UC Berkeley

It has become increasingly clear that the relationship between agricultural productivity and the environment is highly non-linear. In this dissertation I evaluate yield measures commonly used to quantify agricultural productivity and demonstrate that the largest shocks, damages causing crops to become unharvestable, are omitted from the most commonly used productivity measure. I find that these losses are highly non-linear, and that infrequent but extreme damages account for substantial losses to global agriculture yet are often ignored. Accounting for these non-linearities, I examine the impact of climate (temperature, rainfall) and environmental condition (flood, drought) induced changes in agricultural productivity on human health (mortality rates) and the economy (farmer loans). I find that extreme climate conditions have significant impacts on agriculture and that losses in agricultural productivity, in turn, increase mortality rates and the amount of money borrowed by farmers in Southeast Asia. These findings have implications for our understanding of the broader importance of climate vulnerability in agriculture and how governments might value climate adaptation options for agriculture.