UPDATE: the World Bank hosts a discussion on the report that is the subject of this post May 30, 12:00 noon EDT. Link to register here.

Today’s guest post summarizes an April World Bank study of money laundering risk assessments. The first step in preventing money laundering is identifying where it occurs and how likely it is to occur. In short, the risks of money laundering. The Bank study evaluated risk assessments eight governments had conducted in accordance with the methodology prescribed by the Financial Action Task Force. For reasons that will become plain, the post’s author has chosen to remain anonymous.

The title from a new World Bank report on money laundering risks could scarcely be blander: National Assessments of Money Laundering Risks: Learning from Eight Advanced Countries’ NRAs.  The content is anything but. Authored by Joras Ferwerda of Utrecht University and Peter Reuter of the University of Maryland, the report concludes that not a one of the eight money laundering risk assessments examined, all done as the report’s title advertises by “advanced” countries, is worth a damn. Not a one merits a passing grade from the two professors, both highly regarded money laundering experts. What’s worse, despite close to a decade of experience doing such assessments, the two find that no government seems to have learned a thing from the mistakes of others.

This raises a fundamental question about the existing AML regime. How can it be effective if national authorities lack an understanding of the money laundering risks their countries face?

All adopted an approach to risk assessment enshrined in a 2013 FATF Guidancedocument. Which is where the problem begins; for the FATF document is inconsistent with best practice in risk assessment as articulated by the International Standards Organization (here).  For example, a critical issue in conducting a risk assessment is the identification of “threats.” But the NRAs are all over the place when it comes to defining just what a threat is. Canada characterized money laundering “threats” as any kind of group that might launder money (organized gangs or professional money launderers), while Singapore characterized specific crimes as threats. 

FATF now requires that every country show a good understanding of the distribution of risks, preferably through a published National Risk Assessment.  Ferwerda and Reuter analyzed the national risks assessments prepared by the governments of Canada, Italy, Japan, Netherlands, Singapore, Switzerland, the UK, and the USA; all are rated by the IMF as systemically important countries.  And they are surely amongst the countries with the most sophisticated financial regulatory and research communities. 

Each country relied heavily on expert opinion, solicited without any awareness of the standards of practice for eliciting expert opinion (here).  Most provided just the broadest hints of the methodology that they used.  The Netherlands was the only one that described its approach in sufficient detail that it could in principle be replicated.  Finally, almost none provided what the FATF asked for, an assessment of where risks are significant and where they are modest. 

Risk assessment is difficult in any field.  Though there is a well-developed profession of risk assessment, each area requires adapting the methodology to that area’s unique data limitations and institutional arrangements.  Money laundering is a novel field for risk assessment; data on money laundering events are hard to find.  So there is no shame in not having strong NRAs the first time around (here).  What is harder to justify is the casualness of the exercises.  Why not lay out the methods so that other countries can learn from each other’s mistakes and so that the field will develop? 

One gets the sense that NRAs are simply a box-checking exercise to meet an international requirement.  Most are done just before a FATF Mutual Evaluation Report as the new FATF self-evaluation itself reports (here).  There is little evidence that the assessments affect policy decisions, though that is the whole point of doing them.  Ferwerda and Reuter suggest that the closed world of AML is part of the problem; professional risk assessors have played no role either in the development of the FATF methodology or in its operationalization at the national level. 

How can the international risk-based AML strategy succeed if even the most developed countries fail to research and explain their risks well? The FATF needs to clearly explain what it expects from a national risk assessment. A variety of research methods is healthy, but when countries are not researching the same topic and each country comes up with their own meaning of relevant terms, learning from each other is not possible. Countries must be transparent about how they analyze risks and involve risk assessment professionals so that at least the bare minimum of academic standards for analysis are met. The National Risk Assessment exercise is in its early stages. There is no shame in stumbling at the starting gate, but without intervention, all will just continue to stumble.