This dissertation, as a partial fulfillment of the requirements for the degree of Doctor of Philosophy in the dual-title program in Agricultural, Environmental and Regional Economics and Demography, is comprised of three essays that offer empirical investigations into issues of high prevalence of childhood nutritional deprivation in Nepal and into the ever-expanding off-farm employment of farm families in the United States. Essays are titled: (1) Maternal human capital and childhood stunting in Nepal: a multilevel modeling approach, (2) Community environmental contexts and childhood underweight status in Nepal: a multi-level modeling approach, and (3) Government farm
payments and off-farm labor response of principal farm operators: a spatial analysis of U.S county-level data.
The first essay examines the influence of maternal human capital: maternal education, maternal health and community maternal education on childhood stunting, a form of long-term child nutritional deprivation, in Nepal. Community maternal education that captures community spillover effects of maternal education is a new innovation adopted in this study of child nutritional deprivation. This study adopts the multi-level modeling approach to account for the unobserved heterogeneity of the childhood stunting outcome at the household and community levels. Data for this study are from the Nepal Demographic and Health Survey (NDHS), 2001. Maternal own education is negatively related to childhood stunting, especially with higher levels of education. Interestingly, the spillover effect of community maternal education stands out to be negative and robust on the childhood stunting outcome, even if mothers of children are uneducated. Further, results provide evidence of intergenerational transmission of genetic endowment from mother to child: mother’s height is negatively related to childhood stunting, regardless of mother’s educational attainment and place of residence. Key policy implications drawn from the findings, among others, are that the public health policy of alleviating long-term nutritional deprivation among children should further emphasize (1) promoting education among women and mothers, and (2) not only child nutritional health but also maternal long-term health, especially in geographically-disadvantaged areas.
The second essay also utilizes the multi-level modeling approach applied to nationally representative NDHS data. This study focuses on community environmental contexts, especially season and altitude of current place of residence as potential determinants of childhood underweight (short-term nutritional deprivation) among preschool-age children in Nepal. These factors have been largely overlooked in past studies on childhood underweight status. Results provide unequivocal evidence of seasonal variation in childhood underweight status. The odds of children being underweight during the unfavorable season are much higher than during the favorable season. Wealthier households appear to be able to better cope with the environmental stresses associated with unfavorable season. Overall, the caste/ethnicity of household heads is found to moderate the influence of altitude on childhood underweight status. However, in urban communities, the odds of children being underweight significantly decrease with altitude.
Findings suggest that public health policies towards achieving the Millennium development Goals of alleviating short-term childhood nutritional deprivation in Nepal should account for seasonal variation in prevalence of childhood underweight status. Public health and food security interventions targeting child nutritional health should be concentrated in pre-harvest months. Similarly, lower altitude urban areas where underweight prevalence is higher should be the focus of public health and food security interventions.
The focus of the third essay is on participation and scale effects of U.S. government farm payments in total and by type on off-farm employment decisions of U.S. principal farm operators working off-farm for at least 200 days a year. Additionally, this paper for the first time ascertains if there exists a spatial dependency in off-farm employment decisions in the U.S. The paper adopts a recently-developed technique, the Feasible Generalized Spatial Two-Stage Least Squares (FGS2SLS) estimator, to address a system of equations with spatial lagged dependent variables and spatial dependency in error structure as well as endogenous explanatory variables. The county-level data are primarily from the 2002 U.S. Census of Agriculture. Results provide evidence of spatial dependency of off-farm employment among principal farm operators in the U.S., suggesting the need to correct for spatial bias in any off-farm employment studies in the U.S. Overall, participation and scale of government farm payments are negatively related to off-farm response of principal farm operators, but the extent of effects varies substantially between participation and scale of government payments. Further, different types of government payments have different impacts on the extent of off-farm response of principal farm operators. Interestingly, the off-farm employment impacts of government farm payments by type are associated with the magnitude of income transfer, suggesting policy consideration of strong caps on government farm payments.