Racial bias, colorism, and overcorrection joint with Alex Krumer, Rosa Lavelle-Hill, and Tim Pawlowski
This paper examines whether increased awareness can affect racial bias and colorism. We exploit a natural experiment from the widespread publicity of Price and Wolfers (2010), which intensified scrutiny of racial bias in men's basketball officiating. We investigate refereeing decisions in the Women's National Basketball Association (WNBA), an organization with a long-standing commitment to diversity, equity, and inclusion (DEI). We apply machine learning techniques to predict player race and to measure skin tone. Our empirical strategy exploits the quasi-random assignment of referees to games, combined with high-dimensional fixed effects, to estimate the relationship between referee-player racial and skin tone compositions and foul-calling behavior. We find no racial bias before the intense media coverage. However, we find evidence of overcorrection, whereby a player receives fewer fouls when facing more referees from the opposite race and skin tone. This overcorrection wears off over time, returning to zero-bias levels. We highlight the need to consider baseline levels of bias before applying any prescription with direct relevance to policymakers and organizations given the recent discourse on DEI.
A performance-weighted allocation mechanism for repeated contests joint with Mats Duys, Tim Pawlowski, and Elias Tsakas
In repeated strategic interactions with a final reward—such as employee promotions or grant allocations—a key challenge is designing mechanisms that encourage sustained high effort while ensuring equitable outcomes. Traditional methods, like lotteries or outcome-contingent rewards, often fall short, failing to maintain engagement and inadvertently discouraging early underperformers. We propose a performance-weighted allocation mechanism inspired by dynamic rating systems, which continuously adjusts each agent’s probability of receiving the reward based on cumulative performance relative to others. This adaptive design maintains engagement by linking reward probability to performance history, ensuring maximal effort as a subgame perfect equilibrium (SPE).
I4R discussion paper series: A comment on “Agricultural Diversity, Structural Change, and Long-Run Development: Evidence from the United States” joint with Zhanna Kapsalyamova, Wietse Mesman, and Victor Smirnov
We replicate Fiszbein (2022), who finds that greater agricultural diversity in 1860 raised long-run population density and income, and during the Second Industrial Revolution, spurred industrialization, patenting, and the growth of knowledge- and skill-intensive manufacturing. Using the authors’ replication package, we reproduce all results without substantive code modifications. In contrast, when rebuilding the analysis dataset from the provided partially raw inputs, we identify variables that are not generated by the variable construction script but appear in the final curated dataset, namely, the 60-square-mile grid identifier used for clustered standard errors. We also revisit an interpretive assumption underpinning the mechanism linking agricultural diversity to knowledge accumulation: while individual agricultural product production plausibly entails domain-specific know-how, agronomic skills and technologies can significantly overlap. Allowing for this, rather than treating each crop as fully distinct and thus diverse, could attenuate measured effects or alter the interpretation.
Making decision-making more approachable for K-12 students using tangible sports examples
Forthcoming...