Continuing risks in the housing market

By Felix Salmon
June 3, 2010
Wcw explains the economics of the housing bubble, in a smart comment:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Wcw explains the economics of the housing bubble, in a smart comment:

As for borrowing the money, I think the folks who did so did absolutely the right thing. When someone offers you a government-subsidized non-recourse loan on an appreciating asset, they’re giving you a put option. The more you borrow and the less you put down, the more that option is worth. These borrowers maximized their leverage. The market turned, and they exercised their put options. They are not the dumb ones, they are the smart ones. The dumb ones are the ones who gave away the farm lending this money.

It’s well worth noting, here, that there was an extra necessary ingredient: fudgy accounting on the bank side. If the banks had marked those put options to market, they would either have never made the loans in the first place, or they would have noticed their balance sheets eroding long before it was too late to raise new capital.

But they still aren’t marking those options to market. Everybody in America who has a mortgage has the option to default on that mortgage. That option is worth some non-zero sum, and in aggregate, across all the homeowners in the country, that option is worth billions, if not hundreds of billions, of dollars. And if homeowners own an option worth billions, then lenders have a liability of exactly the same magnitude.

The biggest lenders, in this regard, are Fannie and Freddie. But many banks are in the same position.

Look at the question another way. The likes of Mike Konczal have been looking for a while at second liens, and asking tough questions about how much they’re really worth. But let’s zoom back and look at the first liens, and think of it like this:

Let’s say you’re a good, creditworthy borrower: a tenured university professor, say, earning $200,000 a year. You ask for a $500,000 30-year amortizing loan at a 5% interest rate, which would involve you paying back about $2,900 a month — well within your earnings. The bank would laugh you away. But ask for the same loan in mortgage form, and suddenly it’s no problem. Why? Because the loan is secured.

Clearly, the existence of the security radically changes the economics of the loan. And equally clearly, the value of banks’ mortgage security — the houses the mortgages are borrowed against — has been falling fast. We know that the value of a home is the best predictor of future default, and that mortgages become much riskier when the homeowner doesn’t have any equity. We also know that roughly 25% of homeowners with mortgages have negative equity.

So it’s pretty obvious that even without taking into account a single actual default, the value of banks’ performing mortgages has been falling fast. And it’s equally obvious that no one has even attempted to quantify the numbers involved here — just how much has the value of banks’ performing mortgages fallen? And how does that number compare to the banks’ own equity?

This is one of the reasons why I’m still very nervous about markets generally: while there are lots of big sovereign and geopolitical risks out there, it’s far from clear that the big mortgage/housing risk is even behind us yet. And I would be much more reassured if there were some numbers marking banks’ mortgage books to some kind of market or model. Even if those numbers didn’t turn up in the formal GAAP accounts, it would be good to know that someone was at least keeping an eye on them. Especially when prices could easily plunge again, if and when the government stops artificially propping up the market.

13 comments

Comments are closed.