The program was derided by Democrats and Republicans in Congress, some of whom called it “bizarre,” “unbelievably stupid” and “offensive.” Rumsfeld himself said he canceled the program “an hour after I read about it.” ( Wired ,July 2003)
Commonly, government programs engender partisanship and opportunism. Political actors are more successful to capitalize on such initiatives are controversial. This effect is only magnified when the program is headed by a polarizing figure. One prevalent example of this was DARPA (Defense Advanced Research Projects Agency) used Prediction Markets to gather intelligence on future geopolitical events. Once more contentious questions such as terrorist attacks and assassination attempts ended up being addressed, the program began to be publicly criticized.
PAM (Policy Analysis Market) implemented by the Information Awareness Office, a counter-terrorism project ran by DARPA. PAM operated like a future exchanges market for predicting the likelihood of geopolitical events. Including but not limited to terrorist attacks. At various phases of the program, participants (consultation firms, colleges, think tanks) were provided a sum of money to “wager” on the likelihood of certain political events happening (p.77). Those with accurate answers were awarded a larger sum of money.
E.g.) Phase I: Participants were provided $100,000 by the IAO to wager and awarded $750,000 for accurate predictions (p.77).
Mirroring the model used in both past and future prediction markets. Dating back to Robin Hanson first pioneering prediction markets while consulting on project Xanadu in the late-1980s, these markets have always been an incentive-driven phenomenon. It is one thing to claim certainty, but it is another to be willing to die on that hill. Especially when money is on the line. Effectively aligning incentives towards accuracy and rigorous research versus armchair speculation. The objective being the firm, organization, or government department hosting the market with aggregate a large cache of quality information (p.76).In the field of counter-terrorism having averaging consensus from a variety of sources is crucial to avoid engaging the wrong target. Such mistakes will incur costs much greater than monetary losses.
As groundbreaking and innovative as PAM was invariably the program garnered some criticism that eventually devolved into outright censure. Academics and bureaucrats “betting” on the aptitude of terrorist activities and political revolts transpiring may be unsettling from a prima facie standpoint. Particularly if taken at face value with no further analysis. Arguably the criticism of PAM intensified due to the IAO’s controversial director, John Poindexter. Poindexter rose to infamy from his involvement in the Iran-Contra scandal of the Regan administration. Even though all of the insiders of the project acknowledged that Poindexter had little involvement in PAM (P. 6, footnote 7), most of the backlash was directed at him. The fury of pundits, media outlets, and the general public caused Poindexter to resign in the summer of 2003. Leaving the PAM project permanently defunct.
The advocacy and opposition to the implementation of PAM as a means of aggregating intelligence on sensitive matters is no doubt a complex maze of ethical and pragmatic arguments. The use of prediction markets for gathering information for defense planning is just like another government policy, the impact is not neutral. Meaning that keeping or eliminating the program will create disparate consequences. Typically favoring one subset of economic agents over another. Individuals will bear the “expected costs” (p.38) imposed by the impact of the policy. For example, a government program may create jobs for individuals that are politically connected. However, this is generally at the expense of the taxpayer. Vice versa, abolishing a program will eliminate jobs for the clerks and managers operating the department. The impact of policy always affects some individuals positively and others negatively. All political policies involve the transfer of benefits from one party to another.
Considering the non-neutral nature of policy, it would be justifiable to apply Bruce Yandle’s concept of Bootleggers and Baptists to the political pressure to abandon the PAM program. Yes, there were some ethical concerns regarding the prospect of having people “wager” on terrorist attacks. It would be naïve to believe that all the opprobrium was motivated by morality. Much how skilled consultants can profit from participating in a Prediction Market, many actors can also do so by dismantling such a program. Beneficiaries ranging from media outlets to opportunistic politicians. The political opportunism was multilayered including enemies of the Bush administration, the Republican Party, and even direct adversaries of John Poindexter. Proving an opportunity for democrats to temporarily shed their anti-patriotic veneer, to admonish these “conservatives” for making light of national security threats. Yet, the credulous public seldomly questions this moral browbeating. On the surface, these criticisms sound valid. Since when have politicians previously disinterested in national security matters are suddenly deeply invested in the integrity of defense intelligence? As Machiavelli pointed out in The Prince appearances are more important than actual principles in politics (p.42).
It is exceedingly difficult to designate one side of the coalition as a pure Baptist in the public outrage campaign surrounding PAM. The self-interest of the media, politicians, resident experts within the government is glaringly obvious. The potential for Dual-Role Actors (economic agents that benefit materially, but simultaneously sincerely believe the moral argument) in this coalition dynamic exists. However, is muddied by the perverse incentives to use strawman, ad hominem, and other logical fallacies to denigrate the program. The adversaries of PAM had a lot to gain through defaming the program. Not a whole lot of utility to extract from testing the validity of the results. Since the average constituent is not going to care too much about the granular details of the program. Rather be fixated on their visceral reaction to the ethical considerations of “betting” terrorist attacks.
Regardless, of whether moral advocacy is misguided or ill-informed, nevertheless, it is still a normative position. The average citizen happens to be the proverbial Baptist in this coalition dynamic. Any expression of disgust or moral indignation was sincere with little to no observable benefit from ending the program (dispersed costs, concentrated benefits). Even if the public’s concern was stoked by the slanted framing of the program, it still does not lessen make their concerns any less earnest. In the absence of further context, a group of contractors and academics participating in a gambling pool predicting terrorist attacks does sound grotesque. Since gambling is considered a form of entertainment appears to trivialize the severity of contentious situations that could result in the loss of lives. For the honest concern for these moral considerations, the average voter is our Baptist.
One great irony was that one of the academics deeply involved in the project narrowed down the reasonable ethical concerns in a peer-reviewed paper years after PAM had been dismantled. It was none other than prediction markets pioneer Robin Hanson. Hanson citing the following as prevalent concerns of the program:
- “…The first concern expressed—that of replacing professionals with amateurs..” (p.82)
- “…The second fear expressed was that bad guys would be willing to make losing
trades to mislead us..” (p.82).
- “..The third main fear expressed was that bad guys might be rewarded for doing bad things..” (p.83). E.g.) Al-Qaeda’s meddling with airline stocks in the 9/11 attacks.
Hanson tactfully addresses all these concerns explaining how much of these concerns are the result of misconception. Like how the media coverage of the program generated several misconceptions regarding the function and purpose of PAM.
Several various individuals and groups stand to benefit from a sensationalized portrayal of the PAM program. One of the more salient examples would be the media. Media outlets are a business much like another, the incentive is to maximize profits. Logically this premise is cogent to anyone with even a small amount of exposure to economics. This controversy emerged in the primordial era of social media (Myspace being founded in 2003). The internet did exist but did not present any true competition to televised and print news media. For media outlets to have a story as jarring as the government funding a macabre gambling bracket trivializing serious events, instant goldmine. That is the type of story that sells publications. It has all the elements of a good conspiratorial techno-thriller. One only needs to consider the success of Tom Clancy to know how stories of geopolitical/government intrigue are lucrative. It could be argued that the media is merely the messenger, if they happen to profit from the event, it is a natural consequence of the event. How the information is presented and sways public opinion. If news reports are worded in a manner that is hostile towards the program, this will influence public opinion. Creating a feedback loop, inciting the ire of the Baptists while concurrently profiting. This would be an excellent example of the Bootleggers tacitly inciting the indignation of the Baptists.
Another subset of Bootleggers would be the politicians who spoke out against PAM. A book could be written about the political motives guiding the strategy condemnation of the program by various politicians. As previously mentioned, the layers of political opposition operate on a continuum of scale. Varying from individual grudges, contention between political factions, and even opposition to the sitting president at the time (George W. Bush). Despite the complexities of various political considerations, speaking out publicly about a controversial government program fosters a positive public image. Especially for politicians who were affiliated with the Democratic party. During the Bush administration, Democrats were perceived as being soft on terrorism. At a time where terrorism was a hot-button issue, speaking out against counter-terrorism measures was tantamount to political suicide. The whole PAM debacle presented an opportunity for a clean slate. An opportunity to capitalize on a misstep made by the Bush administration and to feed into the fears of the public. Paralleling the Bootlegger –Baptist feedback mechanism generated by the media. See below for a shining example of such sanctimonious posturing:
For instance,” Mr. Wyden said, ”you may think early on that Prime Minister X is going to be assassinated. So you buy the futures contracts for 5 cents each. As more people begin to think the person’s going to be assassinated, the cost of the contract could go up, to 50 cents.
‘The payoff, if he’s assassinated, is $1 per future. So if it comes to pass, and those who bought at 5 cents make 95 cents. Those who bought at 50 cents make 50 cents.’ (Senator Ron Wyden (D), NYT July 2003).