This dissertation studies the economics of post-secondary education in Germany. Considering that young adults in Germany generally take up either higher education or vocational training (or sometimes both in a sequential manner) after they finish secondary education I focus on three fundamental questions: (i) "Does higher education pay off for the individual and the state?", (ii) "How strongly do earnings expectations influence the individual's choice between higher education and vocational training?", and (iii) "What are the distributional effects of higher education funding?"
Importantly, I analyze these questions from a lifetime perspective, i.e. considering the whole life cycle of an individual instead of focusing on one particular point of the life cycle (a certain age, for instance). In addition, the perspective taken in this dissertation is forward looking, in the sense that it takes the perspective of the individuals of a young cohort and their projected life cycles. While it seems self-evident to consider a forward-looking lifetime perspective to answer the questions of interest, such a perspective has rarely been taken in the literature. Most likely a main reason is that, by nature, observable life cycle data (until retirement, for instance) do not exist for younger cohorts. Hence, in order to take a lifetime perspective of a younger cohort one needs to generate ``artificial" data reflecting a plausible life course of currently young adults. Here, a dynamic microsimulation model on the basis of the German Socio-Economic Panel (Goebel et al., 2018) is developed. The dynamic microsimulation model sequentially simulates an individual's life cycle in terms of several key variables such as employment and family formation (Li and O'Donoghue, 2013). This model is the foundation of the empirical work in this dissertation.
The first chapter, "The Private and Fiscal Returns to Higher Education – A Simulation Approach for a Young German Cohort", explains in detail how the dynamic microsimulation model works. Essentially, it first estimates transitions models for the variables that are to be simulated and then uses the estimated parameters to simulate the individual life cycles from one year to the next. In addition, it contains a tax-transfer calculator that models the German tax-transfer system and allows to compute taxes, transfers, and social security contributions. Using the dynamic microsimulation model the first chapter then estimates the private and fiscal returns to higher education. We distinguish between gross and net income and different degrees of income pooling within households. For a typical biography, we find large positive internal rates of return (IRR) for both the individual and the state. At the same time, however, we also find that a substantial share of individuals would incur negative net present values (NPV). Chapter two, "The Decision to Enrol in Higher Education", studies the question how strongly the choice to enter higher education depends on the expectations of future income. Using the dynamic microsimulation model from chapter 1 I forecast an individual's expected life cycle given a specific educational choice. In addition to the dynamic microsimulation model and the SOEP data, I use the starting cohort 4 of the National Educational Panel Study (Blossfeld and von Maurice, 2011) that follows 9th graders until after secondary school graduation. This allows me to estimate an educational choice model where individuals maximize lifetime utility by choosing between higher education and vocational training. Using the estimated parameters from the decision model I simulate the introduction of tuition fees and graduate taxes. I find that such reforms would only induce few people to change their educational decisions.
The third chapter, "Higher Education Funding in Germany – A Distributional Lifetime Perspective", analyzes the distributional effects of higher education funding. For this I first compare the quantitative importance of different funding instruments, ranging from free tuition to subsidized health insurance for students. The analysis shows that free tuition is, by far, the most important instrument. However, there is a large heterogeneity by how much a student benefits from free tuition depending on her field of study. To connect the amount of benefits an individual receives from higher education funding, particularly free tuition, to the expected lifetime income of an individual, I use the dynamic microsimulation model and simulate the individual biographies. Finally, I use the decision model of chapter 2 and extend it to the case of multiple alternatives (with fields of study and vocational training being the alternatives). Using the estimated parameters I simulate how the choice between the fields would change under different tuition fee schemes. In line with the results of chapter 2, I find that the tuition fees would barely change the individuals' educational choices.
