Life-cycle Cost of Overconfidence: Evidence from Maternity Leave Reforms
(Job market paper)

This paper investigates the life cycle costs of overconfidence in future employment possibilities, focusing on females who experience child-related career breaks. To estimate these costs, I develop a novel strategy to identify expectations about employment prospects within a life cycle model of female labor supply and human capital accumulation. Reactions to a discontinuity in the future expected value of non-employment caused by the end of an employment protection allows for identification of expectations. In addition, reforms that exogenously vary the length of this protection period permit to separately identify each of expectations, job-arrival rates, and preferences. In line with suggestive evidence, the estimated life cycle model indicates that expectations are substantially biased: on average women expect the half-yearly job arrival rate to be twice the actual rate. This overconfidence prolongs the average child related career break by eight months, resulting in a larger share of mothers staying non-employed beyond the protection period. The implications of forgone wages and human capital are large, since overconfidence decreases life-time earnings from employment by 14%.

Identification of Time Preferences in Dynamic Discrete Choice Models: A New Interpretation of the Exclusion Restriction

Although time preferences are a crucial factor for decisions in dynamic settings, most empirical approaches take the discount factor as known, usually only assuming the economic agent to be an exponential discounter. Incorrect assumptions about the time preferences can lead to false conclusion from counterfactual policy analysis. Additionally, without accounting for potential time-inconsistencies, possible effective policies, such as commitment devices, might be ignored. Building on the previous literature on identification for dynamic programming discrete choice models, I offer a new interpretation of the exclusion restrictions necessary to recover time preferences from choice data. In settings, in which the decision maker faces a probabilistic restriction of the choice set depending on the previous choice, exogenous variation of the restriction probabilities identifies time preferences. I argue that these models are common, especially in labor economics. The newly derived exclusion restriction are easy to interpret and various examples are provided when these are satisfied.

Time Preferences and Female Labor Supply

(with Peter Haan and Luke Haywood)

We estimate a dynamic life cycle model of labor supply with a strong focus on women's time preferences. We extend the dynamic discrete choice model of female labor supply to accommodate for potentially time-inconsistent behavior. For the identification of the time preferences we exploit natural experiments: we use several parental leave reforms in Germany which extend parental leave from one to three years. Preliminary results provide evidence for significant time inconsistent behavior. Child related career breaks are on average longer when women exhibit time-inconsistent behavior. Our approach allows us to shed light on the importance of time preferences in explaining important labor supply choices.