Royal Bank of Scotland (RBS) recorded last year the biggest loss in British corporate history. The bank had made huge acquisitions at inflated prices, with a depleted capital base and poor risk control. Shareholders have seen their holdings almost wiped out. RBS survives owing to a government rescue costing some £20 billion. The taxpayer now holds a stake of almost 70 per cent.

Yet RBS is preparing to pay annual bonuses to thousands of City staff. The payments may run to hundreds of millions of pounds. Lord Mandelson, the Business Secretary, understandably declares: “They have got to consider how it looks and how it seems when those mistakes and losses have been made.”

It looks like a reward for grievous failure; and the appearance does not deceive. The real economy is suffering from a crisis born in the financial sector. Demand and investment are collapsing; household incomes are being squeezed. The human costs are hardship, unemployment and eviction. Banks have in the past set their compensation levels by looking at the amounts that other banks pay out. RBS remains locked in that insular culture. This is socially untenable. When a bank is the beneficiary of taxpayer funds, it is a matter of simple prudence and equity that it should be seen to consult more widely. But the objection to RBS's bonus policy is not only about appearances. There is a wider problem of misaligned incentives in the banking sector.

The case for the defence is that risk-taking and enterprise ought to generate large rewards, and banking is an important part of a modern economy. And while there will be pressure on the Government to use its stakes in the banks to impose a cap on executive pay, that temptation ought indeed to be resisted. There is a genuine issue about how far a ceiling on pay might merely encourage mediocrity.

The problem in the banking sector is, however, that bankers' pay is not genuinely decided by competitive forces. In the boardroom, it is more usually decided by committee. More widely, bankers' pay and bonuses have not been tied recognisably to performance. The profitability of, for example, Northern Rock in the boom years was unrelated to management skill: it was generated instead by a grossly irresponsible strategy of recycling short-term borrowings.

The practice of paying bonuses in the form of stock options rather than cash does not resolve the problem of linking pay and performance. During the boom, stock options produced large rewards but imposed no penalty for failure. If the stock price of the bank declined, the options might even be repriced. This was no case of reaping the rewards of legitimate risk-taking, for there was no cost of failure. The destructive outcome of the pricking of a huge asset price bubble is being borne by the public, not by the bankers. The banks' bonus system has thus produced perverse incentives, with huge social costs.

There will always be populist objections to large rewards. But the controversy over RBS's bonus payments is more serious. It ought to serve as a catalyst for reforms in the banking sector. A banking cycle may last a decade, whereas bonuses are paid annually. Rewards and long-run profitability must be aligned more closely. Bad business decisions by the banks must in future have only private costs, not costs to the taxpayer.