"If insurance is priced right..."

Aka if we priced insurance at the right level. You'd likely see people stop buying homes on some of these flood plains. As insruance would cost too much.

The premiums for flood insurance can be pretty steep, and I would be extremely surprised if the fund operated at a loss. In my case, the annual premium is a little over 1% of the covered value, even though the historical record high (back in the 1960's) would probably only require about 10% of the total coverage purchased to replace (in this case, mostly from a few out buildings that are a little lower than the house itself).

Then you're extremely surprised.

"Currently, the NFIP’s debt totals $24.6 billion, nearly six times the program’s total annual receipts of $4.3 billion"

https://www.cbo.gov/system/files/115th-congress-2017-2018/re...

And Planet Money, as always, has a great take:

https://www.npr.org/sections/money/2017/09/29/554603161/epis...

After doing some more digging, it turns out that roughly 20% of NFIP policies are subsidized, and the up-to $30b debt was authorized by Congress in 2012 due to the shortfalls.

80% of the policies (the non-subsidized ones) are thought to be actuarially sound, barring unintended consequences of out-of-date flood maps.

So, I'm not terribly surprised. Were it operating like a traditional insurer, it might even be doing fine (especially if it had some other reforms such as eliminating or having better oversight over WYO policies).