This article reports tests of aggregation over consumer food products and estimates of aggregate food demand elasticities. Evidence that food demand variables follow unit root processes leads us to build on and simplify existing tests of the Generalized Composite Commodity Theorem. We compute food demand elasticities using a method of cointegration that is shown to apply to a convenient but nonlinear functional form. Estimates are based on consumer reported expenditure data rather than commercial disappearance data. Copyright 2005, Oxford University Press.

No abstract

No abstract

This article compares estimates of disaggregated market food demand responses to the Supplemental Nutrition Assistance Program benefits based on exact nonlinear aggregation to responses based on linear aggregation. By accounting for income inequality, nonlinear aggregation implies that only the households that receive benefits contribute to market demand responses. In contrast, linear aggregation presumes all households receive benefits and thus contribute to the market demand response. The consequence is that nonlinear market estimates are smaller than the linear estimates by roughly the fraction of households that receive benefits.

No abstract