Do Court-Appointed Attorneys Achieve Better Outcomes for Defendants of Their Own Race?

This study asks whether court-appointed attorneys achieve different outcomes for low-income defendants based on whether they share the same race. Using administrative data on more than 17,000 misdemeanor cases in Travis County, Texas, the authors examine quasi-random attorney assignment to compare results for Black and White defendants. They find that Black defendants represented by White attorneys are 14–16 percent more likely to have their charges dismissed and 15–26 percent less likely to be jailed than those represented by Black attorneys. These effects are not explained by attorney experience, case type, or other observable factors.

Why This Article Was Selected for The Policy Scientist

This article addresses a policy issue of broad importance: how race shapes the functioning of the indigent defense system, a key gateway to the criminal legal process. The topic matters because representation quality influences case outcomes, future labor-market prospects, and long-term system disparities. The study offers a timely contribution as jurisdictions reassess equity in legal institutions. The data are unusually rich administrative records with quasi-random attorney assignment, supporting credible causal inference. The methods are rigorous, and the findings may generalize to similar assigned-counsel systems, though jurisdictional variation remains a limitation.

Full Citation and Link to Article

Mikdash, M., & Oh, S. (2024). The Role of Race in the Legal Representation of Low-Income Defendants. American Economic Journal: Economic Policy. Advance online publication. https://doi.org/10.1257/pol.20240685 

Central Research Question

This study asks whether the racial pairing between indigent defendants and their court-appointed attorneys affects case outcomes in a large urban misdemeanor system. Specifically, the authors examine whether defendants experience different probabilities of dismissal, incarceration, or probation when represented by an attorney of the same race versus one of a different race. Because the assignment of attorneys in Travis County, Texas operates through a rotation (“wheel”) system that creates quasi-random matches, the authors investigate whether this institutional feature reveals systematic and meaningful differences in outcomes that can be attributed to defendant–attorney racial pairings rather than case characteristics, attorney skill, or other confounding factors.

Previous Literature

The article builds on several strands of scholarship examining race and discretion in the criminal legal system. Prior research has identified racial disparities in policing, such as differential use of force, search behavior, and citation patterns. Related work documents racial gaps in prosecutorial decision-making, judicial sentencing, and parole board decisions, though findings are mixed and context-specific. A smaller literature evaluates whether individuals benefit from having decision-makers of their own racial or demographic background, including studies of teachers, physicians, and parole officials. Evidence from those domains often shows positive effects of same-race interactions, such as improved communication or higher perceived legitimacy. By contrast, relatively little is known about race within the indigent defense system, despite its central role—roughly 80 percent of criminal defendants rely on appointed counsel. Research on defense attorney quality has focused more on institutional structures, caseloads, compensation systems, and attorney effort, with limited attention to race. The present study fills this gap by leveraging quasi-random attorney assignment to isolate the causal effect of the defendant–attorney racial match. It also contributes to broader discussions of how frontline actors mediate legal outcomes across racial groups and whether well-documented disparities originate partly from defense-side dynamics rather than solely prosecutorial or judicial choices.

Data

The authors assemble administrative data on more than 17,000 misdemeanor charges in Travis County, Texas, from 2013 to 2022. These records include charge type, defendant demographic characteristics, disposition outcomes, and sentence lengths. The dataset also incorporates attorney assignment information from the county’s wheel system, which records the sequence and timing of attorney appointments. Because Texas does not track attorney race, the authors manually code this information by linking attorney names to websites, professional photos, and State Bar profiles. They restrict the analysis to cases involving Black and White defendants and attorneys to avoid complications in the race coding of Hispanic individuals and to address non-random assignment of Spanish-speaking defense counsel. Additional variables include attorney experience, law school ranking, and monthly caseload, which allow for exploration of quality-based mechanisms. The dataset is unusually detailed for misdemeanor defense, providing granular charge-level outcomes and enabling the authors to measure attorney effort using proxies such as motions filed, compensation claims, and case duration. The richness and completeness of the administrative records significantly strengthen the validity of the empirical strategy and limit concerns about missing or mismeasured outcomes.

Methods

The authors estimate a difference-in-differences model using the quasi-random assignment of attorneys as the identifying device. Because the wheel system assigns attorneys in a fixed rotation conditional on filing date, court, and eligibility constraints, the match between a given defendant and attorney is plausibly exogenous once the authors condition on month-by-year-by-court fixed effects. This structure enables comparisons of Black versus White defendants across Black versus White attorneys while holding constant factors that jointly influence attorney availability. The main specification includes controls for charge type, day of week, defendant demographics, criminal history, and in some cases attorney fixed effects. Standard errors are clustered at both the defendant and attorney level to address within-cell dependence. The authors also conduct a set of formal balance tests to validate the quasi-random assignment assumption, showing that defendant attributes, case attributes, and attorney characteristics do not predict whether a defendant is paired with a Black or White attorney. Additional robustness checks include dropping cases with multiple attorneys, varying the coding of Hispanic status, adding rich sets of interaction terms between case characteristics and attorney race, and examining whether results are driven by a small subset of attorneys. They further estimate empirical Bayes–adjusted attorney fixed effects to visualize the distribution of attorney performance by race. Although this is an observational setting, the quasi-random assignment design is a strong causal inference approach that approximates the logic of an experiment.

Findings/Size Effects

The results show clear and consistent effects of defendant–attorney racial pairing. Black defendants represented by White attorneys are 14–16 percent more likely to have their charges dismissed compared with Black defendants represented by Black attorneys, controlling for observable case characteristics. In addition, Black defendants assigned to White attorneys are 15–26 percent less likely to receive a jail sentence. These are sizable effects given overall dismissal and incarceration rates in misdemeanor courts. White attorneys perform similarly for White and Black defendants; the differential effect arises because Black attorneys achieve lower dismissal rates and higher incarceration rates when representing Black defendants, relative to White attorneys’ performance with Black defendants. The authors rule out several alternative explanations. The effects do not appear to be driven by differences in attorney experience, law school quality, caseload, or charge mix. They also find no evidence that different-race pairings alter attorney effort as measured by motions, case duration, or supplemental compensation. Moreover, the pattern is not limited to a few outlier attorneys; instead, the distribution of White attorneys’ effectiveness shifts upward for Black defendants as a group. Sentence lengths for probation or jail, conditional on receiving such sentences, do not differ meaningfully by racial pairing. Finally, having a different-race attorney does not increase the likelihood of going to trial or affect one- or two-year recidivism rates, though estimates for recidivism are imprecise. The authors also conduct a parallel analysis for Hispanic defendants, though assignment in that subgroup is not random. Those exploratory estimates suggest the opposite pattern—Hispanic defendants fare worse with non-Hispanic White attorneys—but selection concerns limit interpretation.

Conclusion

This study provides the first causal evidence on how race shapes the effectiveness of court-appointed defense attorneys. In contrast to patterns in policing, teaching, or medicine—where same-race pairings often generate superior outcomes—the findings here show that Black defendants obtain more favorable results when represented by White attorneys. The effect is substantial in magnitude and appears to be persistent across the distribution of attorneys rather than concentrated among a few. The authors discuss several potential explanations, including attorney–prosecutor interactions, courtroom dynamics, or changing professional behavior in response to broader racial-justice movements. Although the mechanism remains uncertain, the pattern raises important questions about how defendants are represented within high-volume misdemeanor systems and whether structural features of indigent defense unintentionally disadvantage certain groups. The findings may extend to other jurisdictions that rely on rotation-based appointment systems, though contextual factors such as prosecutorial culture, attorney demographics, and institutional norms could affect generalizability. Overall, the article offers a rigorous and consequential contribution to research on race and legal institutions, highlighting an underexamined component of the criminal adjudication process and demonstrating that defense-side dynamics can meaningfully shape case outcomes.