A Behavioral Economics Approach to Discrimination

How should we measure discrimination? Economics has long relied on the analysis of taste-based discrimination and statistical discrimination, using a well-established empirical toolkit centered on audit studies, correspondence experiments, and regression-based decompositions. These methods have produced landmark findings, yet they face persistent challenges: identifying the precise mechanisms behind differential treatment, capturing the multidimensional nature of real-world discrimination, and accounting for the behavioral complexities that shape both discriminators and those who are discriminated against. This paper reviews the traditional methods used to measure discrimination in economics and the social sciences, and argues that behavioral economics offers a powerful complementary toolkit. Drawing on insights about mental models, implicit biases, adaptive preferences, and social norms, behavioral approaches can illuminate the micro-foundations of discriminatory behavior in ways that standard methods cannot. The paper illustrates these arguments with examples from my own recent fieldwork, five policy-oriented behavioral studies conducted in Ecuador, Spain, Colombia, Pakistan, and Peru, to explain how intersectional discrimination can be studied in the field.