The first principle of insurance reflects the fundamental lesson of the tragic California fires: you can’t get something for nothing. If expected losses from wildfires this year are $N, and the expected losses materialize, then it will take $N to compensate those whose homes have been destroyed by fire. Either $N must already be on hand, or the difference between the dollars on hand and $N will have to be raised going forward. There is no insurance trick, or mystery, or device, or legerdemain that can circumvent this fundamental, sobering truth. The question is where you get that money. Ultimately, if you don’t have, or get, all the money to pay for fire losses, then the homeowners pay, in whole or in part, through a massive diminution of their net wealth because of the destruction of their homes.
If somehow fire risk could be radically reduced, the problem could be solved. But I take it as axiomatic that this is not going to occur in the short term, and probably not much in the middle term. Major fire losses will persist. In the future, therefore, the losses will have to be paid to, or shouldered by, homeowners.
The proportion in which the potential sources of compensation are expected to contribute can be adjusted in the future. But the potential sources will still be the same: The money can come from fire insurance policies for which actuarially accurate premiums were charged. It can come from the shareholders of (stock) insurance companies or the policyholders of mutual companies in the form of payment of coverage claims under policies with excessively low premiums. It can come from taxpayers, directly or indirectly, in the form of state-provided insurance backups, federal disaster aid, or from a federally-financed reinsurance program that might be enacted. Or occasionally it can be paid by tortfeasors, such as power companies, if they happen to be responsible. It can also come from some combination of these sources.
There are pros and cons to each approach, and each potential combination of approaches. And there are practical obstacles to each. It is important to candidly recognize, however, that what confronts us is not the prospect of discovering some ingenious new policy or insurance solution. Rather, as is often the case, we face a set of unpleasant choices that no one with both a heart and a mind will find comforting.
Fire Insurance
There are two principal sources of fire insurance. Fire insurance for individuals is provided by homeowners insurance, which covers not only the peril of loss or damage by fire, but also a whole series of other causes of loss, such as wind, theft, explosion, and so forth. Fire insurance for entities, including businesses, is provided by Commercial Property Insurance. In relevant respects, the coverage that both kinds of policies provide is identical. In fact, contemporary individual and commercial policies provide “all-risk” (or “open-peril”) coverage, under which all risks of “direct physical loss” are covered, subject to a series of express exclusions. Accordingly, there is every reason to think that the overwhelming majority of policyholders whose property was damaged by the California fires are covered for their losses, up to the amount of coverage they purchased.
Fire insurance goes back a long way. It was introduced after the Great Fire of London in 1666. It was sold in the colonies before the Revolution. And it was a risky enterprise. The new insurers had no actuarial data on the probability of fire or on the magnitude of risk posed by different properties. They knew intuitively that stone and masonry buildings posed less risk than wood buildings, but until there had been a lot of fires, they could not know the precise difference, and therefore could not meaningfully calibrate the level of premiums they charged to cover these risks. So they risked insolvency if they underestimated risk and charged too little for coverage. Eventually they solved this problem by pooling their data on losses to create a statistically more reliable basis for setting premiums, arguably at a cost to competition.
The early fire insurers faced a second challenge too: correlated risk. In the densely constructed towns and cities of colonial America, fires could easily spread from building to building, causing catastrophic losses. Sound familiar? Major fires in Charleston and New York put a number of the early fire insurers out of business. Today, correlated risk is still a threat, as the California experience confirms. The market is not structured (and could not effectively be structured) to allocate the homes in the same neighborhood to different insurers. So a few insurers can end up covering portions of a large area and many homes. To address this challenge, individual insurers now reinsure some of their risk, thus spreading their exposure into the global financial markets. It is likely that many of the insurers who pay claims for losses resulting from the California wildfires will be reimbursed for some of their payments by reinsurers.
The Current Challenge
Because you can’t get something for nothing, the cost of insuring against the risks covered by homeowners insurance varies with the degree of risk posed in any given place at any given time. Due to climate change, fire risk and wind risk have increased in many places. If you bought your property with a mortgage, you are required by contract to have homeowners insurance protecting the lender’s interest.
It is deceptively easy to argue that premiums should be level across large areas, such as all of California. That way people in fire-prone areas would not have to pay prohibitively high premiums. But then people in lower-risk areas would pay higher premiums than they had been paying — premiums that are not commensurate with the lower risk of fire they face. And if, as a result, some people’s premiums were artificially low, then they would have a distorted incentive to live in or continue living in high-risk areas. In any event, if one or a few insurers decided to be “public-spirited” and charge level premiums, other insurers would sell coverage to lower risk policyholders at lower premiums, and the public-spirited insurer would lose money. These profit incentives ensure that doesn’t happen.
It is true that premiums are only predictions of loss, which can be inaccurate. In any given year, an insurer or insurers may overestimate or underestimate their future losses. But as losses rise in the aggregate over time, these variations wash out and premiums rise overall — and, if aggregate losses rise in considerable amounts, premiums follow. That has happened in many areas of the country in recent years.
A Role for Regulation
Even in competitive markets, premiums increase as the risk of loss increases. But if there are flaws in a market, state regulation can attempt to address them. If an insurance market is not efficient or fair, then regulation can try to mandate change. Whether that will work depends on the incentives created and the escape hatches left open. For example, suppose a regulator mandated the statewide, level-premium approach I described above. That resembles what the Affordable Care Act did for health insurance: The ACA prohibited varying individual health insurance premiums except based on age, territory of residence, and tobacco use (and whether a policy is being purchased for an individual or a family). Low-risk policyholders couldn’t avoid subsidizing high-risk policyholders because of the ACA’s requirement that all citizens purchase health insurance (the “individual mandate”). By the same token, most homeowners could not simply decline to buy homeowners insurance because of what I would call the “mortgage lender’s mandate” that they buy it. A no risk-rating approach like that of the ACA, however, would be politically unpopular with the large segment of homeowners who would see their premiums increased under that approach.
As has apparently happened in California, homeowners who have paid off their mortgages have sometimes gone “bare” — i.e., uninsured. And those who have partly paid down their mortgages may have bought only the amount of coverage necessary to cover the lender’s equity interest, going bare on the homeowner’s equity. And some people just decide to stop buying insurance because they find it too expensive. That’s what happens when premiums rise to a point that they are not affordable to some. Many of the nonrenewals of insurance reported in the media are probably just that — decisions by homeowners to go bare. Those whose homes have now been destroyed by fire will regret their decision not to renew. But many if not most of the nonrenewals could be framed as insurers’ decisions: because of insurers’ desire to reduce their exposure in high-risk areas; because of regulatory limits on premium levels that threaten profitability; or to play “chicken” with regulators in an effort to obtain future approval of premium increases greater than would otherwise be approved.
Premiums can be kept artificially low through regulatory mandates to encourage reluctant homeowners to buy insurance — but only for so long. At some point, insurers just decide to get out of the market rather than being forced to operate at a loss. Before that, losses are effectively shouldered by shareholders; after that, future losses will be allocated to homeowners. It is true that regulators can require those who sell auto insurance, which is profitable, to continue selling homeowners insurance if they want to keep their licenses to sell auto insurance. Maybe California could do that for a prolonged period, because California is such a big state with a lot of auto insurance revenue. But that’s not feasible in smaller or even medium-sized states. Regulate too strenuously and you will lose insurers and render the market in your state far less competitive, which likely results in higher premiums in the long term. And do that for too long and too strenuously in California, then auto insurers will leave even that state. So there is a practical limit on what regulatory suppression of premium levels can accomplish without throwing the baby out with the bathwater.
When regulators approach that limit, states can create facilities that provide insurance that is otherwise unavailable or too expensive for some. The legal form of these entities varies. They are sometimes called assigned risk pools or joint underwriting associations. They provide the insurance in question to the individuals the facilities are designed to cover. Economically, these facilities charge private insurers the cost of paying the losses in excess of premiums that are covered by the facilities, in proportion to each private insurer’s market share. So, in the end, whether the private facilities and the insurers backing them up lose money or break even depends on whether the state facility charges high enough premiums. Charge enough, and the facility is a going concern, but typically only when premiums are about as high as what the private market charges. Keep premiums lower because of political pressure, and something has to give. Either the facility fails or additional state backup is necessary.
State-Funded Backups
To address these problems, a state can be an “insurer of last resort.” If the state makes up losses incurred by a state-run insurance facility out of general revenues, then in effect the taxpayers — some of which also are policyholders and some which are not — are subsidizing the premiums of policyholders covered by a state-run facility. Florida, for example, has a state “plan” that provides homeowners insurance to a sizable fraction of homes on or near the coast that private insurers often won’t cover. California has a “FAIR” Plan that issues homeowners insurance to those who cannot obtain it in the private market, often because they are in high fire-risk zones. But the maximum coverage limits are three million dollars. If you owned a five million dollar home in Pacific Palisades, the FAIR Plan would not cover the full value of your home.
Insurance Zoning?
We could get to the point at which locating homes in certain areas is so risky that few insurers would cover fire or hurricane risks in those disaster-prone areas. At the same time, providing state backup insurance for the full worth of high-value homes probably would be politically objectionable. Why let people locate in these areas and place the risk of loss at least partly on the taxpayers? Resolving that issue would create obviously different interest groups, and the politics of the outcome would be uncertain. Without full state backup insurance available, as uninsured risks increase, only the very wealthy can afford to live in the high-risk areas. Income segregation of this sort already occurs in many places, but with insurance still available, the segregation is incomplete. There were apparently losses in Pacific Palisades not only among the very wealthy, but also among the middle class. Over time, however, as insurance becomes less and less available and more and more expensive, income segregation in high-risk areas may approach one hundred percent.
Conclusion
We should not think the insurance problems posed by the California wildfires can be solved going forward if only we knew more about insurance or if only we could fix the flaws in the market for fire insurance. Undoubtedly, knowing more about insurance would be good for policymakers. And solving the flaws in the fire insurance market that do exist could generate some benefits too. But ultimately, the main problem is likely to be that there is more risk of fire loss in California (and elsewhere) than homeowners, or taxpayers, or shareholders, or some combination of them, can comfortably pay to insure.
* David and Mary Harrison Distinguished Professor of Law, University of Virginia School of Law.
The post About the California Fires appeared first on Harvard Law Review.