Estimating money laundering flows with a gravity model-based simulation

Abstract

It is important to understand the amounts and types of money laundering flows, since they have very different effects and, therefore, need different enforcement strategies. Countries that mainly deal with criminals laundering their proceeds locally, need other measures than countries that mainly deal with foreign illegal investments or dirty money just flowing through the country. This paper has two main contributions. First, we unveil the country preferences of money launderers empirically in a systematic way. Former money laundering estimates used assumptions on which country characteristics money launderers are looking for when deciding where to send their ill-gotten gains. Thanks to a unique dataset of transactions suspicious of money laundering, provided by the Dutch Institute infobox Criminal and Unexplained Wealth (iCOV), we can empirically test these assumptions with an econometric gravity model estimation. We use this information for our second contribution: iteratively simulating all money laundering flows around the world. This allows us, for the first time, to provide estimates that distinguish between three different policy challenges: the laundering of domestic crime proceeds, international investment of dirty money and money just flowing through a country.

How Big Are Illicit Financial Flows?

“…suspicious of money laundering makes it possible, for the first time, to empirically analyse why money laundering happens more between certain countries and less between other country pairs. Ferwerda et al (2020) find with their panel data regression that money laundering flows follow the so-called gravity model. This model is related to the gravity equation in physics and shows that money laundering flows happen more between bigger countries (size attracts) and when countries are closer to each other (measured in kilometres or cultural, like common language and common religion).…”

“…Figure 5.2 gives an overview of relevant concepts that are part of the over-arching concept of illicit financial flows and estimated by the studies this chapter focuses on. Money laundering (as estimated by Ferwerda et al 2020) is by definition illegal and seems to be mostly happening in the financial sector. Not all money laundering is part of the broad concept of tax gap (defined as unpaid taxes by Murphy 2019), because some criminals actually prefer to pay taxes in an attempt to give a more legitimate appearance to their criminal proceeds.…”

Are Bankers “Crying Wolf”? Type I, Type II Errors and Deterrence in Anti-Money Laundering: The Italian Case

“…19 We are aware of the fact that a work more focused on estimating the incidence of money laundering should use more sophisticated methodologies than ours. See, for instance, Ferwerda et al (2020). 20 See beginning of Sect.…”

Tax Dynamics and Money Laundering:

“…First, to estimate the amount of money being laundered, diverse techniques are used (see for a good overview Unger, 2009, Section 1), ranging from case studies, surveys and interviews to the currency demand (Tanzi and Shome, 1993), the dynamic multiple indicators multiple causes or DYMIMIC (Schneider, 2006), the trade-pricing (Zdanowicz, 2009), and the gravity model (Ferwerda et al, 2013;Walker and Unger, 2009) approaches. A recent paper using the latter approach which also uses data retrieved from iCOV estimates all global money laundering flows at the bilateral level (Ferwerda et al, 2020). Although the estimates differ as much as their approaches, money laundering is a large global phenomenon that requires a multi-disciplinary and multi-national approach (Buchanan, 2004).…”

“…Analysing the impact of money laundering on society and estimating the amounts associated with it are given considerable scholarly attention (e.g. Ferwerda et al, 2020). Yet, the estimations on how much money is laundered show large differences (Ferwerda and Unger, 2021).…”