Moral hazard – the incentive to take risks that comes from insurance or from governmental bail-outs in the case of failure – must be reined in. Usually, a crisis does the job, as people lose money and then behave more cautiously. But this time was different. So many institutions were helped that moral hazard reinforced itself. The biggest banks continue to be perceived as “too big to fail”.

Conventional wisdom initially held that systemically important institutions should be identified and put into their own tightly regulated category. But, as legislatures and regulators get closer to drawing up new banking rules, they are having second thoughts. One complaint is that this would create a binary distinction. Banks in the charmed circle would behave with greater over-confidence and raise finance more easily from the market. The scale and complexity of the biggest banks would militate against successful regulation, even by more intrusive regulators. And size may be a red herring. Force banks to stay small and they miss out on economies of scale and attract fewer gifted executives. Note the rash of US small bank failures.

Such thoughts prompted the International Monetary Fund’s proposal last week that a levy on banks’ assets to pay for a future crisis-prevention scheme should have a “broad perimeter”, meaning that all financial institutions – not merely those too big to fail – should pay. This makes sense. But how to deal with the obvious moral hazard now afflicting the biggest banks? Raising capital requirements, desirable in itself, might spur disposals. Industrial logic might help as Citigroup, among others, has reached the point of diseconomies of scale. But it is hard to avoid a nasty feeling that moral hazard will only be flushed out by another crisis.

E-mail the Lex team confidentially

Copyright The Financial Times Limited 2018. All rights reserved.