Demand for Food in the United States: A Review of Literature, Evaluation of Previous Estimates, and Presentation of New Estimates of Demand

Abstract

Many findings and policy recommendations in the academic literature are influenced by published estimates of elasticities of demand for food. However, the quality of these estimates is diverse and depends on modeling choices and assumptions, including the functional form for demands, types of data used, separability structure, food defi nitions, and statistical techniques used to estimate the models. In this monograph, we make three contributions to the empirical literature on demand for food in the United States. First, we evaluate the elasticities of demand for food from previous studies using the mean absolute error in elasticity-based predictions of quantity responses to actual past changes in prices and total expenditure. Second, we estimate elasticities of demand for aggregate food products using annual and monthly data under various alternative assumptions about functional form. We evaluate how well these new estimates of elasticities of demand predict quantity responses to actual price and expenditure changes, both absolutely and compared with previous estimates from the literature. Third, we estimate two sets of elasticities of demand for disaggregated fruit and vegetables: one that is conditional on the total expenditure on fruit and vegetables and a second that is conditional on expenditure on goods. To facilitate and provide context for these empirical contributions, we begin the monograph with a succinct statement of the relevant theory that underpins demand models, some specifi c discussion of separability and aggregation assumptions and their implications for the interpretation of demand elasticities, and a review of issues more generally that arise in empirical demand analysis.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 67 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.