The UK banking sector needs first to be sanitised and then to be regulated. The former is a pragmatic matter, of identifying a workable, hopefully near-optimal route to recapitalising and re-privatising Britain’s collapsed banking conglomerates (Lloyds-HBOS, RBS and Northern Rock). But regulation needs to rest on a correct understanding of first principles — something of which recent statements by UK financial authorities have shown little sign.

The rationale of bank regulation emerged from 19th-century experience. The developing pivotal role of banks in the economy meant that failures could cause damage beyond merely the depositors, staff and proprietors. At the same time, competing bank managements were evidently willing to embrace excessive levels of risk. Corrective action was called for, with the central bank as lender of last resort a) guaranteeing banks’ liquidity to meet any deposit withdrawals, and b) simultaneously imposing prudential limits on their risk-taking.

Inherent in this proven — but not always acknowledged — framework are four further points. First, effective bank regulation is inseparable from restraint on competition. The socially optimal intensity of competition in banking is low. During the past 35 years the main impact of enhanced competition in UK banking has been ballooning of real estate prices on one side and of personal indebtedness on the other, with little accompanying change in the proportion of owner-occupied housing. Alistair Darling’s fresh enthusiasm for more high street competition in banking is altogether misguided. So is the earlier response from the authorities in Brussels, with their knee-jerk reaction to supposedly anti-competitive aspects of the Government’s rescue operations.

Second, it is absurd for a central bank governor to complain that his hand is forced because of the size of banks (“too big to fail”). The policy of preventing bank failures long antedated their present size. Indeed, that size is partly a result of the policy itself, not least because regulation is in some respects easier when institutions are fewer in number.

Third, this is not to say that banks should never be broken up, or prevented from making takeovers, or have their businesses segmented. Quite the contrary. The most obvious instance in postwar Britain which should not have been permitted and which needs to be reversed is the amalgamation of banks and building societies, and more generally the unbridled participation of commercial banks in the domestic mortgage market. But the purpose of such reversal and similar measures must be to enforce regulation, not to encourage competition or to permit bankruptcies.

Lastly, banks can be adequately regulated only on an itemised and case-by-case basis and by the national central bank. In other words, the central bank needs to crawl all over a bank’s business, both on and off-balance sheet; and if it doesn’t like what it sees, it must require the resignation of the bank’s chairman and/or chief executive. Formulaic devices such as capital adequacy ratios may assist, but cannot carry ultimate responsibility.

Deposit insurance is a snare and a delusion. As the American example shows, when things get hot the insurance system is overwhelmed and becomes part of the problem.

As for the FSA, it may inadvertently succeed in provoking the odd bank closure or two; but to be sure of preventing them it depends, as we all do, on the Bank of England. There are plenty of non-bank financial institutions to keep the FSA busy.

The author is an emeritus fellow of Christ Church, Oxford

Times Recommends

• Lies and cover-up at MG Rover exposed by report
• High oil prices push manufacturing costs up
• Primark founder steps down as chief executive