Essays on taxes and consumption

• Ano de defesa: 2022
• Autor(a) principal: Matos, Mariana Milhomem
• Orientador(a): Costa, Carlos Eugênio Ellery Lustosa da
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Palavras-chave em Português: Optimal income taxation; HSV taxes; Consumer Intertemporal Optimization; Permanent income hypothesis
• Link de acesso: https://hdl.handle.net/10438/33558

This thesis consists of three independent essays in applied microeconomics. The first paper estimates the marginal social welfare functions for Brazil using two functional forms for the agents preferences: quasi-linear in consumption and isoelastic in consumption and in labor. We use a dual approach of the standard optimal income tax model, the logic is inverted, we estimate a social welfare function that makes optimal the actual marginal tax rate schedule. Finally, we check if the revealed social preferences satisfies the conditions of the model, so the inversion is well behaved. We find that with quasi-linear utilities, the revealed social preferences are not decreasing for some values of productivity. When we introduce income effects, we introduce curvature in the utility for consumption and then the revealed social preferences are well behaved and Paretian. The second paper evaluates the welfare consequences of replacing the Brazilian current joint tax system in which spouses’ incomes are added to calculate taxable income with one for which spouses’ incomes are averaged to determine taxable income. In Brazil, spouses have the option to file individually or jointly. We find that a reform in the Brazilian current joint tax system is welfare-improving. Changing to an income tax system in which spouses’ earnings are averaged to determine taxable income increases the social welfare of the economy. The final paper investigates consumption behavior and optimality using microdata available from the Consumer Expenditure Survey. With a longer and more recent sample, we confirm the life cycle and time series patterns found by Attanasio and Weber (1995) for the United States through a synthetic panel and cohort analysis. Furthermore, we follow their approach of modeling the Euler equation with demographic and labor supply variables, but relaxing the implicit assumption that consumers use only one asset to smooth consumption over time. We do so by aggregating returns and using Mulligan and Threinen (2010) measure of return to aggregate capital to test the life cycle-permanent income hypothesis. Our results provide further evidence in favor of the theory and show that ignoring the portfolio optimization of households may overestimate the Elasticity of Intertemporal Substitution.