Nutrition And Dietary Diversity In East African Pastoralist Households: Mental Accounting, Market Failures And Intrahousehold Allocation
In this thesis I explore the determinants and intrahousehold allocation of nutrition in pastoralist households in northern Kenya and southern Ethiopia. Little economic research has been done on pastoralist households, especially on these topics. This work helps to fill that void, while also testing some of the common assumptions made in modeling the relationship between nutrition and income more generally. In Chapter 1 I review some of the literature that provides background or motivation to the questions of this thesis. In Chapter 2, I explore the nutrition-income relationship for pastoralist households in East Africa. Previous estimates of income elasticities of nutritional demand have ranged from zero to close to unity. However, these estimates are always based on nutrition?s relationship with total income. One possible explanation for this wide range is that nutrition may respond differently to different sources of income if, for example, agents engage in ?mental accounting?, the practice of treating distinct income sources as not fully fungible. Estimating income-nutrition elasticities with total income may mask these differential responses and result in very different income elasticity estimates depending on which income source changes. I find that differential nutritional responses across income sources do exist among the pastoralist households studied. Possible explanations for this result are market failures for certain commodities, intrahousehold bargaining and mental accounting. Tests show that neither markets failures nor intrahousehold bargaining fully account for the differential responses observed. Thus it appears that mental accounting indeed plays some part in explaining the nutritional patterns evident in this sample. In Chapter 3, I explore intrahousehold nutritional allocation in the pastoralist households studied. A number of previous studies have compared income or price elasticities of resource allocation across demographic cohorts as a way of inferring intrahousehold welfare disparities (or the lack thereof). However, elasticities have very different welfare implications depending on the direction of income or price changes and thus cannot be used to make definitive welfare comparisons. To control for this problem I estimate cohort-specific income elasticities separately for when income is below or above the household?s intertemporal mean income. Statistical tests show that individuals do exhibit asymmetric responses to changes in above- and below-mean income. Furthermore, I find that household heads appear to disproportionately bear the nutritional burden when household income is below its mean while other household members appear to disproportionately enjoy nutritional gains when household income is above its mean. Stochastic dominance tests on simulated cohort-specific nutritional distributions show that adult daughters are systematically better of than other household members, sons are systematically worse off, and there seems to be little difference between male household heads and their wives.
dissertation or thesis