Abstract Behavioral public policies (BPPs) often affect individuals unevenly: the same intervention can produce systematically different outcomes across socio-economic groups, genders, or other population characteristics. Such heterogeneity raises a distinctive fairness concern. This article develops a mechanistic account of intervention fairness grounded in the ideal of impartial treatment. I argue that impartiality is violated when policy effects are heterogeneously moderated across a population without adequate justification. Building on a causal-mechanistic framework, I identify moderators as the key drivers of unequal treatment and distinguish three potential justifications for heterogeneous effects: randomness, lack of social salience, and desert. Desert, I argue, is limited in the BPP context to prior contributions to—or prior distributions of—the intervention’s target outcome. On this basis, the article proposes a sufficient definition of BPP unfairness: interventions are unfair when they either ignore deserved heterogeneity or produce unjustified moderation by socially salient, non-random factors. Finally, I translate this account into a practical, fast-and-frugal decision tool to help policymakers diagnose unfair behavioral interventions in concrete contexts.
Till Grüne‐Yanoff (Mon,) studied this question.