Essays on the Estimation of Food Supply and Demand

Abstract

I develop and apply structural methods for estimating supply and demand responses of food products and agricultural commodities. Chapter 1 adapts recent developments in dynamic discrete choice econometrics to estimate a model of land use change. The empirical approach is relatively easy to implement, relying on linear regressions. I apply the empirical approach to US crop supply using satellite sensor data. Taking dynamics into account leads to considerably larger long-run elasticities, and suggests that the land-use impacts of biofuels production are larger than previous studies have found. Chapter 2 considers the intensive margin in crop supply responses. I derive and implement an indirect approach to estimating yield-price elasticities which is considerably more precise than the standard approach. I find that yields for US corn, soybeans, and wheat respond very little to short-run price variation, suggesting that extensive crop supply responses (i.e., land use change), is the dominant component of crop supply responses, at least in the short run within the US. Chapter 3 analyzes the fluid milk market, including non-dairy milks like soy milk, as a case study in understanding the potential for greenhouse gas mitigation through dietary change. I estimate a random coefficients model of demand for milk products using retail data, and find that taking unobservable heterogeneity into account suggests that there is substantially less substitution between milk and soy milk in response to price changes. A static demand model cannot explain long-term changes in milk consumption patterns, for soy milk consumption rose dramatically in the early 2000's without any substantial changes in relative prices.