« Back to Results

Advances in Modeling Human Capital and Education

Paper Session

Friday, Jan. 4, 2019 2:30 PM - 4:30 PM

Atlanta Marriott Marquis, L503
Hosted By: Econometric Society
  • Chair: Christopher Taber, University of Wisconsin-Madison

Understanding School Competition Under Voucher Regimes

Cristian Sanchez


I study how different voucher designs and policies affect school competition in education markets, and their consequences for students' school choices and welfare. To that end, I build a structural model of demand and supply for primary education in markets that allow the use of vouchers. Unlike previous studies, I not only model schools' pricing behavior, but also their program participation decisions, a discrete strategy that adds an additional layer of complexity to the solution of the supply side game. I fit my model to rich administrative data from Chile, and leverage its simultaneous implementation of universal (available to all students) and targeted (available only to low-income students) vouchers to estimate my model's parameters. Counterfactual simulations indicate that universal and targeted vouchers affect school competition differently. On the one hand, a higher targeted voucher attracts more schools to participate in the targeted program. However, high-quality schools join only if the subsidy is sufficiently high. On the other hand, a higher universal voucher induces schools to lower the top-up fees they charge to parents. Specifically, a $1 increase in the subsidy translates into a $0.58 decrease in the average fee. Finally, policies that favor the universal voucher are more mobility- and welfare-enhancing in the aggregate, but increase the welfare gap between low- and high-income students, relative to policies favoring the targeted voucher.

Estimating General Equilibrium Effects of Major Education Reforms

Michael Gilraine
New York University
Hugh Macartney
Duke University
Robert McMillan
University of Toronto


This paper sets out an approach for credibly estimating general equilibrium sorting effects of major education reforms. We develop our strategy in the context of California's vast class size reduction program, and use multiple differencing to identify the reform's direct and indirect effects on a common scale (using test scores), along with their persistence. The estimates reveal a large direct class size effect and an even larger indirect effect via changes in school demographics. Further, both effects persist, generating a substantial longer-run policy impact. Our approach is applicable more broadly to large-scale reforms that cover a subset of school grades.

Investing in Children’s Skills: An Equilibrium Analysis of Social Interactions and Parental Investments

Francesco Agostinelli
University of Pennsylvania


This paper studies the effects of social interactions on the dynamics of children’s skills. I build a dynamic equilibrium model of child development with two key ingredients: peer groups forming endogenously and parental investments responding to the child’s social interactions. I estimate the model via simulated method of moments using a dataset of U.S. adolescents. Exploiting within school / across cohort variations in potential peers’ compositions, I identify the degree of complementarity between parents and peers in skill formation. I find that the environment where children grow up permanently shapes their developmental trajectories through the effects of social interactions. Moving a child at age 12 to an environment where children have 1 percentile higher skills at age 16, on average, improves her skills rank at age 16 by 0.63 percentiles. The effects are in proportion to the exposure time throughout childhood, with an average effect of 0.48 percentiles if the child is moved at age 15. As model validation, I show that these results track the estimates of the exposure effects of neighborhoods in Chetty and Hendren (2016a). Decomposing the exposure effects, I find that peers alone account for more than half of the overall findings, while the school and the neighborhood quality account for the remainder. Finally, I evaluate the effects on child development of a policy that integrates a sizable fraction of low-skilled children into a high-income environment. I find: (i) the effects depend on the endogenous formation of new peer groups; (ii) the policy generates dynamic equilibrium effects on parental investments and social interactions, which, if ignored, would lead to policy predictions for children’s skills of approximately seven times smaller.

The Determinants of Teachers’ Occupational Choice

Kevin Lang
Boston University
Maria Palacios
Boston University


Among college graduates, teachers have both low average AFQT and high average
risk aversion. Using a dynamic optimization model with unobserved heterogeneity, we
find that the low mean AFQT score among teachers primarily reflects a low return to
other skills, correlated with AFQT, rather than a low return to cognitive skill within
teaching. The compression of earnings within teaching attracts relatively risk-averse individuals.

Were it possible to make teacher compensation mimic the return to skills and
riskiness of the non-teaching sector, overall compensation in teaching would increase.
Moreover, such a shift would substantially reduce the utility of many current teachers,
making the process of reform challenging. Importantly, our conclusions are sensitive to
the degree of heterogeneity for which we allow, and even a model with no unobserved
heterogeneity appears to fit well within sample. It would be easy to conclude that
allowing for two or three types fits the data adequately. Formal methods reject this
conclusion. The BIC favors seven types. Ranking models using cross-validation, eight
types is better although the improvements from going from six to seven, from seven to
eight types and from eight to nine types are noticeably smaller than those from adding
an additional type to a lower base. Importantly, the results of policy exercises are very
sensitive to the degree of heterogeneity included in the model.
JEL Classifications
  • J2 - Demand and Supply of Labor
  • I2 - Education and Research Institutions