Exploring the causes of intergenerational welfare recipiency

This dissertation presents and estimates a model that examines the economic causes of intergenerational welfare recipiency. Previous literature in this area has found that daughters of welfare recipients have a higher incidence of receiving welfare than daughters of non-recipients. My dissertation measures the possible economic causes underlying this phenomenon by estimating a model of intergenerational human capital investment and welfare participation. Whereas past literature focused on measuring the extent of intergenerational welfare recipiency and on correcting for factors (such as unobserved heterogeneity) that might affect the accuracy of the results, my dissertation focuses on examining the individual welfare participation choices and on postulating a possible link between mother and daughter\u27s behavior that goes beyond habit formation. The advantage of this approach is that all factors affecting the welfare participation decision, including the offer wage and welfare guarantees and taxes, can be taken into account when examining the indirect and direct effects of the mother\u27s welfare participation on the daughter\u27s welfare behavior. The welfare participation decisions are modeled in both one-period and multi-period contexts. The one-period model describes welfare participation as a one-time decision over a number of years. The purpose of estimating this model is that it focuses on the tradeoff between leisure and human capital investment in the daughter. The multi-period model estimates the degree to which the intensity of the mother\u27s welfare dependency affects human capital investment in the daughter. Two models are estimated for each generation: (1) a structural version of the one-period multinomial probit model and (2) a tobit that estimates that multiperiod model. The structural model is estimated using a method of simulated maximum likelihood. Estimation results show that the direct effect of the mother\u27s welfare participation on her daughter\u27s welfare participation preferences is small and statistically insignificant, and that indirect effects explain much of the observed intergenerational correlation