This dissertation comprises three independent chapters in applied microeconometrics.

This dissertation comprises three independent chapters in applied microeconometrics.

In the first chapter, we investigate the effects of earlier and longer exposure to compulsory education on students’ academic performance. Previous studies were limited because they often used variation in only one variable, school entry age or duration. We overcome this limitation by using Brazil's 2006 school reform, which lowered the compulsory school entry age and increased its duration. Furthermore, we examine the heterogeneity in the effect of the reform based on students’ preschool educational status, thereby providing different policy implications for countries where preschool education is either limited or widespread. We find that students who were exposed to primary education a year earlier without any prior education exhibited a larger increase in test scores in the short run (5 years), but the students who attended preschool showed a more persistent increase in test scores in the medium run (9 years). These findings provide new evidence that earlier and longer exposure to compulsory education can generate positive effects on student performance, both in countries with universal and insufficient availability of preschool education.

The second chapter of my dissertation examines the effect of compulsory preschool education on maternal labor market outcomes using Brazil's 2009 school reform. I use a regression discontinuity analysis based on the preschool eligibility rule, which required students to start their first year of preschool education if they had turned four before March 31st. I find significant effects of preschool enrollment on maternal outcomes if 4-year-old children were the youngest members of the households and there were no other relatives present. When children were eligible for preschool entry, mothers not only increased hours of working but also were more likely to take formal jobs that guarantee employee rights and benefits. This study is the first paper to examine the effects of preschool enrollment using a school reform in a developing country.

In the third chapter of my dissertation, I examine the effect of the Zika virus outbreak in Brazil on fertility and female labor market outcomes. The Zika virus outbreak provided strong motivation to delay pregnancy as the Zika virus infection can cause serious birth defects like microcephaly. However, due to the high frequency of unintended pregnancy in developing countries including Brazil, determining whether women indeed delayed pregnancy was not certain. Using the variation of suspected microcephaly cases across states, I find that more suspected microcephaly cases provided incentives for women to delay pregnancy. This trend was more pronounced among younger and more educated women. Despite the fertility decline, my findings provide little evidence that female labor market outcomes were altered by the outbreak of the Zika virus.

This dissertation presents three essays on agricultural productivity and its relationship with farm size and poverty. Chapter 1 addresses the relationship between farm size and productivity, a recurrent topic in development economics. We clarify the common productivity measures used in this literature, their relationships, and their advantages and limitations. Second, we argue that total factor productivity, not land productivity, is the appropriate indicator for most policy questions. Lastly, using a pseudo-panel of Brazilian farms spanning the period 1985-2006, we provide new evidence on the inverse relationship between farm size and productivity. The inverse relationship between size and land productivity is alive and well. The relationship between total factor productivity and size, in contrast, has evolved with modernization during this period. An inverse relationship between farm size and land productivity is neither necessary nor sufficient for an inverse relationship between farm size and total factor productivity.

The hypothesis of a dynamic farm size – productivity relationship is extended to the context of Mexico in Chapter 2, identifying the relationship in a panel of family farms. We find a time invariant inverse relationship between farm size and both land productivity and total factor productivity. Stochastic frontier analysis reveals that, while technical change is expanding the frontier and technical inefficiency is growing for the entire sample, these changes are more pronounced for larger farms. An inverse relationship along the productivity frontier is disappearing in the wake of Mexico’s NAFTA-era reforms to agricultural policy, yet this change has not affected the farm size – total factor productivity relationship due to growing technical inefficiency.

Chapter 3 conducts a counterfactual analysis of the contribution of changing land productivity to poverty alleviation on the farm. Stochastic frontier analysis enables a parametric decomposition of changes to the land productivity distribution in a panel of Mexican family farms. Using the decompositions, the contribution of productivity channels to poverty alleviation are estimated. The counterfactual analysis suggests that raising land productivity through intensification and technical change would be a more pro-poor approach than through increases in technical efficiency.

This dissertation comprises three studies on Pakistan in which I examine the allocative efficiency of groundwater across farm-level constraints; the optimal management of groundwater given differences in agricultural tenure; and the effect of a set of policies on the utilization of groundwater. In the first chapter, I estimate the allocative inefficiency of groundwater in Pakistani agriculture using a panel dataset of rural households and show that the utilization of groundwater varies across a set of farm-level constraints (tenure, farm size, access to surface water and location on a watercourse). In the second chapter, I examine the long-run trend of groundwater depletion in Pakistan’s Indus Water Basin under common-pool resource management—the status quo—and under optimal management. I develop a dynamic optimization problem to illustrate long-run steady states of groundwater pumping under different management, hydrologic, economic and tenure assumptions. The analysis shows that the benefits of optimal management exceed the benefits of common property management, and that the small share of sharecropping does not have an important effect on the results. In the third chapter, I use a panel dataset of rural households—the same dataset used in the analysis in the first chapter—to examine the effects of two water policies—increasing access to surface water and increasing the reliability of the supply of surface water (as measured by being located higher up on a watercourse)—on the allocative efficiency of groundwater and land productivity. The results show that farms allocate groundwater more efficiently (over utilization decreases) as the share of total farm area with access to surface water increases while increasing the reliability of surface water supply does not appear to improve the utilization of groundwater. Increasing the share of total area with access to surface water has a modest effect on land productivity. My research emphasizes the relationship between groundwater conservation and the institutional environment of farms in Pakistan’s agricultural sector, and helps to inform the larger discussion on the effective governance of water resources in the region.

Chapter 1 explores the long-run effects of inequality within Brazil. I exploit variation in the climate to instrument for the local distribution of land in 1920 in a two-stage least squares instrumental variables framework. My instrument is an index quantifying the suitability of the climate for plantation versus smallholder agriculture. IV estimates show that higher levels of historical land inequality are associated with lower local public welfare spending, both historically (in 1923) and in recent times (from 1995-2005). Historical inequality is also associated with lower levels of development, as measured by the local Human Development Index. In Chapter 2 I show that the construction of Afghanistan’s largest modern infrastructure project, a two thousand mile highway known as the A1 or Ring Road, led to a surge in opium production within the country over the past decade. I first exploit spatial and temporal variation in district-level poppy cultivation and proximity to the road to show that districts with better access to the highway witnessed a significant increase in cultivation. I then utilize three rounds of the large-scale National Risk and Vulnerability Assessment Survey to show that improved access to the highway led Afghan farmers to substitute away from legal crops towards illicit opium production. In Chapter 3 I analyze local economic growth within Brazil over the 1920 to 1996 time period. I first find that local per capita economic growth displays unconditional divergence, in that localities which were relatively richer in 1920 grew faster over 20th century. Disaggregating local GDP into its agriculture, industry, and services components reveals long-run divergence to be present in all three sectors of the economy. After conditioning growth on a comprehensive set of controls, including those for geography, inequality, and human capital, I find no evidence that localities have converged in their long-run growth rates.

Chapter one examines the long-term impact of the state prohibitions of alcohol in the United States. Between 1851 and 1920, thirty-four states enacted statewide prohibitions of alcohol at different times. Making use of the variation in state prohibition as a natural experiment, the long-term labor market outcomes of cohorts exposed to state prohibition during the critical early development period is examined. Female cohorts exposed to alcohol-reduced environments during the critical interval from the prenatal period and up to three years of age are shown to have increased labor force participation and increased income in 1960. The results are mainly driven by exposure in the prenatal period. No results were found for male cohorts, which might be explained by the selective prenatal mortality of the frailest male cohort members unexposed to state prohibition. Chapter two provides an impact evaluation of the Brazilian National Land Credit Program. Making use of a panel dataset and a pipeline control group, the chapter evaluates the impact of the program on the outcome variables of agricultural production and earned income, using both difference in differences and individual fixed effects models. Because beneficiaries acquired land at different times, the heterogeneous effect of additional years of land ownership is investigated. The findings suggest that the program is successful in increasing beneficiaries' agricultural production and earned income, but only after four years of land ownership. Once the repayment of the loan is taken into consideration, however, the benefits of the program largely go to making debt payments and improving the net worth of the beneficiaries rather than to raising current household expenditures. Chapter three evaluates the impact of the Brazilian National Land Credit Program on the heights of beneficiary children. Making use of a family fixed effects model, the program is shown to significantly increase the height for age z-scores of beneficiary children exposed to parents' land ownership in the second, third or fourth year of life. The increases in height for age z-scores are likely attained through the mechanism of increased nutritional security with the acquisition of land through the program.