Interim Decision-Making Markets
Predictive decision making can be used to predict not only events that are contingent on a particular decision, but also what a decision might be. We have already seen that prediction markets can forecast decisions (see Chapter 4). The merger analysis example can be viewed as predicting a decision, namely, the ex post measurement of consumer surplus. Often, it might be useful to make a decision based on a prediction of a future decision. For example, sometimes the legal system must establish an interim policy pending a final decision concerning a particular legal issue. In these cases, it often, though not always, will make sense to have the interim policy be the expected final policy. In this case, a prediction market can be used to forecast the final policy decision, and that prediction can serve as the decision during the interim period. Consider, for example, approval of pharmaceutical drugs by the Food and Drug Administration. Under the current regime, pharmaceuticals cannot be sold until the FDA approves them,41 and critics complain that the long approval process sometimes leads to unnecessary deaths and hardship.42 A prediction market could be used to gauge whether each drug ultimately will be approved, and provisional sales of the drug could be allowed pending the FDA decision. For example, the rules might provide that a drug that has at least a 90 percent chance of eventual approval can be sold immediately but would need to be pulled off the market if the probability later fell below 50 percent or if an adverse decision were ultimately reached. There are dangers associated with such a regime: patients and others might stockpile a drug, leaving the FDA powerless to keep the drug away from patients later. But where there is a high probability of eventual FDA approval, and for drugs that are unlikely to be stockpiled, this regime nonetheless might improve on the existing one. One benefit is that the FDA would not need to rush formal approval. An alternative, of course, is simply to allow the FDA to decide case by case whether to allow particular drugs to be sold on an interim basis, but it is difficult to make such decisions without full consideration of the merits of the research studies. Interim prediction markets also might be used to help courts decide whether preliminary injunctive relief is appropriate in a case in which a permanent injunction is sought. In the current system, a judge who eventually will determine whether a permanent injunction should be granted makes a preliminary assessment. This assessment is based on factors including the danger of irreparable harm from an absence of preliminary relief and the probability that the party seeking the injunction eventually will prevail on the merits.43 Judges understandably do not want their preliminary injunction decisions to appear lawless, and they thus often consider the factual and legal issues at stake in great detail, despite the preliminary nature of a decision.44 To the extent that a preliminary decision requires almost as much effort as a final decision, it might not achieve substantial cost savings. At the same time, there is a danger that judges will be hesitant to change their minds, except perhaps by relying on issues that they did not consider at the preliminary injunction stage, for fear of appearing inconsistent. Using prediction markets to forecast whether a permanent injunction will be issued, on condition that the court ultimately resolves the case, could greatly simplify a court’s task. A court would still have to consider issues such as irreparable harm, but in many cases that is far more straightforward. An additional benefit is that prediction markets can take into account not only what will happen at the trial court but also whether that decision will withstand appeal. This would limit the need for appeals of both the preliminary injunction and the eventual decision on the permanent injunction. Prediction markets also might be used in the criminal context to determine whether a particular defendant should be allowed to remain free pending a verdict. For example, an estimate of the probability that the defendant will be convicted and sentenced to prison might be deemed relevant to a determination of whether the defendant should be allowed to remain free pending trial. Naturally, there are other relevant criteria, such as the risk of flight, and the existing bail system already functions as a kind of predictive decision making regime, with bail bondsmen serving both as predictors and as pursuers when a defendant fails to appear.45 Moreover, a prediction market might be useful in determining whether a defendant can stay free pending an appeal. In our current system, a defendant’s recourse is to the trial court judge, and relying on a prediction market would avoid having a trial court judge assess the probability that his own decision will be reversed on appeal. Prediction markets might help forecast not only executive or judicial decisions but also legislative decisions. Consider, for example, the problem of states that frequently miss the deadline to complete their annual budgets, leading to predictable disruptions in public services.46 The legislature could provide that in the absence of agreement, all programs should continue to receive funding at existing levels, but this would provide a bias to the status quo. An alternative approach would be to use prediction markets to assess whether programs will receive funding and to fund them tentatively at the predicted rates. Market participants would have incentives to assess the preferences of existing legislators and monitor decision making concerning specific issues before final decisions are reached. Naturally, this proposal and the others would need considerable refining before they could be implemented. The point, however, is that prediction markets might provide an approach to interim decision making that does not require the ultimate decision making body actually to make decisions, thus overcoming a major problem with interim regimes.
Leave a Reply