Two weeks ago the idea of nationalising the world's banking system seemed absurd. Now the UK has offered to subscribe for preference shares giving it a majority stake in most of the UK's banks. It looks like many will not take advantage of that offer and the USA looks likely to follow suit and offer a similar arrangement. But what was clear from Friday's stock market trading is that this is still unlikely to stop the crash in share prices.

There is good reason for this. The preference share option that the governments of the UK and USA are pursuing is not a solution to this crisis. As has been the case throughout the crisis these governments are acting too timidly and too late to address the issue they are facing, and those in the market and the public at large realise that.

In this case those in the market realise that, as the Financial Times noted on 11th October "Deep down .. bankers do not expect the crash fundamentally to alter the way that Wall Street works". This realisation is based on the simple fact that although these governments are taking majority stakes in these banks they are refusing to accept the responsibility that goes with that ownership, and enforce change up on the banking system. The public have the same instinctive belief. The result is that the markets do not expect bankers to start lending to each other again whilst the public expect bankers to continue their abusive practices that have both brought them into disrepute and brought our economies to its knees.

There is only one solution to this situation: all the major banks have to be properly nationalised: in other words, control will have to pass to the government. It would be best if this took place simultaneously in all the major financial markets but if it does not then the UK will have to take a lead, single-handedly. Only by taking a bold initiative to ensure that it takes control of the banks and imposes its will upon them will the government be able to break the impasse that has been created by the paranoid aversion to risk taking that has crippled our banks under the direction of those who wish to avoid both criticism and the loss of their bonuses in equal parts. And it is only by showing the necessary initiative to remove those who have crippled these banks that the government will inspire confidence in the general public that it is taking the necessary action to restore calm to the markets and protect their savings.

Nationalising the banks does, however, create the spectre of a cost to society in solving this problem, and at the same time poses the strategic question of what the government would expect to do with these banks once it had acquired them. In practice the two questions can be related as part of a coherent strategy that achieves a multitude of objectives, including isolating the toxic aspects of banking that require government support from those elements that remain viable, freestanding and can be undertaken within the marketplace, and which can be restored to a stock market flotation within a very short period time, so providing the essential indication of support to our financial institutions that will be essential if they are to recover from this debacle.

Explained simply, the government could nationalise our banks on any morning it chooses. The cost would be a small element of the total price of support for the financial markets (especially if, as seems likely, HSBC could be excluded from the deal). When nationalising it would require that the capital of these banks be split into two forms of ordinary share. Those already in issue would be called the B shares. It would not acquire these: they would remain as minority shareholders in the nationalised banks. They would have secondary rights because these shares are, for all practical purposes, unmarketable at this moment. The government would acquire the A shares with its money, because they would give it control.

However, it would be quite unreasonable that no compensation be paid to the holders of the B shares in the long term. I do not think this is an appropriate moment to suggest overthrow of the stock market. It will have a long-term role to play in our economy. The question is, how should that compensation be paid? My answer is that in the short term nothing should be paid bar a low rate of interest on the fixed value of the shares frozen on nationalisation, so long of course as the banks in question keep operating. The capital of those shares should then only be repaid out of the value realised from these banks post nationalisation and it is the aim of realising this capital value that suggests the appropriate strategic direction the government should adopt for these banks once they are in state ownership.

Every bank that should now pass into state ownership has within it strong and viable businesses that do not meet state support. For example, Royal Bank of Scotland owns the Directline and Churchill insurance operations. Many own viable leasing companies. There is absolutely no reason why these need to be state-controlled. Options available for businesses of these sorts included in the nationalisation include sale to a third party (which is, however, highly unlikely at this point of time due to the state of the financial markets) or return to private ownership through flotation in their own right as independent businesses. This is entirely plausible. They have strong cash flows, simple and readily comprehensible income streams, and are relatively low risk, all of which should make them attractive at this moment. Nor do they need to raise capital flotation. Sufficient will allocated to them by the banks out of which they will be spun prior to reflotation.

So what should happen is that these operations should be refloated on the stock markets, to provide those markets with essential indications of support, by treating the shares in the new floated companies as being subscribed for by way of repayment of the value locked into the B shares in the banks that have been nationalised, and which represent the residual capital in those companies owned by their third-party shareholders at the time of nationalisation.

So, and for example, if a company was to be floated out of a nationalised bank and the company to be floated was valued at £1 billion at the time of flotation and the bank had been worth £5 billion at the time of nationalisation then the shares in the newly floated company would be allocated in the ratio of one new share for each share held in the bank at the time of its nationalisation, but if each share at the time of nationalisation had been worth, for example, £5 then £1 would have been deemed to have been repaid by the grant of the share on flotation of the new company. This would be done until the entire value of the B shares had been repaid.

I believe that this would be possible: for example RBS had net shareholder funds of £53 billion according to its 2007 audited accounts, and even if this figure was wildly inaccurate (and that is plausible) I believe there must be sufficient worthwhile assets left to repay its current shareholders who stand to be nationalised given that the market value of their shares is now less than £12 billion.

Of course, that does leave the question of what would be left in the banks if this spin off of viable assets took place, and the answer would undoubtedly be the more toxic assets, including much of the mortgage book, many of the derivatives and the more problematic banking arrangements (it not being implausible that the more straightforward domestic banking arrangements might actually be spun off as new banks aimed at the deposit taking markets quite early in the period of government ownership). I can live with that for three reasons.

First, these are the assets that will require state support whilst the solvency issues they create are resolved.

Second, I suspect that before this rump is created the asset sales from the banks will be sufficient in value to begin to repay some of the government funding as well as having compensated existing shareholders in full. That will be a useful way of creating confidence in the long term viability of this process.

Third it is in these areas of activity that radical reform of banking practices is required if further toxic activity is to be avoided, and a significant period of public ownership may be required before sufficient re-education and reformulation of banking regulation can have taken place to ensure that process can be successfully undertaken. This will require, for example, a radical review of the future of the whole of western banking's approach to speculative trading, the use of hedge funds, derivatives, offshore, securitisation and many related issues such as the reporting of corporate risk, its auditing, means of incentivising those who work in banks and more. I do not pretend that any of these can happen in the short term and as such some time must elapse before it will be either safe or possible to return some reformed establishments willing to engage in risk based banking activity to the private sector once more.

But what is absolutely clear is that unless those who are now appointed to take control of the banks (and there must be those who do so) have a strategy such as this in mind all we will be doing is giving enormous sums of money to bankers to lose, and if these is one option that is politically unacceptable this is it. Which is why the almost total absence of comment about what we will do once the banks have been nationalised is for me one of the most worrying features of this weekend.