Thursday, March 03, 2011

Health Care, Incentives, and Complexity

-- Posted by Neil H. Buchanan

Discussions of public policy issues are, of necessity, carried on in abstract and impersonal terms. Most economists, for example, have never experienced long-term unemployment; but they must somehow try to understand that problem, to devise ways to reduce it. Moreover, personal experiences can be idiosyncratic, rather than illustrative; and the person experiencing them can have emotional responses that might make him less objective than he ought to be.

Even so, personal experiences can also bring unseen problems into the light, and bring into focus problems that have been acknowledged but poorly understood. For example, calls to have members of Congress file their own taxes, or to have governors register their own cars with their states' DMVs, are based on the belief that there is nothing quite like lived experience to break through a person's apathy about others' plight.

Health care is one of those issues for which seeing how the other 95% live might well lead to changes in policymakers' and analysts' attitudes about what is acceptable. Set aside the question of whether we should have a universal single-payer national health plan, or a public option, or (at the other end of the spectrum) replace traditional Medicare with vouchers. Set aside also disagreements over the constitutionality of the 2010 health care law. Focus instead on the current system of health insurance in the United States.

As I tell my students in Law and Economics, no matter what one thinks about concepts of efficiency, rationality, market clearing, and all the rest, the one thing that is always worth studying is the incentive effects of various policies and practices. Some of my recent experiences have clarified for me some important issues in the health care debate, especially as they relate to incentives and people's likely responses to changes in those incentives.

I am fortunate enough to have employer-provided health insurance coverage. It is not by any means lavish, but I am given a menu of coverage options each year, with different premiums; and I have not yet been kicked out of an insurance plan for pre-existing conditions or any of the other horror stories that are all too common in our current chaotic system of health insurance. Every college and university at which I have taught has provided a similar package of choices, so this is something with which I have had ample opportunity to become familiar. (Being a Contracts professor also makes me unusually likely to pay attention to the details of such things.)

Five years ago, I experienced a scary episode while biking in Manhattan. I collapsed on the street and was rushed by ambulance to Roosevelt Hospital. As it turned out, I was fine, but they quite sensibly ran a series of tests and kept me in the hospital overnight. I was living in Manhattan at the time, and my employer had thousands of employees living in New York City. Even so, it turned out that my care was incorrectly coded as "out of network" (and thus vastly more expensive to me) -- a mistake that, once entered into the system, required Herculean efforts (and persistence) to correct. I soon embarked on an 18-month-long odyssey through the bureaucratic nightmare that is health coverage in the US.

The logic of the in-network/out-of-network distinction is not obvious, but the theory is that health insurers can negotiate lower costs with certain providers. If a patient chooses not to use those providers, then the theory is that the patient should pay more. Incentives, right?

A few years after the biking incident, I was visiting friends in snowy, icy Ithaca, New York. At one point, our car became stuck on the ice, and I walked on the ice to try to push the car. Without warning, I slipped and hit the back of my head on the ice. I have never felt such intense pain. After a few minutes, however, I gathered myself and told my friends that there was nothing seriously wrong. What was really going through my mind, however, was: "If I go to a hospital and nothing's wrong, how much money and time will this cost me, especially because I have to assume that all local providers (hundreds of miles from my home) are out of network?" Despite the risks of a possible concussion, and ignoring the entreaties of my friends, I toughed it out -- and went to sleep that night knowing that doing so could end my life. I won that bet.

Fast forward to December 13, 2010. By that point, I had been living in Ithaca for over a year. Standing in line at a coffee shop, I was overwhelmed by weakness and nearly lost consciousness. As I was spiraling downward, I kept thinking that I needed to avoid using medical services, because the hassle is just too great. Surely, I thought, I can tough this one out, too. Finally, however, I had no choice but to ask someone to call 911. The ambulance soon arrived, and I went to the ER. I was treated and released the same day.

The bills started arriving almost immediately. In the end, this event cost me about $1,200 out of pocket. I finally called my health insurer to find out why it had cost so much. I discovered that the annual deductible applies to everything, even emergency care, and that patients pay 20% of in-network expenses beyond the deductible, and 30% of out-of-network expenses. Even so, the agent was confused, because I had already reached my deductible limit, yet I was charged the full $750 deductible. After some investigation, he discovered that my plan actually had two annual deductibles, one for in-network and one for out-of-network. The ambulance service was out-of-network, so I paid the full deductible for that, plus 30%.

Back to incentives. First, why in the world is emergency care even subject to deductibles? If the idea behind deductibles is to discourage people from treating expensive things as cost-free, then surely that logic does not apply to a person who is choosing between calling an ambulance or possibly dying. It is possible, I suppose, that there are people who would happily call an ambulance for a minor issue. Can that really outweigh the disincentive for people like me, going forward (now that I know about this rule), to inadvisedly refuse to call an ambulance to save a thousand dollars? Moreover, even if one wants to discourage people from using emergency services, how is one to know about the in-network/out-of-network status of the ambulance that will arrive? Everything else during that incident turned out to be in-network, which is sensible, given that I lived there. The ambulance was not. I guess I was supposed to find out whether the ambulance was in-network and then send them away when I learned that they were not, to wait for a second ambulance.

The larger issue, however, is how incentives are supposed to work with something as complex as health insurance. The various details of the health care plans from which I could choose are available on-line, somewhere, but the page from which one makes choices lists only monthly premiums. With so many moving parts -- in-network deductibles, out-of-network deductibles, co-pays, coinsurance, exclusions, and so on -- how confident are we that people can process the information necessary to make rational choices, even if they were to click through to the details of the various plans? I only found out about many of these details after the fact, because (and only because) I was willing and able to take the time to wade through the paperwork and spend time on the help line -- talking with someone who admitted that he could not figure it out, either!

Over the last few years, my co-pays for doctor visits and medications (also subject to an annual deductible) have gone from $5, to $10, to $15, to $35. David Leonhardt, the economics columnist for The New York Times, wrote in an article yesterday that a system with no or low co-pays "leads to lots of medical treatments that don’t make people any healthier, and to huge costs." (I will have more to say about that column in a future post.) Maybe so (or maybe not), but a system with high co-pays and hidden costs encourages people to choose not to see doctors or take medications -- which also will not make people any healthier, and can lead to huge costs later. At least co-pays are a cost that people know up front. Still, however, it is not at all obvious that this is the best way to reduce over-treatment.

What is obvious is that health care costs are being shifted onto families, seemingly at an accelerating rate. I am still fortunate enough to have health insurance coverage (until, perhaps, I really need it), and I am also fortunate enough that the completely unknowable $1200 cost of this most recent episode was not financially debilitating. Harnessing the power of incentives, however, is not as simple as raising people's costs. Even for those who read the fine print of their health care plans (and the alternatives), the actual import of the various provisions often cannot be known in advance. The flipside of incentives, moreover, is perverse incentives. The current system seems to be running a surplus of those.