Summary
Judging on collegial courts is a collective enterprise. Practicing judges never tire emphasizing the vital role of deliberations with their peers in their decision-making process. Research in social psychology and behavioral economics has documented a multitude of factors that affect the quality of deliberation. However, the vast majority of scholarly work on judicial behavior focuses on how legal, ideological, and strategic concerns influence the choices of individual judges.

This dissertation analyzes judging as a group process. Across three articles, I examine the extent of disparities between groups of judges and identify characteristics that shape the collegial decision-making. The dissertation relies on a new dataset documenting 21,163 decisions on criminal law appeals at the German Federal Court between 1990 and 2016. It combines data on decisions with auxiliary information on the judges’ biographies, the workload of lower courts, and panel attributes.

In the first article, I derive a new estimator quantifying inconsistency from observational data. Inconsistency in judgments is a critical indicator of a fair and just legal system. If judges exchange information and opinions on cases productively and efficiently, different groups should arrive at similar conclusions when presented with the same case facts. The results reveal that more than one out of six decisions would have been decided differently if assigned to another panel.

The following article identifies the familiarity of judges with each other as a crucial determinant of the deliberation process. I demonstrate that the average shared experience of judges in the group increases the likelihood of dissenting opinions. Judges with little previous time on the court and judges that replace colleagues temporarily are more likely to conform to the other judges’ opinions. Finally, I explore the effect of status and social hierarchy on the group’s decision-making. I find that chief judges exhibit leadership that makes them almost twice as powerful as associate judges. Examining heterogeneous effects, I show that chief judges influence the outcome of judicial deliberation in different directions but have a quite homogeneous positive effect on the probability that judges schedule a main hearing.

The results have several implications for research on judicial behavior and policy-makers concerned with the quality of adjudication. First, the dissertation’s focus on the group-level offers a fresh perspective on judicial decision-making that takes seriously the criticism that judges and legal scholars have expressed towards conventional explanations of judicial behavior in the social sciences. Second, the large dataset serves as a valuable resource for future research that aims to analyze judicial behavior in civil law countries or for increasingly popular comparative studies. The dissertation also provides concrete policy advice to improve judicial decision-making. The empirical analyses identify case characteristics associated with high inconsistency, which allows policy-makers to apply targeted counter-measures. Finally, the findings suggest that group composition on collegial courts should aim to increase the familiarity of judges with each other.