Who would have thought public-private partnerships would generate so much interest? The explosive politics of National Air Traffic Services, rail franchises, the PPP on the tube and the search for a Third Way financing for Railtrack have moved centre stage. The passions and the anguish have preoccupied many a national newspaper front page.

Private finance is nothing new. The University of Buckingham was an early example of bringing private money and management into an area of life that had been a public sector monopoly. State supported students went to learn alongside privately financed alumni, all in a privately funded University.

The Dartford bridge was an important watershed, breaking down Treasury resistance to private money for a near monopoly tolled route as part of a wider nationalised road network. It was most successful, with the toll paying for the bridge in a much shorter time than anticipated. The bridge was then passed free back to the public sector.

Both Buckingham and Dartford worked because real risk was transferred to the private sector, and both projects turned out to be supplying something people would pay for. Few wanted to divert from the M25 and drive on worse roads to a different river crossing. Enough students liked the idea of going to Buckingham.

The more recent problems have emerged when the transfer of risk is less clear. This government is more in favour of public-private partnerships. Partnership is a friendly word, but it often means a lack of clarity about the responsibility of the various partners, or a huge complexity in trying to allocate blame.

National Air Traffic Services is a classic case. The two strong options were to privatise it properly, establishing a wide base of shareholders and giving it a sound enough balance sheet to withstand possible adversity, or leave it in the public sector. Many of the government's supporters thought a regulatory monopoly service was best kept public. They believed Andrew Smith, now Chief Secretary to the Treasury, when he said “Our air is not for sale”. Worried by this the government created a company where the state became the most substantial minority shareholder, with 49 per cent, and the employees held the balance, with a consortium of users, the airlines, holding the rest.

There were two big weaknesses in the structure. If air traffic declines, the main private sector shareholders are badly affected in their main businesses at exactly the time that NATs itself might need more shareholder capital. The government, although only a minority shareholder in a limited liability company, could not allow NATs to go bust, bringing dislocation in our air space. The UK has international obligations to meet, as well as having a responsibility to airlines and passengers.

There are similar dangers with the proposed Tube scheme. The government has come up with an incredibly complicated proposal, which entails partially privatising the track whilst keeping charge of the trains. The private sector has declined to tender a price for the full duration of the scheme and has understandably written in a range of protections to the contracts.

The decision to put Railtack into administration will be an expensive one for taxpayers, both directly and indirectly. If the government had negotiated with Railtrack it could have secured reductions in dividend – a primary government objective – and probably influenced the composition of the Board. It already had a decisive say on future developments through the influence of the Strategic Rail Authority. Now the government has to send just about any amount of cash the Administrator says he needs, as it would not be practical to terminate the Administration. Someone has to run the railways, hold the licence, and keep them going. It also means that other private companies, thinking of doing business with the government, will want more contractual protection, and the price of raising PFI type money will go up to allow for the greater uncertainty.

Meanwhile, we should become very sceptical about the government's own accounts. There are more and more hidden liabilities. There was no warning in the government figures that the taxpayer might assume all the financing risks of Railtrack for a time. This year it will mean around £3,500 million of expenditure, much of it money that old Railtrack would not have received from the government. Many private finance initiative projects could end up with the government having to pick up the bill for unfinished business. We need a fuller statement of these hidden liabilities in government accounts, with an estimate of total potential exposure to avoid Enron style accounting practises.

Private finance is a good idea if real risk is transferred and the price is sensible. It is easiest and best where there is a separate revenue stream from service users or customers, as in the Dartford or Buckingham examples, or where there is a proper privatisation, as with British Telecom. If the government insists on partnership, keeps a public monopoly and has to take over if things break down it is just a dear way of the government borrowing money. The decision on Railtrack has raised the price of raising private equity and bonds, and made the private sector more wary of accepting risk. The government would be well advised to heed the warnings before entering into more of these complex and muddled deals. There is no point in the taxpayer underwriting very fancy bond or bank lending arrangements when the government can borrow in its own name at far cheaper rates. There must be real risk transferred, without recourse again to the taxpayer, to make it worthwhile.