The wrong kind of bail-out?

An excellent column by Sebastian Mallaby looks at the unfolding Fed-Treasury plan and finds it wanting:

The plan is being marketed under false pretenses. Supporters have invoked the shining success of the Resolution Trust Corporation as justification and precedent. But the RTC, which was created in 1989 to clean up the wreckage of the savings-and-loan crisis, bears little resemblance to what is being contemplated now. The RTC collected and eventually sold off loans made by thrifts that had gone bust. The administration proposes to buy up bad loans before the lenders go bust. This difference raises several questions.

The first is whether the bailout is necessary. In 1989, there was no choice. The federal government insured the thrifts, so when they failed, the feds were left holding their loans; the RTC's job was simply to get rid of them. But in buying bad loans before banks fail, the Bush administration would be signing up for a financial war of choice. It would spend billions of dollars on the theory that preemption will avert the mass destruction of banks. There are cheaper ways to stabilize the system.

In the 1980s, the government did not need a strategy to decide which bad loans to take over; it dealt with anything that fell into its lap as a result of a thrift bankruptcy. But under the current proposal, the government would go out and shop for bad loans. These come in all shapes and sizes, so the government would have to judge what type of loans it wants. They are illiquid, so it's hard to know how to value them. Bad loans are weighing down the financial system precisely because private-sector experts can't determine their worth. The government would have no better handle on the problem.

In practice this means the government would make subjective choices about which bad loans to buy, and it would pay more than fair value. Billions in taxpayer money would be transferred to the shareholders and creditors of banks, and the banks from which the government bought most loans would be subsidized more than their rivals. If the government bought the most from the sickest institutions, it would be slowing the healthy process in which strong players buy up the weak, delaying an eventual recovery. The haggling over which banks got to unload the most would drag on for months. So the hope that this "systematic" plan can be a near-term substitute for ad hoc AIG-style bailouts is illusory.

I'm a little reluctant to second-guess the proposal put together by Bernanke and Paulson because I don't know everything the Fed knows about the fragility of the credit markets and the urgency of the case. But I agree that the RTC analogy is wrong, and the column is surely right about the problems the Fed-Treasury plan faces. The article goes on to mention separate alternative proposals by Charles Calomiris and Raghuram Rajan. Both stress the need to recapitalise the banks. Calomiris would do it through government purchases of equity, Rajan through mandatory rights issues and a prohibition of bank dividend payments.

You can read fuller statements of these interesting proposals here and here on Martin Wolf's FT economists' forum. (Be sure to read Willem Buiter's comments on each article as well.) These ideas definitely have attractive features--but, to put it mildly, they are not without difficulty and involve complications of their own. For instance, Rajan says:

I suggest restricting the rights requirement only to well-capitalised entities. This may seem like penalising shareholders of well-performing companies. But in fact these are institutions that could use more capital very profitably in buying underpriced assets, and taking over weaker financial companies. Authorities could also reward these companies by facilitating acquisitions, possibly through favourable tax treatment. By contrast, forcing weak companies to issue rights risks tanking an already fragile share price, and is not a risk worth taking at this juncture.

Agreed: but how do we define a "well-capitalised entity" for the purposes of this mandate? If the bar is set too low, the "risk not worth taking" in that last sentence comes into play. Calomiris says:

To ensure that MPS [Matched Preferred Stock--his proposal for government purchases of equity] is only supplied as truly needed from a systemic standpoint, and to limit any abuse of the taxpayer-provided subsidy, the private sector would also be required to act collectively to help recapitalize undercapitalized banks, and share the risks associated with recapitalizing banks.

Specifically, to qualify for MPS assistance from the government, a bank would have to first obtain approval from "the Syndicate" of private banks (including the major institutions who would benefit from the plan as well as others who would benefit from the reduction in systemic risk) to commit to underwrite common stock of the institution receiving MPS in an amount equal to, say, at least 50 per cent of the amount of MPS it is applying for (at a price agreed between the Syndicate and the bank at the time of its application fro MPS). The Syndicate would share the underwriting burden on some pro rata basis. To support that underwriting, the Syndicate would have access to a line of credit from the US government (and from other countries' governments, if non-US banks participate in the MPS system)... For banks participating in the MPS plan that are based outside the US, foreign governments would have to provide the MPS investments. Presumably, those foreign governments would also provide the credit line commitment to the syndicate for its underwriting of common stock.

Much as I like this plan in principle, I don't think I would celebrate simplicity as one of its chief virtues.

It will be interesting to see whether Congress insists on a debate of these and other alternative strategies, or concentrates merely on larding the Paulson-Bernanke approach with additional subsidies for distressed home-buyers.