Sunday, October 12, 2008

Insurance at Risk

The information access revolution taking place over the past 10-15 years has had a number of unintended consequences. Certain professions have been marginalized as with many technological innovations. The rise of do-it-yourself and sharing among amateurs have allowed us to undertake things with more confidence and without the need to consult an expert in many cases. Personal privacy has been put in doubt and the concept of privacy will have to be re-evaluated. Another unforseen change is an undermining of the notion of "pooled risk," a principle that modern insurance is based on.

Insurance companies offer to pay for personal loss of a subscriber only if the pool of buyers is large enough to sufficiently cover the costs of a catastrophe and if the estimated frequency of such a catastrophe is low enough to ensure that claims do not exhaust the accumulated subscriber payments.

But today we know (and can know) a lot about people including an increasingly accurate estimate of their chances for encountering misfortune. With medical records, increasingly sophisticated actuarial tables, public health data, genomics and other information sources that can more accurately predict the likelihood of certain occurrences, insurance companies will become more discriminating in the policies they write and their premiums will more accurately reflect the chances that a subscriber will one day make a claim. Insurance policies will soon be based on data sources that were unavailable just a few years ago.

Computer models which estimate rates of crime, traffic, health problems and which also assess the monetary level of potential damages from these are getting better all the time. I suspect that soon they will be so good as to almost render the notion of pooled risk inconsequential. The uncertainty accompanying risk will be reduced so much that the marketplace for insurance will become nearly unsustainable.

For example, most of us know today that the vast majority of lifetime expenditures on health care take place within the last year of a person's life. The hospitalization, home care, therapies and drugs that characterize the chronically ill octogenarian's last year dwarfs expenditures made throughout his or her life. Of course, there is the possibility of catastrophic illness throughout our lives (such as cancer or permanent disability from accidents) but if we could eliminate those freak occurrences, we would all delay health insurance purchases until we were at retirement age since we probably won't need much in the way of benefits until the last years of our lives. (Health insurance is an odd case and I only use this example to illustrate the way that emerging information resources could guide decisions on insurance purchasing.) In fact, government regulation is the only thing keeping the health insurance market afloat as many insurers would eagerly cancel policies for those who develop a predisposition to illness. Such as the elderly.

Taken together with the burdensome economic situation that many insurance companies are facing, it may be that we can say goodbye to insurance as we have known it for many years. After all, if we can know so much about a person's chances for encountering misfortune, then the risk is removed for both parties. Only those who learn that their chances of having an accident or illness are high will purchase a comprehensive policy but insurance companies will only issue policies with large coverage to those whose chances of an accident or illness are very low.

This assumes that information flows freely; it is likely that corporations (insurance companies and their finance parent companies) will have an advantage in the collection and processing of this information.

But that's a topic for another post.

No comments: