DP20112 Characterizing the Returns to STEM: Marginal and Policy-Relevant Treatment Effects
We estimate heterogeneous returns to STEM education by leveraging regional differences in the relative distance to technical versus general universities as a cost factor driving college major choice. On average, individuals experience substantial wage gains from pursuing a STEM education. The declining Marginal Treatment Effect (MTE) curve indicates positive selection on gains, suggesting that those with the lowest resistance to choosing STEM benefit the most. This positive selection arises from heterogeneity in the conditional expectations of both potential STEM and non-STEM wages. Through policy simulations aimed at increasing STEM enrollment and estimating the corresponding policy-relevant treatment effects, we demonstrate that the effectiveness of such policies critically depends on the characteristics—both observable and unobservable—of the individuals they affect. Furthermore, we highlight how these policies should be designed to both increase STEM enrollment and generate positive returns for targeted groups, particularly women.
