MIT Economist Seeks Facts in Health-Care Policy DebateJeanna Smialek
There’s no shortage of theories about health-care policy. Amy Finkelstein wants facts.
Her newest research, based on data from Oregon, builds on her agenda: measuring the effects of health programs with scientific rigor. She and colleagues found that Medicaid coverage increased emergency department visits by 40 percent, according to the latest findings in their continuing study, released yesterday in the journal Science.
“We’ve shown that there are real benefits but also real costs to Medicaid,” said the Massachusetts Institute of Technology economics professor. “My fervent hope, and I don’t think it’s entirely naïve, is that this will lead to a more informed public-policy discussion.”
Finkelstein’s pursuit of policy-relevant evidence has placed her at the pinnacle of her field. When the American Economic Association awards committee selected her for its 2012 John Bates Clark Medal, it called her “the leading scholar in health economics.” Winners of the medal, awarded annually to an economist under the age of 40, include 12 Nobel Prize laureates and two White House chief economists.
“People get excited about the research that she’s doing, the way that she tackles problems,” said James Poterba, president of the National Bureau of Economic Research and Finkelstein’s colleague at Cambridge-based MIT. “She is one of the leading health economists in North America, in the world.”
Finkelstein’s love of hard evidence started young: She is the daughter of Ph.D. biologists. She was more interested in using the world as her laboratory, so she studied government at Harvard University, graduating with a bachelor’s degree in 1995. In an interview, she said she wanted to synthesize history, her favorite subject, with how society operates -- yet the lack of proof frustrated her.
“My dad would read my term papers and say, ‘Sure, that’s a nice argument, but I could just argue the other side,’” said Finkelstein, 40. After taking former Labor Department chief economist Lawrence Katz’s “Social Problems of the American Economy” her senior year, she found her niche.
“I realized I was more interested in what the impact of a policy is, rather than how it came to be enacted,” she said. “I had drunk the Kool-Aid at that point and wanted to do economics.”
Finkelstein completed a master’s degree at Oxford University on a Marshall Scholarship, which sends recipients for graduate study in the U.K. From 1997 to 1998, she worked as a staff economist at the White House Council of Economic Advisers under then-chairman Janet Yellen, nominated to be the first female head of the Federal Reserve in its 100-year history.
“If you were willing to work hard and had something to contribute, you could work on anything,” Finkelstein said of serving on Yellen’s staff. After trying out a range of topics, she developed an interest in what she calls her “first love”: insurance markets.
“This seemed to be an area where private markets might not work well, where there might be scope for welfare-improving public policy,” she said. “There was interesting data and interesting empirical work to be done.”
She said she chose MIT to pursue a Ph.D. because Poterba was reputed to be “the world’s best adviser,” and earned her doctorate in 2001. She is the third-generation woman in her family with the degree, along with her mother and grandmother, who earned a Ph.D. in comparative literature in Poland.
She became an assistant professor at MIT in 2005 and full professor in 2008. That same year, she was named co-director of the NBER’s Public Economics Program and co-editor of the Journal of Public Economics, positions she still holds.
Much of her recent research has focused on Medicaid, public health insurance for the poor. She and co-authors, including Harvard’s Katherine Baicker, have been analyzing Oregon’s expansion of the program since 2008.
The state had 10,000 additional Medicaid slots available, and used a lottery to select from a waiting list of 90,000. That provided a once-in-a-lifetime opportunity to isolate the impact of newfound coverage, Baicker said. It isn’t ethical to create an experiment that denies health insurance to a control group, and trying to measure the effect of coverage without such a comparison clouds the findings.
Results from the first year, published in 2012, show that Medicaid increased health-care use among recipients. It also basically eliminated catastrophic health costs, according to research published in 2013. Medicaid didn’t make a measurable difference on blood pressure or cholesterol, though positive screenings for depression fell. Both studies received widespread media coverage.
“The headlines are almost like a Rorschach test of people’s ideological bents,” Finkelstein said. “No one is arguing about whether what we found is what we found, they’re arguing about what it means.”
The newest Oregon research, which included 25,000 low-income adults, showed that new Medicaid coverage led to more frequent emergency-room visits, which meant an extra $120 in annual spending per covered individual.
The evidence refutes a “widespread hope that expanding Medicaid coverage might reduce use and potentially save money,” Baicker said. While no research can be perfectly generalized, she said this is the “first-ever” randomized trial examining Medicaid’s effect on emergency-room use. In randomization, the only difference between two groups is the variable being evaluated.
Promoting such controlled studies in economics is a priority for Finkelstein, along with using research to bolster or refute theories about health care.
She’s found evidence to support the idea that long-term-care insurance markets are impaired by so-called asymmetric information: people knowing more about their own risk than companies, and those who are higher-risk being more likely to buy insurance. Her research showed the effects are balanced out as people who are more risk-averse and less likely to use the insurance also enroll.
Many politicians and the press approach a policy as if “there are many different opinions” about what it does, “but at the end of the day, there is a right answer to these questions,” said Raj Chetty, a Harvard economics professor who has written studies with Finkelstein and received the 2013 John Bates Clark medal. “What Amy is doing is helping to foster that type of work.”
Finkelstein’s keen interest in economics has helped shape her personal life. In her second year of graduate school at MIT, she met then-Harvard graduate student Ben Olken at a lecture by Steven Levitt -- who would go on to publish the bestseller “Freakonomics.” The pair struck up a conversation about economics and carried on the banter via e-mail.
“I finally bit the bullet and said, ‘Can we continue this discussion? Are you free for dinner Friday night?’” Olken said he could do lunch Thursday. “Which, you know, was quite a downgrade,” Finkelstein said. “But he quickly realized the error of his ways and recovered from there.”
Finkelstein and Olken, now also an MIT economics professor, married in 2005 and have a 6-year-old son and 3-year-old daughter.
Even as she raises a family and pursues her research, Poterba said Finkelstein dedicates time to helping MIT’s future economists shape their careers. Heidi Williams, a fellow health economist, and Nathan Hendren, whose thesis built on Finkelstein’s research, are among her proteges.
“She’s incredibly dedicated to her students; she read countless drafts of my thesis,” said Hendren, who is now an assistant professor of economics at Harvard.
Finkelstein also is spreading her emphasis on controlled studies. Together with mentor Katz from Harvard, she started the North America branch of the Abdul Latif Jameel Poverty Action Lab, or J-PAL, last year.
Globally, the decade-old center promotes policy-relevant, randomized research. It was founded at MIT by professors Esther Duflo, Abhijit Banerjee and Sendhil Mullainathan, now at Harvard, and has regional offices and affiliates worldwide.
The new branch will help develop controlled studies, train researchers, secure funding and disseminate results that focus on U.S. and North American policies, Katz said.
J-PAL champions Finkelstein’s goal of promoting the kind of empirical research in economics, particularly health economics, that’s standard in other sciences -- such as the biophysics her father still teaches at the Albert Einstein College of Medicine in New York.
“I like ideas, and the art and science of trying to figure out which ideas make sense, theoretically and empirically,” she said. “Through randomized evaluations, we can create important and powerful lessons.”