Does inequality matter for business cycles, and through which channels? The dispersion in earnings and incomes evidently varies over the cycle—but do these fluctuations in inequality simply reflect the underlying business cycle dynamics, or is the cyclicality of inequality itself a mechanism that shapes how the US economy experiences expansions and recessions?

To answer these questions, we estimate a dynamic stochastic general equilibrium (DSGE) model with heterogeneous households, which includes all the ingredients that the literature has shown to be necessary to fit the aggregate data and augment it with a stylized description of household heterogeneity encompassing heterogeneity in marginal propensities to consume MPCs; arbitrarily correlated (endogenous) income processes (and thus income inequality); and income risk, which triggers self-insurance, precautionary savings motives.

We estimate the model using (i) the same aggregate data customarily used in the DSGE literature, as well as (ii) information on the cyclical evolution of the cross-sectional standard deviation of labor income and post-tax income from the Current Population Survey and (iii) a priori information based on microeconomic studies on aspects of the model that affect the interaction between its cross-sectional and time-series behavior (MPCs and higher moments of the income distribution).

We use the estimated model to quantify the role of several aspects of inequality in the amplification and propagation of business cycles. We do so through counterfactual simulations that eliminate features of the model with heterogeneity that the literature has identified as potentially relevant for the transmission of shocks. In a “no-inequality” benchmark, for example, business cycle fluctuations are less pronounced than in the data: the standard deviation of GDP growth is 27 percent lower, while that of de-trended hours worked 36 percent lower. We show that most of the reduction in volatility in the more equal economy is due to the elimination of the steady-state level of inequality, rather than its cyclicality.

Our preliminary results suggest that the striking cyclical pattern of inequality is a reflection, more than a driver, of the observed features of aggregate business cycles. While the cyclical variation of inequality appears of secondary importance for the transmission of aggregate fluctuations, its long-run level is key. According to our model, inequality shapes business cycles mostly through the effect of risk on the behavior of savers; this risk is quantitatively mostly driven by the decline in consumption associated with falling to the bottom of the income distribution. We thus plan to investigate the role of unemployment risk and extensive margins as an underlying factor driving precautionary saving over the cycle.