Insurance is a great human invention, but people should understand how it works. In particular, they need to understand the incentives insurance can create, both for the good of society and for ill.
We face those questions after recent flooding in southeast Minnesota. Several communities were hard hit. Hundreds of homes were damaged. Many were total losses. Few people had flood insurance. Federal and state government will provide some help, but many affected households will bear most of the losses themselves. Some will lose most or all of their net worth.
It is a natural human impulse to want to help such people. It is also a natural human impulse to wonder why people living in towns that have experienced flooding in the past did not take federally subsidized flood insurance that was available. Both impulses entail economic issues.
One response showed up in a recent letter to the editor. It argued government should require that all homeowners insurance include flood coverage. Is this a good idea?
Virtually all homeowners insurance covers damage from fire and wind, theft losses and liability for falls or other injuries connected with the property. Such coverage may include deductibles, co-payments and limitations. You pay extra if you have a swimming pool or a $10,000 diamond. You get a discount for a security system or a steel roof, and so on. But few policies cover flood damage. Why not?
The standard answer is that flood damage is associated with “adverse selection” to a degree that other covered perils are not. If flood coverage is offered as an option, people living in areas prone to flooding will buy it. Those who live high and dry won’t. The pool of covered houses is much more likely to experience damage than the average of all houses.
For the insurers to break even, they must charge high premiums for flood coverage. More people opt out and eventually the only people covered are those with a high probability of flooding.
This is essentially what happened with private flood insurance. Companies offering it found few takers. People with low likelihood of being flooded feel coverage is unnecessary. People in danger of flooding look at actuarially necessary premiums and decide to run the risk.
That happens even when a government program absorbs part of the cost. Some recent flood victims were misinformed by their insurance agents. Others just thought about the $400 higher annual premiums and decided to chance it.
Why, then, don’t we make coverage mandatory? If no one can opt out, then no one will face prohibitive premiums. Anyone flooded out won’t be limited to charity or government handouts.
This ignores “moral hazard.” One function of insurance is to spread costs of losses across society so that no single individual suffers an overwhelming loss. But insurance also functions to motivate prudent behavior. If everyone gets flood coverage by statute and insurance companies cannot charge different rates according to risk, it reduces incentive to avoid building in flood-prone areas.
That is precisely what federally subsidized flood insurance has done. It encouraged people to build on barrier islands or low-lying coastal areas and in flood plains. The net effect has been to increase the amount of property society loses to floods. Mandatory flood coverage would make millions of homeowners pay higher premiums to cover the risky behavior of a few.
I suspect the letter writer imagined that flood coverage could be extended to everyone without anyone paying a higher premium. Insurance companies simply should be forced to accept lower profits. That may be a pleasant delusion, but it is a delusion.
Policyholders ultimately must pay not only all losses paid, but also the administrative costs of the insurance industry. Yes, income from the money insurance companies put into stocks, bonds and real estate helps pay losses. But if households or businesses did not pay premiums, they would get the interest and dividends themselves.
The argument that fire and wind damage don’t motivate adverse selection or moral hazard but flood damage does is not as clear-cut as some think. Universal fire coverage motivates building in fire-prone areas like California’s Santa Ana Mountains just as flood insurance motivates building on the Outer Banks of North Carolina.
Yes, people in the dry chaparral pay higher rates than those in Brooklyn, just as people in low-crime areas pay less than where theft is rampant. But premium differentials often fail to fully reflect risk differences and some cross-subsidy commonly occurs.
© 2007 Edward Lotterman
Chanarambie Consulting, Inc.