When should you protect the innocent even though you may, in the process, let off the guilty? A pacifist would presumably say "always" because war, by its very nature, leads to the slaughter of innocents (or, in our world of media spin, gives rise to "collateral damage"). Those in favour of zero tolerance for all forms of crime might have to say "never" because no judicial system is ever going to weed out those who truly are guilty from those who are wrongfully accused. And then, to make things more complicated, there are those occasions when saving the life of one innocent person requires the loss of another (think human shields or life-threatening child-birth).

When policymakers consider such issues, they junk their natural moral absolutism and turn to the ethics of Utilitarianism, typically defined as the "greatest happiness of the greatest number". Utilitarianism is, of course, far from perfect, a point made forcibly by Aldous Huxley in Brave New World, but, for those who govern, it has the advantage of allowing decisions by numbers. Policymakers can think in terms of costs and benefits and reach a conclusion which offers the illusion of objectivity.

Policymakers' responses to the sub-prime crisis and the possibility of a credit crunch will, ultimately, depend on their views about the protection of the innocent and the punishment of the guilty. Admittedly, the choices won't be quite as wrenching as those faced in other spheres of public life but it's already becoming apparent that the Utilitarian calculations shaping policy decisions are beginning to shift. At first, the attitude towards the sub-prime problems was one of zero tolerance. There would be no bail-out because both lenders and borrowers should have known better. In a matter of days, though, policymakers' attitudes began to evolve.

When things first started to go wrong, the crisis was lodged firmly within the banking system. Many banks and other financial institutions owned collateralised debt obligations (one of a variety of structured products). The values of some of these structures were underpinned by sub-prime debt. These underpinnings clearly weren't terribly strong. It wasn't long before faith in asset-backed collateralised debt obligations was gone and, with it, liquidity dried up. No one wanted to lend to anyone else, because no one knew what exposures others had to CDOs and, indirectly, to the riskier end of the US housing market.

Liquidity crises, though, do not discriminate well between those who deserve to go bust and those who don't. Although, at first, the attitude from central bankers was from the Dirty Harry school of thought ("You gotta ask yourself one question – "Do I feel lucky?" – well, do you, bank?"), this approach quickly mellowed as the dangers of a liquidity crunch became all too clear. Central banks injected liquidity, the Federal Reserve cut its penalty discount rate and the liquidity crisis began to ease.

The debate has since moved on. The big fear now is the onset of a credit crunch. In simple terms, a credit crunch can be regarded as a tightening of monetary conditions for a given level of policy rates. In other words, lending standards in the financial system tighten independently of any decisions made by the central bank. For the man or woman on the street, this may imply higher borrowing costs, bigger required down-payments for the purchase of a property, a lowering of maximum income multiples for mortgages, tougher foreclosure conditions and so on. Put another way, even if interest rates aren't particularly high by historic standards, the ability to borrow (and to renegotiate terms on a favourable basis) is reduced.

This, in turn, may have a series of negative effects on the broader economy. If people cannot borrow so easily, demand for the new car or the loft conversion may begin to decline. Tighter mortgage terms may mean people have to spend more of their income repaying their mortgage, leaving less available for the trip to the local restaurant or ball game. In time, these downward multiplier effects may lead to rising unemployment and, eventually, the onset of recession.

It's for this reason that President Bush chose to step in last week. His proposals to expand the role of the Federal Housing Administration in order to help with the refinancing of sub-prime mortgages and his call for lenders to show greater flexibility towards some of their more distressed customers are recognition of the potential severity of a possible crunch (and its associated loss of votes): "I've made this a top priority to help our homeowners navigate these financial challenges so that as many families as possible can stay in their homes."

Soothing words, perhaps, but they don't quite deal with the underlying policy problem. Central bankers (more so in Europe than in the US) have been worrying for a number of years about the degree of excessive risk-taking within the financial system, whether reflected in rapid money supply growth, incredibly high correlations of returns across both asset classes and geographies, the innovation of structured products or the emergence of private equity (which, supposedly, turns base metal into gold). We may now have the makings of a credit crunch but, arguably, its foundations were laid during the earlier credit "bubble" – when financial conditions were unusually lax for a given level of official interest rates. Perhaps central bankers were also in the wrong, refusing to raise interest rates quickly in the light of the US housing bubble (and similar housing bubbles elsewhere), all too happy to argue publicly that the gains in house prices were simply a reflection of welcome financial innovation rather than of lax lending standards.

If a credit crunch develops, it will ultimately reflect a revised view of the risks associated with the various products that sprang up in the earlier credit boom and which did so much to secure a healthy growth rate and a high level of employment in the US. At the moment, there is no desire to provide any bail-outs for the institutions (or, indeed, the reckless borrowers) which now are seen to be at the epicentre of additional risk-taking. As Ben Bernanke, the chairman of the Federal Reserve, put it, "It is not the responsibility of the Federal Reserve – nor would it be appropriate – to protect lenders and investors from the consequences of their financial decisions."

Fine words and, for the most part, entirely understandable. A bail-out of the reckless, the gullible and the unscrupulous should not really be part of public policy. It creates the classic moral hazard problem, allowing people to believe that the public purse will always be opened to "reward" bad behaviour (the financial market version of this is the "Greenspan put", the idea that interest rates will always be cut to ward off financial distress, thereby encouraging people to take stupid risks). At the end of the day, though, people know that the economy, or the banking system, is "too big to fail". After all, the Savings and Loans industry was bailed out from 1989 onwards, even though its problems resulted, once again, from the behaviour of the reckless, gullible and unscrupulous.

In wartime, policymakers have to accept the death of innocents. In peacetime, society demands the protection of the innocent. Sometimes, though, it's not possible to provide that protection without bailing out the guilty. The moral hazard problem won't easily go away. After all, there will always be those asking themselves Dirty Harry's question – "Am I feeling lucky?"

Stephen King is managing director of economics at HSBC: stephen.king@hsbcib.com

Five of the world's most important central banks had joined together to promise they would rescue the stricken credit markets, but investors can't decide between euphoria and more despair.

The credit markets didn't suddenly thaw, banks didn't instantly start lending to each other again, the crucial inter-bank interest rate that measures these things stayed stubbornly close to its highs. But yesterday's gloomy predictions that the interventions are failing are as premature as Wednesday's elation.

What exactly do we mean when we ask, will it work? There are two main measures. First and foremost, the central banks are trying to avoid any more banking blow-ups such as the one at Northern Rock in September. That has been a particularly acute concern as the end of the year approaches, an important moment in the accounting calendar when banks may need to top up collateral for loans and recapitalise other parts of their business.

Secondly, and more broadly, the central banks hope to thaw the frozen credit markets and bring down market interest rates. Banks and the money market funds that grease the financial system have become more reluctant to lend to each other, because they have their own requirements for the money and because they fear their rivals may not be able to pay them back. That has a cascading effect through the credit markets: businesses have found it harder and more expensive to borrow cash to fund their operations, mortgages are harder to come by, the wider economy is threatened.

On Wednesday, central bankers unveiled a raft of creative ways to encourage institutions to come to them for funds instead of relying on each other, in the hope that this will, eventually, reduce the fear that has paralysed the markets. Where normally the central banks would hand out loans at a predetermined, punitive, interest rate via what is called the "discount window", they will next week begin offering more than $100bn (£49bn) of loans by way of an auction. As well as reducing the stigma, it should also reduce the interest rate on borrowings from the central banks, and that might be enough to tempt in participants looking for a bargain.

Andy Hornby, the chief executive of HBOS, said he did not understand the market's negative reaction yesterday. "I only see upside in it. It is good to see concerted action. These are worldwide markets. All wholesale markets are by implication global."

He predicted that banks will participate in the auctions, the first of which are being held by the ECB and the Fed on Monday, with the Bank of England following on Tuesday. "There will be general demand for more," Mr Hornby said.

John Ewan, the director of the British Bankers' Association, said that other members were also cautiously optimistic about the new plan.

"Banks don't want to use the discount window because, whatever the reason they use it, they can be perceived as having trouble funding themselves. And as we know, it is a short step from the perception of Bank X being in trouble to Bank X really getting into trouble," Mr Ewan said, making a nod to the run on Northern Rock.

"The sums involved in the central banks' intervention are relatively small in comparison to the size of the losses suffered on sub-prime mortgages, which are estimated in the hundreds of billions of dollars. But central banks have left themselves room to intervene again in the coming weeks if it is necessary. More of the same would be welcome, but the members I have spoken to are happier than they were two days ago."

Jonathan Wilmot, the fixed-income researcher at Credit Suisse, said he suspected the Fed was keen to encourage arbitraging. In other words, that banks would participate in the auction in the hope of raising funds at a low level, then lending them straight back out to other institutions at what is currently a high inter-bank lending rate, known as Libor. That extra supply would then drive Libor down closer to the target interest rates that the Fed, the Bank of England and others use to manipulate their economies.

But the Libor interest rate did not crash yesterday, disappointing many who had expected that the central banks' announcement on its own would be enough to ungum the credit markets. In the eurozone, the cost to borrow for three months remained completely unchanged at 4.95 per cent, 95 basis points more than the ECB's benchmark interest rate of 4 per cent, compared with 57 basis points a month ago. The difference averaged 25 basis points in the first half of the year, before losses on securities on US sub-prime mortgages contaminated credit markets. Sterling Libor dropped a modest 12 basis points.

In a note to clients, Mr Wilmot wrote that reducing the rates at which banks lend between each other "will depend on the willingness to lend and not hoard new deposits received [from central banks]. The jury will remain out on this measure until we actually see how banks behave with this money."

The BBA's Mr Ewan concurred. "There is not likely to be a moment where we can say 'gosh, chaps, it's almost like June again'. Libor is more likely to come drifting down rather than drop suddenly."

Mr Wilmot and his colleagues at Credit Suisse yesterday also highlighted other ways that the central banks' new plan will alleviate pressure on financial institutions. By widening the types of assets that can be used as collateral to include some types of mortgage-backed credit derivatives (called collateralised debt obligations, or CDOs), the Federal Reserve will be reducing the unwanted supply of these riskier types of assets in the market.

Best of all, the face value of many CDOs is higher than they would actually fetch in the market, so handing them to the Fed as collateral for a loan is much better for a bank than selling them. At the very least, banks are delaying crystallising their losses, and may avoid a big loss all together if prices rebound in the new year.

John Lonski, an economist at the credit rating agency Moody's, is another with a "jury-is-out" viewpoint, but he suggests that credit markets are gripped by an irrationality that might not properly be shaken until the outlook for the US economy is clearer. Junk bonds are pricing in a significantly sharper rise in corporate debt defaults than seems likely, he says, and the value of mortgage-backed derivatives suggests a higher rate of defaults by borrowers than seems possible. And yet, markets continue to demand big interest rate cuts by the Fed to stimulate the whole economy.

In other words, the markets simply are not listening to central banks.

"We continue to see a high level of risk aversion in the banking sector, but it does not appear to be reflected in the observable fundamentals.

"It's a battle. The Fed clearly does not think the outlook is as dire as the financial markets believe," Mr Lonski said.

"Of course, we are looking at the deepest cutbacks in financial services sector employment since the early Nineties, so perhaps it is understandable that Wall Street has a gloomy view of the prospects for 2008."

Off-balance-sheet rescue plan

The casualties of the credit crisis are littered all around – among them, rescue plans meant to ease the crisis.

Less than two months ago, Hank Paulson, the US Treasury Secretary, was trumpeting an idea meant to prevent tens of billions of dollars of losses on the secretive off-balance-sheet vehicles created by investment banks such as Citigroup, Barclays and others.

Citigroup said it was leading a consortium of financial institutions to create an $80bn-$100bn fund that would buy up mortgage-backed debts that few other Wall Street investors now want to buy. The plan was thrashed out at meetings organised by Mr Paulson, but even his intervention could not rouse significant support.

All week, Citigroup and its two partners, Bank of America and JP Morgan Chase, have been trying to drum up money for the fund, but it is clear that it will have to be scaled back. Executives at two banks, Sumitomo Mitsui of Japan and the US-based Wachovia, played down the importance of the fund, while many others behind the scenes have turned down a chance to participate.

The aim of the fund, called the Master Liquidity Enhancement Conduit (M-LEC), is to prop up dozens of off-balance-sheet vehicles – known variously as structured investment vehicles, SIVs. SIVs raise money at low interest rates in the short-term debt markets to buy longer-term investments, mainly mortgage-related securities, which pay a higher interest rate. The value of the longer-term assets has collapsed, however, and SIVs are finding it hard to find outside investors to refinance them. The M-LEC is meant to step in as a buyer of last resort for assets that might otherwise have to be sold at fire-sale prices.

Analysts say that banks are finding other, albeit more painful, ways to resolve their leaking SIVs. HSBC has bailed out its vehicles by taking the assets on to their balance sheet.

Stephen Foley