Working Paper: NBER ID: w18838
Authors: Jeffrey Grogger
Abstract: Social experiments frequently exploit data from administrative records. However, most administrative data systems are state-specific, designed to track earnings or benefit payments among residents within a single state. Once an experimental participant moves out of state, his earnings and benefits in his state of origin consist entirely of zeros, giving rise to a form of attrition. In the presence of such attrition, the average treatment effect of the experiment is no longer point-identified, despite random assignment. I propose a method to estimate such attrition and, for binary outcomes such as employment, to construct bounds on the average treatment effect. Results from a welfare-reform experiment considered to have sizeable effects appear quite ambiguous after accounting for attrition. The results have important implications for planning social experiments.
Keywords: No keywords provided
JEL Codes: C33; C9; I38
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| attrition due to migration (F22) | identification of average treatment effect (ATE) (C22) |
| observed data (Y10) | lower bounds on treatment effects (C22) |
| attrition influenced by treatment (C22) | identification of average treatment effect (ATE) (C22) |
| terminal runs of zeros (C29) | imputed attrition indicators (J63) |
| treatment status (I12) | employment outcomes (J68) |