Thursday, November 11, 2010

Thorny Issues of Insurance Availability and Renewability

If the individual mandate gets struck down, as I argue could well happen in my post below, then other mechanisms will be needed to avoid the more egregious problems that tend to occur in health insurance markets.

I highly recommend two articles that address some of the issues in thought-provoking fashion: Cochrane, "Time Consistent Health Insurance," Journal of Political Economy 1995; and Patel and Pauly, "Guaranteed Renewability and the Problem of Risk Variation in Individual Health Insurance Markets," Health Affairs, Aug 28, 2002.

The Cochrane piece casts health insurance as having two components: the insurance of health expenses in a period, and then the insurance against premium increases in the future. His basic argument is for time consistent insurance contracts that get "marked to market" each year. If an insured has gotten sicker during the year, so that their risk-based premium increases for the future, then the "second" insurance contract pays an amount equal to the present value of those premium increases. Presto -- the consumer has the money to pay the increased premiums. The problem of having insurance companies raise premiums when you get sick has disappeared, through the magic of mark to market contracts! (As I note below, symmetry requires the consumer to pay money to the insurer if the consumer's health improves over time, but Cochrane shows that the cash flow implications of this can be easily dealt with.)

The Patel and Pauly article argues for a similar in spirit but practically different arrangement to prevent the problems of premium increases upon illness -- the idea of guaranteed renewability, and at a price that does not reflect any changes in health. So long as you stick with your existing insurer, they must renew you each year, and at prices that do not reflect your benefits history. In fact, as Patel and Pauly show, most states regulatory schemes already required this, and historically much insurance had this long term feature (see below my point on term life insurance).

Both of these papers show that one thorny issue, the prevention of having health insurance premiums skyrocket in price upon becoming sick, can be relatively easily solved, with a minimum of regulation. But in reading these papers, many ideas and concerns come to mind.

Think of your own term life insurance, if you have such insurance. I do, and it is guaranteed renewable for a term of 20 years. The price each year goes up only with my age -- an event that nobody can insure against! So I am protected against my insurer raising my price if they were to discover that I had a heart attack. Am I in some sense "stuck" with my company? Yes, at least if I am no longer healthy, because it would be hard for me to switch insurers at that point. But at least I still have my existing insurance. Is there the possibility that another company will try to induce healthy people away from their existing carriers with policies that reflect their lower risk? One would think so, but I am hardly inundated with such offers. One suspects that perhaps those kind of selection problems are not as serious as the textbooks make them sound.

A concern I have with the Cochrane idea is the ability for consumers and insurers to reap the benefits of specific investments in health. Suppose I do all I can to improve my health -- exercise, diet, all risk factors. Cochrane dismisses the role of behavior on health, but the thinking on that has evolved. Just look at obesity and diabetes. Much of my investments in health will be unobservable to others. As my insurance policy only gets marked to market on the basis of observable risk factors, I am left with no reason to improve my health. In fact, since the contract requires the consumer to pay money to the insurer if they get healthier (remember the consumer gets money from the insurer if they get sicker) I can envision negative incentives to invest in one's health. Of course, existing insurance policies also fail to give me much incentive to increase my health. And if an insurer were to spend money on consumers to improve their health (think employer-based insurance) there is no way for the insurer to recoup those investments if the consumer switches carriers. One yearns for a mechanism to make the new insurance carrier pay the old one for consumer-specific investments in health.

Neither the Cochrane or Patel/Pauly ideas covers the problem of people gaming an insurance system by not buying any insurance until they are sick. Cochrane might respond with, "well, they pay their risk-adjusted price at that time". The problem with that is that society will not tolerate the outcome of sick people facing prohibitively expensive insurance.

If the individual mandate is struck down, and the problem of gaming the system is quantitatively important to the overall insurance market, then there has to be some significant cost to people who wait until they are sick to buy insurance, enough to make such behavior unlikely in the aggregate. And this needs to be done without a mandate to buy insurance enforced via a penalty for not buying it. A reasonable approach might be a low quality Medicaid type insurance policy priced at a significant portion of income for those who have not maintained health insurance, public or private, over their lifetime. Since children are now on their parents' policies until 26, most children will automatically have continuous coverage through that age. All they would need to do at 26 is buy a longterm policy a la Cochrane or Patel/Pauly and maintain it until they hit Medicare age.

The Individual Mandate and Insurance Markets

I believe there is at least an even chance that the Supreme Court will strike down the individual mandate in the new health care bill. Here is a paragraph from the Court's decision in US v. Lopez, the first case in decades that did limit the powers of Congress under the commerce clause of the Constitution:
We pause to consider the implications of the Government's arguments. The Government admits, under its "costs of crime" reasoning, that Congress could regulate not only all violent crime, but all activities that might lead to violent crime, regardless of how tenuously they relate to interstate commerce. See Tr. of Oral Arg. 8-9. Similarly, under the Government's "national productivity" reasoning, Congress could regulate any activity that it found was related to the economic productivity of individual citizens: family law (including marriage, divorce, and child custody), for example. Under the theories that the Government presents in support of § 922(q), it is difficult to perceive any limitation on federal power, even in areas such as criminal law enforcement or education where States historically have been sovereign. Thus, if we were to accept the Government's arguments, we are hard-pressed to posit any activity by an individual that Congress is without power to regulate.

This was the case on the Gun Free School Zones Act, and the Government argued that it had the power to regulate guns in school zones under the commerce clause because: a) guns increase the cost of crime, which affects commerce among the states; and b) guns and crime affect national productivity, which also affects commerce. The Court states quite clearly that these arguments would justify any police power at the Federal level and therefore cannot be right.

In terms of the individual mandate, the parallel argument would be that if inaction -- failure to buy a product -- is considered to be reachable under the commerce clause, then the regulatory power of the Congress would be expanded into a region that would further make the commerce clause infinitely powerful. In Lopez, the Court was willing to draw a line in front of an activity that was not sufficiently related to interstate commerce. I can certainly see another line being drawn, a very bright one it would seem, at the distinction between inactivity and activity.

If the mandate is struck down, then some creativity will be required in insurance markets in order to maintain some of the accomplishments of the health care bill, in particular the avoidance of universally-hated insurance policies such as restrictions on pre-existing conditions and indiscriminate premium increases. My next post will get into some of these issues.