Did Household Consumption Become More Volatile?

Abstract

I show that after accounting for predictable variation arising from movements in real interest rates, preferences and income shocks, liquidity constraints and measurement errors, volatility of household consumption in the US increased by 23.5 percent between 1970 and 2004. The increase was lower than that of volatility of family income.However, nonwhite households and household with less than 13 years of education, for whom there was no differential increase in income volatility, experienced significantly larger increase in volatility of household consumption. The effect of race and education remained significant even after income, working history, marital status, family size and composition were controlled for. Substantial differences in wealth and access to credit markets point to the main reason for this divide. Pretis for their excellent research assistance, and the University of Edinburgh for its financial support.

Consumption Volatility, Liquidity Constraints and Household Welfare

“…These results confirm that the differences between demographic groups shown in Figure are statistically significant: consumption volatility is 7 points higher for black and Hispanic households than for white households, and is 1.5 points higher for households whose head had less than 13 years of education. Unlike Gorbachev (), we do not find a statistically significant difference in the trends of income and consumption volatility for white and black or Hispanic households. We find support for previous findings that income volatility is U shaped: it is high at a young age, falls during the mid‐years and rises again later in life.…”

“…Volatility of family income increased significantly between 1980 and 2004 for an average household. This finding is consistent with the most recent study by DeBacker et al (2013), who use a confidential panel of tax returns from the IRS to show that family income volatility increased between 1987 and 2009, as well as earlier studies (Keys, 2008;Gottschalk and Moffitt, 2009;Gorbachev, 2011;Dynan et al, 2012), based on PSID data, who find that household income volatility increased between 1970 and 2006.…”

“…() use a confidential panel of tax returns from the Internal Revenue Service (IRS) to show that family income volatility increased between 1987 and 2009. This reinforces the finding in earlier studies (Keys, ; Gottschalk and Moffitt, ; Gorbachev, ; Dynan et al ., ), based on panel study of income dynamics (PSID) data, that household income volatility increased between 1970 and 2006 . Whether this increase in income volatility affected household welfare, however, depends on whether it led to a comparable increase in consumption volatility.…”

“…Building on Gorbachev's (2011) work, which uses an Euler equation approach to separate unpredictable changes in consumption from predictable changes stemming from observable taste shifters, interest rates, and life-cycle factors; we estimate consumption volatility as the square of the Euler equation residuals. Gorbachev (2011) addressed liquidity constraints only by estimating Euler equations on a sample of households with positive net worth, assuming that these households are unconstrained while those with zero or negative net worth are constrained. 2 In this article, we use direct information on household access to credit from the Survey of Consumer Finances (SCF) to identify the degree to which households are liquidity constrained, and predict the likelihood that individuals are constrained in our PSID sample.…”

Has Labor Income Become More Volatile? Evidence from International Industry-Level Data

“… Gorbachev (), for instance, uses the residual, unpredictable component of consumption growth as a measure of consumption risk; Shin and Solon () analyze the residual change in earnings, adjusted for macroeconomic and cohort factors. …”