During the Great Recession of 2008, the federal government tried to stoke the sputtering economy by lowering the cost of borrowing for banks. One aim of the policy was for the banks to pass on the reduced costs to their credit card customers by offering higher credit limits. The customers, the reasoning went, would then use this extra credit to spend more and help reignite the economy. Did the strategy work?
Neale Mahoney, Assistant Professor of Economics and Neubauer Family Faculty Fellow at the Booth School of Business, recently evaluated the outcome of this policy by examining data from more than 160 million credit card users during the years 2008 through 2014. This included holders from nine of the nation’s largest banks. In preliminary findings outlined in a new working paper, he suggests that the effect of this policy to drive more spending in the marketplace—known as a bank-mediated stimulatory policy—was muted at best.
The weak stimulus effect can be explained by a marked difference between the spending habits of credit card consumers with lower credit ratings, or FICO scores, relative to those with better credit records, Mahoney showed. He calculated that for every additional $1,000 of extended credit, low FICO score cardholders would spend an additional $540 over a 12-month period, while high FICO score cardholders would spend only $30.
Revolvers vs. Transactors
Mahoney showed the low FICO score cardholders with credit ratings ranging 660 and below behave as “revolvers.” They tend to keep a monthly balance and pay against it. High FICO score users—those with ratings of 740 and above—are more likely to be “transactors,” who pay off their credit card balances relatively quickly. Mahoney found there were not many cardholders with payment patterns that fell between these two tendencies.
This difference in spending habits highlights the problem at the crux of this policy: the banks have little incentive to raise credit limits to low FICO score customers because they are more likely to default on these additional loans. But the low FICO score holders are the ones who spend.
“If you are writing stimulus checks, you want to send your checks to these (low FICO score) people because they are going to go out and spend them and stimulate the economy,” said Mahoney.
Mahoney showed that reducing borrowing costs for banks led to an increase in credit limits overall, but these higher limits were most frequently passed on to high FICO score credit card holders, who hardly spent the extra money at all.
“We see these (findings) as revealing the basic microeconomic factors that throw sand in the gears of policies which try to target households through banks,” said Mahoney.
In light of the findings, Mahoney suggests policymakers revisit the intentions of bank-mediated stimulatory policy, but he stops short of making specific recommendations.
“We don’t want to say that bank-mediated stimulatory policy is ineffective. There is a broad set of objectives that the Fed has when they reduce banks’ borrowing costs,” notes Mahoney. “I think (these results) should at least get us to think about other policies to complement these or perhaps to offset them if we are really trying to increase spending at the bottom of the distribution.”
Mining Insights from Massive Data
This paper was coauthored by Sumit Agarwal of the National University of Singapore, Souphala Chomsisengphet, of the Office of the Comptroller of the Currency, and Johannes Stroebel of New York University. It expands on Mahoney’s previous work about the effectiveness of the government’s fiscal policy on consumer financial products following the Great Recession.
In a paper on the impact of the 2008 legislation known as the CARD act that was published earlier this year in The Quarterly Journal of Economics, Mahoney and the same coauthors showed that the policy was effective at reducing overall borrowing costs for consumers. At the time the CARD Act was passed, there was broad concern that the mandated reduction in fees would simply be offset by the banks in other ways, such as by charging higher interest rates. Mahoney showed that this was not the case and that the Credit Card Act saved consumers approximately $11.9 billion per year in extra costs.
In that paper, Mahoney drew on the same universe of credit card holder data as the current work on bank-mediated policy. Mahoney sees these papers as just the starting point of investigating the massive data set of credit card holders over the course of the 2008 financial crisis.
“I hope this will be a series of interesting, potentially important and definitely fun papers on credit cards, and, more broadly, on consumer financial products in the United States,” said Mahoney.
— by Jennifer Roche