Two federal/state deals that aren't working well

Lee County, flood insurance, and data

Two recent developments involving state/federal deals show that a great deal of physical climate risk is being shifted to the federal government—which means everyone is on the hook.

Lee County, Florida.

Jake Bittle (author of a terrific book, The Great Displacement) and Tom Frank (of E&E News) reported earlier this month that Lee County officials are blasting a FEMA attempt to increase the amount homeowners pay for federal flood insurance there. Here's a colorful quote from Frank's story:

"'It's almost like revenge politics,'" [Board of Lee County Commissioner] Cecil Pendergrass said. 'Our citizens and taxpayers are being held hostage here and having to pay more.'"

For more than a year, FEMA has been sending letters to Lee County asking officials to confirm that homes that sustained "substantial damage" from 2022's Hurricane Ian—meaning properties that experienced damage equal to at least 50 percent of their value—were being elevated as they were repaired. But Lee County hasn't sent FEMA this information. 

Many local officials don't want this burden of making homeowners raise their houses. Officials in Lee County weren't checking homes for substantial damage unless "'a resident wanted to do it'" or unless a building permit was applied for—which doesn't always happen. Bittle's story says that officials generally don't proactively assess whether substantial damage has been sustained by homes following hurricanes. Waiting until someone asks for this assessment is "more or less standard across Florida and other hurricane-prone states."

This means untold numbers of homes remain at risk of flooding, over and over again, as extreme events become more frequent and more intense. Those homeowners are expecting that the federal government will rescue them through insurance payouts and disaster relief. Which means all of us will be rescuing them. 

A federal/state deal was made nearly 60 years ago, and now the bill is coming due.

At the heart of the National Flood Insurance Program is this tradeoff: If local communities discourage uneconomic development in floodplains, the federal government will run a flood insurance program. By "uneconomic," the drafters meant development whose full cost to society (including the public costs of disaster relief and rehabilitation) exceeds the benefits of building in risky places. A crucial part of this scheme is that local governments will have major responsibilities, including "avoiding, to the fullest possible degree, use of high hazard areas for uneconomic activities," in the words of the 1966 task force recommending this approach.  

This deal was set up because the federal government was confronting unbearable flood relief costs. Following a mass exit of private flood insurers in the mid-1920s, damages from floods had continued to rise. The eventual scheme created to reduce these unbearable federal expenditures required communities to adopt and enforce strong floodplain management strategies. In exchange, residents in these communities could purchase federally backed flood insurance.

The 1966 task force report includes this prescient paragraph:

A flood insurance program is a tool that should be used expertly or not at all. Correctly applied, it could promote wise use of floodplains. Incorrectly applied, it could exacerbate the whole problem of flood losses. For the Federal Government to subsidize low premium disaster insurance or to provide insurance in which premiums are not proportionate to risk would be to invite economic waste of great magnitude.

The idea in 1966 was that everyone building in risky areas should agree in advance to pay the annual "costs of flood plain occupance" that took into account the entire burden on society of their "locational decision," in exchange for indemnification for loss. This yearly payment would avoid the expense of building zillions of expensive levees and seawalls. Seawalls that made economic sense could still be built, the 1966 task force reasoned: 

An occupancy charge-indemnification fund or flood loss insurance could be used in lieu of an uneconomic structural or other type of measure, and to complement an economic flood protection measure.

Well, the "occupancy charge" idea vanished, but the idea of federal flood insurance stayed in place, as did vestiges of the notion that communities would have to police floodplain use if they wanted to get access to that insurance.

FEMA is allowed to provide flood insurance only to “those States or areas (or subdivisions thereof)” where “adequate land use and control measures” have been adopted that “are consistent with the comprehensive criteria for land management and use developed” by the NFIP. 

The federal side of this deal has been truly painful: NFIP owes $20 billion to the US Treasury, and would owe even more if Congress hadn't canceled $16 billion of its debt in 2017. Another reauthorization of NFIP is coming up in September 2024. Congress has been kicking the can down the road for years—this will be the 31st (or so) short term reauthorization of NFIP since 2017.

Now FEMA is trying to make sure that communities uphold their end of the bargain. It's trying to avoid shifting more costs to the federal government than necessary following flooding and storms.

The Lee County story represents a very mild form of enforcement: FEMA is saying, in effect, "You can still buy federal flood insurance, Lee County. But unless houses that are substantially damaged by hurricanes are elevated as they're rebuilt, you won't get a discount of 20-25 percent. We give that discount only to policyholders in areas that take extra steps to reduce flood damage." That's it. That's what Lee County, with its filled land next to the Gulf of Mexico, is calling "revenge politics" and "hostage"-taking. 

Those discounts have been in place since 2007, so residents got used to them. The average discount per homeowner is about $370, according to Frank's reporting.

FEMA is also gingerly, slowly raising its rates to match the risks in the area. And homeowners may have to elevate their homes if it turns out they rebuilt in flood zones without doing so. The whole system is getting more expensive for homeowners, which means price signals may begin to match reality. But the larger story is that there is already a great deal of flood insurance exposure on the books of the US Treasury, and, thus, on all our books. 

Data call by the federal government. Lenders, mostly nonbanks these days, are handing out mortgages in risky places. As I've written in the past, these mortgages are then swiftly sold to Fannie Mae and Freddie Mac. The nonbanks check for presence of insurance, and the GSEs apparently don’t double-check. Indeed,  Don White, the senior managing director and chief risk officer at PennyMac Financial, one of the largest providers of mortgages in the US, told Maria Volkova of National Mortgage News recently that climate change is seen as a "priceable and manageable risk." Then he went on: "If the insurance industry were to collapse it would become much more of a crisis for lenders...but as of right now, it seems like it's managed," White said. 

Well, we don't know much about what's going on in the insurance industry. We hear stories. There are 30,000 policyholders getting nonrenewal notices from State Farm in California. American National Group is leaving California and eight other states. Most days there’s a story about insurance troubles nationwide.

But this is anecdotal. Surely Fannie Mae and Freddie Mac should be informed by Treasury policy based on high-resolution knowledge of what's actually happening in homeowners' insurance markets around the country.

Without data, Fannie Mae and Freddie Mac can't look ahead. They can’t do more than check boxes to check for insurance—they can't know which zip codes are seeing exits and nonrenewals by insurers. The Treasury is the lead shareholder of the GSEs, with about $310B invested, and a deep interest in their continued solvency and reform, but its policy will based on sand if it isn't informed by good data.

Recently the Federal Insurance Office, part of the Treasury, moved carefully to call for data about insurer practices and prices. There was pushback from the industry. The association of state insurance commissioners is now saying, "We'll do that!" Even this watered down governance isn't going to work: Louisiana has said it won't comply, and Texas and Florida are already saying they're reluctant.

What a mess. This is setting up a systemic risk for the entire economy. The GSEs have risky mortgages on their books but don't have the longterm insights they need. Surely the tradeoff here—we'll soak up your mortgages if you get us information—was sensible. But it's not being honored by the states.

Both of these "local" issues—adequate control of floodplain development, adequate insurance backing up mortgages—are simultaneously places where local governments obtain revenue and support voters' wealth. So naturally there's a desire to ensure the game keeps going. The federal government, for its part, is right to notice that the risks of this behavior are being shunted away from the states.

These federal efforts aren’t "revenge politics." They do, though, signal that a very brittle system is being pushed ever closer to the brink.