The Judge Pool Behind the Venue Effect
The Delaware forum-shopping result becomes a mechanism rather than a correlation only if venue changes the judicial environment that handles the case. This chapter examines that link: how the study measures whether a judge resembles the Delaware pool, why within-court assignment makes judge comparisons informative, and what happens to the venue estimate when judge characteristics are taken into account. It then tests the prediction that firms shop less when their home courts become less congested or more Delaware-like.
A judge-quality proxy that can be compared across courts
A court is not a single decision maker. It is a pool of judges whose experience and approach may differ, even when they operate under the same federal code. To study whether that difference matters, the authors construct a case-specific measure called the Judge DE Gap. It is not a legal finding that one judge is “better” in every sense. It is a proxy for how closely an assigned judge’s observed case outcomes resemble the average observed outcomes of Delaware judges.
The proxy combines four dimensions. The first is the rate at which the judge’s cases are converted from Chapter 11 to Chapter 7. The second is the rate at which unsecured lenders receive a positive recovery. The third is the logarithm of case duration. The fourth is the judge’s experience, measured by the number of cases handled before the focal case. For each dimension, the authors compare a judge’s leave-one-out average with the corresponding Delaware average, take the absolute difference, and combine the four differences using the first component from a principal-component analysis. A higher Judge DE Gap means the judge looks less like the Delaware benchmark; a lower gap means greater “Delawareness.”
The leave-one-out step protects the comparison from a circular result. If the outcome of the case being studied were included in the judge’s own score, a case that lasted longer or converted to Chapter 7 could mechanically make that judge look more or less like Delaware. The authors go further by leaving out cases filed by firms headquartered in the same county as the focal case when calculating the judge’s averages. That helps avoid mechanical correlation when several related filings or subsidiaries appear in the data. The score is therefore designed to describe the judge based on other observed cases, not to smuggle the focal outcome into the explanatory variable.
The construction has a useful strength and an important limit. Its strength is that it measures several dimensions of case handling at once. A judge who resolves cases quickly but converts many of them, for example, is not reduced to a single simplistic label. The four measures also connect the mechanism to the outcomes that matter in the venue debate: liquidation, time, recovery, and experience. The principal-component step supplies a way to summarize the distance from Delaware across those dimensions rather than pretending that any one of them is the whole concept.
Its limit is that the Delaware average is a benchmark, not a complete definition of judicial expertise. The recovery component is noisy, and the measure cannot capture every quality relevant to a firm, a worker, or a creditor. Because the leave-one-out averages vary with the case being evaluated, the same judge can have somewhat different gaps across cases. “Delaware-like” is thus a statistical description of observed outcomes, not a normative ranking of judges or a claim that Delaware’s practices are correct in every dispute.