We provide causal evidence of skin tone discrimination using professional football (soccer) as a natural laboratory. Leveraging a computer-vision measure of skin tone and quasi-random variation in shot outcomes near the goal frame, we implement a Difference-in-Discontinuities design comparing narrowly scored goals to narrowly missed attempts. We find that Light-skinned players receive significantly larger boosts in post-match ratings than Tan- and Dark-skinned peers for identical actions. These disparities appear in both algorithmic and human-assigned evaluations and are concentrated in the subjective component of ratings. Season-level analyses reveal that biased evaluations translate into lower market valuations for darker-skinned players, despite equivalent performance. Evaluative bias, rather than differential treatment in contracts, emerges as a key driver of economic inequality in this high-information labor market. Our findings show how skin color discrimination can persist even in environments with transparent outcomes and extensive performance data.
COLLABORATION
Analyses, survey insights and a practical guide mark a key milestone in the Erasmus+ project supporting data-driven vocational training across five European countries.
