There has been a small but increasing amount of attention being given to the “good bank” idea: instead of creating a government entity to buy toxic assets from existin banks – or nationalizing existing banks, removing their toxic assets, and then reprivatize them – why not create brand new, good banks with the same government money, enabling them to lend money unencumbered by previous bad decisions, and then privatize them? (Willem Buiter floated this idea on January 29, and Paul Romer has a similar proposal in the WSJ, although I’m proud to say that Nemo, who has his or her own blog, raised it in a comment on this blog two weeks earlier.)
Romer suggests using government capital to create new, healthy banks that can essentially compete with the existing banks, which can then be treated under existing rules and regulations – if they become insolvent, they get taken over; some of their liabilities (like FDIC-insured deposits) are guaranteed, and some aren’t – and that’s that. Buiter goes a step further and recommends taking away banking licenses from the legacy bad banks and making them institutions that just run off their existing assets, in part by selling their good assets to the new good banks.
These are elegant solutions because they get to where we want to be – healthy banks – yet avoid the problem of overpaying for toxic assets (which can happen in many forms, including non-recourse loans and asset guarantees), and the alternate problem of having to take over legacy banks (which will be politically difficult, given the antipathy that “nationalization” generates in the U.S.).
This approach faces some serious complications, however. First, there is the question of what happens to the legacy banks. Once such a program is announced, their stock prices should go to zero, since those prices are now predicated on friendly government intervention. That may not be so bad. More significantly, no one will want to lend them money, at least not without an explicit government guarantee. So either there will be a bank run – in which case we’ll have to figure out if they are still Too Big To Fail – or the government will have to guarantee all the liabilities that aren’t guaranteed already, which creates a big taxpayer liability. Of course, maybe this is just the same taxpayer liability we run into down any path we look, so this isn’t necessarily a fatal flaw with the plan.
Another complication is this: How do you get a few sufficiently large banks off the ground and running quickly? Romer proposes creating wholesale banks (the kind that only have businesses as customers) because they can be created more quickly (they don’t need huge branch networks), which would then buy good assets from legacy banks. Buiter also says that the new good banks will buy assets, such as deposit accounts and presumably branch networks, from the legacy banks. The legacy banks certainly won’t want to sell their good assets – have you noticed they’ve only been talking about selling the bad ones? – but Romer thinks that faced with being shut down by regulators they’ll have no choice. As a risk-averse person who found starting a small software company a huge endeavor, the idea of trying to create multiple large banks at once out of an office in the Treasury Department seems daunting, but arguably we couldn’t do much worse than the banks we’ve got now.
And who will be running them? Both Buiter and Romer bring up the problem of having the government owning the few healthy banks in the economy and thereby making lending decisions. But both correctly note that we are already committed to significant government involvement in the banking sector. And one of the problems to date has been the contortions the Paulson Treasury Department went through to try to minimize government involvement. Given the current state of the economy and the banking sector, we might do worse than having some lending decisions made by bureaucrats, influenced by political considerations. We also run a similar risk under the nationalize-clean up-privatize scenario. In any case, there are structural mechanisms that could be taken to try to limit government influence, such as creating independent boards of directors (presumably staffed with upstanding citizens from the financial sector, though where you’d find them I’m not sure – I would nominate John Bogle if he still wants to work) with sufficiently long terms to insulate them from Washington.
I may not have done the proposals justice, so read them yourself and see what you think. But before you get too excited, remember that there is nothing like this in Geithner’s Financial Stability Plan, so it is unlikely to happen.
Update (2/18/09): Since Paul Romer is one of the people whose work I discuss here, I want to point out a comment he left below. You can read it here.