Estimating the Volume of Illicit Financial Flows – Definitely Not a Science Probably Not Yet an Art

Ainsi parlait Michel

There’s no central authority that keeps track of and publishes details of illicit financial flows.  No national central banks of crime.  No international (criminal) organization akin to the OECD or the IMF.

No major listed criminal enterprises that report their annual results of operations including sector information, costs of doing business, including those related to bribes and threats from which we might construct estimates.

By their natures these flows are undisclosed.

Parties to illicit transactions, the intermediaries they use, and parties they co-opt don’t self-report for obvious reasons.

But you do see figures for these flows.

For example, UNDOC states: 

“The estimated amount of money laundered globally in one year is 2 - 5% of global GDP, or $800 billion - $2 trillion in current US dollars.”

You’ll often see that latter number USD 2 trillion cited in press reports.  Here’s one from January this year in which Bloomberg states that “shady transactions continue to reach as much as $2 trillion a year.”

That wording implies that amount of money laundering is capped.  Apparently, once they reach USD 2 trillion in a year, criminals have to stop money laundering.  Unclear how this information is communicated.

Let’s stop for a minute and reflect.

Amounts and Percentages

UNDOC states the amount of money laundering in one year is estimated as a percentage of global GDP.  It then goes to give a range of USD estimates.

What do those estimated amounts work out to in terms of global GDP?  To USD 40 trillion.

According to World Bank data, that’s roughly the estimated world GDP in  2003.

Are we blindly repeating 16 year old estimated amounts?

Referring to the same World Bank source above 2017 world GDP was some USD 80.886 trillion and in 2018 some USD 85.791 trillion.

That would make

1. 2017 money laundering USD 1.687 trillion to USD 4.044 trillion and
2. 2018 money laundering USD 1.716 trillion to USD 4.290 trillion.

Now it could perhaps be that money laundering is not a growth business.  It’s capped at USD 800 billion to USD 2 trillion.  More Sears than Amazon.

AA doubts that.

So, is the answer that we just need to update the amounts?

It’s not that simple.

Origin of the Estimated Range

Before we do, we should know where and when this 2% to 5% estimate came from.

As near as AA can tell, it was first mentioned in a speech by then IMF Managing Director Michel Camdessus in 1998:

“While we cannot guarantee the accuracy of our figures—and you have certainly a better evaluation than us—the estimates of the present scale of money laundering transactions are almost beyond imagination—2 to 5 percent of global GDP would probably be a consensus range.”

Note the words “we cannot guarantee the accuracy” and “would probably be a consensus range”.

Clearly, models are estimates so they are not 100% accurate. Hard to quibble with that statement, though it does serve to warn that one should treat the model’s results with caution.

But “would probably be”.  On its face, that means we really don’t know if it is a consensus or not.  Or who the parties to the consensus might be. Or how they achieved consensus.

Or whether Mike pulled this out of his hat.

Age of the Model

But let’s assume there was a formal model of some sort, which is unclear, and this range is not based on graph drawn on a cocktail napkin in a bar somewhere.  Or a discussion in a bar.

Are we working with a model from 1998?  Should we be?

Is the world the same as it was some 20 years ago?

Results of the Model

The model has a range from 2% to 5%.

At the risk of understatement (which AA delightfully accepts, the risk not the understatement), that’s a wide range.

Imagine you came to AA Investment Advisors (AAIA) and asked AA, the firm’s Chief of Research and Head Strategist, what the value of a single share of Company XYZ and of Company DFE were worth.

If you got the answer about Company XYZ from “USD 20 to USD 50” and Company DFE “from USD 40 to USD 100”, what would you think?

You’d probably not think this was particularly useful, nor something that should be relied on for your investment decisions. Or that the nature of these stocks made it impossible to perform a more precise valuation.

Now if you went to Goldman Sachs and got the same answer, what would you think?

Would you think that AA wasn’t much of a financial analyst, but the analysis of the good folks at the Goldmine was spot on and highly useful simply because they were at GS?

In responding, don’t overlook the fact that  you're unlikely to be able to get a decent cup of Turkish coffee at all of GS's global offices.  But at all of our one office in the world at AA Investment Advisors, you can get what is probably one of the best cups of Turkish coffee in the world.

100% organic Arabica coffee hand-ground fresh for each cup using only the finest Turkish grinders. As long as you promise not to spill it on Madame Arqala’s antique furniture or rugs. There are some severe penalties for such infractions as AA can testify.

Hopefully not, you should think the same in both places and with respect to this model.

The right conclusion is that model is pretty much a rough estimate.  Another understatement.  Probably given the task it has set for itself:  to estimate IFFs that are deliberately hidden.

Don’t be dazzled by the source of the model.  “Aristotle says” or “the IMF, UNDOC, UNCTAD, FATF say” doesn’t make it true. Nor confirm precision where none is possible.

Don’t be More Royalist than the King

If you’re still not convinced by AA, if you dig a bit deeper, you’ll see that UNCTAD believes less in the model than many outsiders appear to.

In UNCTAD’s 2018 Annual report go to Goal 16 and look for the dropdown menus.  Pick the one about illicit financial flows.  There you’ll find this quote:

“In close cooperation with the United Nations Office on Drugs and Crime (UNODC) and the United Nations Economic Commission for Africa (UNECA), UNCTAD is working on developing a measurement framework for Goal indicator 16.4.1. This is a complex project that involves defining and designing measurement tools to capture both illegal and illicit activities which, by their very nature, are hidden deliberately.  As a co-custodian of the Goal indicator 16.4.1, we are striving to define, estimate and disseminate statistics on IFFs in the context of developing economies in Africa, some of the most affected by this developmental challenge. Through a series of implementation guidelines, pilot activities and technical assistance, by 2020 NCTAD, UNODC and UNECA will have developed the capacity to measure IFFs in several participating countries in Africa.  The result will be the capacity to more accurately estimate IFFs in participating countries. The lessons learned will inform the national monitoring of IFFs and will guide policy actions in affected countries to curb these flows. As such, it will also increase the likelihood of developing countries achieving the 2030 Agenda for Sustainable Development.”

This post marks the start of series on Illicit Finance.

In following posts, we’ll take a closer look at  difficulties in modeling IFFs, in measuring the probability of corruption, etc.