Stability of U.S. Consumption Expenditure Patterns: 1996–1999

Abstract

A cornerstone of macroeconomic analysis since publication of Keynes's General Theory in 1936 has been a strong belief in a stable aggregate consumption function. At a micro level, there has been an equally strong belief in invariant individual tastes and preferences. The usual approach in testing for structural stability is to examine consumption, expenditure, or demand functions estimated over different time periods for evidence of changes in marginal propensities to consume, price and income elasticities, and other parameters. This paper takes a different approach. Rather than analyzing stability (or its absence) in terms of invariance in behavioral parameters (i.e., the coefficients in consumption, demand, or Engel functions), the focus is on direct relationships amongst exhaustive categories of U. S. consumption expenditure, using household expenditure information from the on-going quarterly BLS Consumer Expenditure Surveys. Sixteen quarters of data for 1996 through 1999 are analyzed. The results provide strong empirical evidence in support of structural stability in underlying consumption relationships that account for about 85 percent of the variation in U. S. consumer expenditure. Some (speculative) thoughts relating this structural stability to common underlying cultural and genetic factors are offered in conclusion.