It's a risk crisis, not an insurance crisis

Talking (briefly) about NFIP

This past summer, a raft of stories came out about private insurers exiting high-climate-risk states like California, Florida, and Louisiana. Insurers are also narrowing coverage and raising both premiums and deductibles.

As premiums climb, more homeowners in the US, at least 12 percent of them, are going without insurance altogether. Half of those "bare" households have incomes of $40,000 or less a year. Because insurance can help communities and households recover more quickly from disasters, and because so much of the US economy is driven by spending on housing, the inaccessibility and unaffordability of insurance poses a threat to the stability of the entire economy. 

The difficulty of making a reliable profit from pooling risks in an increasingly risky climate context led to those exits you read about this summer. If climate risks were independent, random, and rare events, an entire market might be insurable. But once these risks became correlated and increasingly fat-tailed, with extreme events happening seemingly everywhere, transferring those risks to others became too expensive to make financial sense. This past summer made this obvious.

To keep property markets moving, some states have launched their own property insurance programs. Today, the solvency of these "residual" or "insurers of last resort" is questionable: Florida and Louisiana had to borrow hundreds of millions of dollars to pay off claims this year.  

We’ve gone through this no-private-appetite, government-steps-in sequence before. More than fifty years ago the private sector left the flood insurance market. Since then, FEMA's National Flood Insurance Program (or NFIP, nickname: ENN-phip) has made federally backed flood insurance available to communities that agree to enforce floodplain management ordinances to reduce future flood damage. More than 95 percent of flood policies in the US are provided through this program by the federal government, distributed by private insurers—who get a cut of the premiums that are paid but take on no liability.

Anyone with a federally-backed mortgage whose house is in a FEMA-mapped floodplain is supposed to carry NFIP coverage, which pays out a maximum of $250,000 per household. The bargain behind this program—you get insurance, subsidized by all US taxpayers, if your community limits development in risky areas—has not played out the way its framers intended. Today NFIP is viewed as essentially a subsidy for building and living in and around inaccurately-mapped risky areas. (I cannot say enough good things about Carolyn Kousky’s marvelous book, Understanding Disaster Insurance. Run, don’t walk.)

NFIP's premiums were kept artificially low and standardized between 1968 and 2021, which meant that the charges for its coverage both didn't reflect actual risk and resulted in perverse cross-subsidies from poorer to richer households. It also meant that NFIP swiftly became insolvent when it had to pay heavy claims.

After Hurricanes Harvey, Maria, and Irma wreaked havoc in 2017, Congress bailed NFIP out in 2018, transferring $16 billion to the program from the general treasury. Right now, NFIP is $20.5 billion underwater. When there was a gentle effort in 2021 to begin the process of matching risk to premium for its nearly 5 million policyholders, ten states sued. Which is bananas, in a word. Then again, the whole program is full of incoherences and oddities that I won't burden you with right now.

It's enough to understand that NFIP is such a legislative hand grenade that it's been the subject of 23 temporary authorizations since 2017. Reforming it is too politically difficult for any politician hoping for reelection to take on. 

NFIP's current lease on life runs out at the end of this month. Sen. John Kennedy (R., La.) introduced a bill last week to allow the program to limp forward for another year. He told Jean Eaglesham and  Katy Stech Ferek of the Wall Street Journal that "The only thing worse than what we have is nothing," but despite the senator's frankness his effort was blocked. 

All this is why I want us to celebrate Len Shabman of Resources for the Future. He has been thinking about and working on NFIP for years and he's not giving up. In fact, he's optimistic. I heard him speak last week and I admired his doggedness. 

Shabman points out that private insurance carriers did bear risks associated with selling NFIP policies until 1978, when they got pass-through status. Perhaps the original partnership between the federal government and private insurers, under which everyone acted in a pool, could be resuscitated, he suggested.

He said that risk-based pricing of NFIP premiums, actuarial pricing, is essential and must be possible. It has to be looked at rigorously by some professional, non-governmental organization, he suggested. Shabman pointed out that the "what to charge" question can be treated separately from the "how much to charge which household" question, and should be.

When it comes to affordability, he noted that the statutory language behind NFIP says only that charges have to be reasonable and fair. In the original 1968 legislation, Congress evidenced its understanding that some property owners would be charged less because they didn't have the means. The foregone payment, though, was supposed to be automatically made good, filled in, by the Treasury. He thinks that filling-in process ("equalization") should be brought back to life, and all the reasons to charge people less should be explored. There could be additional good reasons for lower premium payments beyond the means of the household.

As for NFIP's chronic insolvency, Shabman told the audience that the original program had contemplated a "catastrophic loss backstop" in the form of, effectively, public reinsurance. Also, state regulators could require flood insurance by default in any homeowner policy; people could opt out of it, but the default setting would bring more homeowners into the pool and lower the insolvency risk.

Shabman's essential point was that "Congress has to help." Flood insurance, done right, could smooth miserable outcomes for millions of people.

I wish him well.

My two cents, as someone somewhere faces up to NFIP: Congress should also ensure that there are nationwide standards for riskiness that govern the distribution of federal funding generally, as it did in the Coastal Barriers Resources Act decades ago. (Hat tip to Alice Hill for pointing to this precedent.) We need objective triggers that signal that infrastructure will be gradually decommissioned and residents bought out if they want to be. Go, NIST.

What we really need to do now is lower risk, not just tinker with transferring it from one entity to another. But that's a subject for another post.