The Economist reports on the Vicker's Commission's proposals for reforming banking. They appear to do it better than we.
There are two approaches to dealing with failures of large institutions which I will call regulatory and structural. The regulatory approach is to publish vague rules and empower bureaucrats with broad powers to second guess decisions made by private actors. This has the virtue of recognizing an important aspect of the failures of large institutions--you can never tell where they are going to come from or what is going to be the proximate cause. That means or at least suggests to some that you cannot tell in advance what the right thing to do is so the best you can do is have private actors report a lot of data to regulators before or as the private actors make decisions so that the bureaucrats or regulators can make sure the private actors are not taking any undue risks.
There are obvious problems with this approach. For one thing why would we think the bureaucrats are any better at determining what is a bad risk than the private actor. If they were, why wouldn't they be the private actors themselves? Moreover, the endless stream of decisions that must have the regulator's blessing creates a potential for corruption and graft as the power to say no is converted into just plain old power. The very fact that the problems are hard to predict suggests that the regulators are as likely or more likely to be blindsided by events as the business leaders they are meant to correct.
The other approach is what I am terming the structural approach. By structural I mean a simple rule or set of rules that can be written into specific provisions of law and would have the effect of either making disaster less likely or making disaster more survivable if it does occur. In the case of banking regulation structural approaches would be increasing capital requirements or mandating reduced size.
The Vicker's commission has chosen the structural approach. They recommend an increase in capital requirements to 10% of reserves and restructuring of banks to make their investment and retail parts survivable on their own. These make sense to me. I would to see a more general requirement that the banks be limited in size compared to the economy or to the market, but they are certainly going in the right direction.
The regulation approach puts a committee of regulators on the bridge and gives them the power the second guess the captain. The structural approach is the promulgate a rule: have as many places on life boats as you have passengers. Which ship would you rather try to cross the Atlantic on?