Insuring Against Insurance
NEW HAVEN -- Most people who save and invest do so for their entire lives. But most of the institutions upon which they rely for their investments and savings are geared to the short term. This mismatch causes fundamental problems.
An excellent example is homeowners’ insurance. Almost universally in the world today, homeowners’ insurance is short term. Typically, it is renewed annually, which means that it does not cover the risk that insurance companies will raise rates at any future renewal date.
Yet we have seen major changes recently in homeowners’ insurance rates. For example, the average homeowner premium in Florida soared from $723 at the start of 2002 to $1,465 in the first quarter of 2007. Such rapid increases represent a risk that is on the same order of magnitude as many of the damage risks that the policies are supposed to address.
In a study presented in early May at America’s National Bureau of Economic Research, the economists Dwight Jaffee, Howard Kunreuther, and Erwann Michel-Kerjan called for a fundamental change in policy aimed at developing true long-term insurance (LTI) that set insurance premiums for many years. Unless we do that, homeowners are unsure from year to year whether their insurance policies will be canceled or that their premiums will skyrocket unexpectedly as they have in coastal regions of Florida where there is hurricane and flood risk. As the authors point out, for insurers to even consider a long term policy they must have the freedom to charge premiums that reflects risk.
Urbanization itself is also a source of risk, as evidenced by the recent earthquake in China, which has cost at least tens of thousands of lives. Moreover, global warming appears to be increasing the intensity of storms. Some scientists attribute the intensity of Cyclone Nargis, which struck Myanmar, killing more than 30,000, to global warming.
Of course, we do not know for sure that these risks will mean higher insured losses in the future. Population growth in coastal areas may not continue to imply more risk exposure, since choice lots may already be getting somewhat more scarce, and further development may favor more central areas. And urbanization, if done right, leads to better catastrophe planning and stricter construction standards, which might actually reduce risks. In fact, long term insurance may encourage homeowners to invest in risk reducing measures because the premium discounts they will obtain for taking this action will justify incurring the cost of investment.
The course of global warming, and its impact on future storms, is also the subject of considerable uncertainty. Meteorology is not an exact science, and we cannot predict the precise extent and impact of environmental initiatives, though progress in weather forecasting might also reduce the damage caused by hurricanes.
Data presented by Roger Pielke in the Natural Hazards Review in February shows that actual insured losses caused by the most important hurricanes since 1900 followed a U-shaped curve. The most damaging hurricanes (scaled for the size of the economy) to hit the United States occurred both in the early twentieth century and most recently – the worst being the 1926 hurricane that struck Miami, Florida.
Just as no one anticipated this U-shaped pattern of losses, so future losses must remain essentially unknown. This means that the problem is not a certain increase in homeowners’ insurance losses, but rather a risk of increase. Paradoxically, that is a good thing, because it means that risk-management technology can be used to mitigate the problem.
To see why, consider health or life insurance. If genetic information ever enables an exact prediction of each individual’s ultimate date of illness and death, that information, if commonly available, would make such insurance impossible to get (no one would insure someone about whom it is known that he will suffer from the harm to be insured against). It is just the same with homeowners’ insurance: because the risk of losses is uncertain, it can be sold by those who feel most concerned about it to others who can better bear it.
Insurance regulators are certainly well aware of the risk of future increases in homeowners’ insurance premiums. But trying to address these risks by limiting such increases doesn’t work well, because if insurance companies are not making any money, they will withdraw from the market. Nor can this problem be solved by imposing fees on insurance companies that withdraw from the market in response to premium caps, because the companies will eventually learn to consider the possibility of such fees even before entering an insurance market.
Sometimes, governments have become directly involved in providing insurance. In the US, for example, Florida’s state legislature created the Citizens’ Property Insurance Corporation in 2002. But replacing private insurance with government insurance plans is far from optimal. Like other forms of saving and investment, it is better that insurance be allocated by a market, rather than in a political arena.
The beauty of the LTI plan proposed by Jaffee, Kunreuther, and Michel-Kerjan is that it would allow market forces to determine long-term (20 years or more) insurance premiums. The premiums would be set so that there would be no reason for insurance companies to withdraw from the market in response to greater risk. Homeowners can rest assured that they can continue to insure their property at known rates.
Moreover, the premiums would provide price signals that would guide new construction. In areas where scientists think that there is a likelihood that greater risks will prevail in coming years, high insurance premiums would provide a market incentive to curtail development. Everyone would end up better off.
Robert Shiller is Professor of Economics at Yale University, Chief Economist at MacroMarkets LLC and author of Irrational Exuberance and The New Financial Order: Risk in the 21st Century.