Social Security Benefits, Consumption Expenditure, and the Life Cycle Hypothesis

Abstract

This paper examines the impact of changes in social security benefits on aggregate consumption expenditure. Under the null hypothesis, there should be no contemporaneous effect at the monthly frequency because increases in benefits have always been announced at least 6 weeks prior to payment. The paper develops overwhelming evidence--contrary to the null--that benefits have affected aggregate spending. The results have strong implications for several important issues, including Ricardian equivalence, government policy irrelevance, and the excess sensitivity of consumption to changes in income.