Magic mirror in my hand … How trade mirror statistics can help us detect illegal financial flows

• DOI: http://doi.org/10.1111/twec.12840
• Author: Mario Gara,Enrico Tosti,Michele Giammatteo
• Published date: 01 November 2019
• Date: 01 November 2019

3120

|

wileyonlinelibrary.com/journal/twec World Econ. 2019;42:3120–3147.

© 2019 John Wiley & Sons Ltd

Received: 17 September 2018

|

Revised: 20 March 2019

|

Accepted: 11 June 2019

DOI: 10.1111/twec.12840

ORIGINAL ARTICLE

Magic mirror in my hand… How trade mirror

statistics can help us detect illegal financial flows

MarioGara1

|

MicheleGiammatteo1

|

EnricoTosti2

1Analysis and Institutional Relations Directorate,UIF (Italian Financial Intelligence Unit) ‐ Bank of Italy, Rome, Italy

2DG Economics, Statistics, and Research - Statistical Analysis Directorate,Bank of Italy, Rome, Italy

KEYWORDS

illicit trade flows, mirror statistics, money laundering

1

|

INTRODUCTION

An importer in Country A purchases goods from an exporter in Country B and requires that the

goods be delivered to its branch in Country C.1

The importer settles the invoice of the exporter by a

wire transfer. The importer then invoices its branch for a significantly higher value, including a wide

range of inflated administrative costs, which in fact are added so as to allow for the transfer of funds

of illegal origin. The branch settles the inflated invoice by depositing funds into its parent's bank

account.2

The scheme which has just been described, taken from a real‐life case, illustrates how the physical

movement of goods through the trade system can be exploited by criminals as an efficient channel for

disguising the unlawful nature of the proceeds of their activities and integrating them into the legal

economy. Indeed, influential international organisations competent in the field of money laundering,

such as the Financial Action Task Force (or FATF), have long started looking at this phenomenon,

typically referred to as trade‐based money laundering (TBML henceforth), since it has reportedly gar-

nered relevance as a conduit of cross‐border flow of ill‐gotten funds, alongside the use of the financial

system and the physical movement of cash.

Accounting tricks offer a wide range of techniques granting wide enough a room for manoeuvre for

producing international financial flows bereft of inherent economic rationale but on paper. As in the

case illustrated above, under or overinvoicing (depending on the desirable direction of the funds to be

1 The views and the opinions expressed in this paper are those of the authors and do not necessarily represent those of the

institutions they are affiliated with. We wish to thank for their useful comments Giuseppe De Feo, Silvia Fabiani, Domenico

J. Marchetti, Claudio Pauselli, two anonymous referees, seminar participants at UIF, 2016 SIDE‐ISLE Conference in Turin,

Eurostat meeting on Illegal Economic Activities in National Accounts and Balance of Payments (March, 2017) and the 2017

UIF‐Bocconi Workshop “Quantitative methods and the fight against economic crime.”

2 The case is taken from Financial Action Task Force (FATF) (2006).

|

3121

GARA et Al.

transferred) or false invoicing altogether3

can be usefully deployed in order to create artificially in-

flated payments outgoing from a goods importing country or to curb otherwise much higher incoming

transfers accruing to an exporting jurisdiction.

The most obvious underlying purpose of shenanigans of this kind is possibly connected to what can

be defined, with an understatement, as tax optimisation policies, aiming at lowering the tax burden

of a company in a high‐tax rate country and raising it in taxpayer friendlier jurisdictions. This type of

flows can be taken to be illegal only to the point that the underlying tax conducts amount to a criminal

behaviour, which is not necessarily the case in all jurisdictions. Alternatively, the financial wedge be-

tween the actual value of the goods being exchanged and the corresponding movement of funds may

be connected to the proceeds from the supply of illegal goods and services (such as drugs, weapons,

human beings trafficking and kickbacks) or to create cash reserves that can be consequently further

transferred with no legal obstacles and put to different uses.

Possibly such misalignment between what is owed and what is actually paid in connection with a

given exchange of goods can be reflected in national trade statistics since either accounting or custom

documentation presented in different countries may not necessarily coincide. That could be visible by

comparing mirrored bilateral trade data (so‐called mirror statistics) measuring the exchange of goods at

some level of details between a country and each commercial partner. Thus, mirrors, in trade statistics as

in fairy tales, may turn out to possess extremely powerful properties in detecting menaces of some sort.

Alongside purposeful misevaluation of goods, another possible source of discrepancies between

two countries' mirror statistics is misreporting, which can either refer to the type of goods being

exchanged or their country of origin/destination. Quite tellingly, in the case illustrated above the cir-

cumstance whereby the country the goods are shipped to differ from the country where the buyer

resides may potentially lead to the misalignment of the recorded partner country for this particular

trade. Misalignment due to incorrect reporting can be due to the inefficient reporting system of the

countries involved in a trade flow (more likely if either is a developing country) or to different goods

classification criteria (which is less frequent when countries are regular commercial partners or are

parties in multilateral trade treaties). Alternatively, misreporting can also be deliberated, which may

entail the same opaque underlying motivations attached to misevaluation.

Misreporting in all its shapes may pursue a wide array of objectives, in addition to those which

have just been mentioned: an importer may declare the shipment of a different type of goods from the

one actually delivered in order to pay lower tariffs; an exporter may indicate an incorrect country of

residence of the commercial partner so as to bypass commercial embargoes of some kind. The purpose

of this work is that of analysing only those cases in which reporting hoaxes can be used as a conduit for

ill‐gotten financial resources as opposed to those instances in which invoicing or reporting tinkering

pursue other illegal goals.

Based on previous works in this field, we estimate a linear mixed model aiming at identifying

the main determinants of mirror statistics discrepancies. The latter are specified, with reference to

Italian trade flows for the 2010–13 period, at a level of 6‐digit classification for each partner country.

Explanatory variables include those accounting for inefficiencies in the reporting system in the part-

ner country (which is often linked to the level of economic development) and possible misalignments

of product classification (due to the lack of trade agreements or to infrequent trade flows).

Our work improves with respect to the existing literature in three main directions. First, import

and export values are adjusted for cif/fob discrepancies with point data based on the Bank of Italy's

3 False invoicing may be arguably applied much more easily to intangible products, such as services, but less so to physical

goods that can be weighted, counted and measured some way or another. Indeed, an analysis of the service sector is indicated

as one of the potential further avenues of our research in the concluding sections of the paper.