Will the US Make Money on its Bailout?

Jim Surowiecki is quite sure that Treasury's bailout plan, or at least the $250 billion part of being spent on recapitalization, is an investment rather than an expenditure:

I realize that, given the way the U.S. budget is accounted for, it's accurate to say the $250 billion package is increasing the deficit. But it'd be good to see some acknowledgement that in this case, "spending" that money is going to make the government richer, not poorer.

Richer? I doubt it. The US has past history here: the S&L bailout, which was smaller than this one, ended up with a cost to the government 3.2% of GDP.

Bank bailouts in other developed countries have had similar fiscal results: Norway's bank crisis of 1987 cost the government 8% of GDP, the Finnish bank crisis of 1991 cost 11% of GDP, and Japan's bank crisis from 1992 onwards cost a whopping 20% of GDP. If you consider South Korea a developed country, its bank crisis of 1997 cost more than 26% of GDP. And bank bailouts in developing countries can be much more expensive still.

All these figures come from a 2000 World Bank report entitled "Controlling the Fiscal Costs of Banking Crises"; its authors, Patrick Honohan and Daniela Klingebiel, write that

Fiscal costs are systematically associated with a set of crisis management strategies. Our empirical findings reveal that unlimited deposit guarantees, open-ended liquidity support, repeated recapitalizations, debtor bail-outs and regulatory forbearance add significantly and sizably to costs.

Sound familiar?

Adrian Ash was on this back in March, citing I think a different version of the same paper:

On average, the World Bank economists found, "governments spent an average of nearly 13 percent of GDP cleaning up their financial systems" as a result of the bailout programs they tried to implement.
"Indeed, each of the accommodating measures examined," they continued, ..."appears to significantly increase the costs of banking crises."

None of this is to say that Paulson's recapitalization plan is a bad idea. But the probability that it's going to end up making a profit is pretty low, if past experience is any guide.

Update: Surowiecki stands his ground.