The Use of Expected Future Variables in Macroeconometric Models

36 Pages Posted: 16 Jul 2004 Last revised: 15 Dec 2022

See all articles by Ray C. Fair

Ray C. Fair

Yale University - Cowles Foundation; Yale School of Management - International Center for Finance

Date Written: September 1984

Abstract

A more sophisticated expectational hypothesis than is traditionally used in the specification of macroeconometric models is tested in this paper. Economic agents are assumed to use a vector of variables Z(t) in forming their expectations for periods t+l and beyond. These expectations may or may not be rational in the Muth sense. The results provide some evidence in favor of the more sophisticated hypothesis, but they are not strong enough to allow much weight tobe put on the hypothesis as yet. The evidence in favor of the hypothesis is strongest for households' response to future wages and prices in their consumption and labor supply decisions and for the Fed's response to future inflation rates.The sensitivity of the policy properties of my macroeconometric model to the more sophisticated hypothesis is also examined in the paper. The properties are not sensitive for a policy action in which government expenditures are changed. They are somewhat sensitive for an action in which personal tax rates are changed. In the latter case the properties are also sensitive to whether or not the policy action is anticipated.

Suggested Citation

Fair, Ray C., The Use of Expected Future Variables in Macroeconometric Models (September 1984). NBER Working Paper No. w1445, Available at SSRN: https://ssrn.com/abstract=328686

Ray C. Fair (Contact Author)

Yale University - Cowles Foundation ( email )

Box 208281
New Haven, CT 06520-8281
United States
203-432-3715 (Phone)
203-432-6167 (Fax)

HOME PAGE: http://fairmodel.econ.yale.edu

Yale School of Management - International Center for Finance ( email )

Box 208200
New Haven, CT 06520
United States
203-432-3715 (Phone)
203-432-6167 (Fax)

HOME PAGE: http://fairmodel.econ.yale.edu

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
29
Abstract Views
722
PlumX Metrics