We show that the OLS and fixed‐effects (FE) estimators of the popular difference-in-differences model may deviate when there is time varying panel non-response. If such non-response does not affect the common-trend assumption, then OLS and FE are consistent, but OLS is more precise. However,...

This paper uses the differential timing across counties of the removal of restrictions on Sunday alcohol sales in the state of Georgia to determine whether the change had an impact on employment and hours in the beer, wine, and liquor retail sales industry. A triple-difference analysis finds...

We show that the OLS and fixed‐effects (FE) estimators of the popular difference-in-differences model may deviate when there is time varying panel non-response. If such non-response does not affect the common-trend assumption, then OLS and FE are consistent, but OLS is more precise. However,...

This paper uses the differential timing across counties of the removal of restrictions on Sunday alcohol sales in the state of Georgia to determine whether the change had an impact on employment and hours in the beer, wine, and liquor retail sales industry. A triple-difference analysis finds...

This paper examines the causal effect of a change in the nomination scheme of the city manager from appointment by the local council to election by the citizens on urban development policies. Using the fact that the timing of the reform was as good as random in municipalities of the German state...

I construct an immigration policy index which is heterogeneous across destinationorigin country pairs and variant over time. This index is based on three types of entry visa restrictions: visa required, visa not required for short stays and visa not required at all. When estimated in levels,...