Twenty years after it all began, the meaning of privatisation is now clearer. Privatised businesses have fared better than nationalised ones but most have not thrived in the private sector. And the plc model does not work for monopoly services.
This July is the 20th anniversary of the announcement of the decision to privatise British Telecom. It was the start of one of the most radical experiments in public policy, with major consequences in Britain and around the world.
Privatisation is often seen as an ideologically based programme, devised and driven by a powerful leader, motivated by a combination of intellectual conviction of the benefits of free markets and hatred of the power of organised labour. Certainly no term is more indelibly associated with the Thatcher era than privatisation. But there was no master plan when the Conservatives came to office in 1979. Privatisation had been debated in opposition and substantially rejected. The Tory manifesto was correspondingly modest in its commitments. Aircraft and shipbuilding-acquired by Labour only three years earlier-would be denationalised. A Tory government would dispose of the assets of the National Enterprise Board, and undertake “a complete review” of the activities of the British National Oil Corporation.
Privatisation was, in fact, the product of a conflict between adherence to macroeconomic targets for public sector borrowing and the need for large scale investment in digital switching equipment in telecommunications. The government’s ability to commission such investment without breaching its fiscal limits was constrained by the Ryrie rules, a scheme of public sector accounting designed to prevent the previous Labour government classifying subsidies to bankrupt state industries as investment. Less radical alternatives-such as a financial instrument nicknamed Buzby Bonds-did not work, and so the proposal to sell 51% of BT was born.
Driving through the plan to sell BT required exceptional political will. In the early stages, privatisation was generally resisted by the managers of the to-be-privatised entities (as analogous plans are today resisted by managers in health and education). Those involved had not understood that – through higher salaries, share options and greater autonomy – they themselves were to be the principal beneficiaries. Later, managers came to be enthusiastic advocates of privatisation although they continued to oppose any restructuring of their businesses.
The capital market implications were also daunting. The sale of even a half share in BT was six times larger than any previous issue on the London stock exchange. Initially, there was little enthusiasm for the sale from investment institutions. They failed to recognise the potential for growth in telecommunications in the 1980s as much as they exaggerated it in the 1990s. The government turned to the retail market and the idea of selling BT shares to private investors succeeded beyond expectations. There were over 2m applications. These were scaled back so that the maximum allocation in the open offer was 800 shares costing around £1,000.
The sale of BT set the style and tone for subsequent privatisations. And between 1984 and 1996, when the programme of major sales ended, the largest public flotations – of television, gas, airports, airways, steel, water, electricity generation and distribution, rail and nuclear energy-raised total revenue (at 2000 prices) of around £50 bn. There were also numerous smaller disposals and secondary offerings.
The pragmatism of the origins of the programme was reflected in the variety of justifications given. Sometimes privatisation was applauded for enhancing business efficiency, sometimes the contribution to public revenues was emphasised; sometimes the widening of share ownership the declared objective. In the end, the programme acquired a momentum of its own. The fact that it had happened was deemed evidence of its success. For the huge lobby group of advisers and consultants who profited from the transactions, this was true. I shouldn’t knock it: I benefited too.
Privatisation – the world itself was newly coined – became a generic term for a range of market oriented reforms in the public sector. The sale or transfer to private companies of state assets was just one component. The word was also used to describe liberalisation or the introduction of competition. And privatisation also involved disengagement – the creation of a more arms-length relationship between government and industry, most often by interposing an autonomous regulatory agency.
It was apparent early on stage that not only were the sale of public assets and the liberalisation of markets not the same thing, but that they might often conflict. The privatisation of BT probably slowed the introduction of competition into British telecommunications. The deal between Dennis Rooke and Peter Walker which paved the way for the privatisation of British Gas ensured that no restructuring of the company took place and that liberalisation – and the break up of the business – came about only after a long regulatory battle.
What was less apparent in the early stages of the programme was the extent to which disengagement from day-to-day operating decisions was separable from the transfer of assets. The Conservatives achieved big changes in British Steel, British Airways and British Leyland through disengagement rather than through privatisation. The combination of managerial freedom and hard budget constraints – the knowledge that government was genuinely willing to allow these companies to fail – transformed the atmosphere in these businesses. Indeed in all these cases it was improvements in the efficiency of the companies which made privatisation possible, rather than the other way round.
But disengagement does not come easily to politicians. Alan Milburn recently resurrected Aneurin Bevan’s remark about the NHS: that the sound of a bedpan falling in Tredegar Hospital would resound in the Palace of Westminster. Such high level intervention is politically irresponsible. It means responding to well publicised incidents not real problems: and in so doing undermines the authority of operational managers. If ministers are responsible for the dropped bedpan, no one is responsible for anything.
It became clearer after 1997 that privatisation did not necessarily mean disengagement. The Tories had an ideological commitment to take the state out of business. New Labour did not. The new government issued an early consultation paper on utilities and followed with a bill. This included some adjustments to the electricity market and a merger of gas and electricity regulation. But it also contained a variety of proposals designed to stimulate favourable soundbites: measures to strengthen consumer councils, to allow ministers to rearrange price structures if they thought they disadvantaged poor customers, declarations of support for environmental objectives.
Backbenchers got the message and tabled hundreds of similar amendments. No constituent’s bedpan would be dropped unheard. But the industries affected lobbied hard and Stephen Byers was forced to abandon most of the bill, excluding water and telecommunications altogether from its scope.
Labour in opposition had been wrong-footed by privatisation. It had opposed each measure, though with dwindling conviction. no one realistically expected renationalisation, but acknowledging the permanence of privatisation had not been a political option either. That first utilities bill was a consequence of the failure even to discuss, far less to resolve, the implications of the revolution Thatcher had achieved. But now, as some of the unfinished business of privatisation reaches crisis-in railways, air traffic control, posts and the London Underground-the agenda is too pressing to ignore.
The most enduring consequence of privatisation has been the emasculation of public sector trade unionism. Since the closure of Saltley by striking miners in 1972 numbers employed in mining have fallen by 95%. Since 1980 employment in steel has fallen by more than 75%, in railways by two-thirds, in electricity by half, and in water by almost as much. Output in both coal and steel has fallen, but by far less than manpower, with extensive closures of high cost production. In rail, electricity and water, employment has been reduced with little effect on output. Many people – including managers – had known that these businesses employed too many people. But no one imagined that savings of this magnitude were possible.
There can be no justification for running state industries with twice the number of people needed to make them function efficiently. Poor control of labour costs fed weaknesses in most other aspects of management. The electricity industry-widely regarded as the best run nationalised industry-employed far too many skilled workers, and sucked too much scientific talent into the job of making Britain’s misconceived AGR nuclear programme work. Savings similar to those made in Britain have been made in publicly owned businesses in other countries, often without the sale of assets. The hard budget constraint is the transforming factor: without it, internal consensus is always a more pressing concern.
It is true that low paid, low productivity jobs are usually better than no jobs, both for society and the individuals concerned. Many of the people who left steel or mining never worked again. It is not a ridiculous policy for the public sector to offer jobs to people who would otherwise be unemployed. But it proved hard to achieve this legitimate objective without undermining effective management and creating an organisation where the interests of employees are put ahead of the interests of users of services.
How many people are needed to provide public services? It used to puzzle me that managers in either public or private sector, told to reduce their workforce by 10%, could always do it. Were so many sectors really so overmanned? In utilities like water or railways, a high proportion of the workforce is employed to stop things from going wrong, or to fix them when they have gone wrong. In these environments, deciding how many people are needed is a matter of fine judgement. Once there were too many. But, as each regulatory review leads privatised utilities to reduce their workforce further, there comes a point at which there are too few. An incident like the Hatfield rail crash was inevitable. Sooner or later, similar incidents will occur in other utilities.
Perhaps rather surprisingly, given the claimed success of privatisation, privatised businesses have not thrived in the private sector. Owning privatisation stocks was once a way in which anyone with a few hundred pounds to spare could be confident of adding a hundred or two to it: buying these shares at flotation was almost a guarantee of profits. But holding these shares for the long term was not. The performance of privatised businesses has fallen short of market averages by a very substantial margin.
Even those companies which at first sight seemed to be clear successes of privatisation- British Steel, British Telecom, British Airways and Cable & Wireless – were ultimately to be disappointments. British Steel, now Corus, became one of the best managed and lowest cost steel producers in the world, but that has proved barely enough for viability in a world awash with low price steel. British Telecom and Cable & Wireless have struggled, and failed, to find a coherent business strategy. British Airways raced ahead of other European airlines by concentrating on cost control, marketing, and, above all, customer service. But other airlines could get things right as well, and eventually did. Steel, telecom, airways did better after privatisation than as a nationalised industry, only to discover that better than before was still not good enough.
Most privatised firms attempted to use the reliable but slow growing cash flow from their core businesses to diversify. The record of such diversification has been very poor. The most ludicrous excesses of companies attempting to leap free of the shackles of regulation have happened elsewhere. Enron was once a plain Texas utility. Jean-Marie Messier if Vivendi sought to launch himself as an international media mogul, complete with Manhattan apartment, from the humble base of France’s leading water company.
Still, leaders of Britain’s privatised businesses have had similar instincts, and the result has been a long series of acquisitions, followed by write downs and disposals. As I write, Anglian Water has announced losses in its construction business and Northern Ireland Electricity issued a profits warning prompted by its losses in software. Managers prefer spending time closeted in boardrooms with investment bankers, or jumping on planes to distant cities to supervising sewer replacement. But these activities are not profitable for shareholders or useful to customers.
The best outcome from utility privatisation would have been the emergence of private companies with specialist capabilities in the management of these businesses. Managing utilities well is hard, and not often achieved. Particular management skills are needed to handle a highly decentralised workforce and to engage in endless negotiation with government. At times, there have been indications that such focussed companies might emerge. Vivendi still has unmatched technical competence in water and sewerage treatment. But the lure of the takeover trail is usually too much: Scottish Power, possibly the most successful British company in the development of utility management skills, is now struggling with the consequences of its costly US acquisitions.
All privatised companies start from behind. If you were setting out to establish a successful private sector business, you would not begin with a water or electricity board or a government department. You might well do better to start with nothing at all: Vodafone, which began operations at the same time as BT was privatised, is now bigger, more efficient, the quality of its management more highly regarded.
Many of those who reached the top of privatised companies must have felt like lottery winners. Having joined the foot of an administrative ladder in one of the less glamorous divisions of a local authority or government department, they found themselves feted as executives of large international companies, with salaries and share options to match. It is not surprising that not all of them proved up to the job.
Customers have on balance gained from privatisation, but not hugely. An index of “prices of privatised goods” since 1980 has generally tracked the movement of prices generally. Labour costs in these industries have fallen, often substantially. But capital and other costs have risen. And each industry has its special factors. Technology has reduced the price of telecoms equipment; EU environmental directives have increased water supply and sewerage costs. Electricity and gas prices have fallen somewhat, rail and bus fares have risen. There has been some unwinding of cross-subsidies to high cost customers, but fears that privatisation would lead to a loss of universal service or to swingeing increases in charges for poor households have proved unfounded.
There is a general pattern of price increases up to privatisation, followed by stabilisation, and most recently by reductions. This might seem to demonstrate how bad nationalisation was, and how successful privatisation has been. But the trends in prices in these different phases of evolution are mainly the result of government and regulatory policy. The period of most effective cost control seems to be the one immediately preceding privatisation.
The trend was for privatised industries to be fattened for sale, and floated with a benign regulatory regime which made it easy to outperform the targets set and report larger than expected profits. Thereafter the regulatory screw tightened. Returns were pushed down to levels more in line with the cost of capital and targets for future cost control became harder to achieve. This regulatory action has produced big price falls in telecoms, gas, electricity, and even patchily in water bills.
Firm evidence on customer service is harder to come by than it should be because, as too often in Britain, statistics are published to support a story rather than to inform debate-definitions change, series are discontinued. Improvements have been clearest where businesses have faced competition, as in telecoms and airlines. But privatised utilities also have given much more attention to customer service. Privatised companies have come to see that the viability of a regulated business depends on its acceptability to public opinion in a way that had no parallel under nationalisation. Moreover, most workers in a service business prefer the pleasure of empathising with customers to the authoritarian satisfaction of explaining to them that they cannot have what they want. Once the overall direction is changed, the atmosphere can change rapidly. This has proved to be as true in the Inland Revenue as in gas supply.
The revenue implications of privatisation seemed to matter a lot at the time. The programme originated, after all, in concern at levels of public sector borrowing. Yet these financial issues now seem minor. From 1979 to 1997, privatisation receipts averaged less than 2% of government revenue. Yet the concern with the national accounting implications of policies, from which the policy of privatisation was born, continues to exert an influence on policy today. Incredibly, the Office for National Statistics is influential in the restructuring of Railtrack: what it will and will not categorise as government ownership and government borrowing is an important consideration. Ministers are now required to describe higher public spending as “investment.” At the same time, taking genuine investment in public sector assets off balance sheet continues to be a public policy objective.
Enron has reminded people that accounting guidelines should be seen as statements of principle, rather than rules to be worked round. This is particularly true when government sets the accounting standards itself. The borrowings which finance new sewerage treatment works for a private utility are obtained from more or less the same people and repaid by exactly the same people as those who would have provided and repaid the borrowings under public ownership, and the macroeconomic consequences are thus more or less identical.
Nor did privatisation have much impact on the structure of share ownership. Many Prospect readers who go through the effects of their parents or grandparents will find share certificates for 800 shares of British Telecom, 500 shares of British Gas. Memories of a curious ritual of the Thatcher years: the advertising campaigns, starring Sid and the Loch Ness Monster, the oversubscriptions, the frenzy of the first day’s trading. But most share certificates drifted into the hands of the same insurance companies and pension funds which own shares in other British companies. There is no big difference between the structure of share ownership in privatised businesses and that of large companies in general.
A big problem in assessing the effects of privatisation in Britain is determining the standard against which the policy is to be assessed. Nationalisation was not a success. The model of public ownership pursued by the 1945 Labour government was the autonomous public corporation. The model was London Transport-which had served the capital well before the second world war-and, in Herbert Morrison’s phrase, the leaders of the new businesses were to be “high custodians of the public interest.”
But the new nationalised industries were, in the main, ill managed. An ambitious and poorly controlled investment programme for the railways coincided with a rapid decline in traffic: this was the period in which car ownership was extended to a mass market and road transport became the major means of freight distribution. But the railways were only the worst example of what the Treasury came to see as a failure of control on a scale which threatened the whole structure of public finance.
The government response was steadily to erode the autonomy of public corporations, stripping them of borrowing powers and intervening more and more frequently in operational decisions and investment plans. A Tory government reacted to the railways’ problems by putting an ICI executive in charge, Richard Beeching. He prepared an ambitious restructuring plan which involved closing most of the branch network. The whole saga contributed to a negative public attitude towards nationalised industries in general.
Relations between civil servants and politicians on the one hand and managers of nationalised industries on the other became increasingly abrasive. Government deprived managers of autonomy without itself taking responsibility for the performance of the business. The morale and competence of the management of the nationalised industries declined, this decline justified further detailed supervision and so on. By 1979 any change in organisation was likely to be an improvement, and privatisation generally was.
The subsequent effects of privatisation are as much a consequence of the fact of change, rather than the effects of any specific change. Managers are more confident, organisations whose primary concerns were technical excellence have learned about marketing, customer service, and financial control. Without privatisation in the broad sense of the term, none of these shifts in culture would have happened. All of them could have been achieved without the sale of the businesses, and in some cases were, but it is unlikely they would have been implemented had asset sales not been on the agenda. Still, of the three components of market oriented reform, disengagement and liberalisation were more important contributors to cultural change than the sale of assets itself.
There is, in fact, a striking similarity between the evolution of nationalisation from the 1950s to the 1970s and the evolution of privatisation from the 1980s to the present day. The process begins with an initial confident establishment of autonomous businesses. But after that euphoric phase performance fails to meet expectations and it is recognised that some political control is unavoidable. That leads to increased central control over pricing and investment. Within the industries themselves, there is resentment that promised autonomy has not been delivered. Relations between government and the businesses become increasingly strained. That was the history of nationalisation and privatisation.
Not, it should be acknowledged, across the board. Businesses that were once publicly owned divide into two main groups. Some provide commercial products in markets which are, or could be, competitive-airlines, electricity generation, steel. Some are necessarily monopoly providers of public services-water supply, electricity and gas distribution, rail networks, the London underground, air traffic control. A few straddle both categories, or fall between. These latter activities are neither competitive businesses nor natural monopolies: they offer goods and services whose main benefits go to those who use them but their activities nevertheless have implications for a wider public-airports, bus services, operating trains.
Where competition is possible, it is usually desirable. When government allows the businesses it owns to operate as competitive businesses in competitive markets, their performance is often good (as in British Steel and British Airways from 1980 to privatisation). But if they are to operate as competitive businesses in competitive markets, they might as well be privately owned. State ownership merely allowed the government to interfere in operational matters, usually unhelpfully (as in promoting an absurd expansion of British Steel). Retreat by government from these competitive activities has usually been desirable and is likely to be permanent. Most of the businesses concerned have ended up merged into well-established private companies. The recent acquisition of Enterprise Oil by Shell leaves few independent, competitive businesses among those privatised companies.
The polarisation of competitive and monopoly elements has been accelerated by a recent trend for companies with both competitive and natural monopoly businesses to separate them. British Gas has hived off its main regulated businesses into Lattice. Most water companies have now ring fenced the natural monopoly business. Similar evolution is likely to follow in electricity distribution and in telecoms.
Within a short time, therefore, the industries that were once publicly owned will have been split more or less completely into two groups. One will have been wholly absorbed into the private sector. The other are those services which remain natural monopolies. The latter activities are central to the natural economic infrastructure. They will always require regulation. It is for this group that a rethinking of policy which recognises both the successes and the failures of privatisation is required.
The plc model has not worked especially well for natural monopoly businesses and in several cases-railways, air traffic control, water -is currently unwinding. There are several basic problems. The first is that the plc structure lacks legitimacy. In a regulated market economy, authority in economic matters comes either from success in a competitive market or from accountability to a political process. Private companies which are monopoly providers of public services have acquired legitimacy from neither of these sources. So their activities are resented, their failures seized on and exaggerated.
This matters more for businesses that deal directly with the public-like water supply or Railtrack-than for those that do not-such as air traffic control. But it undermines both. Francis Fukuyama’s description of political legitimacy identifies the issue: “Legitimate regimes have a fund of goodwill that excuses them from short-term mistakes, even serious ones. In illegitimate regimes, on the other hand, failure frequently precipitates an overturning of the regime itself.”
This describes precisely the process which destroyed Railtrack. Railtrack was a badly managed business, but so was British Rail before it. The tracks may have been less safe under Railtrack than under British Rail, but there is no evidence of this from accident statistics. Yet however inept British Rail was, it was our British Rail. Railtrack was not seen in a similar way and there was no tolerance for its incompetence. Hostility grew. When things went wrong, the organisation panicked. In doing so, it destroyed itself.
Railtrack’s crisis had a precursor in Yorkshire Water’s supply failures in summer 1995. There was the same public intolerance of mistakes. Drought under public ownership was the fault of nature; drought under private ownership is the fault of the water company. The company’s management was toppled, and the confidence of the industry shaken.
Railtrack’s failure exposed a second big problem for the plc structure: what happens when private utility provision fails. The legislation governing rail privatisation set out a special administration procedure, designed to secure a smooth handover to a new operator. Similar provisions exist in some other industries, such as water and air traffic control, although in others (electricity, gas, telecoms) the regime relies on general insolvency law and the quick-footedness of the regulator.
For Railtrack, the special administration rules have not worked as hoped. The administrator’s obligations to creditors, and the holding company’s duties to shareholders, have worked against the overriding public interest in the speedy injection of new management and refinancing. The administration procedure, which is designed for failed private companies in competitive markets, is insufficiently adapted for failed private companies in monopoly public services. The priority in these latter cases is not the realisation of assets for the benefit of creditors and shareholders, but better management of the service. The need for resolution to these issues has forced the government to make a payout to Railtrack shareholders. Political intervention in these situations is inevitable, but at present it lacks a coherent basis: the legal structure of the plc conflicts with the necessity of the public service function.
A third problem with the plc structure is that the cost of capital is too high for a regulated business. The corporate entities which result are not sufficiently transparent to enable the stock market to see through to the nature of the underlying business. This became apparent after the last, toughest round of regulatory reviews. Companies have discovered that ring fenced utility businesses with high levels of debt can be financed much more cheaply than the capital structures which were imposed on businesses at privatisation. This amounts to a rejection of privatisation by the market itself.
This issue interacts with a fourth problem. Regulatory reviews are a kind of game in which the companies present inflated investment plans and operating cost projections, expecting the regulator to beat them down. In the heyday of privatisation, it seemed that this process could be eliminated, or at least made objective, by RPI-x mechanisms of price control. These dreams evaporated. Regulatory negotiation in practice has involved discussions between companies and regulators about how long sewage outfalls need to be in specific locations and what standards of security are required in a national transmission network. These discussions are necessarily fraught whether the structure is public ownership or a regulated business. The information needed to make these decisions best is held by operational managers. But their incentives do not coincide with a broader public interest. Utility managers want to build and invest. They are often in love with the technology of their industry. They want to provide ample margins of security, to protect themselves against mistakes. In a privatised company, they also want to please investors, boost their share options, and pay for diversifications.
These negotiations worked badly under nationalisation. Frustrated managers dealt with civil servants whom they felt, with justice, were more concerned with process than efficient outcomes. The substitution of honest regulators who are genuinely concerned to struggle towards the right answers is a big improvement. In the early stages of utility regulation this worked well. The first appointees to regulatory posts were impressive people. Divergences in their style provided a variety of experiences which helped the process of regulation evolve. Being the first regulator of an industry, with the opportunity to define the structure of regulation and set the tone of regulation, is an interesting job: being the second or third in line, less so. The political aspects of the job have become more wearing. Whatever the reasons, it has proved difficult to maintain the calibre of regulatory appointees. It is now time to establish a single regulatory agency, with a strong chief executive, and a brief to develop the professionalism of its staff. The individuals concerned should expect to move between agencies and industries: sometimes gamekeeper, sometimes poacher. This obviously requires freedom from civil service structures and pay scales.
But better regulation is not enough. The problems go deeper. The overriding requirement for monopoly utilities is to devise a vehicle which is not a plc but is sufficiently divorced from political control to preserve the advantages which have been derived from privatisation: most importantly, the greater efficiency and customer sensitivity which have come from managerial autonomy. The freedom to drop bedpans, and the obligation to pick them up afterwards.
The restructuring of Railtrack is a major opportunity to work through the problems which have emerged from privatisation and to establish a model which will be applicable not only to other contested businesses-such as the postal service – but also to other monopoly utilities and, perhaps, to other areas of public services where privatisation is unthinkable or unworkable. Political involvement in these industries is inevitable. To pretend otherwise simply drives it underground, making it informal and irresponsible. There needs to be ultimate political responsibility, but the wise minister will exercise it extremely sparingly, and may wish to tie him or herself to the mast to avoid the siren voices of the Today programme.
Whether this political involvement in a utility happens in the private or public sector is not that significant. If we say that the German utility RWE owns London’s sewers, we are obviously using the word own in a very different sense to the one we use when we say that British Airways owns its planes. RWE cannot levy charges for its assets, dispose of them, or use them in any different way without the consent of government, and no one wants it to be otherwise.
Governance arrangements are the key, not ownership. The structures which are in place for Glas Cymru (Welsh water) and are proposed for Railtrack-an executive board responsible to a body which is representative of stakeholders-make good sense so long as the involvement of stakeholders focuses on appointment and reappointment and does not extend to operational issues. This structure cannot be shoehorned within the framework of a plc. The supervisory board needs a legitimacy it cannot acquire from either self appointment or political appointment: and the primary legal duties of such a board should be to public service rather than investor interests or political masters.
This public interest orientation is important to the motivation of those who deliver the services. Most of all it is essential to establish the legitimacy of the organisation which delivers them. Even now, the meter reader often says he comes from the gas or electricity board, making a statement about both his attitude and his expectation of yours. A public interest orientation also implies that diversification should be confined to related businesses. It cannot be stressed too often that while managing monopolies is easy, managing monopoly utilities well is hard.
The maxim guiding state involvement in industry should be “competition where possible, regulation where necessary” and although competition with natural monopolies is impossible, competition in the provision of functions to them is desirable. Supply competition now exists in gas and electricity and should develop in telecoms. This should not be confused with substantive competition, because the proportion of the value of the product which is subject to effective competition through this route is tiny. It is the analogue of competing checkouts at the exit to the store. New public service corporations should, like Glas Cymru, aim not to be big employers, subcontracting most of their functions to other businesses and thus prompting the emergence of specialist, competitive utility management companies.
These public service corporations should, however, assume directly the responsibility for financing their functions, and for capital investment programmes. Since the reality is that all these businesses have an assured revenue flow from taxpayers and captive customers, customers and government should reap the cost savings which follow from spelling out that fact. There does, however, need to be a layer of debt on which companies can reduce or defer payment without provoking insolvency. The alternative of bouncing government or regulator into providing whatever revenue the business finds it needs, leads inevitably to further political involvement in the business on ill-defined terms.
Railtrack, the postal service, air traffic control and the London Underground are not one off problems, requiring ad hoc solutions. They are all aspects of the general problem of finding structures for the efficient provision of public services. Nationalisation of utilities, although not a successful policy, was not adopted through sheer perversity. The issues which led to it have not gone away. The opportunities to find better long term solutions are still there to be taken.