John Hempton says that I'm misrepresenting his views. He doesn't think that the banks should just be allowed to muddle along indefinitely: they should be subjected to price discovery (eg the Geithner plan) and then nationalized if they're found to be insolvent:
I never advocated that muddle through is the right policy. I just happen to think that muddle through will work for most banks because most banks are not that insolvent.
I'm not convinced that the Geithner plan will be particularly great when it comes to price discovery, because correlation is at 1 right now, meaning that differences between good and bad assets get elided, and also because the plan is structured so that you can make a profit even when you overpay. I'm also not convinced that the biggest banks are as solvent as John thinks they are, but he and I have been back and forth on this already, and there's no point in repeating all that.
On the other hand, Ezra Klein has five excellent reasons why it's a good idea to bring in private investors all the same, including this one:
Getting private investors to buy-in to this stage of the recovery plan might make it easier to ensure their cooperation with later stages. It may effectively co-opt the participants. Nationalization is one example. If TIAA-CREF vouches for insolvency, the market is likelier to think nationalization an act economic necessity than political capriciousness.
Brad DeLong says something similar:
I suspect that at least one of the reasons the Obama administration is not nationalizing the banks right now is named "Voinovich": getting 60 votes in the Senate for bank nationalization is no easy task with this Senate until the administration can demonstrate that it has gone the extra mile.
I'm not sure I buy it: this plan is not a necessary precursor to nationalization, and the government has given no indication whatsoever that nationalization is its last resort. Yes, it might be a tiny bit easier if this plan fails. But that can't possibly be a good way of getting there from here.