Justice of the Beholder
Partisan Bias in Juror Decision-Making
Jesse Rhodes, Tatishe Nteta & Douglas Rice
Journal of Law & Empirical Analysis, forthcoming
Abstract:
How does partisan identity shape perceptions of guilt? In this paper, we examine whether a hypothetical defendant’s perceived political party identification influences jurors’ beliefs about guilt. Among both Democrats and Republicans, we find a striking pattern of bias in favor of copartisan defendants and against out-partisan defendants, with the greatest effects among those that are most affectively polarized. In explaining their decisions, both Democrat and Republican respondents utilize similar themes, but employ them differently by the perceived partisanship of the defendant. Our results shed new light on the significance of partisan bias in American society, demonstrating how partisan allegiances distort jurors’ evaluations of guilt and innocence.
LLMs Are Bad Judges. So Use a Classifier Instead.
Jack Kieffaber et al.
Harvard Working Paper, July 2025
Abstract:
Large Language Models suffer from prompt variance — meaning they’ll give you totally different legal answers depending on how you phrase your question. Jonathan Choi demonstrated this recently when he asked ChatGPT five legal questions, each rephrased 2,000 times, and watched as the bot spat out different answers every time. When you tell somebody that AI is going to replace the judge, the lawyer, and the legal system in the next twenty years, Choi’s article has become the go-to rebuttal; it’s the crown jewel of the “AI bad” genre. Choi’s absolutely right that LLM’s are bad judges. And, if every AI was an LLM, your biglaw job would be safe. But there’s this other type of AI — it’s a classifier, and we built one called Arbitrus. We put it through a mini-Choi test and it mopped the floor with the competition, delivering perfect consistency across all prompt variants with zero hallucinations. We’re going to tell you how it works, why it’s better than an LLM and, ultimately, why it’s better than you. And we’ll close with the frank assertion that a judge’s highest telos — in any legal system — is amendable consistency. That means the Choi test, simple though it may seem, isn’t just a “gotcha" for chatbots; it's the defining test of judicial qualification.
Examining the Effects of New York’s Bail Law on Pretrial Recidivism: A Controlled-Interrupted Time Series Analysis
Stephen Koppel & René Ropac
Criminal Justice Policy Review, forthcoming
Abstract:
In January 2020, New York implemented a bail reform law restricting judges’ discretion to set money bail for certain offense types. We used a controlled-interrupted time series (CITS) design to estimate the reform’s impact on pretrial recidivism. Leveraging the reform’s offense-based eligibility criteria, defendants were separated into treatment and control groups. By comparing recidivism pre- and post-reform and between the treatment and control groups, we were able to minimize confounding from coinciding changes. We evaluated the new law’s effect on multiple recidivism measures: any re-arrest, felony re-arrest, and violent felony re-arrest. In addition, we conducted subgroup analyses for high-risk defendants with recent criminal history. We found a statistically significant increase in violent felony re-arrests among the subgroup of individuals with recent criminal history.
How Court Mitigation of Jail Overcrowding Affects Homicides
Richard Boylan
Journal of Law and Economics, August 2025, Pages 561-584
Abstract:
I analyze court orders issued between 1970 and 1988 that were intended to address jail overcrowding. I find that these court orders led to a 21 percent reduction in jail populations but resulted in a 15 percent increase in homicide rates outside of jails. Recent scholarly work suggests that jail incarceration rates could have been reduced without significantly impacting crime. Thus, my findings suggest that court orders may be an ineffective method for reforming jails.
In-court identifications affect juror decisions despite being unreliable
Jacqueline Katzman, Elaina Welch & Margaret Bull Kovera
Law and Human Behavior, August 2025, Pages 376-386
Method: Participants (N = 422 following exclusions) watched a mock criminal trial that varied the nature of the out-of-court identification (none, poor prior lineup, good prior lineup) and whether the eyewitness identified the defendant during trial (present, not present).
Results: Both in-court and out-of-court identifications independently affected verdicts, irrespective of whether the out-of-court identification was good or poor. In-court identifications, despite having little to no evidentiary value, increased the likelihood that witnesses rendered guilty verdicts. In contrast, participants were sensitive to variations in the quality of the out-of-court procedure; participants who heard evidence about an identification obtained through a suggestive out-of-court lineup rated the prosecution’s case as weaker and the identification as less fair than participants who heard evidence about an identification obtained through a nonsuggestive out-of-court lineup.
Measuring How Much Judges Matter for Case Outcomes
Ryan Copus & Ryan Hübert
Journal of Law and Courts, forthcoming
Abstract:
A large empirical literature examines how judges’ traits affect how cases get resolved. This literature has led many to conclude that judges matter for case outcomes. But how much do they matter? Existing empirical findings understate the true extent of judicial influence over case outcomes since standard estimation techniques hide some disagreement among judges. We devise a machine learning method to reveal additional sources of disagreement. Applying this method to the Ninth Circuit, we estimate that at least 38% of cases could be decided differently based solely on the panel they were assigned to.
Settlementality
Jessica Bregant, Jennifer Robbennolt & Verity Winship
Journal of Empirical Legal Studies, forthcoming
Abstract:
Is settlement of civil litigation good or bad for the legal system? Should we have more or less of it? Should settlement have a private or public purpose? Legal scholars and lawyers have considered these questions for decades, but very little is known about the views of ordinary people. This Article reports results from a nationally representative survey of over 1000 U.S. adults about what respondents thought settlement should look like. Respondents indicated, for example, the extent to which they agreed or disagreed with statements like these: “A settlement between two parties is nobody's business but their own.” “Settling parties are more interested in money than justice.” Our findings convey nuanced lay intuitions about civil settlement, including mixed views about secrecy and settlements, and a tendency to want settling defendants to admit fault and apologize. We also discover that a striking number of lay observers prefer private resolution through contract to oversight by judges. Indeed, a considerable majority thought that at least 75% of disputes would be resolved by settlement in an ideal world, with 100% being the most common choice. The view that more cases should settle was both striking and complicated, highlighting the tension between private resolution of disputes and the public function of litigation and courts.
Do Representative Payments Matter? An Empirical Study
Brian Fitzpatrick & Colton Cronin
Journal of Empirical Legal Studies, forthcoming
Abstract:
For many decades, courts have awarded the representative plaintiffs who bring class actions an extra payment when the actions recover something for the class. It has long been thought that the payments are necessary to induce a class member to step forward and serve as a representative, and, without them, many class actions would go away. Indeed, we show that the payments had become all but ubiquitous in non-securities class actions. In 2020, however, the United States Court of Appeals for the Eleventh Circuit became the only circuit to hold the payments unlawful. We test whether the Eleventh Circuit's decision caused a decline in class action filings there relative to other circuits. Much to our surprise, we did not find good evidence of a decline. We surmise that class members may be willing to serve as representatives without the extra compensation for selfless or principled reasons. Although, if true, this might not entirely defeat the case for representative payments, it certainly complicates that case because some believe the payments are not costless. As a result, our analysis may give courts and commentators additional reason to rethink their support for prevailing practices.
Keeping Invention Confidential
Colleen Cunningham & Aldona Kapacinskaite
Management Science, forthcoming
Abstract:
This study investigates the use of a prevalent but rarely studied form of intellectual property protection: trade secrecy. Building on existing survey evidence of firm-level, cross-sectional use of secrecy, we document the effect of stronger legal protections for trade secrets on the project-level use of such secrets. Our setting is the U.S. oil and gas hydraulic fracturing industry, from 2014 to 2018, in states where firms are required to disclose fracturing fluid ingredients to regulators except for substantiated claims of trade secrets. We examine how the enactment of the federal 2016 Defend Trade Secrets Act (DTSA) affects well-level trade secret use across states with varying levels of pre-DTSA protection. We find substantial increases in the use and novelty of trade secrets. Further, wells with trade secret ingredients are, on average, more productive. However, the DTSA exerts limited additional effect on trade secret–related productivity. Supplementary tests address alternative explanations, show no evidence of intellectual property substitution, and provide additional support that we are capturing policy effects. Our results provide rare empirical evidence on actual trade secret use and enhance our understanding of how appropriability shapes use of trade secrets and associated inventive activity.