Veranstaltungsreihe: IAB-Colloquium

Racial gaps in student loan accumulation and repayment are substantial. Using data from the Beginning Postsecondary Students survey, we document that Black students are more likely to borrow than White students, and they accumulate larger student debt conditional on borrowing. Black borrowers are also more likely to be enrolled in income-based or extended repayment plans, so they have lower average monthly payments and pay off their debt more slowly. Nevertheless, Black borrowers are 2-4 times more likely to default on student loans. To what extent can initial conditions and lifecycle financial circumstances account for these racial differences in student loan repayment and default? We construct a lifecycle consumption-savings model that captures observed heterogeneity in initial wealth and student debt, as well as unobserved heterogeneity in parameters governing initial human capital and lifecycle human capital accumulation. The model produces earnings dynamics, labor supply choices, human capital accumulation, and financial asset accumulation that are consistent with lifecycle data. We use our model to quantify the relative contributions from each of these channels to the racial default rate gaps over the lifecycle. We aim to use our model to advance policy proposals that can mitigate racial gaps in student loan default.

We quantify aggregate effects of changes in the potential benefit duration (PBD) in Poland using administrative data containing the universe of unemployment spells over more than two decades. Individual workers’ PBD depends on the county unemployment rate relative to the national average in the previous calendar year. We exploit this sharp discontinuity with RDD estimates and construct impulse response functions to estimate effects of a longer PBD at the county-level. After 12 months, the effect of a PBD of 12 vs. 6 months is an increase in the log stock of all unemployed of 0.03 and an increase in the log stock of the directly affected by 0.1. In contrast, we find no evidence on spillovers on indirectly affected unemployed and no effect of PBD on labour market tightness. We document that inflows into unemployment respond strongly to PBD changes. A decomposition of the effects of a longer PBD on the stock of unemployed shows that the effect on inflows is more important than the one on the exit rate.

This paper documents and theoretically explains a nexus between globalization and wage inequality within plants through internal labor market organization. We document that the dominant component of overall and residual wage inequality is within plant-occupations and, combining within-occupation task information from labor force surveys with linked plant-worker data for Germany, establish three interrelated facts:

1. larger plants and exporters organize production into more occupations,
2. workers at larger plants and exporters perform fewer tasks within occupations, and
3. overall and residual wages are more dispersed at larger plants.

To explain these facts, we build a model in which the plant endogenously bundles tasks into occupations and workers match to occupations. By splitting the task range into more occupations, the plant assigns workers to a narrower task range per occupation, reducing worker mismatch while typically raising the within-plant dispersion of wages. Embedding this rationale into a Melitz model, where fixed span-of-control costs increase with occupation counts, we show that inherently more productive plants exhibit higher worker efficiency and wider wage dispersion and that economy-wide wage inequality is higher in the open economy for an empirically confirmed parametrization. Reduced-form tests confirm main predictions of the model, and simulations based on structural estimation suggest that trade induces a stricter division of labor at globalized plants with an associated change in wage inequality.