Essays on education and labor economics

Resumo

This thesis contains three independent papers. Below follows a brief description of each article. The first article is given by a joint work with Bruno Delalibera, where we study the allocation of public expenditures across educational stages in a developing country. We construct a general equilibrium model that features heterogeneous agents, credit restrictions, basic and tertiary education, public and private educational institutions. We calibrate the model’s parameters using Brazilian data. Simulations show three of the model’s features. First, economic inequality is mainly explained through endogenous educational decisions, instead of exogenous ability shocks. Second, abolishing public educational institutions contracts GDP by 5.5%, decreases welfare of households in the first income quartile by 7%, and does not affect significantly the welfare of the remaining households. Third, borrowing constraints are tight for the poorest agents in the economy. In the main exercise, we find that reallocating public expenditures from tertiary towards basic education to mimic Denmark’s allocation of public expenditures across educational stages decreases GDP, aggregate welfare and the Gini coefficient by 1.5%, 0.2% and 2%, respectively. The second article is co-authored with Nezih Guner and Cezar Santos. We start by motivating with the fact that a large number of laborers in developing countries encounter poor working conditions. Corruption episodes related to labor inspections, some involving political figures, occur regularly in such environments. We investigate the effects federal deputy electoral results on labor inspections in firms that donate to electoral campaigns in Brazil from 2002 to 2014. To estimate a causal effect of electoral results on inspections, we use a regression discontinuity approach in close elections. We find no causal evidence that federal deputy electoral results affect variables related to labor inspections. This result is unchanged when we estimate heterogeneous effects as a function of labor inspection authorities’ position type, depending on which inspections can potentially be more or less influenced by political forces. Lastly, the third paper is co-authored with Pedro Cavalcanti Ferreira and Cezar Santos. We provide a new education quality index for states within a developing country using 2010 Brazilian data. This measure is constructed based on the notion that the financial returns obtained from an additional year of schooling can be seen as being derived from the value that market forces assign to this education. We use migrant data to estimate returns to schooling of individuals who studied in different states but who work in the same labor market. We find very heterogeneous educational qualities. In fact, Brazil displays cross-state educational quality variation almost as large as that observed across countries. We compare our index with standardized test scores, educational outcome variables, and public expenditure per schooling stage at the state level, producing new evidence related to education in a large developing country. We conduct an education quality-adjusted development accounting exercise for Brazilian states and find that human capital accounts for 26%-31% of output per worker differences. Adjusting for quality increases human capital’s explanatory power by 60%.

Ficha do documento

Tipo

Tese

Ano

2018

Instituição

Fundação Getulio Vargas

Fonte

Repositório da FGV

Idioma

Inglês

Acesso

Acesso aberto

Identificador

oai:repositorio.fgv.br:10438/24813

Temas

EconomiaFinanças PúblicasGovernançaEducação

Palavras-chave

EducationEducational stagesPublic expendituresGeneral equilibriumEstágios educacionaisDespesas públicasEducaçãoEconomiaInvestimentos na educaçãoTrabalho