Of course, the railways were not and are not a part of the welfare state. Nevertheless, the more general point might be that private or market provision does not work.

In which case, presumably those who say 'what about the railways?' can think mainly, or only, of the railways as an example of private supply 'not working'. They appear to accept that the vast array of other products and services provided by the private sector are fine (to just start a very long list: house-building, supermarkets, manufacture of computers, manufacture of digital cameras, clothes shops, development and printing of photographs...). Presumably they also accept that the many other industries that were privatised have gone well, such as provision of gas, electricity and water, air transport, telephone services and road haulage.

If so, their argument must consist of saying, "Look, out of hundreds of industries, here is one which did not go well. That proves that private supply is no good." The argument only needs to be stated to demonstrate its absurdity.

They would also have to go on to argue that state provision, in contrast, has never gone wrong. They would have to maintain that - in admirable contrast to private provision - there is not a single state-supplied service which has failed. They would therefore have to say that state provision of dental services, council housing, care for those with cancer, postal services and, in the days, the state provided them, telephone, gas, electricity and water services, are and were absolutely fine. In fact, of course, all those services are or were bad.

So the mention of railways by no means invalidates the argument that private, charitable and market provision of services and products has a vastly better record than that of the state. Private provision is not perfect. Marks and Spencer is not perfect. Privately manufactured computers are not perfect. It is folly to to expect perfection in human affairs. But private provision is demonstrably much better than state provision.

All the above is without even beginning to contest the assumption that privatisation of the railways was an unarguable example of a privatisation failing. It is indeed possible to contest that assumption. I won't put the argument here at full length, but here are a few points:

First and most important, the railways were never fully privatised. They were only semi-privatised. This mattered in certain important ways. One was that the contracts with the train operating companies were for a maximum of seven years at the outset. This was too short a period to make it clearly worth the while of the companies to make major capital investment in new trains and other capital equipment. In Sweden, of all places, where I understand a more wholesale privatisation of rail took place, the record of the privatisation has been far better. In Switzerland, where there are some private railways, they work very well and reliably. In Norway, recently, I travelled on a superb private railway line.

Second, even the semi-privatisation of British Rail resulted in massive investment and many improvements such as far better provision of booking services. I remember in the days of British Rail, one could spend half an hour trying to get through to an office which would tell you the times of trains. After privatisation the information service become very fast and efficient. It is true there were some serious and fatal accidents in the semi-privatised years, but the safety record - as measured in the way most used in the industry - was better than previously. Fatal accidents in rail travel - measured over one or even five years - do not happen in a consistent way that accurately reflects the underlying safety level.

Look at the semi-privatised years from another angle, it was a curious kind of 'failure' that resulted in an increase of passenger numbers of some 27 per cent, if memory serves. That was unprecedented in the post-war years. During the decades in which British Rail operated the railways, there was a steady trend of falling passenger numbers.

With best wishes, James Bartholomew