Drugs, Crime, Money, and Development

Remarks at the World Bank Group/International Finance Corporation Financial and Private Sector Development Forum

April 25, 2007

It is a pleasure to be with you today at this Financial and Private Sector Development Forum and to participate on this panel on Drugs, Crime, Money, and Development.

First, let me explain my point of departure in addressing these issues. I am an economist by training. I worked in the US central bank and treasury for about 30 years as an economist. I had some involvement in money-laundering issues, including a minor supporting role in shaping of national and international policy to confront the phenomenon and involvement in a few specific cases. After retiring from government, I coauthored with Peter Reuter a book on antimoney laundering that was published a few of years ago: Chasing Dirty Money: The Fight Against Money Laundering. Peter also is an economist, but he has spent much of his career studying and developing policy with respect to illegal markets. It is the fact that we are both economists that led us to focus in our study on institutions and incentives.

I have divided my remarks this afternoon into two parts. First, I will provide my perspective on the phenomenon of money laundering and its links to crime and on the global antimoney laundering regime. Second, I will offer some observations on the implications of my assessment for the development activities of the World Bank Group.

Money laundering is a crime in most countries primarily because many crimes produce financial gains that criminals have to preserve and disguise before they can be employed. Thus, an antimoney laundering regime is principally a tool to fight those crimes. One clear exception to this generalization is the financing of terrorism where the financial resources generally are not derived from crime but are intended to support criminal activity. (I will set this aspect aside; we have enough to cover.) A less clear exception involves the societal objective of preserving the integrity of the financial system. (I will come back briefly to this aspect because it relates to the activities of this institution.)

Estimates of the scale of international money laundering start at the hundreds of billions of dollars and extend to more than a trillion dollars per year, or upwards of 2 percent of global GDP. The estimates are extremely imprecise, but they do establish that the money-laundering phenomenon is large. However, they are not useful in analyses of progress in fighting money laundering or the underlying crimes because of their imprecision. Moreover, society does not judge the seriousness of the underlying crimes on the basis of the amounts of money they generate to launder.

Money laundering is facilitated by the existence of national and other jurisdictional borders. Money crosses national borders much more easily than information about its possible origins. Not all money laundering involves an international dimension, but the existence of national borders facilitates the process. Consequently, the fight against money laundering is a quintessential international public good, requiring international cooperation to control it effectively. It has been characterized as the dark side of international financial globalization. As my former boss at the US Treasury, Larry Summers, said, "To secure the benefits of the globalized financial system, we need to ensure that its credibility and integrity are not undermined by money laundering, harmful tax competition, and poor regulatory standards." However, money laundering is an issue not only at the global level but also at the village level, which is one of the many challenges.

Antimoney laundering (AML) strategies initially were developed primarily to help fight the sale and distribution of illegal drugs. The implicit model was based on a drug dealer who accumulated a lot of cash. He took that cash to a money launderer (a bank), who kept his money or transferred it elsewhere, including across national borders, so that it could be retransferred many times and the origins of the money could not be traced. The AML strategy involved forbidding the bank from offering its services to the drug dealer. This was expected to drive up the cost of his activity and the price of his product, and to diminish the demand for drugs and profits from drug dealing.

The problems with this model are several. First, criminals have multiple channels to launder money; close down the banks and criminals will use gold dealers or arrange for couriers to carry suitcases of cash across the border.

Second, drug dealing is only one of many underlying crimes that generate proceeds that have to be laundered, and most of the other examples do not involve large amounts of cash. In some cases, in areas such as fraud and embezzlement, the money is already inside the financial system when the crime is committed. Moreover, one is hard pressed to identify stand-alone money-laundering operations. They are either ad hoc arrangements or built into the execution of the crime.

Third, the risk of successful prosecution is low. The United States probably has the highest conviction rate for money laundering of any country in the world. Nevertheless, Peter Reuter and I estimated conservatively the probability of conviction for money laundering at about 6.5 percent. Even if the risk of conviction were as high as 10 percent, against a gross income of $500,000 to $1 million, the returns to the money launderer are large relative to the risks involved. The low risk of conviction and confiscation implies that the effects on the cost of money laundering or the underlying crime are commensurately small. They are even smaller elsewhere in the world where the risk of conviction is much lower.

Notwithstanding these shaky analytical foundations, we have built an elaborate global antimoney laundering regime to help to combat crime. The regime is based on the Financial Action Task Force's (FATF) "Forty Recommendations on Money Laundering and Nine Special Recommendations on Terrorist Financing." The regime is oriented primarily toward preventing financial institutions and nonfinancial institutions from becoming involved with money laundering, an approach derived from the simple model of the drug dealer who needs to launder his cash. Moreover, the prevention pillar of the antimoney laundering structure is only loosely connected with what we call the enforcement pillar of the regime that is addressed at prosecutions and punishments for the crime of money laundering or for the underlying crimes. (As I just noted, conviction rates even in the United States are low.) Why? Governments and their law enforcement agencies face budget constraints. Thus, zero tolerance is not a realistic standard.

The global antimoney laundering regime has been erected to address a serious phenomenon: crime and its facilitation. However, in my view, the structure we have built, and as a former official I do not exempt myself from this criticism, has several drawbacks. The lack of linkage between the prevention and enforcement pillars of the regime undermines its credibility. Moreover, the regime is costly, and a large portion of those costs, in particular for the prevention pillar, are not borne collectively by the public sector but by private sector institutions and their customers. We do not have a good handle on those public and private costs, which is another one of the challenges. ( Peter Reuter and I constructed a crude estimate for the United States of $7 billion dollars a year (as of 2003) or $25 dollar per capita. That amount is not much for the United States, but $25 per capita, even after making a purchasing-power-parity adjustment, would be significant for poor countries.) Finally, the AML regime, again in its prevention pillar in particular, creates distortions. We do not know the magnitudes of these distortions, but we can be reasonably confident about their (negative) signs.

As I indicated earlier, in addition to fighting crime, a secondary objective of the global AML regime is to protect the integrity of the core financial system-principally banks. (Indeed, the unit in the World Bank that is responsible for antimoney laundering is called the Financial Market Integrity Unit.) The rationale is that we do not want financial institutions, whose effectiveness-sometimes their viability-depends upon their good reputations, to be involved with crime and criminals. This modest objective is relatively easy to achieve with a sufficiently robust prevention pillar and occasional reinforcement from the enforcement pillar. The problem in the development context is that distortions are created in the process. For example, by discouraging the use of the core financial system, the AML regime promotes the development of nontraditional financial institutions and the informal economy. This, in turn, negatively affects the efficiency of resource allocation, tax revenues, and economic growth.

I do not want my views about the phenomenon of money laundering to be misunderstood. It is a serious, complicated, international challenge. However, we don't really know what works to control money laundering. We do not have enough analysis of the incentives involved. We do not know the costs of the various elements of the global AML regime. Thus, it is difficult to assess whether we can and should do more to achieve added benefits. I am reasonably confident that the net benefit of what has been done in constructing the global regime over the past 20 years has been positive, but I am equally confident that we have made mistakes and could be more efficient and effective with the resources at hand.

This brings me to the second part of my remarks: implications of my assessment of the global AML regime for the development activities of the World Bank Group. My observations cover five topics: (1) strengthening the global antimoney laundering regime; (2) research on money laundering and best practices in combating the phenomenon; (3) the development of financial sectors; (4) assistance to low-income countries; and (5) dealing with corruption.

With respect to the global AML regime, the Bank, in cooperation with the Fund, the FATF, and FATF-style regional bodies, is already actively involved in the assessment of compliance with the standards of the current global regime. Criticisms have been raised about the consistency of those assessments across bodies and countries. No doubt improvements are being made. Legitimate issues also have been raised about the resource implications of these activities within the Bank and the Fund. It follows from the concerns that I have already expressed about our not knowing what works in combating money laundering that a broader consideration of some of these resource concerns is warranted, in part, as an element of the research activities of the Bank on which I will comment next.

However, my view is that the Bank and the Fund have an important role to play in these assessment activities. They are near universal institutions of global governance. Fighting money laundering and the financing of terrorism is an international "public good." It follows that global institutions should play a role in upholding global standards of best practice.

The Bank also has a role in shaping the global regime and ensuring its consistency across countries, for example, with respect to the identification of the "predicate," or underlying, crimes for which money laundering can be prosecuted on a cross-country basis. The FATF's "Forty Recommendations" designate 20 categories of high-priority offenses that should be covered by national money-laundering statutes. However, in many countries not all of those offenses qualify when the crimes are committed in other jurisdictions. Seven do not qualify in the United States, including sexual exploitation, trafficking in human beings, and counterfeiting. The Bank can be an advocate in this area.

The Bank can also be an advocate for the inclusion of tax evasion in other countries as a cross-border money-laundering offense and for greater international cooperation in information exchange on tax evasion, which has a huge impact on the fiscal positions of developing countries.

Second, the Bank is well placed to conduct research about what works and does not work in combating money laundering. This type of research should be at the theoretical level like Elod Takats' recent International Monetary Fund (IMF) working paper on potential misplaced incentives for financial institutions to over report potential suspicious activities. More research is also required at the empirical level, for example, on the costs of antimoney laundering regimes. Much of the Bank's excellent and varied work on remittances was motivated by, and has implications for antimoney laundering, and that work should be continued.

The Bank could also take the lead in collecting international, case study information about money laundering or to help set standards for, and to coordinate the now limited national efforts to collect, this type of data. The emphasis should be on collecting information about methods, prices, and quantities, as well as on incentives and disincentives, to help devise more efficient strategies for dealing with the activity.

Finally, in connection with the important work that the Bank has done on corruption, Peter Reuter and I have suggested that it should collect detailed information on cases of kleptocracy-high-level official corruption-including names, amounts, convictions, and amounts returned to countries. Thus, the international community would have a scorecard to help measure its progress in this important area. We also pointed to the importance of full ratification of the UN Convention Against Corruption and more important its use in areas such as asset recovery. I am, therefore, gratified by the Bank's recently announced Stolen Asset Recovery (StAR) Initiative and wish it well.

Third, the Bank's work with members on the development of their financial sectors comes into play here. As I noted earlier, the integrity of core financial institutions is crucial to their effectiveness. The effectiveness of the financial system is, in turn, crucial to economic growth and the alleviation of poverty. However, the need to integrate appropriate regimes to combat money laundering and the financing of terrorism into the development agenda is an area where it is important to resist an approach of one size fits all. We know that zero tolerance is not a realistic standard. The conventional prescription is to apply a risk-based approach, but this requires sophisticated differentiation by country and type of financial institution. A bank serving primarily rural clients should not be required to have as sophisticated an AML regime in place as one in the financial center of a country that has a large number of national and international clients. The last time I looked, this type of distinction was not being thoroughly implemented in analyses of AML regimes. Moreover, as noted earlier, in many countries, those regimes focus disproportionately on prevention, which is meaningless without stronger enforcement.

This brings me to my fourth topic, assistance to low-income member countries. From a global perspective, concerns about the interaction of drugs, crime, and money are luxury goods, receiving higher priority in advanced countries than in countries with much lower incomes per capita. Some may bemoan this situation, in particular because they understand correctly that a country without an AML regime that functions at least at a rudimentary level creates a potential, gaping loophole in the global regime.

One answer to this dilemma is to promote extensive technical assistance from institutions like the World Bank and the IMF. However, I suspect that this not enough because the technical assistance is primarily directed at the design of the prevention pillar of national money-laundering regimes not the operation of those prevention systems, much less the enforcement of laws against money laundering and, more importantly, the underlying criminal activities. Resources for training can help, but more is required. Thus, Peter Reuter and I argued for the wealthier countries to subsidize the operation of AML regimes in the poorest countries through global institutions like the Bank.

Finally, I come to the tortured issue of corruption and its links to money laundering. In most jurisdictions, many forms of corruption are criminal offenses, and their proceeds have to be laundered. However, that is only the beginning of the story. The amounts may be large or small. Corruption may involve many types of economic and noneconomic activities; many may be illegal, but some are "only" unethical. Moreover, though national laws are more uniform than they once were, legal standards still differ. In some countries, many forms of capital flight are illegal. In most countries, facilitating tax evasion is not illegal even though the economic costs generally are larger.

These complexities and differences complicate the work of the World Bank Group in dealing with corruption in general and as a source of demand for money laundering. They do not permit the Bank to ignore the phenomenon, and the Bank has not. The Bank has raised the international profile of the issues, including through its research activities. Some of this work is controversial because of its necessary reliance on surveys of perceptions rather than so-called objective measures like amounts changing hands. Efforts have been made to answer critics and deal with deficiencies, and I applaud those efforts. However, it is fair to say that Bank research and analysis of corruption issues have not yet swept the critics from the field. We cannot construct dollar-and-cents measures of corruption and use them to calibrate progress in this or any other dimension of money laundering. This means that measuring progress and effectiveness is challenging, but it does not mean that efforts should not be made. My sense is that the Bank's overall research agenda should pay more attention to constructing measures of progress and to creating a database on cases of money laundering, corruption, and kleptocracy.

With respect to corruption as well as other criminal activities where the proceeds have to be laundered, the real challenge is to determine what works and why. We should be humble about the power and efficiency of our prescriptions. However, we can have confidence in a few principles such as de facto as well as de jure transparency and accountability in the fiscal affairs of government. I commend the Bank's work in this area and its willingness to turn the light of transparency and accountability on its own lending activities.

It is important to resist becoming too discouraged about corruption and money laundering. Eighteen months ago, a former colleague at the US Treasury called me. He was depressed about a front-page story in The New YorkTimes about the alleged corruption and money laundering of a governor of a Nigerian state (Governor Alamieyeseigha of Bayelsa). He felt it signified a lack of progress. My contrarian interpretation was that the story was good news. At a minimum, it increased transparency. The man currently faces a 40-count indictment in Nigeria, so maybe the report signifies increased accountability as well.

My conclusion is that we have come a good distance over the past 20 years or so in combating money laundering. However, our understanding of the activity, and how best to control it as part of an effort to address the underlying criminal activity, is still rudimentary. The Bank Group can and should play a major role in the years ahead in further advancing our understanding of, and developing mechanisms to deal with, the interactions of drugs, crime, money, and development.