Working Paper: NBER ID: w31393
Authors: Kirill Borusyak; Peter Hull; Xavier Jaravel
Abstract: Many studies in economics use instruments or treatments which combine a set of exogenous shocks with other predetermined variables by a known formula. Examples include shift-share instruments and measures of social or spatial spillovers. We review recent econometric tools for this setting, which leverage the assignment process of the exogenous shocks and the structure of the formula for identification. We compare this design-based approach with conventional estimation strategies based on conditional unconfoundedness, and contrast it with alternative strategies that leverage a model for unobservables.
Keywords: design-based identification; formula instruments; causal inference
JEL Codes: C21; C26
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| exogeneity of shocks (F41) | avoiding omitted variable bias (OVB) (C20) |
| random assignment of shocks (C90) | avoiding omitted variable bias (OVB) (C20) |
| controlling for expected instrument (C36) | eliminating omitted variable bias (OVB) (C20) |
| using recentered instruments (C36) | eliminating omitted variable bias (OVB) (C20) |
| assignment of shocks (C78) | outcome variable (C20) |
| shift-share IV estimators (C26) | valid asymptotic inference (C20) |