An error-correction modeling of US consumer spending: are there asymmetries?

Abstract

Purpose – The purpose of this paper is to investigate four possible asymmetries in US aggregate consumption and its major components (durables, non-durables, and services) for the period 1990-2013. Understanding the asymmetric behavior of the components is important since the impact of monetary policy on separate consumer spending categories may differ substantially. Design/methodology/approach – The authors first employ stationarity and cointegration tests to specify and estimate the long-run equilibrium relationship between consumer spending and such variables as disposable income, consumer sentiment, and the expected real interest rate. The authors then specify a structural error-correction model for each spending category to simultaneously investigate such possible asymmetries due to the ratchet effect, psychological negativity bias, interest rate effect, and varying degree of adjustment in eliminating disequilibrium defined as the gap between actual and desired spending. Findings – First, consumption and its major components all display asymmetric behavior consistent with psychological negativity bias. Second, consumer spending on durable goods also displays asymmetries consistent with both the ratchet effect and the interest rate effect. Third, non-durables respond asymmetrically to disequilibrium; consumers adjust (increase) spending on non-durables only when actual spending is below desired spending on non-durable goods. Fourth, services also respond asymmetrically to disequilibrium; consumers adjust (reduce) spending on services only when actual spending is above desired spending on services. Originality/value – This study provides new insight on the asymmetric behavior of consumer spending. The authors believe that the findings should help with macroeconomic policymaking when such indicators as income, consumer sentiment, and expected real interest rates display significant variations.