The argument against an enforced separation of retail and investment banks - as made by the Government and the Lex column in the Financial Times - seems to based on the fact that both types of bank had to be rescued in the recent financial crisis. Northern Rock, a mortgage lender, had to be nationalised here in the UK. And a host of investment banks, which had nothing to do with the "utility" side of banking, had to be propped up in the US to prevent wider economic disaster. So, say these critics, the restrictions of Glass-Steagall would not have prevented the credit meltdown and a return to such controls would not do so in future.
I think this misses the point. As Mervyn King and many others have argued, the problem with the shape of the banking sector in the run-up to the crisis (and the even more glaring problem that exists now) is the incentives in the system. If an investment bank knows it is "too important to fail" it will tend to be less careful with its bets and investments than if it knew the consequences of heavy losses would be a Lehman-style bankruptcy.
It is true that, even with a return to Glass-Steagall, it is possible to imagine circumstances in which retail or investment banks would need - in extremis- to be rescued. Yet surely the crucial point is that investment banks would be less liable to need a bailout if politicians made it clear to them that they are not regarded as an essential financial utility and that they could not rely on injections of taxpayers' capital and unlimited central bank borrowings in times of difficulty. How to make this sufficiently clear without mandating a divorce?
It seems to me that, following the recent meltdown, the onus is on the opponents of Glass-Steagall to demonstrate why allowing these state-guaranteed universal banks to remain intact makes the financial system - and ordinary members of the public - safer. It's not good enough simply to point out that the proposals on the table would not create a perfect world.