Robert Aliber has a peculiar op-ed in the FT, presenting a TARP plan "that should revive the market in mortgage-related securities, greatly enhance bank capital and earn several tens of billions of dollars for the US Treasury". It’s peculiar because just as the world and Hank Paulson is finally coming around to the fact that US banks are facing a solvency crisis rather than a liquidity crisis, Aliber still uses the kind of back-of-the-envelope calculations we saw a lot of in early 2007 to assert that there’s no solvency crisis at all.
The distress in the US credit market reflects that MRS are no longer priced on a rational basis. Rather, a few companies with a desperate need for cash have sold these securities for 25 cents on the dollar; the accounting conventions require that this price is used to value similar securities owned by other banks.
There are 40m mortgages on residential real estate in the US. Ninety seven per cent or 98 per cent of homeowners make their mortgage payments on a timely basis. The median home price in the US is $250,000. Assume 1m homeowners are subject to foreclosure and that the lenders incur a loss of $100,000 each time a borrower defaults. The losses to the lenders then would total $100bn. If 4 per cent of the homeowners with mortgages default, the losses to the lenders would total about $150bn…
US banks already have reported losses approaching $350bn, or more than twice the estimate of eventual losses of $150bn.
Using this reasoning as a starting point, Aliber then proceeds to come up with a clever plan for essentially monetizing the difference between the $350 billion that banks have written off and the $150 billion that they will eventually lose.
But if you really think that US banks will ultimately suffer no more than $150 billion in mortgage-related losses, I’ve got a triple-A-rated super-senior collateralized Brooklyn Bridge obligation to sell you. The reason banks aren’t lending to each other is that they’re convinced the final losses will be much bigger than the ones already taken, rather than much smaller. (Although, in a piece of good news, the TED spread is now back down to less than 400bp this morning. I doubt that means banks are lending to each other yet, but it’s definitely a move in the right direction.)
The fact is that if there’s $13 trillion of housing wealth in the US, and house prices end up falling by 25%, then that’s $3.25 trillion in paper losses. Of course, not everybody bought or refinanced at the high — but many people did. And when real-estate losses start ticking up into the trillions — and remember, this is just residential, we haven’t even started on commercial real estate yet — it’s improbable, to say the least, that mortgage losses will be capped at $150 billion, especially when most of those loans were extended in an era of easy no-money-down mortgages.
And then, of course, there’s credit cards, and all the other loans banks have made which are much more likely to go sour over the course of a long and painful recession. This is why a bank recapitalization plan makes a hell of a lot more sense than concentrating, as Aliber would have us do, solely on bad mortgage assets.