| This dissertation examines the labor and time allocation behavior of small-holding farm households in Kenya and Tanzania.;Chapter 1 uses an agricultural household model to test for separation of a farm household's labor usage from the labor supply of its resident household. Applying this test to a data set of 700 Kenyan households, I find that both household size and household composition are strong determinants of family farm labor usage. Both findings are robust to a variety of empirical specifications. This rejection of separation implies that modelling the consumption side of a peasant household's behavior and assuming the production side to be perfectly competitive may be improper and lead to erroneous conclusions.;Chapter 2 posits a two period model of educational investment decision-making by farm households. I test whether the ability of households to freely hire in and hire out labor affects the amount of schooling children receive. Using a Kenyan survey of rural households (the same survey as in Chapter 1), I find that the amount of land a household farms is negatively correlated with the level of schooling obtained by the household's children. This finding implies that human capital investment decisions are not independent from farm production decisions, and is contrary to the theory of producer-consumer separation.;Chapter 3 analyzes the labor allocation behavior of "mostly-autarkic" farm households throughout Tanzania. I find that, in general, farm labor markets are not well-functioning, but that the presence of communal farming and efforts by Village Councils to allocate landholdings among households does have some positive effect upon the workings of private labor markets. |
