Unnatural Experiments? Estimating the Incidence of Endogenous Policies

The US federal system provides great potential for estimating the effects of policy on behavior. There are numerous empirical studies that exploit variation in policies over space and time. In pursuing this line of enquiry, the issue of policy endogeneity is central. If state policy making is purposeful action, responsive to economic and political conditions within the state, then it may be necessary to identify and control for the forces that lead policies to change if one wishes to obtain unbiased estimates of a policy's incidence. The aim of this paper is to investigate how recognition of policy endogeneity affects attempts to analyze policy incidence. Throughout, we take a specific context -- workers' compensation benefits. We contrast the use of differences-in-differences estimation, where a comparison is made between a group affected by the policy change and a control group, with instrumental variables estimation when political variables are used as instruments. Although conclusions drawn must be confined to the example at hand, we believe that the analysis illustrates why it may be important to consider the implications of policy endogeneity more generally.