This dissertation focusses on three topics related to human capital investments,
microcredit and agriculture in East Africa. The first essay investigates how health shocks
affect farm productivity in the presence of microcredit. It is expected that microcredit
increases agricultural productivity by enhancing allocative and technical efficiency and by
overcoming financial constraints that reduce purchase of inputs. However, microcredit will
have competing uses in the event of a health shock to the household; hence, it is important
to investigate changes in farm productivity due to health shocks with and without
microcredit. Existing studies on the microcredit-productivity relationship do not account
for the effect of uninsured health shocks to the household. A theoretical model is developed
and empirically tested using data from Uganda. The problem of self-selection into
microcredit is addressed by use of an endogenous switching regression model. The results
reveal that uninsured health shocks lower farm productivity. However, microcredit has a
significant mitigating effect on the productivity losses. Microcredit effectively serves as
insurance against health shocks in rural areas where formal health insurance markets do
not exist. Thus, microcredit generates a double dividend in smallholder agriculture by both
improving health status of the farm population and improving agricultural productivity.