Appointed judges more effective, U. research finds

In their article “To elect or to appoint? Bias, information and responsiveness of bureaucrats and politicians,” which was published in the January volume of the Journal of Public Economics, economics professor Matias Iaryczower, graduate student Garrett Lewis and Caltech professor Matthew Shum analyzed almost 6,000 state supreme court rulings on criminal cases from 1995 to 1998. The study did not analyze judicial rulings on civil cases.

According to the study’s authors, appointed justices possessed better information about the cases over which they were presiding, were more likely to change their preconceived opinions about a case and made fewer mistakes in their rulings than elected justices.

The findings could illuminate the debate over the respective merits of appointed and elected judges. University of Toronto professor of economics Yosh Halberstam said the study’s findings also serve as a case study through which to understand whether certain jobs are better performed by elected politicians or by appointed bureaucrats.

According to Lewis, one might expect elected justices to be more moderate, since they will feel pressure to present positions that can appeal to the maximum number of voters, while one might think that judges who are appointed by partisan officials would be more likely to have some initial bias. But in actuality, appointed justices are more willing to gather information about cases and to prevent their personal biases from influencing their rulings, he said.

To measure each justice’s effectiveness, the researchers considered both the justices’ biases and the quality of their information, or their ability to make decisions that the researchers determined to be correct under the law given the facts of the case. They found that appointed justices had a 39 percent higher quality of information than elected justices.

They then assigned each justice a FLEX score, which is the probability that a justice would have voted differently in the absence of case-specific information. They found that appointed justices were 23 percent more likely than elected justices to vote differently in the absence of case-specific information — in other words, appointed justices were more likely to take case-specific information into account. Finally, they estimated how often each justice makes incorrect decisions, using the votes of all the justices hearing a certain case and the characteristics of the case to assign each case a probability of being wrongly decided under the law. They found that elected judges were more likely to make incorrect decisions.

“[Elected] justices are not necessarily the great legal minds, they’re the great political minds, the ones that can get elected in a popular election,” Lewis said, contrasting elected judges with appointed judges.

In the study, the researchers wrote that, though elections are designed to hold the justices accountable, elected justices often “pander” to the public through their rulings.

“The basic thing that we do see is that elected officials, including politicians and judges as well, are responsive to the voters,” Iaryczower said. “There is a lot of theory indicating that sometimes elections induce incentives for politicians to actually make wrong choices, to actually throw away information, to act in a way that makes them look as if they were more competent or as if they were more in line with the electorate, but without actually taking full consideration of the impact of their actions on the policy.”

One of the study’s most innovative features is the model researchers used to evaluate rulings made by Supreme Court justices, Halberstam said. The model incorporates the quality of justices’ information about cases, justices’ tendencies to vote on the basis of case-specific information versus their personal biases and the probability that justices reach an incorrect decision.

“The basic idea [of the model] is that with a really good justice, there’s going to be very little variation in the decisions he or she renders,” Lewis explained. “Given certain characteristics in the case, he or she is going to consistently come down on the same side of the case, whereas a lower-quality justice will be more all over the place.”

“It is a great first step in the right direction,” Halberstam said of the study. He added that political scientists could extend the model developed by Iaryczower, Lewis and Shum by incorporating judicial deliberation into their calculations.