Private Information about Health Risks Can Undermine Insurance Markets
The long-term care insurance ownership rate among those at genetic risk for developing HD (50 percent) is five times the rate of ownership in the general population (10 percent).
Personalized genetic information is increasingly available, and continuing advances in technology are likely to make even more such information available in the future. For example, a perfectly predictive genetic test for Huntington Disease (HD) has been around since 1993. Individuals who learn that they carry the HD genetic mutation and will develop HD can expect to begin to deteriorate mentally and physically between the ages of 30 and 50. They could face about twenty years of intensifying disability and increasing need for care before dying. Such individuals are likely to want to purchase long-term care insurance before their symptoms occur, and without necessarily revealing their risk for HD to an insurer.
In Genetic Adverse Selection: Evidence from Long-Term Care Insurance and Huntington Disease (NBER Working Paper No. 15326), co-authors Emily Oster, Ira Shoulson, Kimberly Quaid, and Ray Dorsey ask how this asymmetry between what individuals and insurers know might affect the markets for individual-payer insurance, and perhaps eventually influence the viability of long-term care insurance. In this study, the researchers compare rates of long-term care insurance ownership among asymptomatic individuals from the general population to long-term care insurance ownership rates among asymptomatic individuals who are at risk for HD. Analyzing data drawn from the Health and Retirement Survey (HRS) and from a 1,000-person prospective cohort study (PHAROS), they find that the long-term care insurance ownership rate among those at genetic risk for developing HD (50 percent) is five times the rate of ownership in the general population (10 percent). Furthermore, among individuals whose genetic testing shows that they are 100 percent at risk to develop HD, 50 to 75 percent own insurance. This strongly suggests that individuals make different insurance ownership choices based on the degree of private information they have. That is called "adverse selection" in individual-payer insurance markets.
The timing of individuals purchase of long-term care insurance is also consistent with HD risk generating adverse selection. The data used here reveal limited evidence of increased insurance among tested individuals before private testing but a large increase in ownership among individuals after testing.
In addition to HD, three other diseases with long disability periods and similar long-term care needs -- Parkinson's, Alzheimer's, and Lou Gehrig's or ALS have some genetic basis. At least some individuals affected by those three diseases could have perfect genetic information prior to any symptoms, an asymmetry that could multiply the effect of adverse selection on private long-term insurance.
As more individuals gain private information about the likelihood that they will require costly long-term care, adverse selection may threaten the viability of private long-term care insurance, at least in its present form. The authors illustrate that if a monopoly insurer were to offer a single premium price in a market with two types of individuals -- those with private genetic information and a very high probability of needing long-term care and those with neither -- and there were relatively few of the higher risk individuals (as is the case with HD), then the insurer could make a positive profit. The risks in the group would be pooled, because some lower-risk people are willing to purchase insurance at actuarially unfair prices. But as the share of higher-risk individuals in the pool increases, the market eventually shifts, abruptly, to selling only to that type of person -- and this shift occurs when the share of the high-risk type is quite small, around 3 percent. The authors conclude that even relatively limited increases in genetic information may threaten the viability of private long-term care insurance in the near future.
-- Sarah H. WrightThe Digest is not copyrighted and may be reproduced freely with appropriate attribution of source.