Turnover and Accountability of Appointed and Elected Judges

Abstract

Each year, more than 90 percent of civil and felony crime cases in the United States are handled by state court judges. This paper investigates two different systems that are used to select and retain these judges. Under one system, when there is an open seat on the bench, the governor appoints a new judge; when the term of the judge expires, he faces an up-or-down (i.e., yes-or-no) majority decision by voters, without facing a challenger. Under the other system, judges are selected and re-elected through competitive elections. We analyze judges' reelection and its effect on their behavior under the two systems, using data from the State of Kansas. In Kansas, both systems are used to select and retain state district court judges. The data shows high fluctuation in reelection rates of elected judges, while it shows no failure of appointed judges in the reelection process. This difference raises three questions that are crucial to understanding these systems: (i) How does the reelection concern affect judges' behavior under the two systems? (ii) To what extent do reelection outcomes of elected judges depend on other factors such as judges' age and tenure or voters' party preference? (iii) Are there any differences in the types of judges selected under the two systems? The goal of this study is to answer these three questions. Our findings are as follows. First, the sentencing behavior of elected judges is an important determinant of their reelection. Hence, elected judges face serious reelection incentives on their behavior, which appointed judges do not face. However, the extent and the direction of the effect are substantially different depending on the political orientation of their constituency. Second, party affiliation and political climate during an election significantly affect the reelection probability of the elected judges. Lastly, our estimates show that appointed judges are more homogeneous than elected judges in terms of their sentencing preferences.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 15 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.