economic Sanctions
January 1989 ISBN 0-88132-068-4 Kimberly Ann Elliott, Gary C. Hufbauer,
9 Whither APEC? The Progress to Date and and Jeffrey J. Schott
Agenda for the Future* Merry Sisterhood or Guarded Watchfulness?
C. Fred Bergsten, editor Cooperation Between the IMF and
October 1 997 ISBN 0-88132-248-2 the World Bank Michael Fabricius
10 Economic Integration of the Korean Future of Chinese Exchange Rates
Peninsula Morris Goldstein and Nicholas R. Lardy
Marcus Noland, editor NAFTA: A Ten-Year Appraisal

January 1998 ISBN 0-88132-255-5 Gary Clyde Hufbauer and Jeffrey J. Schott
New Agricultural Negotiations in the WTO
11 Restarting Fast Track*
Jeffrey J. Schott, editor Tim Josling and Dale Hathaway
Workers at Risk: Job Loss from Apparel,
April 1998 ISBN 0-88132-259-8
12 Launching New Global Trade Talks: Textiles, Footwear, and Furniture
Lori G. Kletzer
An Action Agenda Jeffrey J. Schott, editor
September 1998 ISBN 0-88132-266-0 Responses to Globalization: US Textile
and Apparel Workers and Firms
13 Japan's Financial Crisis and Its Parallels to
US Experience Lori Kletzer, James Levinsohn, and
Ryoichi Mikitani and Adam S. Posen, eds. J. David Richardson
September 2000 ISBN 0-88132-289-X The Strategic Implications of China-Taiwan
Economic Relations
14 The Ex-Im Bank in the 21st Century: A New
Nicholas R. Lardy
Approach Gary Clyde Hufbauer and
Rita M. Rodriguez, editors Making the Rules: Case Studies on
US Trade Negotiation
January 2001 ISBN 0-88132-300-4
15 The Korean Diaspora in the World Robert Z. Lawrence, Charan Devereaux,
and Michael Watkins
Economy
C. Fred Bergsten and Inborn Choi, eds. US-Egypt Free Trade Agreement
Robert Z. Lawrence and Ahmed Galal
January 2003 ISBN 0-88132-358-6
16 Dollar Overvaluation and the World High Technology and the Globalization
Economy of America Catherine L. Mann
International Financial Architecture
C. Fred Bergsten and John Williamson, eds.
Michael Mussa
February 2003 ISBN 0-88132-351-9
Germany and the World Economy
Adam S. Posen
WORKS IN PROGRESS The Euro at Five: Ready for a Global Role?
Adam S. Posen, editor
Automatic Stabilizers for the Eurozone
United States and the World Economy: Foreign
Adam S. Posen
Policy for the Next Decade
Global Forces, American Faces: US Economic
C. Fred Bergsten, editor
Globalization at the Grass Roots
New Regional Arrangements and
the World Economy J. David Richardson
C. Fred Bergsten Prospects for a US-Taiwan FTA
Daniel H. Rosen and Nicholas R. Lardy
The Globalization Backlash in Europe and
the United States Curbing the Boom-Bust Cycle: Stabilizing
Capital Flows to Emerging Markets
C. Fred Bergsten, Pierre Jacquet, and Karl Kaiser
Dollar Adjustment: How Far? Against What? John Williamson
C. Fred Bergsten and John Williamson, editors
DISTRIBUTORS OUTSIDE THE UNITED STATES

Australia, New Zealand, Canada

and Papua New Guinea Renouf Bookstore
D.A. Information Services 5369 Canotek Road, Unit 1
648 Whitehorse Road Ottawa, Ontario K1J 9J3, Canada
Mitcham, Victoria 3132, Australia tel: 613-745-2665
tel: 61-3-9210-7777 fax: 613-745-7660
fax: 61-3-9210-7788 www.renoufbooks.com
email: [email protected]
www.dadirect.com.au

United Kingdom and Europe India, Bangladesh, Nepal, and Sri Lanka
(including Russia and Turkey) Viva Books Pvt.
The Eurospan Group Mr. Vinod Vasishtha
3 Henrietta Street, Covent Garden 4325/3, Ansari Rd.
London WC2E 8LU England
Daryaganj,
India New Delhi-110002
tel: 44-20-7240-0856
fax: 44-20-7379-0609 tel: 91-11-327-9280
www.eurospan.co.uk fax: 91-11-326-7224
email: vinod. [email protected]. ems.vsnl.
net.in

Japan and the Republic of Korea Southeast Asia (Brunei, Cambodia,

United Publishers Services Ltd. China, Malaysia, Hong Kong, Indonesia,
1-32-5, Higashi-shinagawa, Laos, Myanmar, the Philippines, Singapore,
Shinagawa-ku, Tokyo 140-0002 JAPAN Taiwan, and Vietnam)
tel: 81-3-5479-7251 Hemisphere Publication Services
fax: 81-3-5479-7307 1 Kallang Pudding Rd. #0403
[email protected] Golden Wheel Building
For trade accounts only.
Singapore 349316
Individuals will find HE books in tel: 65-741-5166
leading Tokyo bookstores. fax: 65-742-9356

Thailand
Asia Books
Visit our Web site at:
5 Sukhumvit Rd. Soi 61
www.iie.com
Bangkok 10110 Thailand
tel: 662-714-07402 Ext: 221, 222, 223 E-mail orders to:
fax: 662-391-2277 [email protected]
email: [email protected]
www.asiabooksonline.com

52 50
’
 DATE DUE

1 i' ^
7M ft

HIGHSMITH #45102
Originally developed to reduce drug trafficking, efforts to combat money laundering
have broadened over the years to address other crimes and, most recently,
terrorism. In this study, Reuter and Truman look at the scale and characteristics
of money laundering, describe and assess the current anti-money laundering
regime, and make proposals for its improvement.

"This superb study is the most comprehensive ". . . the most seminal study ever undertaken
treatment of money laundering to date and of the effectiveness of global efforts to deal
certainly will become the benchmark by which with the growing threat to US national security
all future discussions of this subject will be and to the world financial system of money
laundering. . .
assessed."
Vice President, Research and Policy, Stuart E. Eizenstat former deputy secretary of the
Kauffman Foundation; Senior Fellow, US Treasur
Economic Studies, Brookings Institution

is a professor in the School of Public Affairs and in the Department

of Criminology at the University of Maryland and editor of the Journal of Policy
Analysis and Management. He was a senior economist in the Washington office
of the RAND Corporation (1981-93), and he founded and directed RAND's
Drug Policy Research Center (1 989—93). He is the author of Disorganized
Crime: The Economics of the Visible Hand (MIT Press, 1983), which won the
Leslie Wilkins award for outstanding book of the year in criminology and criminal justice. He has
served as a consultant to numerous government agencies, including the US General Accounting
Office, the White House Office of National Drug Control Policy, the National Institute of Justice,
and the United Nations Drug Control Program.

, senior fellow since 2001 , was assistant secretary of the trea¬

sury for international affairs (1998-2000). He directed the Division of Interna¬
tional Finance of the Board of Governors of the Federal Reserve System from
1 977 to 1 998. From 1 983 to 1 998, he was one of three economists on the staff
of the Federal Open Market Committee. He has been a member of numerous
international working groups on international economic and financial issues. He
is the author of Inflation Targeting in the World Economy (2003).

1 750 Massachusetts Avenue, NW

INSTITUTE FOR
INTERNATIONAL Washington, DC 20036-1903
Tel: (202) 328-9000 Fax: (202) 659-3225
ECONOMICS www.iie.com