Abstract
In this note, we reexamine an anomaly that appeared in estimates of the post-1973 textbook Keynesian consumption function as first reported in this journal by Shapiro (1988). The anomaly is the nearly unit value of the marginal propensity to consume and the significant negative intercept term obtained in post-1973 estimates. This result is contrary to the earlier estimates of the Keynesian consumption function and to the traditional Keynesian assumptions. We consider the possibility that the anomaly is due to omitted variable bias. By incorporating variables like the real rate of interest, the expected rate of inflation, and lagged wealth, the anomaly disappears. These variables have thus assumed more importance in explaining consumption as the U.S. economy became more volatile in the early 1970s.
Get full access to this article
View all access options for this article.
