Skip to main content
ARS Home » Southeast Area » Auburn, Alabama » Soil Dynamics Research » Research » Publications at this Location » Publication #179213


item Bergtold, Jason

Submitted to: Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 7/24/2005
Publication Date: 7/24/2005
Citation: Bergtold, J.S., Peterson, E.B. 2005. Introducing asymmetric separability in the fast multistage demand system. In: Proceedings of the American Agricultural and Applied Economics Association Annual Meeting, July 24-27, 2005, Providence, Rhode Island. Available:

Interpretive Summary: Government policies affect the food industry in different ways. To help determine what could happen if a particular policy was passed, policy-makers sometimes rely on models that determine how these policies might change the amount of food consumed by the public. The purpose of this project was to examine a specific model, known as the FAST model, looking at processed foods, such as canned foods, deserts and baking goods. This model has the ability to represent how a person spends their income and can be used to predict changes in spending patterns on goods and services. Economists at the National Soil Dynamics Laboratory and Virginia Tech examined a number of different ways to set-up the model that would help improve its reliability in predicting changes in the purchase of processed foods by the general public. The paper provides guidelines for helping to build the model, that can be used by economists and other scientists interested in examining how people will change what they spend their money when prices of food products and/or personal income change. The FAST model with these guidelines can be used by the government to help develop policies that avoid harming the food industry or general public.

Technical Abstract: This paper determines the set of parametric restrictions required to maintain flexibility under asymmetric weak separability for the flexible and separable translog (FAST) multistage demand system. Because there is not a unique set of parametric restrictions that ensures separability and the values of the unconditional price and expenditure elasticities depend on the parametric restrictions imposed, the appropriateness of a chosen set of parametric restrictions should be tested empirically. An empirical example that illustrates how the choice of parametric restrictions affects the estimation results and the functional form of the price and expenditure elasticities is provided.