Flaws in the Ointment
Disaster insurance markets are distorted, economists argue, and could be improved.
Published December 1, 2006 | December 2006 issue
Analysts suggest that private markets for disaster insurance are underdeveloped because of several flaws that affect both demand and supply. In brief, government bailouts and consumer myopia suppress demand, and government provision of insurance at subsidized rates decreases supply by crowding out private insurers. Addressing these flaws—through mandatory disaster insurance, among other steps—could improve private insurance markets. Here are a few details.
What, me worry?
In the view of some economists, consumer attitudes toward disaster protection reduce the demand for insurance. Psychological studies have found that the human mind has trouble grasping the true risks of low-probability yet life-changing events, such as hurricanes, volcanic eruptions and terrorist attacks. In an August 2006 paper, Howard Kunreuther and Mark Pauly of the University of Pennsylvania draw upon 30 years of analysis of insurance purchase decisions to argue that most people don't rationally weigh the cost of insurance against the chance, remote but nonetheless real, that a disaster will occur. Instead, people downplay the risks, assuming that they are too low to worry about.
"There's a tendency to really push these things into the background, to say, 'This is a low-probability event that is below my threshold of concern,'" said Kunreuther, an expert in risk management at the University's Wharton School, in a telephone interview. Consequently, many homeowners and businesses don't even consider buying earthquake or terrorism insurance, or regard it as a discretionary expense that can be cut when money gets tight. That mindset shrinks the potential market for disaster insurance, making it more difficult for insurers to spread risks through diversification. Kunreuther and Pauly note that only about 40 percent of residents in the New Orleans area had flood insurance before Katrina hit, even though the federal government heavily subsidized the policies.
The trouble with subsidies
One problem with government insurance programs is that they're subject to political pressure that undercuts their effectiveness from the start; they often end up operating more like social welfare programs than true insurance pools. If current programs offer any guide, elected officials all too often give in to the impulse to provide a benefit to their constituents. The premiums paid by some policyholders in the federal government's National Flood Insurance Program (NFIP), for example, account for less than 40 percent of the costs the program would bear if flooding occurred in those areas. The California Earthquake Authority, created in 1996 to provide policies for homeowners in earthquake-prone areas, is supposed to charge risk-based rates, but in fact, households pay less than a private insurer would charge. It's likely that insurers with federal catastrophe reinsurance would also pay subsidized rates that wouldn't fully reflect the actual risk of a mega-disaster.
"Government doesn't think that insurance pooling itself is enough of a benefit, fairly priced," said Ron Feldman, a vice president at the Minneapolis Fed whose research focuses on government credit and insurance programs. "So oftentimes it will set the price lower than it ought to be, or establish less stringent criteria for insuring people than the private market would accept."
Subsidizing premiums in particular aggravates the problem of moral hazard—the tendency of people with insurance to expend less effort to avoid risks than they would if they had no insurance. Cheap hurricane insurance gives people little incentive to move away from the coast, or to try to reduce their premiums by making their homes less vulnerable to damage. Artificially low rates can also crowd out private insurers who might be willing to write catastrophe insurance, if not for the presence of the government program.
Poorly structured programs are doomed to fail sooner or later, because the risks they bear are too high for the premiums charged. Then, as is likely to happen with the hurricane-weary NFIP, taxpayers must honor commitments to policyholders. Bailing out bankrupt programs engenders more moral hazard because those who receive disaster relief expect the government to step in after the next disaster.
Policies required; bailouts denied
Kunreuther and Pauly favor a form of government intervention in catastrophe insurance markets—mandatory coverage. If, as insurance surveys indicate, many people dismiss hurricanes, earthquakes and other dangers as unworthy of concern, they won't buy coverage even if premiums accurately reflect the risks they face. Kunreuther and Pauly, with a nod to the Nobel Prize-winning contributions of economists Edward Prescott and Finn Kydland, suggest that people living in hazardous areas should be required to purchase comprehensive natural disaster insurance.
In a groundbreaking 1977 paper on monetary policy, Prescott (now a senior monetary adviser to the Minneapolis Fed) and Kydland show in a number of contexts that setting rules can be more effective than allowing discretion. They use the example of people who choose to live in floodplains, requiring costly public investments in levees and floodwalls to protect their property from floods. As a matter of public policy, the economists argue, it may be better to prohibit people from building in floodplains in the first place. But the government has to enforce that policy consistently and not bail people out if they build there and then suffer a flood.
In a twist on that notion, Kunreuther and Pauly assert that making disaster insurance mandatory would benefit property owners in the long run—and eliminate taxpayer-funded bailouts of the uninsured after a disaster. Private insurers would write the policies, basing premiums on the actual risk of a flood, tornado, earthquake or other natural disaster, and either financial institutions (which already require homeowners' insurance as a condition for mortgage approval) or a government agency would enforce the mandate. Recognizing that risk-based premiums may drive low-income residents from some communities, Kunreuther and Pauly recommend that those residents receive government aid to defray the cost of insurance. They also support some form of federal catastrophe reinsurance, provided at actuarially fair rates, to serve as a backstop for private insurers and existing or newly created state insurance funds.
If government decides that it should remain in the catastrophe insurance business and even expand its involvement, what's needed, Feldman says, is incentives for policymakers to design actuarially sound programs and resist the temptation to bail out disaster victims after the fact, whether they have insurance or not. Such incentives could include mandatory disaster insurance, as Kunreuther and Pauly suggest, or budgetary rules that require government to cover the cost of insurance subsidies and bailouts by cutting other programs.