Partial Retirement, Financial Student Aid, and Labor Market Responses ; Empirical Evidence from Germany ; Arbeitsmarkteffekte von Teilzeitrente und finanzieller Bildungsförderung ; Empirische Evidenz aus Deutschland
This dissertation evaluates policies in two strands of social policy: education and pension politics. Both play a major role in state budgets and in 2013 public spending in these sectors amounted to 8.0% and 8.2% of GDP in OECD countries. More precisely, this dissertation focuses on the analysis of underlying mechanisms in the impact of financial student aid (Chapter one) and partial retirement (Chapters two and three) on labor market outcomes using a dynamic structural life-cycle approach. Chapter 1 investigates time- inconsistent preferences in educational decision making and corresponding policies using a structural dynamic choice model. For this purpose it augments a dynamic structural model by hyperbolic discounting, a behavioral bias that exhibits time-inconsistent preferences. Estimates of this model are compared to the same model but with classic exponential discounting. The estimation is based on a sample of West German students from the German Socio-Economic Panel (SOEP). In line with Chan (2013), Fang and Wang (2015), and Haan et al. (2017) identification is achieved not only on the basis of functional form assumptions but also by imposing exclusion restrictions that affect educational choices indirectly through their impact on the transition probabilities of relevant state variables and reveal information on the discounting behavior. Birth cohort groups as well as regions that were affected by different educational policy reforms are used as exclusion restrictions. Their relevance on the time invested for the attainment of educational degrees is shown. Agents are assumed to face two kinds of uncertainty: (1) there is uncertainty over whether an additional year invested into education will, in fact, be successful and lead to a degree (affected by the policy reforms); and (2) there is uncertainty over the returns to the degree earned when exiting education. The estimation of the structural parameters of the choice model indicates time-inconsistent behavior and provides quantitative evidence for its ...