Under the first system of educational provision the schools would not be part of the public enterprise sector. State ownership combined with the absence of revenue from sales would place the schools in the framework analysed in Module 7. Education would be supplied by what is called in the US, a bureau.
The suggested reform is perhaps a little fanciful but it highlights the problem of classification. On the face of it there is little to distinguish the suggested framework from that of a set of educational ‘public corporations’ selling their output in the market. There are no private and exchangeable claims to the corporations' resources, and revenue is obtained by attracting students.
However, there are two aspects to the situation which appear rather special. In the first place, education is widely accepted as involving social and distributional considerations which place it in a somewhat different category from other goods. It is not clear precisely why this should, in itself, make it necessary to remove ‘merit goods’, as they are called, from inclusion in the public enterprise sector, but as we saw in the text, definitions in some countries do limit the sector in this way.
The second and related point is that the education ‘market’ described in the question is dominated by funds which have been given to consumers by the state. This is not unusual in the sense that resources are often provided to individuals and institutions to bid on private markets. But the complete control of revenue flows to the sector as a whole, it might be argued, makes the situation distinctly different commercially from a market in which the state's presence as a customer is less pronounced. The reform, it might be argued, is an attempt to introduce some additional competitive forces into a sector which is still basically bureaucratic rather than commercial in nature. It represents a sophisticated way of channelling resources to competing bureaucracies.
The possibility that schools could be paid on the basis of a capitation allowance, thus removing the necessity of transferring funds to students or their parents, simply confirms the argument of the preceding paragraph that the market is dominated by state rather than private spending and that this should prevent the classification of educational establishments as public enterprises.
Let us assume that all schools are non-profit enterprises free to set their own fees and determine their own educational policies subject only to the policing of minimum standards by the state. As there are no private property rights in the assets of these institutions it might be argued that this part of the definition of ‘publicness’ is still satisfied. On the other hand the fact that the institutions are independent of the state suggests that, although they represent a form of enterprise that is non-private, they are not state enterprises as normally understood. Most definitions of public enterprise are ultimately concerned with the ability of the central government or local government to exercise control. By this criterion therefore, non-profit schools independent of the state would not be part of the public enterprise sector.
It might still be objected that a vouchers scheme would give the central authorities effective control even over legally independent institutions and that the system would still effectively represent one of competing bureaux. This, however, concerns the boundary between the public and the non-public sectors more generally, rather than that between public enterprise and bureaucratic supply. If consumers are permitted to spend on education whatever they wish, and the state confines itself to providing a certain level of entitlement, then education might plausibly be held to be non-public. The point at which the dominance of state finance determines the character of the entire sector is not one that can be identified with precision. As John Jewkes (1948) Ordeal by Planning, Macmillan, London, noted in a similar context, the ability of economists to notice when differences of degree become effectively differences in kind is important. One would not claim, he pointed out, that the continued existence of half a dozen private droshky drivers in Moscow in 1946 made the Soviet Union a mixed economy. Similarly, if educational spending from sources other than the government is limited to ballet lessons and occasional school trips, one might question the classification of legally independent non-profit schools as outside the public sector.
Our rankings of the institutions by constraints on managers are reproduced in Table A1.5. We should emphasize again that they are not definitive and depend to some extent upon how the question is interpreted. Our reasoning is as follows:
| Bankruptcy | Labour | Product | ||||||||
| Institution | Monitoring | Take-over | constraint | market | market | |||||
| Plc | 2 | 1 | 3= | 1 | ?(1) | |||||
| Regulated plc | 4= | 2 | 4 | 3 | ?(4) | |||||
| Public enterprise | 4= | 6= | 5 | 5= | ?(5) | |||||
| Partnership | 1 | 6= | 1 | 2 | ?(2) | |||||
| Non-profit enterprise | 6= | 6= | 3= | 5 = | ?(3) | |||||
| Bureaucracy | 6= | 6= | 6 | 6 | ?(6) | |||||
Monitoring
This refers to monitoring by ‘owners or the representatives of owners’. We have placed the partnership at the top of the list because of the monitoring by other partners that is likely to exist in this type of organization. This does not, of course, imply continuous observation of effort and activities day by day. Indeed the difficulty or fatuity of such monitoring is sometimes given as a reason for the partnership form of enterprise in the first place. However, peer group appraisal figures strongly in partnerships and this is doubly effective because the appraisal is undertaken by people with a great deal of knowledge of the enterprise and a personal interest in maintaining its reputation.
Monitoring by shareholders in a plc is likely to be less effective because, with dispersed shareholdings, each individual shareholder will find the costs of monitoring will outweigh the individual benefits received. This is an example of the familiar ‘prisoner's dilemma’. Nevertheless a significant minority shareholder or an institutional shareholder may have an incentive to monitor managerial performance. A regulated enterprise will attract less shareholder monitoring if regulation reduces the return to such monitoring. For example, an enterprise facing rate-of-return regulation will not be able to pass on the returns from greater productive efficiency to shareholders and this will obviously reduce the incentive of shareholders to monitor.
Monitoring by government agencies such as the Monopolies and Mergers Commission is important in the case of public enterprise and it is a matter of judgement whether or not this represents a more or less tight constraint than shareholder monitoring in a regulated plc. We have placed outside monitoring of bureaucracies and non-profit enterprises last because of the lack of incentive to monitor combined with the difficulties of establishing agreement about the precise objectives of these institutions.
Take-overs
At one level this could be answered simply in terms of whether a take-over is a legal possibility. We, however, have ranked the plc above the regulated plc on the grounds that the incentive for a raider to take-over a firm will be lower if the benefits depend uncertainly upon the reactions of the regulators. In the limit, of course, a strictly enforced rate of return constraint will give no incentive for a take-over no matter how inefficient the firm has become unless returns have declined persistently below the constrained level.
An even more relaxed interpretation of the question (which does not underlie our ranking in the table) might have asked whether the possibility of institutional restructuring by outside forces was a possibility. In this very general sense, the threat of privatisation could be regarded as a relevant management discipline in the public enterprise with analogous consequences to the take-over. A study by Neuberg (1977) of electricity distribution in the US, which found public firms to have lower costs than private firms, noted that investor-owned firms might see themselves as relatively secure, whereas in a hostile political climate municipally owned firms ‘had to justify their existence on the basis of their competitive performance’ (p. 321). This interpretation might have permitted us to rank the public enterprise above bureaucracy. Even a bureau is not entirely secure, however. For example, it might be argued that the ability of schools to opt out of local authority control in favour of central funding is a way in which one set of bureaucratic managers can be replaced by another if they do not satisfy certain interested agents, in this case school governors.
The bankruptcy constraint
We have ranked the options here not just in terms of logical possibility but also from our impression of the consequences for the controlling managers of bankruptcy. Partners, it might be expected, would have the greatest personal reasons for avoiding bankruptcy and we have therefore ranked the partnership first followed by the plc and the non-profit enterprise. The regulated plc is ranked fourth on the grounds that the managers may have allies in the regulators when it comes to avoiding bankruptcy in a regulated concern. Ranking public enterprise is complicated by the fact that bankruptcy may or may not be a legal possibility. For example, a plc in which the state owns more than 50 per cent of the shares can obviously go bankrupt although we might expect politicians to try to avoid this eventuality by special ‘assistance’. A public corporation on the other hand cannot go bankrupt and its debts will always be honoured by the state.
The managerial labour market
A manager in any of the institutional forms that we are discussing might leave and look for alternative employment. Some, however, may find it easier than others to provide evidence of their efficiency. This is implicitly a question about ways of circumventing the ‘adverse selection’ problem. As mentioned in the text, a senior manager of a plc may find his/her reputation is dependent upon the behaviour of the shares in the company. Information about performance is much more difficult to convey in the other cases. Bureaucrats may face a particularly difficult problem here both because their degree of success or failure is difficult to measure and transmit to others, and because the criteria of success may differ between bureaux. Success in one field may be a less reliable predictor of success in another field than it is thought to be in the purely commercial arena. Wherever financial results are not of the essence, we have made the judgement that reputation on the managerial labour market will be less effective as a constraint on managers. This tends to put non-profit and bureaucratic enterprise at the bottom of the list. There may, however, be interesting exceptions to this rule. An important part of managing non-profit institutions is fund-raising from donors or sponsors, e.g. for educational institutions or other charities. This is a part of management activity in a nonprofit environment where success is judged in mainly financial terms. A market in professional management of this type is conceivable therefore, and may be quite highly developed in countries such as the United States.
Product market competition
Competition in the product market may influence all the institutions. The question is not concerned with whether in practice we observe one type of institution in a competitive market more often than another type. Obviously regulated plcs and public enterprises are frequently in natural monopoly situations and partnerships are not. Rather we are concerned with whether given a certain market environment competitive forces are likely to impinge on management decisions in one form of enterprise more than another. If they do, it is our judgement that the mechanism is through the other constraints which we have already considered. Clearly, competition in the product market is more serious if bankruptcy threatens than if it does not.
It is also necessary to remember the point made in the text that competition is not simply about market structure (i.e. the number and size distribution of the enterprises present) but about the incentive to act competitively, and this may in turn be related to the assignment of property rights. Unfortunately, competitive responses may be related not only to the property rights structure within an individual enterprise, but also to the nature of other competing enterprises. If this is so, our question becomes impossible to answer because of interdependencies between the differing types of enterprise.
We have indicated the extreme complexity of this question by a series of question marks in the table. For the sake of completeness, however, we have recorded a ranking which should be taken even less seriously than the others and which is based on the importance of profits to the enterprise. Non-profit enterprise is ranked above public enterprise and even above a regulated plc on the grounds that, although profits are not an end in themselves, the bankruptcy constraint is important, and downward pressure on prices in a competitive market would force such an enterprise to respond more quickly than would a public or regulated concern.
Output per head will rise faster than total factor productivity if inputs other than labour are increasing in importance relative to labour. If all inputs increase by the same percentage and output increases by a greater percentage, both labour productivity and total factor productivity will show the same rate of growth. The generally larger figures for labour productivity in the table indicate that all the public enterprises had been economizing on labour relative to other inputs over the period.
For some of the enterprises the growth rate of output per head was closer to the growth rate of total factor productivity than for others. The most spectacular difference is that for British Steel while the smallest is that of the Post Office.
The most obvious explanation is that industries in which technological change was slowest were those for which both productivity indicators gave a similar measure of growth. Examples would include the Post Office, National Bus and British Rail. British Telecom by contrast would be an example of an enterprise facing technical changes involving a substantial investment in capital equipment and resulting in a discrepancy between growth of labour productivity and growth of total factor productivity.
Another possibility is that those industries which were the most productively inefficient and wasteful in their use of labour at the beginning of the period were able to improve their performance as measured by labour productivity. This would appear to fit the case of British Steel which in its 1979 Corporate Plan aimed to cut capacity from 35 million tonnes to 14.4 million tonnes per annum. Labour input fell even more dramatically, however, from 196 941 in 1978 to 54 241 in 1986 (MMC 1988b, Table 9.1, p. 104). A similar argument would also apply to British Coal in the period after 1985 which, as was mentioned in the text, saw enormous increases in labour productivity.
All productivity measures requires a measure of ‘output’. An enterprise with low measured increases in total factor productivity can therefore always try to defend itself by claiming that important dimensions of output have been ignored.
In the case of telecommunications, an output measure in terms of number and duration of telephone calls would not take account of the quality of the calls, the reliability of the equipment, the availability of call boxes and so forth. Resources devoted to improving reliability of equipment will therefore not improve measured productivity. Molyneux and Thompson noted that, in 1978/79 58 per cent of reported faults were cleared by the next day compared with 90 per cent in 1984/85, yet this made no difference to measured output.
If public enterprises have social responsibilities these should be reflected in output measures. Suppose, for example, that the output of electricity generating companies is measured entirely in terms of power supplied, investments in environmental improvements such as the ‘scrubbing’ of flue gases will reduce measured total factor productivity.
By reference to passenger miles travelled, total factor productivity was virtually unchanged at National Bus according to Table 8.5. Such an output measure will not pick up improvements in the comfort of buses, passenger services available, safety, convenience and reliability.
