Broadly, my dissertation focuses on changes in the midstream and downstream of the agrifood value chain in Tanzania. The first essay examines the patterns and determinants of household-level consumption expenditure on processed food and meals away from home. I use a detailed food consumption diary from Tanzania to explore the relationship between the budget share spent on more convenient foods, such as highly processed food and food away from home, and income levels. Additionally, I use (i) geo-spatial data to analyze how these relationships change over space, and (ii) detailed labor data to analyze the correlation between men's and women's non-farm labor force participation and the budget share spent on higher value-added foods.In my second essay, I revisit the old debate of whether the poor pay more for food, using the same spatial and food diary data as in the first essay. I find that, surprisingly, the poor generally are not more likely to buy in smaller quantities, the rich are not more likely to buy non-perishables in larger quantities, and that bulk discounts are modest at best for most food products we study. Most intriguingly, we find that the poor do not pay more than richer households.Finally, my third essay uses primary data from maize flour retailers to explore the modernization of the maize flour value chain in Tanzania. I use various measures of value chain structure, conduct, and performance, and I disaggregate by retail type (traditional shops, transitional mini-supermarkets, and modern supermarkets) and town size, to study where changes are occurring. I find a rapid proliferation of maize flour brands, a move toward disintermediation (especially in the secondary cities) and longer supply chains, and an emerging adoption of mobile money by traditional shops in smaller towns. These findings point toward a supply chain in flux.
Read
