Providing political coverage for this year’s Kenneth Arrow Lecture on moral hazard in the health insurance industry is a difficult task because everyone in the room seemed to be convinced that healthcare is not a political issue. This year’s speaker, MIT economics professor Amy Finkelstein, spent her time outlining what we’ve learned about moral hazard since Arrow’s own groundbreaking work in 1963. By Finkelstein’s own admission, Arrow “solved the problem” of moral hazard almost fifty years ago, and researchers spend most of their time figuring out exactly where it comes from.
Simply put, moral hazard is the idea that being protected from risk will make you change your behavior towards that risk. In theory, this can happen either before or after you get sick. A classic example is the non-smoker who picks up the habit once he is covered by insurance. But Finkelstein has found little evidence that people actually behave this way. Instead, most instances of moral hazard seem to come after one is already sick. A groundbreaking randomized study of Oregonian Medicaid recipients showed that instead of behaving more riskily, patients’ use for services went up when they were covered by insurance regardless of how sick or healthy they were. Needless to say, this costs the system money. And the raison d’être of the Oregon study shows why this is problematic; the state was too poor to give Medicaid to all who needed it, and instead doled it out in a randomized lottery. This is great news for social science methodology, but bad news for the poor and sick.
Panelist Jonathan Gruber, another MIT economist and architect of both Mitt Romney’s Massachusetts health care law and Obamacare, rightly pointed out that it’s not just patients who can behave in a hazardous way. Health care providers face distorted incentives, where the cost of treatment (and not the outcome) determines provider income. While this isn’t technically moral hazard, it is proof that, yet again, incentives matter when designing health care policy.
Healthcare isn’t just being used more in the United States; the cost of treatment is also increasing steadily. Finkelstein pointed out that spending growth is heavily tied to improvements in technology, and that as the market for these innovations has grown larger, the rate of growth has picked up. Without meaningful cost controls, having more people insured tends to increase the price of their treatments. Maybe that is why universal systems like those in Europe have tighter cost controls, mostly because the government is directly involved.
Kenneth Arrow, the Nobel Prize winner who originally “solved moral hazard” on paper, was also puzzled by why the US system is not universal while most of its OECD neighbors extend coverage to everyone. In fact all the panelists’ comments seemed to confirm that the health care problem has been ‘solved’ in their eyes, at least on paper. But Finkelstein pointed out that 15 percent of the US population remains uninsured while health care spending has risen from five to 17 percent of GDP.
Clearly, healthcare itself has not been solved in the United States, even if moral hazard has been for almost half a century. The solution has to come from the legislative process, but the political realities of the health care debate were left largely untouched by the speakers and maybe it isn’t the role of economists to pick sides in that particular trench. After all, Jonathan Gruber helped design almost the same plan for both 2012 presidential candidates (don’t tell Mitt Romney about that though). Why have the problems of moral hazard outlasted Kenneth Arrow’s groundbreaking work by so long? The answer is a political system that is prone to stops and starts and half-baked compromise solutions that solve little in the long run. As is often the case in economics when the theory does not fit the facts, economists will just have to wait for the facts to adapt.