Identification in Regression Discontinuity Designs with Multiple Cutoffs

• Activity Number: 64
• Type: Topic Contributed
• Date/Time: Sunday, August 9, 2015 : 4:00 PM to 5:50 PM
• Sponsor: Biometrics Section
• Author(s): Luke Keele*
• Companies: Penn State
• Keywords: causal inference ; regression discontinuity ; heterogeneity

Abstract

We consider a regression discontinuity (RD) design where the treatment is received if a score is above a cutoff, but the cutoff is a random variable as opposed to a known constant. This multi-cutoff RD design is very common in practice, and researchers often normalize the running variable, using the same cutoff on the normalized variable for all observations and estimating a pooled RD effect. We formally derive the form that this pooled parameter takes, and discuss its interpretation under different assumptions. We illustrate our derivations with a study of the effects of a party winning office on the party's future vote share, using data from Brazilian mayoral elections and U.S. Senate elections.