On 6 May 1997, the BBC ran a picture of a youthful, determined, courageous Gordon Brown with the text "...the bank will now be free to decide monetary policy without taking the short-term wishes of politicians into account.''

The newly installed chancellor's first and last interest-rate-setting meeting with the governor of the Bank of England - five days after the "new Labour" government led by Tony Blair had been elected in Britain after eighteen years of Conservative rule - was thus an announcement of the bank's independence.

A decade on, and less than three months after Brown inherited the prime ministership from Blair on 27 June, the bank's behaviour over the crisis surrounding Northern Rock - Britain's fifth largest mortgage-lender - suggests that there is nothing much left of that independence. It seems certain that the bail-out was ordered from the Treasury, now led by Brown's successor as chancellor, Alistair Darling. Take this sequence of events:

* On Wednesday 12 September, Mervyn King (since 2003, the Bank of England's governor) wrote to the Treasury select committee of the House of Commons that bank bail-outs are bad because the implicit safety-net they give bankers "encourages herd behaviour and increases the intensity of future crises". The mechanism by which this happens is that of "moral hazard"': if banks can count on being bailed out, they will have little regard to risk when lending. Their chief function in the global economy, as allocators and judges of risk, is fundamentally damaged (see "The end of gentlemanly capitalism", 13 August 2007).

Tony Curzon Price is the editor-in-chief of openDemocracy. He worked as a consultant economist for more than ten years. Since 1997, he has lectured on economics and energy policy to postgraduates at Imperial College, London, and at the École Polytechnique Fédérale de Lausanne (EPFL)

Among Tony Curzon Price's articles on openDemocracy:

"The ‘as if' economist: Milton Friedman's legacy"(27 November 2006)

"The wisdom of the openDemocracy crowd"(29 December 2006)

"The Economist Redux"(5 February 2007)

"Tony Blair and centralisation"(20 February 2007)

"The reach of economics: a reply to Diane Coyle"(13 March 2007)

"Das Google Problem: is the invisible mouse benevolent?"(20 April 2007)

"The reinvention of scarcity" (13 June 2007)

"Making up minds" (23 July 2007)

"Corporate liability and social interest" (25 July 2007)

"The end of gentlemanly capitalism" (13 August 2007)

"The conditions of quality" (22 August 2007

* On Thursday 13 September, Northern Rock, having failed to find a buyer for itself, formally applied for emergency loans from the Bank of England (which one presumes King would have known was on the cards on Tuesday or Wednesday)

* By Thursday evening, the bank had announced the Northern Rock bail-out, and stressed that the Treasury and the financial services authority (FSA) had been involved in the decision.

Willem Buiter and Anne Sibert, both professors of economics in London, remind us that the "memorandum of understanding" that sets out the nature of the central bank's independence gives the Treasury ultimate responsibility for bank bail-outs, and that these should occur only when there is "a genuine threat to the stability of the financial system" (see "Bail-out that will damage Bank's credibility" Financial Times, 16 September 2007). Thus it is Alistair Darling - and not Mervyn King - who was heard across the airwaves reassuring depositors that their savings are safe and extending the emergency loan.

If Northern Rock's collapse posed a threat of system-wide collapse, the Treasury did right. If not - as I believe - the rescue was politically opportunistic - just as were the interventions in monetary policy in the bad old days before 6 May 1997, which Gordon Brown is so rightly proud to have put an end to.

The return of risk

Northern Rock has suffered from the failure of its basic business model, which involved three simple steps:

1 convince someone to take on a mortgage, with rates fixed for some term

2 find investors willing to advance you the capital for the mortgage, shopping around for the cheapest source of capital ... very often finding that you borrow for a shorter period than the corresponding lending you have extended (encapsulated as the risky strategy of ``borrow short, lend long'')

3 trust that you can get the borrowers to pay you more than the lenders.

A single Northern Rock mortgage, through step 2 of the business model, would often find itself mixed in with all sorts of assets, chopped and diced in "mortgage-backed securities'' and presented in the most attractive form to the ultimate investor with cash. The market for these derived products, many of them containing highly mis-priced slices of assets (including American sub-prime loans) has more or less dried up. You need to offer very high interest rates to convince anyone to swap the safety of cash, government bonds or equities, for them.

The six weeks since early August 2007 have witnessed the sudden return of a phenomenon that since 1999 had virtually vanished: the "risk premium" (that is, the amount that investors are compensated for taking on financial risk). A combination of faith in financial wizardry (the development of new ways to slice and dice risky assets) over the past eight years, reinforced by central-bank policy after 9/11, had led to its disappearance - thus opening a period of generalised recklessness in lending. When Mervyn King wrote to the Treasury select committee on 12 September, explaining that moral hazard needed to be avoided, he was in effect welcoming the reappearance of the risk premium.

The business or the depositors?

Northern Rock's business model, although not necessarily idiotic - after all, it is just a financial intermediary - was relying in these circumstances on the risk premium having disappeared forever. The problem with borrowing short and lending long is that your cost of borrowing can rise without a corresponding increase in your receipts from lending. The return of the risk premium would inevitably do that, and would inevitably threaten Northern Rock's viability.

Also in openDemocracy on a financial crisis's fallout:

Ann Pettifor, "Debtonation: how globalisation dies" (15 August 2007)

Christopher Harvie, "Gordon Brown vs Scotland: the balance-sheet" (17 September 2007) To base a business strategy on the end of risk was unwise: Northern Rock, in a sense, would sink sooner or later. The sinking, in other words, has nothing to do with a traditional "run"' on Northern Rock; the sinking is not caused by a loss of confidence - or rather, both are caused by quite genuine problems with the business.

The special nature of deposit-taking means that deposits are regulated in special ways. Just as we unquestionably expect our doctors to improve our health, and regulate them accordingly, so we expect our savings to, indeed, be safe. So we regulate deposit-takers accordingly. And we operate a scheme of social insurance in case things go wrong: deposits up to £2,000 are guaranteed by other taxpayers, as is 90% of the deposit up to £35,000.

So, given the "fundamental"' nature of the problems at Northern Rock, why did the Treasury decide to save the business rather than save the depositors?

The argument that the panic was contagious and would provoke a system-wide collapse seems unlikely. The trouble the big banks are in today is not the same as Northern Rock's. Banks are asking for stiff premiums on short- term lending, because they worry that they will need the cash. They are adjusting the costs and revenue streams associated with their financing activities in a way that Northern Rock could not do because of the fixed contracts entered into with mortgage-holders. In the market for credit from banks, interest rates have gone up, and a risk premium has emerged around liquidity questions.

The politics of retreat

The evidence of Mervyn King's 12 September letter suggests that the Bank of England's governor would have preferred to see the collapse of Northern Rock, followed by the admittedly painful social-insurance mechanism to compensate depositors. It would have sent a clear signal to bankers that the safety-net of the past had been removed from under their bad investment practices. The management and shareholders of the company would have been seriously reprimanded, and what effort and funds were to go into rescue would go to help depositors.

Mortgage-holders would presumably have seen their contracts sold to others in the liquidation, with the terms of those mortgages possibly changing. It would have been entirely possible for the Treasury to decide to help those whose mortgages had become less attractive as risk was repriced. This would have been a properly political decision, with the case for compassionate redistribution being made explicit.

In other words, a political solution was available which achieved three things:

* preserving the Bank of England's reputation for independence

* sending the right message to financial markets

* doing right by the depositors and mortgage-holders of Northern Rock, while punishing management and shareholders.

To this was preferred a bail-out with diffuse but long-term costs - the cost of reputation to the bank, and of more bad lending in the future, as bankers learn that certain sectors really can't be allowed to go down.

The lesson from the fudges of monetary policy are very resonant. The great sigh of relief from Brown's courageous decision of 6 May 1997 was that no longer would there be a pre-election mini-boom orchestrated by cheap money, leading to a post-election hangover that would last for half the government's term. Those election-cycle booms were a constant temptation to the incumbent politicians, because the effects of changes in monetary policy are diffuse, not intuitively linked to policy action, and the cost - in the form of future inflation - could always be presented as caused by something else. Independent central banks armed with inflation targets certainly eliminated that cycle.

Today, however, we see the same sort of political cowardice at work in the pretence that a Northern Rock collapse would create a systemic risk. That pretence allows the government to avoid the real and hard political questions emerging from the return of the risk premium (especially, how much compassionate redistribution to have towards those whose riskiness had been mis-priced) and to hide behind a technocratic-sounding claim that "system risks"' should be administered to by central-bank medicine.

The government must be wondering whether to call an election in October 2007 - more than two years before one is due, but a good moment to exploit Gordon Brown's lead in the opinion polls and disarray among the Conservative opposition. A messy denouement to Northern Rock would take that option off the table. Brown's hard-won reputation for good economic management would be in doubt, while the question of how to compensate the "victims"' of a bankruptcy would present voters with hard questions of priorities. I assume the desire to maintain the option of an early election will have played some part in the decision to bail out Northern Rock.

There was courage in Gordon Brown's decision on Bank of England independence in the aftermath of a landslide election victory in 1997. He was saying to the people of Britain - citizens, voters, mortgage-holders, consumers - "Let's have a grown-up democracy in which we don't try to fool ourselves about what is happening. Let that reality sometimes be hard, when the bank does something we would rather avoid. But let's accept that it is good, and mature, to commit ourselves to not always taking the easy monetary option.''

The courage of that message would still inspire today. It is a pity that the British prime minister has missed an opportunity to deliver it again.