Alejandro Hope, until recently in charge of organized-crime analysis at CISEN (the Mexican intelligence service) shreds the latest nonsense numbers from the UN Office on Drugs and Crime.
Part of the problem with drug policymaking at every level is its almost complete detachment from reality. In particular, statistics about drug volumes and revenues are more or less invented at random, then passed around like bad pennies.
I have this flagged under “Lying in Politics,” but “lying” isn’t really the mot juste. This is more what Harry Frankfort calls “bullsh*tting”: the statements are false, but not really intended to deceive. Stating them, and citing them, are ritual gestures: socially meaningful, but essentially content-free. (And debunking them is one of the central themes Jon Caulkins, Angela Hawken, and I pursue in Drugs and Drug Policy, whose subtitle should really be, not “what everyone needs to know,” but “what everyone needs to stop believing.”
Translation below the fold.
Footnote More from Alejandro at his new blog (mentioned earlier by Keith). It’s called “Plata o plomo” (“Silver or lead”), the proverbial choice offered to law enforcement officials by Latin American drug traffickers: take our money or face our bullets.
Google Translate does a pretty good job, as usual. Here’s their version, mildly corrected by me (and re-corrected by the author).
UNODC and the Fine Art
of Numerical Sleight-of-Hand
I’ve told this story before, but it’s well worth repeating. In the eighties and nineties, the United Nations Program for the Control of Drugs estimated the global market for illegal drugs at $500 billion. One day, the number dropped to $400 billion. When asked about the figure by the Colombian economist Francisco Thoumi, officials discovered that the UNDCP had 1) taken various estimates, 2) averaged them and obtained a score of 365 billion and 3) rounded up for purposes of communication.
In fifteen years, things have not changed much. The United Nations Office on Drugs and Crime (UNODC, successor organization to the UNDCP) this week released a report on illegal money flows. The findings are striking: the authors report that the criminal activities generate $2.1 trillion (million million) dollars, of which $1.6 trillion is laundered. If we consider only the income generated by drug trafficking and other turns of transnational organized crime, we reach a total of $870 billion dollars, of which 580 billion would go through the money laundry.
How did the UNODC arrive at those numbers? Very easy: 1) took various estimates, of heterogeneous quality, and 2) took a simple average (not rounding). The selected estimates include the following:
* An estimate based on data from 1988! generated by the Financial Action Task Force (FATF), a multilateral organization dedicated to the fight against money laundering and therefore interested in presenting a number as large as possible.
* An estimate of an NGO called Global Financial Integrity, dedicated to issues of money laundering and with similar incentives to the FATF (the larger the number the better).
*UNODC’s own figures about the size of the global drug market.
* A paper whose lead author is an Australian consultant named John Walker (don’t know if you like whiskey).
* Various estimates of varying quality (to put it gently) for six industrialized countries, some of which date back to 1990. After averaging, the authors extrapolated to the entire planet.
Perhaps aware that the procedure could result in some slight credibility problem, the manufacturers of the report also decided to include a bottom-up estimate, starting with estimates of the revenue generated by illegal activities and using the global cocaine market as an example. The result is a death-defying leap of logic with a Dutch triple twist:
The first step is to estimate the size of the global cocaine market at retail prices. Apart from the slight difficulty of estimating the global number of users, the exercise requires retail pricing information on all (or almost all) countries in the world. Outside the United States and one other country in the OECD, there are no price data bases: at best, the countries have some systematic collection of anecdotes. How you build a fairly robust estimate with that kind of information is a mystery that is beyond me.
The second step is even more complicated: it requires estimating, country by country, the income generated by cocaine. That requires, among other things, locating with some precision cocaine flows between pairs of countries, besides knowing import and export prices worldwide, to determine where the value is added. If information on retail prices is bad, that on import/export pricing is catastrophic. The authors use as a source the ARQs (Annual Report Questionnaires): As the name suggests, are questionnaires (on consumption, seizures, prices, etc..) that UNODC distributes each year among the member countries to prepare the World Drug Report. In the case of Mexico, the questionnaires are answered by mid-level officials of the PGR (the Attorney General’s office), apparently by Divine inspiration, without any formal validation process and for the fundamental purpose of getting something down on paper. They can hardly do otherwise, as the country has never had a systematic database of drug prices: the most we have is the occasional anecdotal data provided by one or another drug arrest. You can be sure that the procedure is similar in the vast majority of countries.
The third step is very interesting attempt to estimate the overall number of dealers, wholesalers and retailers. How they do that? Simple: start with estimates from United States and four other countries, take a simple average, and extrapolate globally. It’s that simple: ignore any differences in distribution patterns between countries.
Assuming a Pareto distribution for the market (20% of participants capture 80% of revenues) an estimate requires only simple arithmetic: divide a dubious number (the revenue estimate by country) by a highly questionable figure (the number of participants per country), weighted with an arbitrary distribution criterion. Very scientific.
Then we know how much the narcos earn. Now we have to define what percentage of that income is sent to the laundry. The report’s authors have the perfect solution: assume that all money received by a drug dealer over the country’s per capita income is saved and therefore laundered. Thus, in the case of Mexico, all income of a drug dealer over $9,000 goes to the laundry. To which on must ask: Really? Have you never heard of drug dealers with a taste for extravagance and conspicuous consumption? Of the zoo of Pablo Escobar? How about the diamond-encrusted guns? Or the Bacchanalia of Arturo Beltran Leyva? Do you really believe that a person who is risking his life every day has a high propensity to save? Worse, you never have heard of the reinvestment of profits? As I have already mentioned, the narcos are distinguished by their lack of access to credit and other formal financial instruments (insurance, for example). Therefore, to fund their operations, they should maintain comprehensive financial reserves in highly liquid instruments: cold, hard cash. And besides, what legitimate business would give a rate of return higher than the drug trade? If a cocaine dealer can invest in cocaine, why invest in real estate or CDs? It’s not as if this crowd is highly risk averse.
The last step is to determine where money is laundered, using what is known as a “gravity model.” But that’s a mostly useless exercise, because the whole prior process is a joke in bad taste.
In addition to the generic absurdities, the report contains anthology-quality gems of reasoning. For example, the authors argue that Mexico’s net income from exports of cocaine are two billion dollars. But as the number sounds small, inconsistent with the story of Mexico’s drug lords as business tycoons, the authors decide to allocate another five billion dollars. The excuse for that number is that, according to the DEA, 15% of people arrested for cocaine trade in the U.S. are of Mexican nationality. The report therefore assigns to the Mexican cartels that proportion of the U.S. retail market: rounding up, 15% of $35 billion becomes $5 billion.
The little problem is that there are many Mexicans in the United States involved in the cocaine business that have nothing to do with the cartels (other than as customers). Has the UNODC not heard of the Mexican Mafia (the famous Eme) or the Texas Syndicate or Nuestra Familia or the many other gangs that recruit Mexican nationals north of the border? These gangs are hardly mere appendicies to the Mexican cartels.
To top it off, the report attempts to justify the award of $5B to the Mexican narcos by referencing an estimate of revenues from drug trafficking in Mexico, conducted in 2006 by the Office of National Drug Control Policy (ONDCP). The only drawback is that ONDCP itself disavowed this estimate last year, after supporters of the legalization of marijuana began to used it as the argument for Proposition 19 in California.
Beyond its specific atrocities, this report, like all its kind, faces an insurmountable difficulty: if the dirty money is as abundant as they suppose, why has no one ever found it, at least not in the expected amounts? The same report acknowledges that seizures of illicit assets globally do not reach even one percent of the estimated illegal income, but attributes that fact to regulatory weakness. Maybe, but as it happens almost every country in the world has strengthened its regulatory and monitoring mechanisms against money laundering, and yet the imagined sums are still nowhere to be seen. Is this because of corruption? Maybe, but all around the world? What of the United States, the United Kingdom, Switzerland, Sweden, Singapore, Canada?
Here we face the financial equivalent of the Fermi paradox: the illicit money should be there and we should find it, and yet does not appear in the expected amounts. Shouldn’t we start to suspect that maybe the inexhaustible rivers of illegal money are imaginary? If they exist, they shouldn’t be impossible to find.
I leave for the weekend, thinking.
(No, I don’t read Spanish, but I know what Alejandro is trying to say. The biggest howler is translating “se lavarÃan” (“is laundered,” in the context of criminal money moving) as “is washed,” which isn’t as nearly as good as the live translator at a conference I recently attended who translated the statement “Some of these ideas will need to be corrected in light of later information” as “Some of these ideas will need to be punished …”)