Measuring the gain from labor specialization

We estimate the productivity effects of labor specialization using a judicial environment that offers a quasi-experimental setting well suited to this purpose. Judges in this environment are randomly assigned many different types of cases. This assignment generates random streaks of same-type cases, which creates minispecialization events unrelated to the characteristics of judges or cases. We estimate that when judges receive more cases of a certain type, they become faster, that is, more likely to close cases of that type in any one of the corresponding hearings. Quality, as measured by probability of an appeal, is not negatively affected.