I've been having a look at Nouriel Roubini's estimate of a total of $1.8 trillion of losses in the US banking system, compared to just $230 billion in TARP recapitalizations to date and $200 billion of new money from private-sector sources. And while I might niggle with a few of the numbers he uses to reach that $1.8 trillion number, there's nothing there which is obviously crazy: it's very much within the realm of possibility.
There's no one huge number which accounts for most of the $1.8 trillion, although the single scariest one is probably the $295 billion that Nouriel reckons banks are going to lose on their commercial and industrial loans -- a number which doesn't even include another $35 billion in leveraged loan losses or $127 billion in mark-to-market losses on high-grade and high-yield corporate debt which ended up on US banks' balance sheets despite the fact that it was securitized.
Nouriel uses a naive view of the Markit ABX, TABX, and CMBX indices to work out the mark-to-market value of asset-backed bonds; I'd quibble with that. On the other hand, there are lots of potentially dodgy assets which Nouriel isn't including in his calculations, including sovereign debt, municipal debt, and -- biggest of all -- loans extended by US banks to foreign companies and individuals. We've already been through a round of European banks taking big losses on their US assets; we haven't even begun to see US banks take losses on their European assets, or their loans to formerly-flush companies in commodity-rich countries like Russia, Brazil, and Australia.
It's worth bearing Nouriel's numbers in mind when reading Tim Geithner's testimony from this morning:
The tragic history of financial crises is a history of failures by governments to act with the speed and force commensurate with the severity of the crisis. If our policy response is tentative and incrementalist, if we do not demonstrate by our actions a clear and consistent commitment to do what is necessary to solve the problem, then we risk greater damage to living standards, to the economy's productive potential, and to the fabric of our financial system.
Senators, the ultimate costs of this crisis will be greater, if we do not act with sufficient strength now.
In a crisis of this magnitude, the most prudent course is the most forceful course.
"The most forceful course" is clearly not the incrementalist one of adding to Paulson's TARP here, throwing in a few tax cuts there, announcing some big public works, and hoping for the best. The cost of recapitalizing the banking system alone -- to say nothing of the losses elsewhere within the shadow banking system -- might well be larger than Obama's entire stimulus plan.
Now those costs don't need to be paid for in cash dollars. Government guarantees, whether they arrive implicitly, through nationalization, or explicitly, through loan acquisition and the creation of a ring-fenced "bad bank", can do a lot of good at zero up-front cost. But given the commitment that both Obama and Geithner have made to transparency, I do hope that they're very clear about the real present value of such guarantees -- a sum of money which could easily run into the hundreds of billions of dollars.
But most of all, given the implosion in the Citigroup (C) and Bank of America (BAC)share prices after they got bailed out with government loan guarantees, I hope that Geithner has learned that ad hoc solutions are a recipe for disaster. I look forward to an announcement, soon, of something big, coherent, consistent, and transparent. Not that Geithner gave any hint of having any such thing in mind during today's testimony.