Few issues today are as heatedly debated as health care reform, which has inspired some of the most antagonistic moments in recent American politics (remember “You lie!”?). As a debate, it seems perfectly designed to push us to the brink of collective madness. Beneath the difficult moral argument about what’s fair and right lies yet another difficult argument, an irredeemably wonky one about cost, value, and economic incentives. It’s a deadly combination—the policy equivalent of texting while driving.
Considering how well-developed the moral argument is—not to mention how many billions, even trillions, of dollars are at stake in whatever we do—we know surprisingly little about the real-world effects of making changes to health care systems. The economic debate about health care has been mostly theoretical, or based on models. When it comes to what really happens to patients, doctors, and budgets when health care systems change, the evidence is shockingly sparse. It’s not easy, after all, to experiment on a health care system.
Stepping into that gap is the MIT economist Amy Finkelstein, who recently won the prestigious John Bates Clark Medal, largely for her work on the economics of health care. The medal is awarded each year by the American Economic Association to the best economist under 40, and carries immense prestige in the field. (About 40 percent of Clark winners eventually are awarded the Nobel.)
For more than a decade, Finkelstein has worked to understand the nuances of the health care system—and to add some hard data to what passes for informed debate. “Many, if not all debates in health care are argued out of opinion and anecdote,” says Amitabh Chandra, a professor of public policy at the Harvard Kennedy School. “It’s always a story about how my doctor treated me, or about what happened to my uncle when he had a heart attack. What Amy has done, and done repeatedly, is say, ‘I’m going to use lots and lots of data and give you a more realistic, complete characterization of what’s happening.’ She can give us the number that helps us weigh costs and benefits.”
Finkelstein hs found that changes often have unintended consequences—sometimes for the better, sometimes for the worse. She’s studied the way that changes in one part of the system (an expansion in the number of insured Americans, for example) can create unexpected ripple effects elsewhere (like speeding the adoption rate for new technologies). She’s tried to find out exactly how Americans benefit from being insured: It turns out that they’re not just healthier—they’re also more financially stable. And many of her most important contributions to economics have come from studying the peculiar ways in which markets for insurance differ from markets for other products. (They differ because buyers and sellers of insurance have “asymmetric” information: Patients can hide how unhealthy they really are, hoping to reap the benefits of underpriced insurance, and driving up costs for everyone else.)
Often Finkelstein’s findings don’t fit easily into the narratives our politicians offer. One strand of her research, which looks back at the expansion of health insurance after the introduction of Medicare, suggests that we might be underestimating the benefits of widely available insurance. One example: She has found that, as more patients were covered, hospitals and pharmaceutical companies used their increasingly predictable revenues to invest in innovative care, like cutting-edge cardiac operating theaters, from which uninsured patients also benefited.
At the same time, however, her work has also shown that insurance—any insurance, publicly funded or not—is more costly than many people would like to think: The more insured patients there are, the more the health care system shifts to emphasize expensive care.
Finkelstein’s findings about health care don’t often fit easily into our political narratives—a mixed outlook that one analyst calls the ‘Amy signature.’
This balanced, mixed outlook—a corrective to both sides of the political debate—is what Chandra calls the “Amy signature.” Changes in health care, Finkelstein shows, are hardly ever all good or all bad: Cardiac operating theaters, she writes, have “had an important role in both the rise in health spending and the increase in life expectancy over the last several decades.”
In her best-known recent research, Finkelstein has been studying the health care system in Oregon, where, since 2008, a “natural experiment” has been unfolding: Facing budget constraints, the state has been able to expand its Medicaid program only through an insurance lottery, in which participants end up being insured or uninsured at random. Such a lottery is a gold mine for academic research, allowing comparisons that would be impossible to achieve any other way. A forthcoming paper on the Oregon data, written in partnership with many of the country’s other leading health care economists, reveals good and bad outcomes to insuring patients. On the plus side, as one would expect, the newly insured were healthier, and incurred less medical debt. On the minus side, they used substantially more health care. “There are people who have said that health insurance will save money,” Finkelstein says, “because it will reduce emergency room visits. That would be wonderful, because then there wouldn’t be a trade-off. But our work, at least from the first year in Oregon, suggests that that might not be the case. Health insurance might have many benefits, but it’s not going to save money.”
It’s ultimately up to voters, not economists, to decide which trade-offs are worthwhile—but those decisions will be better, and perhaps even easier, if they are backed up by empirical research. Finkelstein is optimistic about the power of economic insights to add more nuance to a sometimes out-of-control health care debate. And there are many details still to fill in. “There are some things where economists know exactly what to do,” she says. “And there are other cases where we have educated guesses, or where we’re still learning.”Joshua Rothman is an Ideas columnist. He can be reached at firstname.lastname@example.org.