Anyone who follows the government's involvement in the mortgage industry with a skeptical eye should find it utterly unsurprising that the Federal Housing Authority (FHA) might be in trouble. The New York Times has an article today indicating that it may go the way of Fannie Mae and Freddie Mac and require a bailout. I think it could go either way, depending on its several factors.

The NY Times reports on FHA Commissioner David H. Stevens' testimony before a House subcommittee yesterday. He swears the FHA won't need a bailout. The data indicates otherwise:

But he acknowledged that some 20 percent of F.H.A. loans insured last year -- and as many as 24 percent of those from 2007 -- faced serious problems including foreclosure, offering a preview of a forthcoming audit of the agency's finances.



And its portfolio is growing rapidly. The FHA has played a significant part in the government's plans to attempt to prop up the mortgage market since its collapse:

The F.H.A. is insuring about 6,000 loans a day, four times the amount in 2006. Its portfolio is growing so fast that even F.H.A. backers express amazement.



Here's one pessimist also testifying at the hearing:

"It appears destined for a taxpayer bailout in the next 24 to 36 months," Edward Pinto, a former Fannie Mae executive, said in testimony prepared for the hearing. Mr. Pinto, who was the chief credit officer from 1987 to 1989 for Fannie Mae, went further than most housing analysts and predicted that F.H.A. losses would more than wipe out the agency's $30 billion of cash reserves.



Let's do a little math here using that $30 billion in cash reserves figure. According to the article, the FHA portfolio's size is around $675 billion. If 20% of those end up in foreclosure, how big a loss will that be? Obviously, it depends on how much the loss is on each of those mortgages. Those numbers dictate that those troubled mortgages would have to average a loss of around 22% of their value to completely wipe out the FHA's cash reserves.

Could it happen? Assuming the portfolio doesn't deteriorate further (which is optimistic), it entirely depends on what the loan-to-value is for its portfolio. If the mortgage sizes are much higher than what the homes are currently worth, then that would spell disaster.

This means it also likely depends on when those loans were originated -- back when houses were well overvalued or after the real estate market had already declined. If most if its newer originations include refinancing underwater homeowners, leaving the mortgage principal intact, then could also be a problem. But if a large portion of its troubled loans are newish and their values already reflect the market's decline, then its losses could be much more limited. After all, home prices probably aren't going to decrease another 22%. I hope.

So it's hard to know for sure just what kind of predicament the FHA is in. Clearly, its portion of bad mortgages shouldn't comfort anyone. But how steep of losses it will actually incur isn't easy to identify without having detailed data of their troubled mortgages' vintages and loan-to-value ratios. While I certainly hope its commissioner is right that it won't need a bailout, the cynic in me worries.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.