The term public service is old-fashioned but still useful. A public service is an activity whose purpose is to meet the needs of its customers when these needs cannot be appropriately met by the process of profit-making firms operating in a competitive market. Perhaps because competition is impossible, as in electricity transmission. Perhaps because the service is intrinsically uneconomic, as with railways. Perhaps because there is a strong social interest in the quality of the output, as in education or water supply. Or perhaps because commercial motives are incompatible with the ethos that customers expect from the service, as in health.
The most important problem Labour faces in its second term concerns the management of these public services. In education and health, voters want better results and the government is ready to pay for them. But in government there is the fear-even the sickening certainty-that the money will disappear down a black hole. Five years from now spending will be higher, but dissatisfaction with what is provided will be just as great.
The failure of railway privatisation, and the wrangles over the London underground, are also reminders of unresolved problems in the regulation of privatised industries. Privatisation was meant to end direct government involvement in these businesses, but even before Gus Macdonald and John Prescott appeared on television discussing every operational problem on the railways it was evident that this was wrong. Privatisation was a more modest reform than either its supporters or its detractors recognised: simply another phase in the relationship between government and utilities.
The transformation of Welsh Water into the non-profit Glas Cymru is one of many steps on the road back from privatisation to public service. British Gas has already hived off its regulated business to a separate company. It is very likely that other water companies, electricity distributors, and British Telecom will do the same. It is simply cheaper to raise finance for a ring-fenced public service than for a diversified private company. In this fundamental way, the market itself has rejected privatisation.
If privatisation was the Conservative answer to the problems of managing public services, targets and objectives are the Labour response. Hardly a day passes without some new targets, new objectives or new control mechanism being rolled out. These mechanisms will not work. They won't work because public services have complex and multiple objectives, because both the balance among these objectives and the best means of achieving them change constantly, and because the centre has no means of knowing what local targets are really achievable. The consequence is that targets proliferate as those initially specified fail to achieve their purpose, while the people who run the services become cynical and demoralised and devote ever more of their time to lobbying the centre and ever diminishing time to running the controlled activities.
There is ample experience of the failures of this style of planning. If a regime of targets and objectives worked, Russia would have worked. If these mechanisms were a good means of service delivery, India after independence would have moved rapidly towards economic development rather than degenerating into the permit raj.
Still, despite the weaknesses of Conservative privatisation and Labour targeting, both are better than what went before. In nationalised industries and in directly controlled government services a deadening absence of real accountability for performance was combined with arbitrary interference in operational decisions. The latter fatally undermined anything that was left of the former. To the extent that bodies providing public services had any objective at all, they were based on the interests and perceptions of those who worked in them. The very notion of "customers" of public services is a recent one. Traditionally they had subscribers, passengers, users, students or patients. All of these terms convey the sense that the beneficiaries of the service are essentially passive recipients.
It is easy to mock the way in which some privatised businesses changed this terminology in the 1980s and 1990s. Privatised train announcers now remind customers to take all their belongings with them at the next station stop. Yet the shift was not purely cosmetic. I recently watched a French postal clerk pull down a blind labelled guichet ferm? in front of a long queue of customers, her scowl reinforcing the message that her primary obligations were to the French state and her own lunch, not to people waiting for stamps. This happens much less in Britain today as other, more fundamental changes have begun. We thought that electricity was one of the most efficient nationalised industries. We now know that the industry employed twice as many people as it needed and built plants that were monuments to engineers rather than efficient generators of power. Breaking these cultures is a necessary first step in public sector reform, but only a first step.
The next phase requires a clearer conception of what is, and what is not good about markets. Labour's love affair with business has helped perpetuate the idea that the private sector is simply better at doing things than the public sector. But this is not because of the superior abilities of private sector business people. It is not even due to the soothing effect of high salaries and generous share options. Market environments create a structure which gives freedom for diversity of individual experiment, but assesses the results of these experiments, holds people accountable for the results, and quickly terminates experiments when they are seen to have failed.
It is these "pluralist" processes which are the essence of the market's superiority over central planning-a superiority which is as evident in relation to co-ordinated social democracy as over bureaucratic communism. Comprehensive education was a perfectly respectable piece of social engineering which happened to fail. No one should be criticised for embarking on it. What should be criticised is that the experiment was compulsorily undertaken on a nationwide basis, that there was no process for measuring its consequences (as distinct from the extent of its implementation), that there was great reluctance to acknowledge it had failed, and that no one had to bear responsibility for what went wrong.
The same was true of the economic disaster of Britain's nuclear power programme and the related decision to equip a generation of power stations with generating sets far larger than had been previously envisaged. The judgements made were defensible at the time but with hindsight mistaken. What was wrong here too was the scale on which the policies were implemented and the inability to recognise their failure or hold anyone to account for that failure.
Disciplined pluralism-processes of diverse experiment, audit and accountability-is automatic in competitive markets and provides a spur to innovation and efficiency. But how can it be reproduced when competitive markets are not possible? Accepting such pluralism is particularly difficult for politicians. It is in the nature of politics that politicians have views on every subject under the sun. And it is not only that politicians do not know what is the best way to build power stations, teach mathematics, or cure diseases. It is that no one else really knows either, and the way in which knowledge evolves is through a messy process of mostly unsuccessful experiment. This is equally difficult for voters to accept. If we are truly insistent on ending "postcode lotteries" and want the same public services to be delivered everywhere in the country, we need to understand that the likely consequence of the control mechanisms needed to achieve that result is that the standard of services delivered everywhere is very low.
The key requirement is to define and refine the British model of what we might call the public service corporation. Public service corporations cover activities as diverse as water supply, rail services, airport management, museums, hospitals and universities. All public service corporations need mechanisms for governance, for audit and for recovery.
The governance of the public service corporation must itself entrench pluralism. The supervisory board should be representative of a range of interests. But it should act as a board, rather than as representatives of these interests. And it should be incapable of being captured by any one of these interests-whether by local politicians or business people, the people who work in that business, or those who manage it. And that board, like politicians themselves, must stay out of operational decisions. Its job is to appoint managers, and to back them or sack them. Reviewing strategy is fine, so long as it does not relieve executive management of responsibility-which is the reality, if not the legal fiction, of how big private businesses are run. In the provision of public services, democracy should be about accountability for decisions, not the decisions themselves.
That accountability must be informed by external audit. This includes financial audit, but a review of the performance of public service functions is even more important. The purpose is not merely to report on activities but to provide information for comparisons and-as with the financial audit of private businesses-its purpose is primarily to encourage the corporation itself to keep its affairs in order. The creation of a climate of openness and transparency is a principal objective. Such openness is self-evidently appropriate for a public service corporation and a characteristic which few of them today display.
A recovery procedure is necessary: failing public service corporations need to be put into administration-a procedure designed to provide interim management while the board and executives responsible for the failure are removed and a merger or reorganisation is implemented. Financial failure would be one event giving rise to administration, but the principal trigger would be failure in terms of public service delivery. There is no point in allowing public service corporations to go bust. The public-private partnerships we have today require users and taxpayers to pay a premium to the markets for a risk of financial failure which theoretically exists but probably does not arise in practice.
Elements of all these procedures already exist. Almost every existing public service corporation has its board. The Companies Act board of a privatised industry, the trustees of a museum or gallery, the senate of a university, the governors of a school. Some of these work well, most do not. We also have a proliferating range of audit and accountability mechanisms-utility regulators, the National Audit Office, Ofsted, the Quality Assurance Authority.
There has been the least progress in devising recovery mechanisms, but these are developing. We have administration procedures for restructuring failing schools. A special administration regime exists for the water industry. Railtrack is effectively under a form of administration, and it is desirable to have a proper structure for the interventions which are needed to sort out its legacy of mismanagement.
It is tempting to argue that this plethora of ad hoc initiatives needs to be put into a single framework. There would be merit in establishing a simple, coherent legal framework for the public service corporation and the reconstructions now in progress in water, railways and the underground provide an opportunity. Since the early days of privatisation, there has been an evident need for a framework which gave operational autonomy and managerial responsibility in public services but which was not the framework of the plc. There is merit in pluralism here too. The major need is for recognition of how different initiatives relate to each other and for an ability to read across experiences from one area of public service to another. It is on success or failure in public services reform that the 2005/6 election will turn.
