Detecting Illegal Arms Trading

I always love it when I see a paper that only an economist could write. An ingenious new study by Stefano DellaVigna and Eliana La Ferrara definitely fits that description.

The issue they tackled is detection of illegal arms trades that defy United Nations embargoes. Their idea was to use the information embedded in stock markets to tease out indirect evidence of the illegal deals. Legitimate weapons dealers that obey the embargoes are hurt by them, while illicit dealers benefit. So DellaVigna and La Ferrara examined how the stock prices of weapons dealers in different countries are affected by events that raised and lowered the likelihood that the embargo will be rescinded. They found that the stock prices of arms sellers in countries more likely to house illicit dealers (e.g. countries with more corruption and less rule of law) rise when events prolong embargoes, whereas stocks of arms dealers in countries like the U.S. fall in response to the same news.

While this is not the sort of evidence you could use to convict somebody, it is an excellent example of what some are calling the “forensic economics” movement. In a column in last week’s Philadelphia Inquirer, the conservative talk-radio host Michael Smerconish called for forensic economists (myself in particular) to apply our tools to terrorism. More to come, perhaps.

Rita: Lovely Meter Maid

Steven Levitt: "I always love it when I see a paper that only an economist could write." Ah, for the love of Zeus, I'm wringing my hands in anguish here! My dear, dear, potentially misguided Levitt, have we not yet learned to use a modicum of caution when mocking papers? Remember you not, Steven of Levitt, the woesome AC/DC imbroglio of only a day ago? A mere day has passeth, and you are back to this bouncy attitude of savoir faire (shakes head sadly). Oh well. I really wrote this so I could employ the term: imbroglio, so I guess it's all good.

Declan Chellar

“I always love it when I see a paper that only an economist could write.”

I didn't think you were mocking the paper. I thought you meant you actually loved it.

Did I miss something?

Brick Cheney

Can Rita read?

Mark McClellan

The last time I played a game like "Silver Falls", millions of people weren't slaughtered in ethnic cleansing when I quit the game.

Blue Sun

"Forensic Economics..."

Before I read a paper by an economist, any economist, I remind myself of what an economist I used to work with was fond of saying:

"An economist is somebody who can explain tomorrow why what he predicted yesterday did not happen today."


It would be more interesting if they didn't find what everyone already believed but instead found a counter-intuitive relation. You don't need sophisticated analysis to believe and know their conclusion. It would also be more interesting if they could somehow measure the effect - as in anticipated revenues, reported earnings - of displacement of sales from high legal to low legal countries. That would give an indication of the degree of effectiveness of an embargo and the degree to which arms sales would merely be moved to less lawful locales.

George from Oregon

Jeeez Another "Conservative Talk Radio Host"! They sprout like mushrooms living in a environment of their own making.

I think you would do well to check out their financial statements not their stock prices. Stocks move for many reasons. What we need is forensic accounting, not economics. It seems like it would be easy, in just looking at the stock price, to draw false corelations. Just because something seems to "make sense" does not mean that it is anything more than just coincidence.

Chris Jacques

It seems that some people either haven't read Freakonomics, or missed the point of this blog post.

Dr. Levitt doesn't use economics to predict things with precision. Instead, it seems to me that his approach is to explain, or at least explore, events and trends in the past. From these novel observations, one may be able to draw novel conclusions. Of course it's always possible to draw lessons from the past and apply them to the present, but a good economic thinker would realize that all situations are different, all information is incomplete, and that the past can only be used as the most general of guidelines.

Max Power

Not only has market data been applied to detect illegal arms trades, as in this case, but stock market data has certainly been used to evaluate counterterrorism policy. A paper by Asaf Zussman and Noam Zussman in the JEP use stock market data to determine the effectiveness of assassinations as a counterterrorism policy in Israel. It certainly is an interesting approach.

Rita: Lovely Meter Maid

Admittedly, this topic was a (tad, wee, tiny, bit) more (dare I say it)? boring for me, than others, so I guess I (ahem) *jumped the gun* and gladly took to the idea that Levitt was being enjoyably ironic again. Or, maybe I had hoped he was, since I actually didn't read much beyond "I always love it when I see a paper..." Now I see he was sincere. Deeply sincere and indeed, Decidely sincere. Yes, and so, now I am too. Sincere, that is. I'm Older. And Wiser. Yup, and Much More prone to reflect now, rather than just *fire off* a response. At least for today, anyhow...

Ken Hanson

Ease up on Rita. Anyone that can work imbroglio into a post needs to be met with aplomb.

Enough Wealth

Just goes to show that economists, unlike freakonomists, just draw the obvious conclusions. Of course, "...stock prices of arms sellers in countries more likely to house illicit dealers ... rise when events prolong embargoes, whereas stocks of arms dealers in countries like the U.S. fall in response to the same news.". That's exactly what drives stock prices - people buying stocks that they *think* will benefit from news and selling those that they *think* will suffer from the same news. Unless there is a LOT more insider trading going on than most people think, this doesn't mean that stock prices actually indicate which companies WILL benefit/suffer from the news. The only way for such stock analysis to correctly identify companies that trade in illegal arms would be to see what stocks were being bought or sold by "knowledgeable insiders" ie. those that buy illegal arms and therefore know which companies actually will benefit from the news.




Thank you, Chris Jacques!


This is not forensic economics, it is forensic statistics.


This is related to a topic down the page, the sunk cost.

Doesn't the argument made on this link break down if there is some probability of success if the venture is pursued?

Whether or not we think the Iraq war is hopeless now, a few years ago at least we can agree that it certainly wasn't. Was it foolish to pursue a victory in 2004? 05?

What if someone came to you and said "I'll roll a dice, and you pick a number. If you get it right, I'll double your life savings of 100k". Horrible deal, right? Your odds of hitting are 1 in 6 and your payout is only 2:1.

But what if you are a moron and you go ahead and do it anyway, and lose. Now the same person comes to you and says "if you can get 10,000 more dollars, I'll give you another roll of the dice to try to win the 200,000 dollars (your savings + his match). Do you do it? Of course, because now your payout is vastly superior to your odds.

In short, this sunk cost "fallacy" assumes the endeavor is hopeless. Its quite difficult to prove something like that, isn't it? I wonder how many times history is riddled with people who kept pursuing a goal and it DID pay up in the end. Of course, for the ones that quit early, they were inevitably going to fail, right?



this is the greatest blog thta i ever read, this is about the illegal bomb seller and more dangerous thing.

good blog

keep on working


Here! Here! At the risk of rather self-agrandizingly meeting myself with aplomb of my own marshalling, I hereby convoke my uniquorate panel in adjudgement of the meritocratic, if onanistic, manner in which panygiric be fawned upon the blogasider's elocutive invocation of "imbroglio", lumpenly, brazenly.


Re: Jeeez Another “Conservative Talk Radio Host”!

By jove, I think you've cracked it, George. How silly of us all not to have already considered simply walking into the nearest office of NamibianArmsULike and checking their sales invoices?

Blinded by the Comma


Besides taking a lesson in reading comprehension you should also enroll in punctuation and grammar. Let the rest of the readers comprehend your points!

Bill W.

What I'd like to know more about why Rudy Giuliani still keeps crediting himself with a drop in crime in NYC when, if I recall from a couple of years ago, Freakonomics credited the drop to Roe vs. Wade. Which made much more sense--with figures, even, to back it up.

As to forensic economics (would that be Forenomics?--or Forcenomics--pron. fork-enomics??):

Rather than watching the stock market in these matters (the gamblers' pit), it would make more sense to get the "underground" numbers for:

The illicit drug business: Agriculture, harvesting, grain elevator guy (or whoever stores up the raw stuff, like poppy-sap in Afghanistan), processing, transport, wholesale and retail.

Match up w/ estimates of bulk arms sales (weapons, ammo, accessories).

Match up w/ annual increases in funding for (one could almost giggle if it weren't serious) "the war on drugs"--this in terms of:
1. Public funding of drug interdiction efforts
2. Public funding of miscreant warehousing operations (now being privatized as "Prisons R Us"
3. Public funding of the warehousing feedstock, or judiciary, w/i ts attendant prosecutorial and defensorial impedimenta.
4. Tax subsidies (as taxes foregone, perhaps).
5. Private support for the "I apologize to the people I hurt the most Spas, Inc."
6. Analysis of the percentage of funding governmental secret organizations get from licit (public funding, via state and federal legislatures) and illicit (public funding, via the underground economy) sources.

Put all this together in a series of singular regression analyses (multiple regressions would be too complex, so just do 'em one at a time, seriatim--doing multiple regressions is like trying to hurry the process of building a baby by having 9 ladies dancing (for a month), rather than letting Ms. Partridge alone for 9 months in her pear tree. Seems to make sense until you try it w/ your plimsoles on the ground. Or espadrilles in August, is it? (I'm using a borrowed computer that doesn't have "define, double-click, lookup in dictionary" access to the OED--which always seems so funny, that OED, which is newer than the NED--which you hardly hear about these days in lexicon-land.

All I'm sayin's you gotta get real when yer talkin' black economy, black bags, black ops, black hoods--and the inevitable blackguards who make it all happen.

Which reminds me, while I'm on this Shandy-esque series of multiple digressions, just who is it that owns the Federal Banks? Are they really federal--as in, "owned by the US government," or are they actually privately-owned but with a very clever corporate name, "Federal," to make it sound like they're publicly owned, when they're not.

If they're private, then isn't there a law that prevents folks from incorporating businesses so's to make 'em sound "official-like"? On the other hand, if they're totally private, then they don't have to rely on Patents & Trademarks 'n' all as gov't agencies to defend/regulate their incorporations.

Anyway, who does own the federal banks, or "The Fed"? To trace gun-running, etc., would these outfits have to be factored into the "things to do" list of considerations you'd have to look at the figure out where the guns were going? (Well, guns, tanks, missiles, ammo--the whole 9 cubic yards)?

Which reminds me of another nagging question: To whom was the chopped-up steel from WTC 1, 2, 3, 4, 5, 6 & 7 sold? Who paid for the removal of it from the WTC? The loading of it onto barges/ships? To whom did the sales price innure (Silverman, or Silverstein, whomever--the guy who got back Billions of bucks in insurance payout for his millions in premiums (VERY good foresight, there. His occulist should go public.).

Anyway, there's probably not much higher a calling than Forensic Economics, IMHO. And if Freaky to boot, so much the better.
(as they'd scribble, back in the day)