The Policy Analysis Market: Using Economics as Counterterrorism

By Christopher Comas

The silver lining to the various internal crises plaguing our nation is that the threat of terrorism, once at the forefront of America’s collective consciousness, has faded into the background.

No comments

By Christopher Comas

The silver lining to the various internal crises plaguing our nation is that the threat of terrorism, once at the forefront of America’s collective consciousness, has faded into the background. Our once salient fear of terrorist crises has been upstaged by a politically fractured nation, wounded by its own ancient gun laws, and troubled by an epidemic of sexual misconduct. We sometimes forget that America and the world are positioned against overwhelmingly malevolent and influential international terrorism elements that have led to unforeseen attacks in Boston, San Bernardino and Orlando. Other countries are grappling with a continent-wide network of extremist influence and the threat of potential attacks. Turkey, France and Belgium have faced horrible acts of violence. Perhaps an  article like this which examines one particular counter terrorism measure is too late, even irrelevant. If, like many, you do not think that counter terrorism should be among our primary focuses at the moment– an admittedly fair contention– I hope that I can show you that the Policy Analysis Market, or PAM, is a really cool idea that could be put into practice with a few changes.

PAM was a counter terrorism tool proposed by a subdivision of the Department of Defense (DoD), the Defense Advanced Research Projects Agency (DARPA). DARPA wondered if they could create a marketplace that would help produce knowledge for counterterrorism. To do so, PAM would have traded on terrorism information derivatives; that is, bets on pertinent terror contingencies such as whether or not Saddam Hussein would be caught by 2007. If PAM behaved according to established economic theory, then the trades made on these securitized informations would reveal the dominant and most likely outcome. After all, quality information is the basis of effective counterterrorism.

The theoretical basis of PAM is grounded in the efficient market hypothesis, which holds that existing share prices will always incorporate and reflect relevant information. Proponents of this theory cite the aftermath of the challenger crash in 1986. A panel was convened in congress after the disaster to determine what mechanical parts had failed in the course of the launch, and which of NASA’s 4 major contractors would be held responsible. The panel took months to reach a conclusion, but market trades in the initial days after the crash accurately determined the responsible firm by targetedly devaluing its equity.

The efficient market theory holds that market foresight is borne of agglomeration not expertise. While many individual members of the market might be completely misinformed or misguided, the aggregate of their knowledge tends towards accuracy. This phenomenon is known as the dumb agent theory. Where human judgement is fallible in isolation, markets aggregate the collective knowledge of droves of self-interested traders.

The idea of PAM was to make terrorism related events and relevant regional statuses securities. For example, the recognition of Palestine  as an independent state in a particular span of time, the likelihood of a regime change in Syria, or data indices that track economic health, military disposition etc. DARPA’s Policy Analysis Market (PAM) would have offered futures contracts for such contingencies, which are essentially wagers on the status of these events.

The Government would have acted as the market maker, a kind of middleman who holds futures contracts in the absence of an immediate buyer, thus enabling on-demand trading. Uncle Sam would be disposed to incur losses in this capacity as the market shifted to new equilibriums in order to gain information. To sustain these costs PAM would have required a steady stream of subsidies.

DARPA sold PAM as a lightning rod of counterterrorism information, a way to efficiently aggregate knowledge from vetted experts into a theoretically accurate market consensus. DARPA intended to offer trading licenses to 1,000 select experts, eventually expanding participation to a cap of 10,000 participants. How these traders would have been selected is not entirely clear. Financial means were likely a component of the selection criteria, but it’s implied that participants would have some degree of expertise and perhaps ‘special’ knowledge of terrorist related activity. Joseph Stiglitz, a Columbia economics professor, mocked the selection process in an editorial in the Washington Post published days after the program’s demise, asking “shouldn’t these people be investigated rather than rewarded?” In any case, trades in the market were limited to $100 dollars so the potential for outsize profit was minimal.

While PAM would have helped assemble expert opinion from the public and private spheres, it was also intended to serve as an easily accessible gathering point for government agencies. The issue of interagency communication is salient in conversations about counterterrorism; how do the CIA, the DoD and the FBI all circulate the knowledge they’ve individually generated? PAM would have satisfied this concern, providing an easily accessible internal bellwether of the leading perspectives on the current key issues.

In July of 2003, the government’s plans for PAM suddenly collapsed in the span of 24 hours. Senators Byron Dorgan of North Dakota and Ron Wyden of Oregon held a press conference denouncing the market’s ostensibly dubious ethical underpinnings. “Can you imagine,” Dorgan asked, “if another country set up a betting parlor so that people could go in—and is sponsored by the government itself—people could go in and bet on the assassination of an American political figure?” Their invectives launched a barrage of indignation. Critics labelled PAM the ‘terrorist futures market.’ Senate Democrat majority leader Thomas A. Daschle condemned it as a “plan to trade in death.” Skeptics also raised the possibility that terrorists could make loaded trades in the market for profit; the DoD could conceivably be offering guaranteed profit to terrorist organizations by offering them a bet they could always win by attacking or abstaining.

Initial criticisms were largely the work of Democratic Senators looking to burn the Bush administration in effigy. Unfortunately for DARPA, Republican Senators crossed the aisle the very next morning in denouncing the program. The Bush administration then disavowed the proposal almost instantaneously, and DoD cancelled the project one day after it was made public. DARPA’s head, Admiral Poindexter, was forced to resign.

The furor over the ethics of the proposal is largely unfounded in retrospect. Much of the indignation was predicated on blatant misunderstandings and mischaracterizations of PAM’s manifest form. Chief among these fallacies was the claim that DARPA was encouraging ‘betting on death’ by–as critics said–offering contracts on assassinations of political figures and other acts of violence. Such complaints amount to straw man arguments. The market was never intended as a tool tailored to predicting granular acts of terrorism or isolated instances of violence. While betting on the deaths of certain key political figures was at one point part of the conversation concerning the market’s validity, PAM was always conceived primarily as an aggregate of knowledge on key political, economic and regional stability indicators.

It’s also unlikely that terrorists would’ve been able to gain access through DARPA’s screening process. If they did, or if they worked in concert with an approved trader, the potential for arbitrage profits would have been attenuated by the $100 trading cap. Furthermore, terrorists are already free to make trades prior to perpetrating an attack. Many have speculated that terror elements shorted the market prior to 9/11. In sum, critics peered down the slippery slope, but ignored PAM’s cautious stance 10 feet back.

The concerns, like the cancellation, were misguided. Protesters’ scruples were founded in deep-seated misunderstandings. Meanwhile the instrumental merits of the plan were, with a few qualifications, impressively robust. If the DOD made certain technical and rhetorical adjustments to account for some of the practical shortcomings of its initial formulation, PAM would have been a major boon to counterterrorism efforts. With the right improvements, it still could be.

My argument is that a market of this nature would be effective and preclusive of these ethical perils if it were maintained as an internal federal agency instrument, rather than one open to external non-government traders. Writing in the now defunct academic journal Strategic Insights, Economics professor Robert Rooney notes that “assuming DARPA’s ultimate aim is to derive better forecasts of political instability, terrorist attacks and the like, one approach might entail…an internal, interagency version of PAM traded on exclusively by governmental analysts (and perhaps a select outside group of academics, businessmen, and contractors.)” Aggregating opinion and expertise in an online marketplace would effectively cut through bureaucratic barriers. Current wisdom would be discernible in one snapshot of market standing.

Careful and precise rhetoric is also a keystone in the fight to restore PAM. No, there are not futures securitizing the assassination or death of certain political figures abroad. Yes, there are bets on the stability of foreign regions and, yes, this means traders would be betting on potential coup d’état and the like; but, how is this any different than conventional markets falling in anticipation of political instability? Aren’t trades made every day that are predicated on evaluations of the stability of the US government?

Ultimately, PAM in this capacity would be extremely helpful so long as the DoD treated it with a dose of sobriety. The market is not always right–the efficient market hypothesis is a theory. Markets are correct on average (it seems) but not in every individual case. A little bit of veracity wouldn’t hurt anyone.


Share a comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s