As a congressional committee was approving a reform proposal last week, the program was nearly $18 billion in debt to the U.S. Treasury. That's about $160 per U.S. household - an amount about six times greater than what NFIP collects in annual premiums.
"You're never going to get the $18 billion back," Robert Hunter, who ran the program in the 1970s and who has been a vocal critic, said last week.
In testimony before Congress in April, Craig Fugate, director of the Federal Emergency Management Agency, which runs the insurance program, conceded as much. "FEMA is unlikely to pay off its full debt," he said, adding ominously, "especially if it faces catastrophic loss years."
This is looking like one of them.
After a few relatively benign years in which FEMA was able to pay down some of the debt, the Mississippi flooding alone is expected to result in multibillion-dollar losses. This almost certainly will be the program's worst year since 2005, when it was devastated by Katrina.
Flooding is ubiquitous in the United States, and thousands of Philadelphia-area residents can affirm that it is a persistent local problem.
New Jersey and Pennsylvania rank in the top 10 in NFIP losses in the period of record, which dates to 1978. Yardley, with $27.3 million, is among the top 1 percent for losses among the 21,000 towns in the program. Tiny New Hope, at $13.6 million, is in the top 2 percent.
The Mississippi flooding, however, is particularly ominous for the program's fiscal health. Mississippi and Louisiana are braced for record flooding. Only four states have accounted for $1 billion or more in losses; Mississippi and Louisiana are two of them. In fact, Louisiana is responsible for more than 40 percent of the program's $38 billion in loss payments.
But whatever else happens during the next several days, this won't be a disaster for private insurers. At least 95 percent of the insured losses will fall to the flood program, estimates Michael Barry of the Insurance Information Institute.
Though private insurers collect fees for writing the policies, they don't assume the risk, Hunter said.
So how did U.S. taxpayers wind up standing behind one of the world's largest insurance companies?
The short answer is, by default. Flooding historically has been so common and pervasive that private companies wanted no part of insuring against it. Thus, the U.S. government became an insurer of last resort, and the first policy was sold in 1969.
Flood-hazard areas were identified and mapped, and towns that agreed to participate had to undertake flood-control measures and limit building in high-risk areas.
Unlike conventional insurance companies, which are required to have cash reserves to cover catastrophic losses, NFIP has the U.S. Treasury.
To entice property owners to sign on, the government offered discounted rates of up to 65 percent for buildings constructed before the mapping was completed in the early 1970s.
It was all well-intentioned, but then came the disastrous 2005 hurricane season.
NFIP incurred more than $17 billion in losses that season. Of claims stemming from Hurricanes Katrina, Wilma, and Rita, 60 percent were for properties built before the 1970s.
Congress has since wrestled with ways to overhaul the program, due to expire Sept. 30. On Friday, the House Financial Services Committee approved a bill that would extend the program for five years. Among other things, it calls for a phase-in of "actuarially sound" premiums. It stops short of forgiving the debt, calling instead for a gradual pay-down.
The best solution, Hunter said, would be for the private market to get involved and get taxpayers off the hook.
He said FEMA had addressed critical issues that scared off insurance companies. Among them was the accuracy of government flood maps. Though not perfect, FEMA's new maps are drastically improved.
For now, however, private insurers have little incentive to rush into the floodwaters when they can let the government worry about the risks. Said Hunter, "They got a sweetheart deal."
Contact staff writer Anthony R. Wood at 610-761-8423 or email@example.com.