Dynamic Optimization in Models for State Panel Data: A Cohort Panel Data Model for the Effects of Divorce Laws on Divorce Rates

Tongyai Iyavarakul, Office of the Prime Minister, Thailand
Marjorie McElroy, Duke University
Kalina Staub, University of Toronto, Mississauga

We posit and estimate a model of the effects of state divorce laws on divorce rates that incorporates a basic dynamic insight: the liberalization of divorce laws affects the divorce propensities of couples that married before (who are surprised) and after the liberalization (who selected into marriage despite the change) in fundamentally different ways. Our model distinguish between the right to divorce and the cost of divorce; we recode state divorce laws to make this distinction. Despite the fact that state panel data on divorce rates necessarily confounds these surprise and selection effects, our procedure disentangles them. We find significant surprise and selection effects for the cost of divorce but cannot reject the Coasian hypothesis that the adoption of unilateral law had no effect on divorce rates. Our technique applies to a wide class of dynamic decisions where the researcher must resort to state (county, state, country, etc.) panel data.

  See paper

Presented in Session 137: Causal Inference and Experimental Designs