Household Cash Buffer Management from the Great Recession through COVID-19

Introduction

Cash balances are a linchpin of household financial health. Maintaining a sufficient level of liquidity—which varies with spending needs—allows households to manage spending through short-run changes to income. Over time, savings in the form of cash can also be used for large expenditures or wealth-building investments. The dynamics of checking account balances therefore reflect both consequences of external shocks affecting consumers—e.g., job loss, fiscal stimulus, or unexpected expenses—as well decisions of when and how to save.

COVID-19 brought economic volatility that led to historic swings in the savings rate from a record high in 2020 to very low levels last year before rising steadily in recent months, reflected in trends in cash balances seen in the most recent JPMorgan Chase Institute [Household Pulse report]. In addition to measuring cash liquidity over a longer time horizon, this report analyzes within-household cash management behavior to help make sense of recent fluctuations. We use a sample of de-identified data covering 19 million individuals from 2008 through early 2023.

Our focus is on cash buffers (liquid balances scaled to expenses), interpreted as the amount of time a person could continue their rate of spending without receiving inflows or accessing new credit. The research questions we target in this report are:

1. How have median cash buffer levels evolved for different segments of the U.S. adult population since the Great Recession?
2. To what degree does individual-level liquidity tend toward a stable level over time? How fast does liquidity converge to this average following shocks?

A central finding of this report is the relative stability in cash buffers for over a decade prior to the surge in balances during the pandemic. From 2008 to 2019—a period over which the unemployment rate swung from as high as 10 percent to under 4 percent—median cash buffers stayed within a range of just a few days’ worth of spending.¹ While we also find that many individuals frequently experience large swings in liquidity, these idiosyncratic surges and dips do not normally last very long. This is because individuals tend to spend in excess of income when cash buffers are temporarily elevated and vice versa.²

The ongoing decline in the stock of cash balances that were built up in the pandemic is qualitatively consistent with pre-pandemic patterns, although the drawdown has been somewhat slower than usual, according to our estimates.

We organize the analysis around four findings: