Timo Schenk, Erasmus University Rotterdam

"Inference in Event Studies with Approximately Parallel Trends"

Abstract

Examining pre-trends violations is a common approach to validating the parallel trends assumption necessary for difference-in-difference designs. When parallel trends are rejected by the data, however, applied researchers commonly continue to interpret treatment effect estimates as long as the observed pre-trend violations are deemed sufficiently small. To rationalize this behavior, we recast difference-in-difference designs as relying only on an approximate version of parallel trends, which allows parallel trends not to hold in all realizations of the data. Our reformulation delivers new inference procedures that account for uncertainty about possible deviations from parallel trends.

(with Nikolaj Harmon and Mikkel Sølvsten)

Contact person: Jesper Riis-Vestergaard Sørensen