"We can't say if more severe sentencing is better for society or worse, but our findings show us how political pressure can distort the sentencing process and can lead to starkly different sentences for similar criminals sentenced at different times," says Noam Yuchtman, assistant professor of business and public policy at the Haas School of Business.

Yuchtman and Carlos Berdejó, associate professor of law, Loyola Law School, are co-authors of "Crime, Punishment, and Politics: Analysis of Political Cycles in Criminal Sentencing," forthcoming in the Review of Economics and Statistics.

The study examined the felony sentences of 265 full-time Superior Court judges between July 1995 and December 2006 in the state of Washington, covering three elections in 1996, 2000, and 2004. The authors focused on the most high-profile crimes such as murder, assault, rape, and robbery, which represent 6.7 percent of 18,447 sentences conferred. Case-specific controls included the defendant's age, gender, race, and prior criminal history, as well as an indicator of whether the sentence resulted from a plea agreement. The study also accounted for a number of potentially confounding variables such as changes in attorney behavior, case re-assignment, political cycles of other officials, and seasonal variations; for example, if more homicide sentencing hearings than usual happened to occur right before an election.

Yuchtman and Berdejó find that sentence lengths increase at the end of judges' political cycles, then sharply fall when their next term begins, only to rise again as their next election approaches.

Importantly, they do not find an increase in sentencing severity at the end of terms of judges who are not seeking re-election. Judges only increased the severity of their sentences at the end of a political cycle when they were facing re-election.

The findings also indicated that toward the end of their terms, judges tend to become more calculated in making their sentencing decisions, deviating from normal sentencing guidelines 50 percent more often at the end of the electoral cycle compared with the beginning. These deviations account for a large fraction of the harsher sentencing, suggesting that the influence of politics on sentencing crucially depends on the discretion judges have in sentencing.

History reveals that most judges are re-elected and don't even face competition at the polls. These findings from the state of Washington suggest that just the threat of political competition can affect behavior. "Judges may fear that a lenient sentence for a violent criminal might be turned into a political opportunity for an ambitious prosecutor seeking a harsher sentence," says Yuchtman.

Yuchtman and Berdejó's study helps to inform the debate on whether judges should be elected or appointed. The authors say while they cannot predict whether society would benefit from appointed-only judges across all jurisdictions, their results conclusively determine that sentencing patterns would differ.

Yuchtman says, "When you tell people in other countries that some American judges are elected, they are often shocked. Maybe they're right: we don't like to think of judges as being influenced by external pressure. On the other hand, our results suggest that elections do make judges feel accountable. This is a simple, but important, tradeoff."

Note: If no author is given, the source is cited instead.

• Political Science
• Justice
• Conflict
• Legal Issues
• Educational Policy
• Bioethics