Public enterprises produce public and private goods
About a century ago, public enterprises were still primarily launched and undertaken in order to contribute to public revenues in many European countries. Following a trend which started after the First World War, tax revenues have become increasingly more important than non-tax revenues.
Today profits from public enterprises are a minor source of public revenues. From a strictly budgetary point of view, public enterprises today are often more important for the tax revenues they generate than for the non-tax revenues they provide in profits and sometimes also in fees. Today the charters of public enterprises often call for a whole range of different objectives other than the pursuit of profits. In Germany, there is even a school of experts in public enterprise economics who explicitly reject the idea of profit seeking for public enterprises and, instead, subscribe to the notion of ‘ GemeinwohT as an appropriate description of their activities, thereby putting the whole public enterprise industry into the non-profit sector.A point of theory
In trying to suggest an appropriate technique for analysing the performance of public enterprises thus described, we have to reckon with the many and different purposes governments try to pursue with them. As the preceding discussion may have suggested, the definition of the purpose of public enterprises is mainly negative. It is easier to characterize them in terms of the purposes they are explicitly not suggested as serving, instead of precisely establishing what the purposes are that they are supposed to fulfil. This may be a deplorable situation from the point of view of economic analysis, but we cannot reject the possibility out of hand that the apparent fuzziness of public enterprise purposes in most democratic societies is the very reason why they occupy such a prominent and increasing role in many economies, privatization programmes notwithstanding.
The implication for attempts at evaluating the performance of public enterprises is that a technique of performance analysis has to be found that can serve as an envelope for the fuzzy purposes public enterprises are fulfilling. Instead of clearly establishing performance criteria drawn from applied welfare economics and attributing them to public enterprises, a technique may be preferable that starts from the public enterprise itself and attempts an increasingly more precise picture of the fuzzy set of objectives pursued. While such an evaluative procedure may be less straightforward than the Marchand et al. approach, the hope is that it will eventually be more enlightening in telling us what public enterprises really do, how they do it, and how well they do it.The basic theoretical point of departure for this approach was introduced by Mancur Olson at a National Bureau of Economic Research (NBER) conference in 1973 devoted to the measurement of economic and social performance (Olson, 1973). Olson’s point of departure is the suggestion that ‘governments are in fact as well as by reputation usually inefficient, and that this is mainly because they deal with collective goods and externalities’ (ibid., p. 359). It should be emphasized from the start that Olson explicitly refers to ‘technical (in)efficiency’. His notion of efficiency has to be seen in the context of, for example, the literature on production functions, and not in the sense in which it is used by members of the Chicago school such as George Stigler. (See the interchange on X-inefficiency by Stigler, 1976, and Leibenstein, 1978.) This choice of efficiency concepts implies that public production, even where its methods have been shown to be inefficient for not reaching the production possibility frontier, may yet not be improvable in the Pareto sense because no better alternative is known that can readily be implemented. In terms of a transactions cost approach, this implies that public production, even where it is shown to be technically inefficient according to the performance analysis approaches, may still be efficient because unimprovable, since the transactions costs involved in this particular process of production are prohibitively high in rendering the internalization of externalities infeasible.
Which areas of public production is Olson referring to? His examples include law and order, defence, basic research and pollution control (Olson, 1973, p. 359), areas which form the new primary domain of public enterprises as opposed to the past smokestack industries. It’s now some 40 years since Olson made this suggestion, public enterprises today are increasingly retreating from the production of merely private goods and thriving where they contribute to the production of positive externalities and public goods, a contribution often made as a byproduct of traditional production. Research laboratories with spin-offs for defence, passenger railroads to replace car transport and universities contributing to basic research are all cases in point and mark public entrepreneurship to which all Western states cling, irrespective of privatization programmes they may otherwise be engaged in.
The inherent inefficiency to which Olson pointed is unavoidable because it is inherent in the very definition of a public good. These inefficiencies are of two kinds. To the extent that no-one can be excluded from the consumption of the good, the true preferences for the provision of the public good are difficult to establish. Hence, in assessing the performance of public production, we never know whether too little or too much is being produced. The second source of inefficiency is more difficult to handle. Even where governments provide too little or too much of a particular public good, they may do so perfectly efficiently in terms of reaching the production possibility frontier. The condition of jointness of consumption implies that the marginal calculus breaks down and that measuring of outputs becomes infeasible. As Olson puts it: ‘The very characteristics of a collective good that make it a kind from which non-purchasers cannot be excluded, also make it a kind of which the output is not in the form of divisible units that can be readily counted’ (ibid., p. 362).
Since evaluating the performance of public production requires the attribution of some values to the output, performance analysts face a grave dilemma:
The customer [demanding a private good] is in a position to estimate the value to himself of an additional supply of service from a firm because he can experiment with different levels of purchase.
His estimate of the value of the goods shows up in the marginal revenue the firm receives. By contrast, in the case of a collective good going simultaneously to many individuals, the individual consumer cannot take more or less to see how that effects his wellbeing. (Ibid., p. 363)This second difficulty leaves us with the dilemma that the production functions for public goods and externalities cannot be established and production cannot be evaluated in terms of technical efficiency (ibid., p. 405). Since this will also be true when public enterprises simultaneously engage in the joint production of private and public goods, a different approach will have to be devised.
While Olson’s point is extremely important and very basic, it does not lead immediately to ready-made solutions to improve the efficiency of public sector production. His own suggestions (selective and experimental pricing and spot controls) might help on occasion, but if systematically used would likely invite inefficient responses on the part of the public institution thus controlled (see Peirce, 1977). Not surprisingly, agreement could not be reached at the NBER conference, and Olson’s important paper seems not to have had major consequences so far. Fortunately, a completely different strand of economic literature has developed right to the point that it can be linked up with Olson’s approach.
Improving performance when objectives are vague
The economics of public enterprises and public administrations has traditionally received the attention of public finance theorists. Students used to be trained in methods of assessing the performance of public institutions and ensuring administrative efficiency. Public finance developed as a discipline separate from political economy because its subject matter, the conduct of state institutions, required a combination of economic theory and the art of public management and accounting. In the aftermath of the Keynesian revolution, applications of the Keynesian paradigm became for a while the main
focus of public finance research.
The resulting decoupling between the discipline and its subject matter produced a gap which still remains to be filled satisfactorily in terms of modern political economy. Since a practical need was not addressed by the discipline, a substitute had to arise. In Germany, Ludwig Mulhaupt and his students played a major role in establishing the new sub-discipline of ‘offentliche Betriebswirtschaftslehre' (public managerial economics). The essence of this approach is readily stated. In order to improve the performance of public institutions, modern management techniques are to be applied. The same techniques that have proved valuable in making corporations more efficient should also help public institutions more efficiently reach their objectives. This proposition raises the central question of work done by Gunther E. Braun (1988).What are the objectives of public institutions as compared to private corporations?
Braun’s studies are organized in six chapters. A very short first chapter introduces the theme and summarizes the structure of the book. Chapter 2 discusses the research programme of public managerial economics. Chapter 3 introduces the methodology of comparative studies and Chapter 4 gives a general overview of the comparative analyses of objectives. Chapter 5 provides an in-depth description of the empirical study and Chapter 6 draws the conclusions for the possibility of using the planning methodologies developed for corporations in public institutions. The spectrum of public institutions discussed ranges from applied research organizations such as the Bundesgesundheitsamt (Federal Health Office), Bundesinstitut fur Berufsbildung (a federal institute charged with researching vocational and professional education), the Gesellschaft fur Mathematik und Datenverarbeitung and the Hahn-Meitner Institute (charged with research into computer science and nuclear physics, respectively), to the Federal Department of Agriculture, the Chancellery of the State of Rhineland Palatinate and several municipal welfare offices in North Rhine-Westphalia.
As varied as these institutions are, they have two related characteristics in common: they are public and their objectives are vague. The vagueness of these objectives is throughout established and further detailed in 13 hypotheses, largely corroborated, with the exception of hypothesis 2. These hypotheses are as follows:1. the more an agency is engaged in producing public goods, the less precise its objectives will be;
2. the more the objectives relate to technology and the natural sciences, the more precise they will be. This is the hypothesis that could not be corroborated; objectives relating to technology and the natural sciences are just as vague as those relating, for instance, to social policy;
3. the broader the menu of programmes at the top hierarchical level, the less precise its objectives will be;
4. the broader the service programme of a municipal office, the less precise its objectives will be;
5. the broader the service programme of a bureaucratic agency, the less precise its objectives will be;
6. the higher the external hierarchical level at which an institution operates, the less precise its objectives will be;
7. the higher the internal hierarchical level, the less precise its objectives will be;
8. the more professionalized an organization, the more precise its objectives will be;
9. the closer a public institution is to other actors in the political system, the less precise its objectives will be. The other actors may be various pressure groups, supervisory bodies, competing institutions and so on;
10. the more external participants are involved in decision making, the less precise its objectives will be;
11. the more long term and/or strategic the orientation of an organization, the less precise its objectives will be;
12. the more objectives serve as integrative or public relations devices, the less precise they will be; and
13. the less objectives are related to problems, programmes or projects, the less precise they will be.
Hypotheses 3, 6 and 8 could not be systematically tested with the available material.
Professor Braun uses these results to indicate where and to what extent corporate planning techniques can be tried in public sector organizations. Obviously, the more precise the objectives of public sector organizations are, the more applicable those planning techniques. On the other hand, softer planning techniques may be used where objectives are necessarily vague. The author is careful to point out that it would be wrong to insist on more precise definitions of objectives when the vagueness may be inherent and functional for public sector organizations. The very application of business techniques in public sector organizations can be a political ritual, and the benefits of the exercise should be considered in this light (Braun, 1988, ch. 6, § III). Performance studies commissioned by public authorities may sometimes serve a similar purpose.
The theoretical study by Olson and the empirical study by Braun both point in the same direction. The objectives of public institutions are typically vague and not readily expressed in terms of the received theory of the firm. The production of externalities and public goods is an important function of public institutions, often pursued side-by-side with the production of private goods and services. As a rule, the performance of public institutions cannot be assessed with the same methods used for private firms. The exception to this rule is the case where a public enterprise is explicitly and exclusively run as a revenue source for the public purse. Although the objectives of public institutions seem vague and fuzzy, because they cannot be expressed in clear and simple terms, they can still be established with sufficient clarity to enable the analyst to assess the performance of the institution. While this claim can ultimately only be defended by pointing to a successful performance analysis on which agreement can be reached, the stepwise procedure developed in the following section may at least serve to indicate what can be expected from the alternative approach suggested here.