Terror Trading

Lots of controversy over a Defense Department plan to set up an online futures trading market to help forecast terrorist attacks.

Smart money is that the Pentagon messes this up if it ever does get off the ground, but it beats a certain other DARPA initiative.

"Research indicates that markets are extremely efficient, effective and timely aggregators of dispersed and even hidden information," said the Pentagon statement. "Futures markets have proven themselves to be good at predicting such things as elections results; they are often better than expert opinions."

That's a concept experts in Washington aren't likely to appreciate.

Comments

[Anonymous]

Well, Brad, this is all academic now that the market has been canned. What a bunch of gutless weasels in DC, anyhow.

To respond - my point was that, in order to significantly distort the market, the terrorists would have to trade against the trends. Trading against the trends is very difficult to do without getting caught - the SEC already runs computer programs to catch people doing this in the much larger and more liquid financial markets, because trading against the trends is the fundamental strategy of most market manipulation schemes.

Also, I believe that the point of this market was to move the analysis of intelligence outside of the narrow bureuacratic world and draw on the collective intelligence of a market mechanism. This market would serve as a checkpoint for the wonks, rather than just another collection of wonks.

But, its dead now. So never mind.

[dragoon]

Game over, man.

It's been closed down due to complaints. What a shame. DARPA seems to be pissing off a lot of people nowadays. That anger should have been saved for the TIA project.

[Rex Stetson]

No Future for FutureMAP
The Pentagon has now cancelled FutureMAP-the bizarre but clever "futures trading" approach to terror attack prediction, amid heavy criticism from Congressional Democrats
This is another humiliating PR defeat for the Terrorism Information Awareness Project, who have already had to change their name, remove the Masonic imagery from their logo, and downplay and reduce the scope of their all-inclusive, all-knowing super-database. Sounds like DARPA really needs some good PR help.....
The full story/rant is available on my blog. click on the name if interested.

[Tom Hynes]

I am a big fan of getting information from a market.Check out http://www.tradesports.com You can bet on Osama's capture, whether we will find WMD in IRAQ, Gray Davis's recall, the next Supreme to resign, whether Hilary is the next president, future alert levels. Real money. (Full disclosure: I lost $20 on when Saddam would be gone, won $20 betting against Clay in American Idol). This is much more fun that the Iowa site.

Libertarians should embrace this unregulated market.

[Jon H]

The problem is simple.

Al Qaeda could see the site too. Why would they 'buy high', and choose an 'expensive' target, when they could 'buy low', and choose a cheap target, or a target not listed at all, which according to the market is a target unlikely to be hit?

The theory behind the market idea assumes that the terrorists are bad investors without a web browser. That's the only way it would work if the high-priced futures actually represented likely scenarios.

Instead, the list of futures available would just represent targets and tactics the terrorists should avoid.

[Jon H]

Brad S writes: "As a possessor of an MBA from one of the top business schools in the world"

So does Bush, a serial business failure, so this really doesn't carry the weight it used to.

Nothing personal.

[Jon H]

BTW, above, when I write about Al Qaeda 'buying high' and 'buying low' I don't mean literally buying futures. I mean launching a terrorist attack that corresponds to a scenario with a high price in the market or a scenario with a low price in the market - ie, a scenario the market deems unlikely.

One other thing comes to mind...

Consider how much money Al Qaeda gets from the Saudis. Saudis who are wealthy and probably know something about investing could support Al Qaeda indirectly by creating noise in the market.

Also, I saw somewhere that the participants would have been limited to $100 stakes, total. Which means that nobody would get rich, but it might make it easier to create a pop or drop in the futures for a particular scenario, because there wouldn't be that much money in the system overall.

[Brad S]

A futures market for terror? What the hell? Am I going to be calling the Chicago Board of Trade one of these days, "Yes, I'd like to buy 100 contracts on April wheat, I'd like to sell short 500 contracts on June pork bellies, oh, and I'd like to buy 100 contracts on October Al Qaeda..."

[s.m. koppelman]

Given this new futures market's logo, I reckon the good Admiral was in the room when this thing was being planned.

Does this mean it's now all right to bet on domestic assasinations too?

[Andy]

The hysteria that has followed this story is hilarious; a bunch of wackos and the press are assuming that this market involves real money and somehow terrorists are going to make dough by buying futures on attacks they are planning. Of course, this completely overlooks the fact that this is a simulation/game....DARPA has always used simulations to predict stuff. Oh well, never let a fact get in the way of a good story. Here are a couple examples
Linked by Drudge

CNNs take on this

[s.m. koppelman]

Note to Andy, or, uh, "DJ Big Bird": It's not a simulation. It's a market. With real money.

[Alex]

I guess if I buy catastrophe futures on the CBOT I'm hoping for an earthquake, and if I buy fire insurance, I'm hoping my house burns down.

Andy, it certainly involves real money.

[Brad S]

Andy - naturally, terrorists won't be able to profit from this, but couldn't they use this as a mechanism to mess with our minds? i.e. buy lots of contracts when no attack is planned as a "bluff", or take a short position in lots of contracts for a time when an attack is in the works? I'm just hoping that the Defense Department isn't really going to be relying on this for much other than entertainment value.

[Lefty]

Stiffs.com has a celebrity dead pool already in place. Maybe Poindexter can get a few pointers from them.

[Alex]

Brad, I think none but the world's dumbest terrorist would participate, given the scrutiny market participants will no doubt attract. The purpose is (I think) to aggregate the opinions of third parties informed enough to back them up with hard cash.

[joe]

I take back any remarks I may have made about libertarianism lacking a foreign policy.

Is this a gag?

[Brad S]

Alex - but if this market has any appreciable size, it will likely be touched by countless intermediaries. A terrorist could easily go through multiple intermediaries and make trades in a market such as this.

[Alex]

I don't think there will be any brokers or intermediaries since betting seems to be restricted to market registrants.

[Alex]

I don't think there will be any brokers or intermediaries since betting seems to be restricted to market registrants.

[Rex Stetson]

As a libertarian, I must now confess to being somewhat torn about the Total...er..Terrorism Information Awareness Project. On the one hand, I fear and distrust the loss of my privacy and the creation of an all powerful, all-knowing supercomputer ( no doubt from having repeatedly watched both Wargames and Superman III dozens of times as a child) but as a lover of the free market in favor of privatizing many government functions, I have to applaud this clever idea that is most likely much more efficient and accurate that the analyses of the so-called intelligence "experts" that dropped the ball pre-9/11, the collapse of the Soviet Union, Nigerian Uranium, etc...
My full rant is here if anyone is interested: http://www.rexstetson.blogspot.com/

[s.m. koppelman]

Who needs intermediaries and scrutiny of traders? With an open market functioning with real funds, the data will be good enough for a parallel trading floor to be set up. What will stop offshore betting parlors from offering wagers that track the prices on DARPA's market?

In principle, this wouldn't be much different from dead pools and the dubiously tasteful wagers on mayhem that these firms already offer. In practice, though, it raises the stakes both for gaming the system and drawing insight into US policy from the actions of the "experts" invited to participate with their own money. It's bad for the same basic reasons you don't want to let athletes and coaches place wagers against themselves. Not only does it make it likely someone's going to throw a game, but it also means nobody's going to bet against a coach who bets against herself.

[Keith]

The Iowa Electronic Market (http://www.biz.uiowa.edu/iem/), which sounds like the model for this options market, used to boast that it was more reliable on elections than Gallup. This is a useful tool - participants can arbitrage their perception of national elector psychology against the market perception. The ratios of the different payout conditions is much closer to "expected value" in the technical probability sense as compared to poll results, because it is an economic value.

I for one would be fascinated if valuations in this market did not predict DHS color changes. In periods where they don't track, one or the other group believes something that the other doesn't. It's that kind of arbitrage of sentiment that I believe the DOD wants to capture. But, I think it would have been better off as a political move if this was hosted outside the government.

Will there be enough "real money" in this market for, for instance, insurance companies to hedge against terrorism? I just had a thought - can this market be re-packeaged and re-sold as THE Federal "Insurance of Last Resort"? That might make squeamish congressmen buy in; and also establish the premium cost of any federal "insurance of last resort" at least semi-rationally instead of as an emotional revenue boondoggle. And it would force insurance co. participants to pay a damn lot of attention to the options values if their bail-out is in this market, and only in this market.

Huh. I'll have to think that through some more.

[Lefty]

Actually, it's a new way for Poindexter to finance arms sales to Iraq. Maybe he won't go to jail this time.

[Alex]

Bets are already taken on terrorism, by proxy through insurance stock and commodity options and directly through catastrophe and terrorism insurance... how is this significantly different?

[Jesse Walker]

I take back any remarks I may have made about libertarianism lacking a foreign policy.

Touch?.

[Anonymous]

"Alex - but if this market has any appreciable size, it will likely be touched by countless intermediaries. A terrorist could easily go through multiple intermediaries and make trades in a market such as this."

So what? If the terrorists are trying to game the market to make money off of it, they can only do so by predicting their own attacks. Gee, we wouldn't want terrorists to pass information to DOD about when and where they plan to attack, would we?

If the terrorists are trying to game the market by shorting their own attacks, well, that will show up as an anomaly, too, just as any break in a pattern of trading shows up as an anomaly. In this case, if they do so little trading that it doesn't show up as an anomaly, then it won't be enough to affect the market or corrupt the predictive value of the market, but if they do enough trading to distort the market, the smart folks at DARPA will be able to spot it.

Do you people know nothing about technical market analysis? Geez.

What is the downside here? I haven't heard any objections yet that didn't boil down to "I think its icky" or "I am a flaming idiot about markets and how they correlate and weight diffuse information."

[s.m. koppelman]

Fair point, Alex. I'd say the difference between this and terrorism insurance is that insurance pricing is set by people who read analyst reports and conduct their own analysis. This market, on the other hand, goes straight to the policy analysts and--given the close links between government and think tanks--policymakers themselves. Someone who wants to find the soft spots in US policies and defenses right now would need to pore through reams of data and come up with her or his own analysis. With this market in place, the underlying whims that go into defense policymaking and foreign relations are laid bare in predigested, bite-size form. I was going to say it's like the difference between playing the board game Risk in its original form over the course of hours, or playing the late-80s Macintosh version of the same game in minutes, but really it's like the difference between playing poker and playing poker with see-through cards.

[Alex]

if terrorists want to piss away large piles of enriched-uranium-money merely to misdirect some DARPA project, god bless their pointy heads.

[Neil Block]

..and I thought the floor of the CBoT was exciting to visit. I already contacted the PAM project to find out how to become a trader; it should be an interesting ride.

Markets rock it hard.

[Total Information Analysis]

I can't believe no one here has realized the real model for this venture -- Jim Bell's Assassination Politics.

Bell's idea involved encrypted anonymous ecash though, and did not exclude domestic assassinaitons from the betting pool.

For his trouble in doing Poindexter's thinking for him, Bell was harrassed by the Treasury dept and then thrown in jail for keeping records on the fedthugs stalking him.

[Alex]

I agree w/Keith -- the IEM has been up and running for over 15 years (in the old days via telnet) and is the first "outcome market" I remember seeing.

[dude]

That PAM logo: Poindexter seems to have a fetish for masonic triangles/pyramids beaming some sort of info to/from the globe. What's up with that? Is he trying to troll conspiracy theorists or something?

[Alex]

I was thinking "Dark Side of the Moon"....

[Anonymous]

So when Uday says, "This time I think the Americans are serious. Bush is not like Clinton. I think this is the end." That means it's time to sell Baath, or you'll end up taking one.

http://www.washtimes.com/world/20030727-122655-7094r.htm

[Alex]

Ugh -- CNN reporting the project is scrapped...

[Brad S]

To anon on 7/29/03 11:49 AM

As a possessor of an MBA from one of the top business schools in the world, and as someone who deals with multiple markets every single day, I understand how markets work.

Markets can be moved by an almost infinite variety of factors, many of them measurable and quantifiable, and many of them virtually random. If you show two educated, rational investors the same set of market results, you may get two very different interpretations, and two very different sets of attributed causes.

The same is true with this market. Your post implies that DARPA will be able to determine with 100% accuracy what drives any blip or anomaly in this market. By your logic, DARPA will be able to say, "Well, the market is up for October terror, but that's simply because the government has been feeding the investing public warnings about terror. However, the market is also up for December terror, and that's because true terrorists (or those close to terrorists who would have "inside" information) know something and are acting on it. Really, how does anyone know? Is the market telling us something real, or is it simly crying wolf again? Of course, I guess one could argue that our goverment cries wolf with regards to terrorism quite often anyway.

Bottom line - the problem with this market is that it puts 100% faith in our federal government to accurately interpret and gauge the threat factor inherent in every piece of intelligence information it receives - which is essentially what doomed us to the events of 9/11 in the first place. So although there may not be any real problem with this type of market, I'm also not sure how much it changes our current situation.