Quasi-rational expectations, an alternative to fully rational expectations: An application to US beef cattle supply

The rationale for, and difficulties of estimating rational expectations (RE) models of dynamic economic behavior are contrasted with quasi-rational expectations (QRE) models. A multivariate model of US beef cattle supply, derived from dynamic optimizing behavior, is estimated using QRE. In contrast to the recent influential study of Rosen et al. (1994), our model takes adequate account of the complex and changing structure of the industry, particularly of the shift of finishing on ranch and farm to feedlots, and of the way in which the USDA generates the statistical time series used. Our analysis is based on quarterly and monthly data since 1944 and permits a detailed treatment of the dynamic structure of beef production. Tests for QRE against a limited structural form of FRE are presented.