Understanding the marginal propensities to consume of American households is important for economic policymaking, particularly when the U.S. economy slides into recession. Knowing how readily different households will consume an extra dollar of income helps policymakers design programs that could help the economy bounce back more quickly from recessions.

Differences in the marginal propensity to consume among U.S. households is fairly well acknowledged at this point, but their consumption patterns can change over the course of a business cycle. A new study shows how the marginal propensity to consume changes as the economy enters into a recession. The new paper, released as a National Bureau of Economic Research working paper, is by economists Tal Gross of Columbia University, Matthew Notowidigdo of Northwestern University, and Jialan Wang of the University of Illinois at Urban-Champaign. It looks at how a consumer’s propensity to consume changes as the economy goes from expansion to recession.

Specifically, the economists look at the difference in how much individuals who just had their “bankruptcy flag” taken off their credit records consume because of their increased access to credit. All the individuals in their sample had previously filed for bankruptcy and the flag on their credit records were removed after 10 years. The fact that the flag disappeared 10 years after going through bankruptcy lets the authors show that access to credit was sudden and outside the control of the individual.

Perhaps unsurprisingly, Gross, Notowidigdo, and Wang find that marginal propensities to consume were higher for these individuals emerging from bankruptcy during the Great Recession. These individuals’ average marginal propensity to consume was 20 percent to 30 percent higher during that sharp downturn in 2007-2009 than it was for those who had their bankruptcy flag removed during the subsequent recovery up through 2011. Importantly, the three authors show that the higher consumption rate wasn’t due to a change in the kind of people who were having their bankruptcy flag taken off. Instead, it appears to have been due to changes in their access to credit.

There are limitations to this paper. The three economists only look at the changes in marginal propensities to consume among people who just came out of bankruptcy. So these specific results are probably most applicable for individuals with low credit scores. Whether there are such significant variations for wealthier individuals or individuals with higher credits scores is a question for future research.

But if these results hold up and are valid for a larger section of the population then it should inform policymakers’ design of fiscal stimulus measures once a recession hits. Policymakers also should consider this research when setting policy for so called automatic stabilizers, such as unemployment insurance or the Supplemental Nutritional Assistance Program. Higher marginal propensities to consume during recessions may mean that targeted programs could be more effective than previously thought at boosting demand to turn around a slumping economy.