Working Paper: NBER ID: w7131
Authors: Jinyong Hahn; Petra Todd; Wilbert van der Klaauw
Abstract: The regression discontinuity (RD) data design is a quasi-experimental design with the defining characteristic that the probability of receiving treatment changes discontinuously as a function of one or more individual characteristics. This data design occasionally arises in economic and other applications but is only infrequently exploited in evaluating the effects of a treatment. We consider the problem of identification and estimation of treatment effects under a RD data design. We offer an interpretation of the IV or so-called Wald estimator as a regression discontinuity estimator. We propose nonparametric estimators of treatment effects and present their asymptotic distribution theory. Then we apply the estimation method to evaluate the effect of EEOC-coverage on minority employment in small U.S. firms.
Keywords: antidiscrimination laws; minority employment; regression discontinuity design
JEL Codes: C14; C51
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Federal antidiscrimination laws (J71) | increase in minority employment (J68) |
| Firm size (15 or more employees) (L25) | Federal antidiscrimination laws (J71) |
| Firms just above the threshold (D21) | higher minority employment (J15) |
| Firms just below the threshold (D21) | lower minority employment (J79) |