The University of Michigan
Interview with Katherine Swartz, Professor of Health Policy and Management, Harvard School of Public Health
|Harvard School of Public Health professor and ERIU researcher Katherine Swartz has spent most of the past two decades studying why people lose health coverage, the socio-economic diversity of the uninsured pool, how long people spend in spells without health insurance, and what economic policies could expand coverage to more people. Swartz has focused recently on the individual health insurance market, including how fear of adverse selection leads insurers to avoid people with high medical costs, and what policies could help keep premiums affordable. Swartz has worked to keep the issue of the uninsured on policymakers' radar screens, and she just finished a book reflecting on the
past 25 years of the uninsured quandary. She recently talked to ERIU about inequities in the individual health insurance market and what solutions she favors for addressing them.
You have been looking at some of the issues facing those who have to buy health insurance in the non-group market. Has carrier fear of adverse selection created a different form of competition here?
SWARTZ: Insurers' fear of adverse selection is quite real. The fact that it's not a requirement that we all buy health insurance leads individual market insurers to set up selection mechanisms to protect themselves from the threat of adverse selection. As a result, some people are simply denied coverage or offered a policy with many limits, for example, no coverage for lung conditions if you have hay fever. Anyone over age 45 buying coverage in the individual market starts at a real disadvantage and is likely to pay very high premiums.What conditions keep the market from operating more efficiently and fairly for consumers?
SWARTZ: In the small group and individual markets, insurers presume that those who are most likely to show up at their doorstep are most likely to know something about their own health that the insurers won't know about. The insurer can't know that there is a family member who has heart disease or bouts of cancer and that you've been diagnosed with the same problem. That is what insurance companies fear. They can't possibly learn all the information about you when you apply, which leads to their fear of adverse selection.So we should be sympathetic to the insurance company?
SWARTZ: We need to realize that insurers need to deal with this real, potential problem of adverse selection -- they set up selection mechanisms to try and weed out people they think will be very high cost, and the mechanisms they use aren't tremendously accurate.So what can be done to make the non-group insurance markets more accessible and less expensive to those people who can't access group-sponsored coverage?
SWARTZ: One way to help make private insurance markets more accessible to those who don't have insurance is to use a reinsurance mechanism, where the government would pick up most of the costs of people in the top one percent of the expenditure distribution of the entire population. These are people who spent at least $50,000 a year - for example, people with hemophilia, who can easily spend as much as $100,000 a year on needed drugs; people with various types of cancer, or heart attacks, or people who are victims of car crashes and need very expensive rehabilitation therapies.How can re-insurance mechanisms expand coverage? How would they work?
SWARTZ: Reinsurance is usually activated once a person's annual medical expenses go above a threshold, say $50,000. The federal government could be responsible for some fraction, say 75 or 90 percent, of the costs above the threshold. The insurance company would pay the other 25 percent or 10 percent of the person's costs. This would give the insurer an incentive to continue to manage the care of the people with very high costs.The insurance company would still shoulder some costs?
SWARTZ: Yes. Having the original insurer retain some risk for costs above the threshold that initiates the reinsurance puts a strong incentive in front of the insurance company to manage the care of high-cost people. From the insurer's perspective, it is relieved of 75 percent (or whatever percent) of costs above $50,000 -- that's a lot of money. Those in the top one percent of the expenditure distribution account for 28 percent of all health care expenditures -- so this type of program would relieve insurers of a great deal of risk.So what is the effect on an insured person?
SWARTZ: If the government provides reinsurance to the small group and individual markets, insurance companies will lower their premiums, and younger and healthier people will enter these markets and buy coverage. Their presence will help keep premiums lower than they are now in these markets, and more people who are currently uninsured will be able to afford to buy coverage.Are there effective reinsurance models working now?
SWARTZ: Yes, in New York. Healthy New York (HNY) is a state program with a reinsurance component; the program is targeted at lower income people and low-wage workers. When you compare HNY premiums to premiums in the regular individual market, what New York calls the direct-pay market, the HNY premiums are about half as high. HNY pays 90 percent of an enrollee's annual expenses between $5,000 and $75,000. If the enrollee has costs above $75,000, the insurer is responsible for all of the costs above $75,000. This is the closest I've seen to an effective government-sponsored reinsurance program.So the incentives are different for reinsurance than they are for a high-risk pool?
SWARTZ: Yes. The whole point of reinsurance is to reduce insurers' fear of adverse selection, the fear of having a disproportionate number of people with really high medical costs. If we can do this, then insurers will reduce their use of selection mechanisms that keep people from getting health insurance. Plus, premiums will be lower in the market because the reinsurance program is responsible for the bulk of expenses above the threshold that activates the reinsurance. Reinsurance pays for costs of people who, in fact, had very high expenses. Under a high-risk pool, there is no incentive for insurers to reduce their risk-selection mechanisms or reduce premiums. Insurers still have to predict who is likely to have high costs and then cede the person to the high-risk pool. Because the insurers' process of predicting who will have high costs is not 100% accurate, they will still have an incentive to avoid enrolling people they suspect will have high medical expenses.Why should the federal government assume the responsibility?
SWARTZ: States don't have the financial ability and it is in our national interest to have more people insured. High medical costs could happen to any one of us on a random basis. If part of what we're trying to do is reduce the number of uninsured, it benefits all of us. We pay for their care one way or another.How do we pay for a reinsurance program?
SWARTZ: We can fund it out of a dedicated tax or general revenue funds. It would be a mistake to fund such a program by assessing or taxing the insurance companies themselves. They would go right back to trying to stay away from those with high costs.But wouldn't this be exorbitantly expensive?
SWARTZ: Some estimates are that it would cost anywhere from $5 to $20 billion. We need better estimates of the costs, but if that would allow us to have a private health insurance system in which many more people would be covered, it seems worth it.What are some of the policy challenges?
SWARTZ: Cost, obviously, but the alternative is more costly. We have a lot of younger adults who can't get health insurance where they work or can't buy it on their own. The challenge is to get politicians to realize that this is becoming a very serious problem for large swaths of our population, and will be an even bigger problem in the future when they will really need coverage as middle-aged adults.
You have been studying the uninsured problem for some 25 years. How much has the landscaped changed? Is the problem that much worse than in the early 1980s, when you first started studying it?
SWARTZ: Yes, the problem is much worse now than 25 years ago. Many more people as a percentage of population are uninsured today. When I first started looking at this, using data from 1979, 14 percent of the non-elderly were uninsured. Now it's 18 percent. In terms of numbers, it's an increase from 28 million to 45 million uninsured people. Second, the chance of being uninsured is way up among the middle-income working population. One -tenth of the working-age middle-class population are uninsured, a percentage that has increased since 1980. To be sure, it is a much worse problem for those who are poor, but the fact that it's affecting the middle class makes this a more wide-spread problem than it was 25 years ago.
Funded by The Robert Wood Johnson Foundation, ERIU is a five-year program shedding new light on the causes and consequences of lack of coverage, and the crucial role that health insurance plays in shaping the U.S. labor market.