Working Paper: NBER ID: w16773
Authors: Jin Young Lee; Gary Solon
Abstract: Following an influential article by Friedberg (1998), Wolfers (2006) explored the sensitivity of Friedberg's results to allowing for dynamics in the response of divorce rates to the adoption of unilateral divorce laws. We in turn explore the sensitivity of Wolfers's results to variations in estimation method and functional form, and we find that the results are extremely fragile. We conclude first that the impact of unilateral divorce laws remains unclear. Second, extending Wolfers's methodological insight about sensitivity of differences-in-differences estimation to allowance for dynamic response, we suggest that identification in differences-in-differences research becomes weaker in the presence of dynamics, especially in the presence of unit-specific time trends.
Keywords: No keywords provided
JEL Codes: C23; J12; K36
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Friedberg's estimates (C13) | divorce rates (J12) |
| OLS instead of WLS (C51) | positive impact of unilateral divorce laws on divorce rates (J12) |
| Logarithmic transformations of divorce rate (J12) | positive effect of unilateral divorce on divorce rates (J12) |
| Unilateral divorce laws (K36) | divorce rates (J12) |
| Differences-in-differences research design (C93) | causal impact of unilateral divorce laws on divorce rates (J12) |