Essays in family macroeconomics

Bellue, Suzanne

PDF
BellueDissertation.pdf - Veröffentlichte Version
Download (2MB)

URN:	urn:nbn:de:bsz:180-madoc-653258
Dokumenttyp:	Dissertation
Erscheinungsjahr:	2023
Ort der Veröffentlichung:	Mannheim
Hochschule:	Universität Mannheim
Gutachter:	Tertilt, Michèle
Sprache der Veröffentlichung:	Englisch
Einrichtung:	Fakultät für Rechtswissenschaft und Volkswirtschaftslehre > Makro- u. Entwicklungsökonomie (Tertilt 2010-)
Fachgebiet:	330 Wirtschaft
Freie Schlagwörter (Englisch):	macroeconomics , education , family
Abstract:	This dissertation employs quantitative methods to study questions in family macroeconomics, with a focus on parental decisions and intergenerational mobility. It consists of two self-contained chapters. Chapter 1 is titled ``Why Don't Poor Families Move? A Spatial Equilibrium Analysis of Parental Decisions with Social Learning.'' In this first chapter, I propose a mechanism that rationalizes the observed heterogeneity in parental investment choices across socioeconomic groups in the United States. I develop a quantitative spatial and overlapping generations model in which parents decide on two parental inputs influencing their child's next period human capital: neighborhood quality and parental time. Importantly, I introduce imperfect information and social learning about one parameter of the child skill technology that governs the returns to the two parental inputs. Young agents learn about the unknown parameter using only information available at their neighborhood level. However, because of a selection neglect bias that leads to misinferences from non-representative samples, segregation generates information frictions that systematically distort parents' beliefs and behavior. Specifically, in equilibrium, parents raised in low-quality neighborhoods tend to underestimate the importance of parental inputs, and parents raised in high-quality neighborhoods tend to overestimate it. I calibrate the model using several United States representative data sets. The calibrated model matches targeted and, more interestingly, non-targeted moments regarding parental behavior across socioeconomic groups. On the contrary, a model that matches the intergenerational persistence of income but assumes perfect information cannot rationalize the heterogeneity in parental inputs across socioeconomic groups without assuming highly heterogeneous preferences. I find that parents' beliefs about the importance of parental inputs increase the income Gini index by 3% and the intergenerational income rank coefficient by 12%. Finally, scaling up a housing voucher policy generates, in the long run, and in general equilibrium, a shift in parental beliefs that contributes to the reduction of inequality and to the improvement in intergenerational mobility. Chapter 2 is titled ``Efficiency and Equity of Education Tracking: A Quantitative Analysis'' and is co-authored with Lukas Mahler. In this chapter, we investigate the question of early school tracking in Germany---the allocation of students to different types of schools. We develop a life-cycle overlapping generations model to evaluate the aggregate effects of a policy that would delay the school track decision by four years: from age ten to fourteen. Crucially, we incorporate linear classmate peer effects, non-linear school track instruction pace effects, and age-specific skill shocks into the child skill technology. We show analytically that this technology embeds this idea of learning gains through homogenous peer groups and that it can rationalize reduced-form empirical evidence on school tracking. The child skill technology parameters are one of the crucial elements that drive potential efficiency gains of early school tracking. We then calibrate the model using multiple German representative data sets. Our calibrated model predicts that around 23% of lifetime earnings and 30% of lifetime wealth variation is already explained at age ten, the time of the school track choice. Conditioning on the initial school track choice alone accounts for 12% of lifetime earnings variation and 13% of lifetime wealth variation. Finally, we find that postponing the tracking age from ten to fourteen generates significant improvements in intergenerational mobility. However, these come at the cost of efficiency losses in aggregate economic output. The size of these losses depends on the design of the instruction levels in each school track.