The Political Economy of Public Administration: Institutional Choice in the Public Sector
Murray J Horn
Cambridge University Press, $54.95,
ISBN 0 521 48436 7
New Zealand has had several secretaries to the Treasury with high levels of intelligence. It has had even more secretaries to the Treasury who have displayed great diligence. A few have belonged to both sets. It has not previously had a Secretary to the Treasury who has been able to combine the duties of the office with turning a PhD thesis into a book which advances the boundaries of theoretical knowledge and also contributes to severely practical issues of public management.
“Public choice” is often combined with “new right” as part of the lexicon of demonology which disfigures political debate. Even in more sophisticated discussion it is often treated as the simplistic notion that public officials and legislators are entirely self-interested. Not surprisingly, it is then found to be unsatisfactory when it is expected to provide ready-made and complete answers to policy issues. Murray Horn shows that public choice is a mode of reasoning which includes concepts and techniques which can be used to illuminate both policy issues and public management.
In some respects, Horn returns to a very old question. How can successors be bound? Machiavelli offered reflections along the lines of “the welfare of a republic or kingdom does not rest on having a prince who governs it prudently while he lives but rather on having one who organises it in such a way that it endures after he dies”.
Recent New Zealand history suggests that we should link such thoughts to his related reflections:
And it ought to be remembered that there is nothing more difficult to take in hand, more perilous to conduct or more uncertain in its success, than to take the lead in the introduction of a new order of things. Because the innovator has for enemies all those who have done well under the old conditions and lukewarm defenders in those who may do well under the new. This coolness arises partly from fear of the opponents, who have the laws on their side, and partly from the incredulity of men, who do not really believe in new things until they have had a long experience of them. Thus it happens that whenever those who are hostile have the opportunity to attack they do it like partisans, whilst the others defend lukewarmly, in such wise that the prince is endangered along with them. It is necessary, therefore, if we desire to discuss this matter thoroughly, to inquire whether these innovators can rely on themselves or have to depend on others: that is to say, whether, to consummate their enterprise, have they to use prayers or can they use force? In the first instance they always succeed badly and never compass anything; but when they can rely on themselves and use force, then they are rarely endangered. Hence it is that all armed prophets have conquered and the unarmed ones been destroyed. Beside the reasons mentioned, the nature of the people is variable and, whilst it is easy to persuade them, it is difficult to fix them in that persuasion. And thus it is necessary to take such measures that, when they believe no longer, it may be possible to make then believe by force.
Horn provides alternatives to both prayer and force. His argument is essentially an explanation for two important types of institutional regularity:
1. Distinctive governance, financing and employment characteristics of different organisational forms — like bureaus, SOEs and regulatory agencies — appear to be remarkably stable in the modern bureaucracy.
2. There are important regularities in the relationship among these different forms of organisation and the administrative functions they are asked to perform. When it is possible to sell public sector output, for example, we are more likely to see an organisation with the characteristics of an SOE than a tax-financed bureau; that is, an organisation that is funded by sales rather than taxes is governed by a board that enjoys some independence from the legislature and is less constrained by civil service rules. And, when public and private enterprise are compared, there is a remarkable similarity in the industrial concentration of SOEs across sectors in different countries. (p2)
(A “bureau” is more often referred to as a “department” in New Zealand. The book is genuinely international in conception and is not unduly influenced by its origins as a doctoral dissertation in the United States. Indeed, its Harvard sponsors are to be congratulated on their tolerance of New Zealand empirical material.)
The argument puts legislators in the central position, which has interesting implications for two common beliefs, that public servants do not recognise the importance of Parliament and that public choice treats legislators as a kind of personified counting device which determines which voting choice serves the material interests of the greater number of voters.
Horn rejects both simplifications and conceives the central political problem as delivering a programme and providing some assurance that it will continue to exist in the future. The important decisions include whether legislation should be detailed or should provide scope for discretionary decisions by administrators and what kind of secondary decision-makers should be engaged — courts, specific regulatory agencies or public sector producers. This in turn permits exploration of the roles played by familiar institutional structures such as security of tenure and hierarchical salary structures. It is a subtle argument, drawing together ideas which are often seen as coming from different intellectual traditions in “public choice” and “transaction costs” analysis (and even from the remote discipline of management), and yet a great deal rests on the simple distinction between the “enacting coalition” and the “incumbent coalition”.
More formally, the essential core of public management is seen as the control by legislators of four elements of cost.
The first is legislative cost. Like everybody else, legislators have limited time. Spending time on refining a piece of legislation so that it earns gratitude from some voters without unnecessarily antagonising others means that time cannot be used for other purposes. (Horn is concerned with legislators who know what they are doing, so that he mostly ignores the problems of confused and inexperienced legislators. There is nevertheless some immediate resonance with the complaints of successive leaders of the House that their colleagues expect them to perform miracles in managing the legislative programme of the House.)
Second, there is commitment cost. Legislators will gain more credit from the supporters of a piece of legislation if they ensure that they or their successors cannot reverse the relevant intervention — legislators can gain now by making it more difficult for themselves to act in future. This has been much discussed in recent years under the general heading of “credibility”.
Third, agency cost. Legislators usually need administrators to implement and manage the programmes they mandate but those administrators will not necessarily have the same interests as the originating legislators — there is a familiar principal-agent problem.
Fourth, constituent participation. Constituents can manage for themselves some of the follow-up to legislation but they will vary in the cost they need to incur to do so. A small number of constituents, each of whom has a lot at stake, will be willing to participate directly and we might expect that legislation will leave a lot to the discretion of subsequent administrators. Where legislation has a lot of risk-averse beneficiaries, each of whom benefits only a little, legislators will expect little gratitude for anything not securely tied down in advance.
Horn then uses this analysis (to good effect) to explore the regularities which can be observed in how regulatory agencies, bureaus and state-owned enterprises are used.
The analysis is full of interesting implications. In the case of regulatory agencies the key issue is the possibility of constituent participation. At one extreme the legislation need provide only for self-regulation; that will most obviously be the case with a small group of participants able to check on their competitors. The Advertising Standards Complaints Board seems to fit this picture. But what about telecommunications? Shouldn’t it also provide a clear example of self-regulation at work? The argument seems to imply that we simply have to be patient while key players in the industry learn how to manage their own interests. That is not implausible but we may wonder how long we have to wait. In the meantime, we can reflect on how well the theory fits the different fates of the pay equity legislation and the public sector reforms as a whole and notice how little justification there is for industry-specific regulators.
A particularly interesting application of the general theory is in explaining why “merit appointment” was widely adopted in the late nineteenth century (whether in the progressive era in the United States or in the 1912 organisation of the public service in New Zealand). Arguments about corruption and meritocracy can take us so far but explicit attention to the importance of “commitment” and preserving reforms from subsequent political attack is an organising principle that reconciles different experiences in a much more satisfactory manner.
This is a highlight of the book, almost a “play within the play”. Horn notes (p96) that a satisfactory theory has to be able to explain:
1. It should be able to account for the widespread persistence of the merit system, especially appointment by competitive examination and restrictions on dismissal. Since the middle of the 1800s the merit system has become firmly established in many countries.
2. It should also be able to explain the other major characteristics of the conditions of employment established by the modern civil service. Why do bureaucrats often receive protection from outside the civil service? Why is their tenure typically so secure? Why is their compensation system structured the way it is, with payment based on a centrally determined number of graded positions (rather than work done) with longevity payments and with a relatively prominent role for pensions?
Horn shows that these frequent and long-standing characteristics of the public sector are not arbitrary. And he does this without resorting to any simple belief that the market will already have exploited all opportunities for material gain. An explanation can be developed in terms of managing the four kinds of cost described above, especially preserving administrators’ commitment beyond the life of the originating legislature.
Similarly, Horn shows that we can use the same elements to explain why state-owned enterprises across many countries are concentrated in particular sectors of the economy.
It is sometimes difficult to repress the reaction that with sufficient ingenuity an explanation can be provided for anything, especially if any collective decisionmaking is involved. We do need to ask not only, “can this institution be understood?”, but also, “when this institution is understood, does it still serve the policy objectives which politicians subscribe to after they have engaged in debate with well-infomed advisers?”
Alternatively, we in New Zealand can ask how do we evaluate the public sector reforms of the last decade and a-half if traditional practices can be provided with such a cogent rationalisation? We should first of all note that the book is a revision of a dissertation which was completed in the 1980s. Its ideas were available to those who were concerned with the implementation if not the original design of the State Sector Act 1988 and Public Finance Act 1989 and they very clearly informed the design of the Fiscal Responsibility Act. This book is not an argument that the public sector reforms have been a mistake.
It is the basis for noting that one aspect of the public sector reforms was to undo the mechanisms by which previous legislators had tried to insulate their solutions from reconsideration. The key to Horn’s analysis, it will be recalled, is the difference in interests between the enacting legislature and the incumbent legislature. But each successive legislature will want to promote its interests against earlier (enacting) legislatures and to protect itself against subsequent legislatures.
An economist will ask how we know when the interests of the incumbent legislature should prevail against those of preceding enacting legislatures and will be inclined to find answers in terms of short-term crises and longer-term declines of relative living standards and so return to standard analyses of the 1980s. A political scientist might ask how “democracy” fits into a distinction between the interests of the incumbent legislature and those of a preceding enacting legislature. Government of, by and for which people? An historian might say this is all very well but ask how we tell when a legislature has made a commitment with the electorate?
The alleged compact of increased tax levels in return for full employment and welfare services was created in the 1960s and applied retrospectively, rather than agreed and implemented in the 1940s and 1950s. The “broken promises” of the 1980s were created by media (and political opponents’) repetition of their view of what “reasonable” listeners would have understood politicians to say rather than direct observation of promises and subsequent actions. So how do we tell the intentions of enacting legislatures?
Horn works hard to confront his theoretical analysis with empirical data. To some extent he relies on “stylised facts” and draws equally readily on the data from the United States or New Zealand or anywhere else. It would be simply wrong to describe his book as abstract, let alone ideological. But it is still concerned with “public administration” (or the “new public management”) at a high level of generality. Those who would understand specific events and processes, such as New Zealand in the 1980s and 1990s, need more than an explanation which operates on such a general level. But they would be unwise not to recognise how much can be explained at a general level rather than by particular features of New Zealand.
Some ministers in the 1980s were fond of reflecting that when a group of ministers met a group of departmental chief executives there was, for the first time in New Zealand’s history, a reasonable chance that the former was more intelligent and had higher educational qualifications. Even though Horn’s predecessor, Graham Scott, was not one of those upstaged in that way, The Political Economy of Public Administration should give pause to any minister tempted to make such a claim now. The intelligence and academic achievement documented here is formidable.
But there is a still deeper point. The most important response to a minister of the 1980s is that he and his colleagues showed that intelligence and academic qualifications do not always translate into political achievement. What is remarkable about The Political Economy of Public Administration is that it gives ample testimony to the compatibility of intelligence and academic qualifications with common sense and judgment. We now know that all those qualities can be found in senior bureaucrats; whether they are to be found among politicians remains an open question.
Gary Hawke is director of the Institute of Policy Studies at Victoria University.
The Political Economy of Public Administration recently was awarded the United States Academy of Public Administration’s Louis Brownlow book award. Murray Horn is Secretary to the Treasury.