Learning or Networking? The Causal Effect of Judges Sitting by Designation

Chang‐Ching Lin & Yun-chien Chang

Journal of Legal Studies2025https://doi.org/10.1086/734721article
AJG 3
Weight
0.37

What the paper says

When lower court judges sit on panels with higher court judges, the former may learn from the latter, enter into personal relationships with the latter, or both. Lemley and Miller’s empirical study on US federal courts found that judges sitting by designation see a subsequent reduction in their reversal rates. Leveraging a unique dataset encompassing close to 1 million cases from Taiwan, in particular 2,591 appeals, our study examines whether Taiwan’s transitory promotion system has led to lower reversal rates and, if so, why. Using a difference-in-differences framework, we find that when cases appealed to appellate courts were assigned to judges who were formerly colleagues of the judges who penned the district court opinions, reversal rates drop statistically significantly (demonstrating the “networking effect”). Without a causal identification design, we show that the overall reversal rates are lower in the period after transitory promotions (consistent with the “learning effect”).

1 citation

Open paper page →

Cite this paper

https://doi.org/https://doi.org/10.1086/734721

Or copy a formatted citation

@article{chang‐ching2025,
  title        = {{Learning or Networking? The Causal Effect of Judges Sitting by Designation}},
  author       = {Chang‐Ching Lin & Yun-chien Chang},
  journal      = {Journal of Legal Studies},
  year         = {2025},
  doi          = {https://doi.org/https://doi.org/10.1086/734721},
}

Paste directly into BibTeX, Zotero, or your reference manager.

Flag this paper

Learning or Networking? The Causal Effect of Judges Sitting by Designation

Flags are reviewed by the Arbiter methodology team within 5 business days.


Evidence weight

0.37

Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40

F · citation impact0.16 × 0.4 = 0.06
M · momentum0.53 × 0.15 = 0.08
V · venue signal0.50 × 0.05 = 0.03
R · text relevance †0.50 × 0.4 = 0.20

† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.