It is the supervision of the banks, not their structure, that is most important

Much of the current debate about the future of banks in Britain takes as its starting point "too big to fail" (or "too important. . ." as the governor of the Bank of England would prefer) with the central concern of avoiding any future call upon public money to rescue a bank in danger of failure. This starting point can be helpfully expanded into "too big to be allowed to fail", presenting us with a problem which has two aspects: "too big" and "fail".

One approach (apparently looked on with favour by the governor and also by the chairman of the Independent Commission on Banking) focuses upon the "too big" aspect of the problem and proposes solving it by imposing a change in the structure of Britain's large universal banks, separating each of them into two smaller units, a retail bank and an investment bank. These, like all banks in future, would have a living will, detailing how their affairs would be settled in the event of failure. This structural approach appears to accept that any bank is liable to degenerate into a stumbling institution which, being smaller, could be allowed to fail because any wreckage could be quickly and easily cleared up, and there would be no cost to public funds. But while such a failure would be smaller, it would hardly be small since the separated parts of British universal banks would themselves still be large banks.

The supervision approach, on the other hand, starts at the "fail" aspect of the problem. It begins by making a clear distinction between a bank as an organisation and its management. It sees a failing bank as a symptom of failing management, and failing management as the real danger to individual bank and system stability. It then looks to a regime of continuous effective bank supervision to curb failing management and to help spread best banking practice. This effective supervision would be entirely different from the light touch supervision which was disastrously ineffective in the run-up to the banking crisis and for which the Treasury, the Bank and particularly the Financial Services Authority all bear a heavy responsibility. This is usually overlooked by those who blame the banks as they are now, even though as organisations they too were victims of the disastrous policies of their then chief executives, most of whom left long ago with their reputations in tatters. Even a moderate level of proper supervision could have greatly eased, perhaps even prevented, Britain's banking crisis.

We now have a stark contrast between the structural and supervision approaches to the problem of "too big to be allowed to fail". The structural approach assumes that the splitting of Britain's universal banks will produce a system in which managers can be left to their own devices. It looks complacently at the possibility that a separated, but still large, part of a universal bank could fall into the hands of bad managers; they would simply be left to apply their lethal talents to the bitter but tidy end. To the structural approach, the important thing is that public money would safeguarded; beyond that, it shows no concern even though such a failure would impose heavy social and financial costs upon employees, customers and shareholders as well as damaging confidence in general. In these circumstances public pressure could well force government to intervene using public money, which would nullify the main objective of the structural approach.

The supervision approach takes a broader, and more responsible, view. It goes to the heart of the problem by seeing sound management as the essential way of preventing bank failure and maintaining system stability. In brief, it aims to ensure this sound management through well-trained supervisors working closely with the banks, scrutinising their operations, bringing additional questioning eyes to bear upon their judgments of risk and broadening their awareness of impending problems. With effective supervision in place, the separation of universal banks into retail and investment banks proposed by the structural approach would add nothing of any value. It would be a drastic half-measure, costly but irrelevant.

A final note. One charge levelled at the major banks is that because they are seen as "too big to be allowed to fail" they gain from the confidence that they will always be bailed out by public money and this gives an undeserved enhancement to their profitability. But where confidence in bank stability is achieved through effective supervision, this is financed by a levy upon the banks to meet the cost of the supervisory authority and any enhancement is then paid for by the banks themselves.

Roger Alford is an emeritus reader in economics at the London School of Economics