U.S. consumers hold trillions of dollars in debt, but they hold it very unevenly. “If you work with credit data and outcomes in the U.S., you see this tremendous variation both across states and within states,” says Paul Goldsmith-Pinkham, an assistant professor of finance at Yale SOM. “There are a lot of reasons for that, but one interesting thing is how variation in health insurance looks similar to variation in debt.”
The question is: are the geographic differences in health insurance coverage and the geographic differences in debt related? And if they are, does that mean that greater access to health insurance can help ease consumer debt? Goldsmith-Pinkham examines the connection between health insurance and debt in a recent working paper, co-authored with Maxim Pinkovskiy of the Federal Reserve Bank of New York and Jacob Wallace of the Yale School of Public Health.
“In an ideal world, you’d design a randomly controlled trial,” Goldsmith-Pinkham says. Such an experiment would assign who gets insurance and who doesn’t and then, down the road, measure the effect on their debt levels. But the universal availability of Medicare offered a reasonable substitute. “We know in this case that the policy takes hold at a specific age—65—and that the people on each side of this line are comparable.”
By pairing demographic and health insurance information from the American Community Survey with credit reports from Equifax, the researchers discovered that a sharp jump in insurance coverage among 65-year-olds corresponds with a sharp decrease in collections debt at exactly the same time. Interestingly, other financial outcomes, like bankruptcy and credit scores, do not have the same tight association with insurance coverage.
The average reduction in debt among all 65-year-olds is roughly $28, but, Goldsmith-Pinkham notes, the size of this number is misleading. “In fact, the share of people at that age who have debt collection is only about 5%,” he says. “If you start talking about the actual reduction for those who have debt, you multiply by 20 for a back-of-the-envelope calculation and get more like $600 per person.”
A closer look at the data revealed a number of notable details. One is that the reductions in collections debts varied widely by geography. In San Francisco, for example, the average per capita decline in collections debt was $8; in Raleigh, North Carolina, where gains in insurance coverage were comparable to those in San Francisco, the average per capita decline was $53. More generally, across the southern United States, the availability of Medicare has uniquely strong effects on collections debt.
When the researchers broke the data out into 741 “commuting zones,” they found that the effect of Medicare on collections debt is more pronounced in regions with greater shares of Black residents, people with disabilities, and for-profit hospitals.
“This suggests there are big things going on above and beyond just getting health insurance,” Goldsmith-Pinkham says. “There could be other dimension to the value of Medicare. Perhaps, for example, it gives people the chance to switch from a bad private plan to a good public one.”
Finally, the researchers found that those with the greatest debt benefitted the most from Medicare. Half of 65-year-olds who owed $10,000 or more moved out of this debt class once Medicare became available; among 65-year-olds who owed $1,000-$2,500, fewer than 1% moved into a grouping with less debt. To be fair, Goldsmith-Pinkham notes, only a tiny share of 65-year-olds have more than $10,000 in debt—roughly .2 percent of the population in their study.
“But, nonetheless, there is an element to which these long tails are getting brought in: Medicare seems to provide significant relief to the small groups of people who owe thousands and thousands of dollars,” Goldsmith-Pinkham says. “We can’t say exactly where they move to because of statistical noise, but we can say the program improves the very worst state of world.”
So Medicare seems to reduce the debt burden for individuals. Does it reduce costs for us as a society? If Medicare eligibility were, say, pushed back from 65 to 60, or 55, would there be overall financial savings in the system? Unclear.
“In this case, when we give insurance to people, the government is paying for it,” Goldsmith-Pinkham says. “The question becomes—and our paper can’t answer this—if we’re moving money around to some extent, then are we doing it in a more efficient way, and so actually creating savings? Or are we seeing more of a redistributive effect, one that is more equity-based, where the burden of payment gets shifted?”