As the United States was hit with COVID-19, Congress passed the $2.2 trillion Coronavirus Aid, Relief, and Economic Security Act in an attempt to soften the blow from widespread lockdowns and business closures that led to soaring joblessness.
The CARES Act—which included one-time cash payments of $1,200 or more to households starting in April—bolstered incomes and spurred spending as promised, although the effect was uneven, suggests research by Northwestern’s Scott R. Baker, Columbia’s R. A. Farrokhnia, University of Southern Denmark’s Steffen Meyer, Columbia’s Michaela Pagel, and Chicago Booth’s Constantine Yannelis. The researchers conducted an almost real-time review of how a significant slice of the population used the direct payments.
The stimulus prompted an immediate, general uptick in household spending, the researchers find, but households with cash on hand tended to save their stimulus checks, while those without cash on hand spent almost half their checks within 10 days.
The researchers tapped newly accessible data from SaverLife, a nonprofit organization that helps families develop long-term saving habits. The SaverLife data provided detailed, high-frequency information including day-to-day inflows, outflows, and balances of anonymized individual bank accounts. This enabled the researchers to analyze the impact of the CARES payments on households, taking into account changes in overall income level, cash flow, and existing liquidity.
Using data on more than 6,000 US households for April, the researchers calculated households’ marginal propensity to consume―the proportion of every dollar received that they spent from the moment they received the CARES payments.
“We wanted to understand the multiplier effect of CARES payments―how when the government gives you a dollar, you spend it and effectively give someone else a dollar, who then goes on to spend it, giving someone else a dollar, and so on,” Yannelis says. “This is how fiscal stimulus works, so you have to look at people’s marginal propensity to consume to assess the multiplier effect.”
The researchers find a sharp and immediate response as payments started hitting bank accounts. Within the first 10 days, households spent an average of 29 cents from every dollar received. The bulk of this spending was on food, rent, and bills, most likely in response to the shelter-in-place directives and supply-chain restrictions. The spending couldn’t significantly benefit the restaurant, services, and hospitality industries because they were largely shut down to slow the spread of the pandemic.
Moreover, household spending patterns were driven largely by access to cash, the researchers find. Families that already had more than $3,000 in the bank didn’t spend a cent of the CARES payments, while households with $500 or less spent almost half of every CARES dollar within 10 days.
Since household liquidity determined the policy’s effectiveness, the researchers say that more-targeted payments could have caused a more powerful multiplier effect. However, the government doesn’t have access to information about household liquidity, they note. And while Congress could have narrowed the eligibility parameters—by not making payments to households with higher incomes, while increasing cash to those at the bottom—that has the potential to distort behavior by discouraging people from saving, Yannelis says.
“Assuming that policy makers are using these cash payments as a means of fiscal stimulus to spur movement in the economy, the findings suggest that CARES payments are something of a blunt instrument,” Yannelis says. “They are untargeted and do not discriminate between households and recipients.”