Jim Surowiecki provides proof that I’ve been writing too much and too hastily about bank nationalization of late:
Jim kicks the debate forwards helpfully, describing the obstacles to bank nationalization in the US — not least the sheer number of banks here. "This would be an incredibly complicated process," he writes, "with massive ripple effects (psychological as well as practical) throughout the entire economy." He also ingeniously uses Willem Buiter’s argument as
mitigating working against US bank nationalization. (Paul Krugman and I took it much more at face value.)
And while I’m on the subject of reasons not to nationalize, there’s also the question of bankers’ pay. How many US government employees do we really want earning more than the president? Already we’ve seen AIG pay out a massive $619 million (and rising) in retention pay, or about $150,000 per employee — how much money would the government end up paying to stop Citigroup employees from leaving? Some AIG employees are getting as much as $4 million to stay, which is ten times the president’s annual salary.
Meanwhile John Hempton and Kevin Drum put forward a pretty concrete idea of how nationalization might work in practice: Hempton calls it "nationalisation after due process". It still involves trying to work out exactly how much the bank’s bad assets are worth, however — an almost-impossible thing to do in this market — and what’s more, it results, once all’s said and done, in a severely undercapitalized bank: in fact Hempton seems to be happy with a bank at just half the required capitalization. (Or, to put it another way, double the maximum allowable leverage.)
Hempton’s an investor in bank stocks, so I can see why equity in highly-levered banks might appeal to him. But from a systemic perspective, this "solution" seems fraught with the risk that the bank in question will simply fail all over again, sooner rather than later.
On the other hand, I do like the proposal put forward by William Wild: leverage a bank’s existing origination and underwriting infrastructure, but fund new loans with ring-fenced new equity provided by outside investors. Here’s his abstract:
This paper outlines a new structure for lending by regulated banks, under
which the Tier 1 capital required to support a new loan is provided by the
borrower’s own equity-holders. In a downturn cyclical environment this
would secure a new, motivated and informed class of bank capital provider
to counter the pro-cyclicality of bank lending. The new structure would be
competitive in terms of cost to borrowers, nondilutive of existing bank
capital and credit risk neutral. It also has the potential to be an
effective instrument of market discipline in economic upcycles and
regulators might consider adopting it as a pillar in any revised bank
This doesn’t address solvency issues at the legacy institution, of course. But it might well help to get banks lending again, which is one of the main things that Willem Buiter was worried about.