Trickle-Down Consumption

Marianne Bertrand (Booth School of Business)

Adair Morse (Booth School of Business)

Abstract

While incomes in the lower and middle portions of the US income distribution have only been rising slowly over the last three decades, incomes in the upper part of the income distribution have risen sharply. At the same time, the average saving rate in the US has been in constant decline since the early 1980s. We ask whether these two trends are related. In particular, we ask whether rising consumption among (increasingly) richer households induces the relatively worse off to spend a higher share of their disposable income. We find evidence consistent with this, suggesting that up to a quarter of the decline in the savings rate over the last three decades could be attributed to rising top income levels. We argue against a permanent income explanation for this finding; we also fail to find evidence that this higher consumption level out of disposable income is driven by upwardly-biased biased expectations about future income. Consistent with our core finding of higher expenditure to income ratios among non-rich households exposed to higher top income levels, we find that households exposed to more spending by the rich self-report more financial distress. Likewise, in a state-year panel, higher top income levels in a state are predictive of a higher number of personal bankruptcy filings in a state. Finally, we investigate the political economy implications of our findings. Looking at both federal and state legislations, we find evidence that, holding ideology constant, legislators that represent areas where income inequality is higher are more likely to vote in favor of policies that increase credit availability or decrease the cost of credit.