When banks voluntarily do principal reductions

By Felix Salmon
July 11, 2011

The holy grail of mortgage modification is principal reduction — the only thing which gets homeowners out of negative equity hell. And one of the big questions is why it’s not more common: it seems to make sense for all concerned, given that a sensibly modified mortgage is likely to be much more profitable for a bank than forcing a homeowner into a short sale or foreclosure and trying to sell off the home in the current market.

Last week the NYT, in a front-page story, found that Chase is actually doing principal reductions — quietly, on some of the most toxic mortgages written during the subprime bubble. But the mechanism was very mysterious — for one thing, the principal reductions were being done on many mortgages which were actually current and in good standing, rather than on mortgages which were careening towards foreclosure.

Philip van Doorn followed up, and my reading of his article — he doesn’t make this explicit — is that there’s actually method to the madness here. In order for banks to offer principal reductions, two criteria need to have been met: (a) they came into the mortgages via acquisition, rather than writing them themselves; and (b) they bought the mortgages at a discount.

Wells Fargo said that even though most of the Pick-a-Pay modifications had resulted “in material payment reduction to the customer,” Wells Fargo had not been forced to make larger provisions for loan loss reserves — which would have hurt earnings results — because of the aggressive write-downs taken when the loans were acquired…

JPMorgan had $24.8 billion in option-ARMS as of March 31 within in its $70.8 billion purchased credit impaired portfolio, acquired as part of the company’s purchase of the failed Washington Mutual from the Federal Deposit Insurance Corp.in September 2008. The PCI loans were written-down to fair value when they were acquired, and as of March 31, JPMorgan said that although it had set aside $4.9 billion in loan loss reserves for all of its PCI loans, “to date, no charge-offs have been recorded on PCI loans.”

It seems that Wells and JP Morgan are happy to do principal reductions only on the mortgages they bought at a discount from Wells Fargo and WaMu respectively; Bank of America, meanwhile, which inherited a bunch of these loans when it acquired Countrywide, is not doing principal reductions, and I don’t think it’s a coincidence that the Countrywide loans were bought at very close to par.

The behavioral psychology here is very easy to understand. No bank wants to admit that it wrote idiotic loans, and write down its own assets from par. Meanwhile, it’s much easier to write up an acquired asset, if the amount you reduce the loan is less than the discount you bought the loan for in the first place.

Economically speaking, however, what the banks are doing here does not make sense. Either writing down option-ARM loans makes sense, from a P&L perspective, or it doesn’t. If it does, then the banks should do so on all their toxic loans, not just the ones they bought at a discount. And if it doesn’t, then they shouldn’t be doing so at all.

The truth is, of course, that banks should be doing principal reductions, and they should be doing them on lots of their loans, rather than just the ones they bought cheap. And the fact that they’re already doing this, entirely voluntarily, on some of their loans is the best possible indication that it makes perfect economic sense to do so on all of their loans. Even if doing so might involve admitting that the subprime crisis still isn’t fully over.

34 comments

Comments are closed.