What Do You Do With a Drunken (and Drugged) Jury?

by robekulick

So my midterms are done! Time to start blogging again. And thanks to Liz, I have a very good one to start this academic quarter with. Liz is in law school and while studying she came across the following case.  Rather than summarize what happened, I’ll just let this summary pulled from the case speak for itself:

“In the interview Hardy stated that he “felt like … the jury was on one big party.” Id., at 209. Hardy indicated that seven of the jurors drank alcohol during the noon recess. Four jurors, including Hardy, consumed between them “a pitcher to three pitchers” of beer during various recesses. Id., at 212. Of the three other jurors who were alleged to have consumed alcohol, Hardy stated that on several occasions he observed two jurors having one or two mixed drinks during the lunch recess, and one other juror, who was also the foreperson, having a liter of wine on each of three occasions. Id., at 213-215. Juror Hardy also stated that he and three other jurors smoked marijuana quite regularly during the trial. Id., at 216-223. Moreover, Hardy stated that during the trial he observed one juror ingest cocaine five times and another*116juror ingest cocaine two or three times. Id., at 227. One juror sold a quarter pound of marijuana to another juror during the trial, and took marijuana, cocaine, and drug paraphernalia into the courthouse. Id., at 234-235. Hardy noted that some of the jurors were falling asleep during the trial, and that one of the jurors described himself to Hardy as “flying.” Id., at 229. “

So what was the conclusion of all of this…the Jury verdict was upheld! In other words, unlike the drunken sailor in the song who suffers all manner of abuse, the Supreme Court decided that basically we shouldn’t do anything about a drunken jury.

For an economist studying the legal system, this case and a zillion other examples begs the question, is the current jury system the optimal method for reaching legal decisions or would a system relying on, for instance, professional juries be better?

There are really two issues to consider when evaluating the jury system from an economic perspective. The first I’ll call the efficiency of the jury system, which corresponds to the question of on average does the jury get decisions right. The second I’ll call the variance of the jury system, which corresponds to how fair the jury system is.

How are these different? Consider two jury systems, one that always makes the right decision and another that has an equal likelihood of over-punishing versus under-punishing. If the second jury system still gets the average decision right, then it is as efficient as the first system, but it subject to higher variance and is less fair. Of course one could also create an example of a jury that is inefficient (gets decisions wrong on average) but has a low-variance (is relatively even-handed about sentences are assigned.

Research on decisions reached by groups suggests that on average, groups tend towards the correct answer. A famous, and simple example, is the oft used example of guessing the weight of a cow at a country fair. Overall, there is high variance in people’s individual predictions about the cow’s weight, but it if you average the decisions, there is a remarkable propensity for the average of the guesses to be very close to the cows actual weight. Of course it would take real research to confirm that this is true of juries, but given that this phenomenon applies to a wide a variety of much more complicated examples (it’s the operating theory behind Intrade) it is quite possible that that the jury system is efficient.

However, what examples like this and a million others  suggest is that there is a large potential of variance in jury decisions due to the variance in the make-up of the juries. Of course it is possible that drunk juries perform just as well as sober juries, but I think most people would assume that this is not the case. I think there is a strong case to be made that a professional jury system would be subject to less variance.  But there is a catch at this point. The theory of groups tends to suggest that the more rarefied the group, the less good it is as reaching the correct decision. If for instance, juries are composed of Professors at Ivy League Universities, juries will then exhibit the biases the Ivy League professors tend to have. So despite the fact that examples like this are sort of mortifying when you consider that your legal fate is sometimes in the hands of people you wouldn’t trust to take your order at McDonald’s, it’s unclear whether overall we would be better off with professional juries. I still haven’t really decided how I feel about this issue.

Oh, and one more question to raise on a separate note. If these people were doing cocaine, shouldn’t that have kept them awake during the trial?