To speak of controlling public expenditure is to assert what remains to be proven. For the Treasury to control public expenditure it would need to control not only the state of the economy as a whole but also spending and revenue-raising by central government, local authorities and nationalised industries. Insofar as this is difficult, the Treasury could maintain the appearance of control if it predicted what would happen to the fiscal system each year, and adopted these predictions as its own policy goals. As long as the outcome was as predicted, the correlation between what happened and what the Treasury proposed would be consistent with a claim of Treasury authority.

But in turbulent times predictability, the holy grail of social scientists, is easier sought than attained. Insofar as the chief determinants of a fiscal system cannot be predicted, then central as well as local governments must make fiscal decisions that are very uncertain of realisation. Instability can become a chronic condition rather than a merely transitory phase of public finance. The object of this chapter is to demonstrate that local government fiscal systems are now best described as chronically unstable rather than incremental in their principal financial characteristics, to explain why instability occurs when the desire for predictability is great, and to consider problems of coping with a fiscal system that cannot be controlled.

Fiscal stress is a symptom that something is wrong in our cities today, and that something should be done to relieve or remove the causes of stress. The problem is that there are more claims upon the public purse for services than there is money to meet these claims. Stress affects the everyday workings of local government, and creates friction in relations between town halls, responsible for delivering many public services, and Whitehall, responsible for many decisions about public spending and service priorities. Ordinary citizens are aware of fiscal stress, simultaneously facing annually escalating bills for local taxes and threats to cut back local services long taken for granted as desirable. Nor is fiscal stress unique to Britain; it is found in the United States and many European countries as well.

The problem of fiscal stress is not simply a dispute about cutting public expenditure or about the need to maintain local services. It is about achieving a balance between the costs and benefits of government. Fiscal stress is the dominant problem facing cities in the 1980s, if only as an unintended consequence of cities seeking to deal with other problems. Local government is a major provider of public policy benefits such as education, housing, roads, social services for the elderly, and police and fire protection. These are usually thought of as ‘good’ goods and services. But providing benefits costs money, lots of money. As the cost of government rises, its need for revenue increases.

Urban change and its costs are no novelty; it is almost exactly a century since Mearns (1883) published The Bitter Cry of Outcast London. Nor is the concern of governments with the costs of urban change new. Indeed, whenever governments have attempted to leave the cities to their own fates, as in the periods 1854–70 or 1921–9, the sheer pressure of events has eventually forced a renewal of intervention. It is already evident that the street disturbances of 1981 constitute a major challenge to present urban policy and it would be surprising if these disturbances did not produce some change of direction by central and local government.

Historically, questions of urban change have tended to dominate the agenda of central-local relations in Britain. But the urban problems of the 1980s are quite unlike those of the 1880s. Victorian urban problems arose from industrialisation and urban growth, from a mismatch between population growth and the development of urban infrastructure, from shifts in the arrangement of the space economy, in particular labour and capital, in favour of cities. The problems of social and fiscal stress in the 1980s arise from the obverse of these developments: urban disinvestment and decline. There is something new about recent urban change, something that is caught by the terms counterurbanisation or the post–City age. There is something new in national policy too, an arguably misplaced concern with economic regeneration to the virtual exclusion of ameliorative social measures.

Introduction

What has come to be known as public choice rests on the basic intuition - initially sketched by Schumpeter - that the methodology of neoclassical economics can be used to analyze decisions and adjustments of participants in the public sector, and in other areas where decisions and adjustments are collective, just as it can in the market sector of modern economies. The working out of this intuition, one imagines, would narrow the awesome gap that separates public- and private-sector economics and eventually produce a unified body of analytical and testable propositions. Some progress in that direction has no doubt been achieved in the last 40 years, but the gap is still very much a reality.

One of the central tenets of neoclassical economic methodology is that the forces that operate to determine any outcome can be classified into two separate, but interacting, sets usually labeled demand and supply. In this book, we focus solely on supply questions by concentrating on decision making and adjustments in governmental bureaucracies and business corporations. We are therefore assuming that the demand side in the private and public sectors can be modeled separately and that those demand models will be consistent with the suggested theory of supply. By the latter, we mean that it will be possible to formulate a theory of the interaction of the two, because the ultimate test of the validity of any theories of supply and demand is how they combine to generate desired outcomes.

Introduction

As we have stated in previous chapters, informal labor services in bureaucracies are bought and sold in networks; in other words, selective behavior originates, takes form, and unfolds in networks. We have also argued that the activities taking place in these networks are best modeled as mimicking market supply and demand situations, hence our references to surrogate markets.

However, markets - whether competitive or not - presuppose the existence of property rights that are generally supported by laws and law enforcement. Although we treat networks as surrogate markets, we must insist that they differ from markets in the fundamental sense that the property rights that they presuppose are not supported by legal arrangements. Instead, they are supported by trust. In other words, whereas market exchange requires law-based property rights, network exchange necessitates trust-based property rights.

The analogy between law and trust should, however, not be pushed too far for at least two reasons. First, as we shall note in Section 4.6, the nature and properties of such market phenomena as externalities, moral hazard, and contingent contracts are different under the two arrangements. Second, many real-world property rights - in both markets and networks - are supported by both law and trust, albeit in sometimes dramatically different proportions. These arguments notwithstanding, it is enlightening to suppose that the distinguishing trait of markets is legally enforceable property rights and that of networks property rights based on trust. In the remainder of this chapter, we will assume that markets are supported only by legally enforceable property rights, whereas networks are supported only by trust.

The motivation for writing this book came from our dissatisfaction with existing economic theories of bureaucracy. In his 1974 monograph The Economic Theory of Representative Government, Breton adopted Niskanen's general approach to bureaucracy. Therefore, he had nowhere to turn for a hypothesis about the relationship between public bureaucrats and politicians. To solve this problem, he assumed that whenever there was a conflict between these two groups, the one with greater power would win. Without a theory of accumulation of power by bureaucrats and politicians, that assumption was, of course, unsatisfactory.

In 1975, Breton and Wintrobe published a note on Niskanen's theory seeking to provide the beginnings of a model on the relationship of bureaucrats and politicians. In that paper, it was suggested that bureaucrats must be viewed as subordinates even if they are also someone's superiors. It was argued that the power of bureaucrats lies not in their formal position but in their control of information, that is, in their ability to distort and conceal it. Further, to the extent that governing politicians can police or counteract the distortions of bureaucrats through the use of redundancy, external data checks, and direct monitoring, the power of bureaucrats would be checked, and they would revert to their formal status as subordinates.

In 1976 Wintrobe completed his doctoral dissertation on bureaucracy. Building on an idea of Tullock, he sought to break with the tradition of modeling bureaucracies - public or private - either as pure systems of authority or as pure voluntary associations such as teams.

Introduction

Both public bureaucracies and corporate (private) bureaucracies are engaged in the production of goods and services. However, if we are to understand bureaucratic production, at the level of decision making it is necessary to distinguish between two types of production decisions: (i) policy decisions, which embody expedient courses of action to be pursued by the bureau or bureaus, and (ii) technical decisions, which relate to the choice of techniques and the combination of factors of production required to implement policy decisions. Because the first type of decision - the policy decision - is by far the more important, we focus on it in this chapter and in the book. Indeed, we assume that governmental bureaucracies produce public policies, whereas corporate bureaucracies produce private policies. To put it differently, we assume that in governmental and corporate enterprises, what is properly bureaucratic pertains to decisions about what will be produced; the question of how it will be produced, although important, is secondary.

There is a long and respected tradition in the public finance literature that identifies the output of governments with public, collective, or social goods, sometimes defined as goods (and services) that are equally available to all and sometimes, rather vacuously, as goods supplied by governments. There can be no useful quarrels about definitions. definitions.

Introduction

In this chapter, we present some applications of the theory developed in the foregoing chapters. We seek to achieve three objectives. First, we hope that the selected cases will help illustrate the logic of bureaucratic conduct in more concrete terms than the abstract arguments of the preceding chapters have done. Second, we present these applications as exercises in comparative statics. All are analyzed as responses of bureaucracies - public or private, depending on the case - to exogenous disturbances and are consequently true comparative static exercises. Third, these cases represent problems that are badly understood by economists. If we are able to explain these phenomena, the power of our model will be underscored.

We first analyze the role of organizational structure as a factor of production, in addition to conventional inputs such as labor and capital, in explaining productivity differentials among firms and the growth rate of productivity. We will also use this application to explain other phenomena, such as the well-known evidence that productivity is procyclical and the hypothesis that political instability (frequent changes in the governing political party) implies that decision-making power tends to fall into the hands of the bureaucracy. Section 7.3 then uses this model of organizational structure to explain how the Japanese system of management works and why Japanese productivity has grown so rapidly since World War II.

Section 7.4 applies our model of bureaucracy to a problem in public sector economics: the origin of wage and price controls.

Departures

In the preceding chapters, we presented a theory of bureaucracy that is logically self-contained and capable of predictions, and that, we believe, explains a good deal of bureaucratic conduct. Although it is also a fairly simple theory, it departs from accepted models and views of bureaucracy in a number of ways that we will mention as a way of summing up.

Our most important departure from conventional theories of bureaucracy is this: We do not model the relationship between superiors and subordinates as relations of authority, but instead as relations of exchange based on trust. In our view, bureaucracies should, at least in part, be conceived as sets of networks (surrogate markets) in which bureaucrats - superiors and subordinates at all levels - trade with each other. It is true that when inefficient services are supplied by subordinates, superiors acquire something they do not want. However, this does not mean that the relationship is not one of exchange, any more than having a transmission replaced in a car that needs a new gasket implies that one has left the world of market exchange.

One could say that our analysis of network trading - in which buyers sometimes get what they want but at other times get something quite different - is applicable to a wide range of transactions in the markets of everyday life, where one generally gets what one expects, but sometimes not. The difference is that network trading is mainly supported by trust, whereas market trading is mainly supported by contracts. But that difference should not be exaggerated, because trust also plays a role in many market transactions.

Introduction

As a label, the word bureaucrat is easily misleading, because a typical bureaucrat is at once a subordinate and a superior. This dual role of bureaucrats implies that it is not possible to formulate an adequate theory of bureaus unless at least three levels are distinguished in hierarchies. Let us call these levels i, j, k and label the bureaucrats at each level Bi, Bj, and Bk. Typical bureaucrats Bj, are at once the subordinates of Bi and the superiors of Bk. In a bureau with N levels, the bureaucrats at the N - 2 intermediate levels are all Bj - bureaucrats; hence the word typical. Moreover, if we recognize that the decision makers at the highest level in the hierarchy are not bureaucrats at all, but politicians in governments and managers in businesses, then our typical superior-subordinate, dual-role bureaucrat is to be found at N - 1 levels in hierarchical structures.

The dual role of bureaucrats plays virtually no part in the general literature on bureaucracy and is seldom recognized in the economic literature on that subject. However, it is central to the discussion that follows and, we suggest, to a clear understanding of the workings of bureaucracies.

A theory of bureaucracy, as we emphasize in Chapters 5 to 7, must deal with the nature and outcome of competition between bureaucrats and between the bureaus that constitute the whole.

Introduction

In this chapter, we develop a theory of the allocation of resources among bureaus. We explain why some bureaus grow and others decline, or to put it differently, how some bureaus can obtain more resources from sponsors over a given period of time, whereas others cannot. We shall not be concerned with the growth or decline of the bureaucracy as a whole; that subject is not addressed in this book. We deal with the relative growth or decline of bureaus in a context in which the size of the bureaucracy as a whole may be constant, increasing, or decreasing.

The starting point for our theory of resource allocation among bureaus is the theory of selective behavior. Viewing selective behavior in terms of supply and demand for informal services, that theory up to this point has been based on the assumption of a fixed or predetermined demand curve. We shall therefore begin by developing a theory of the demand for informal services.

Coupling this theory with that of the supply of informal services - developed in the last chapters and elaborated further in the present one - gives us a complete theory of the quantities of informal services traded in different bureaus and, therefore, a theory of the size distribution of networks. To go from there to a theory of the size distribution of bureaus requires one additional assumption. That assumption, discussed in more detail below, is that in the absence of informal services, the costs of producing policies in different bureaus (the FAC curves) are identical.

Introduction

The existence and extent of competition within organizations - whether public or private (corporate) bureaucracies - is clearly of fundamental importance. In the literature on corporations, there appear to be two classes of models. In one class are those models that postulate competition; in the other are those that (implicitly or explicitly) ignore it completely. Models that ignore competition among bureaucrats (called “managers” in that literature) explain corporate behavior as based solely on the interests of these bureaucrats or managers and lead to fears of excessive growth and power in the corporate sector of the economy. One well-known writer in this tradition visualizes the entire economy as if it were a single giant corporation. On the other hand, models that postulate competition among managers lead to the easy dismissal of these fears and to an understanding of the growth of the corporate sector as prima facie evidence of its efficiency rather than just a lurid reflection of managers' interest in that growth.

Turning to the literature on governmental bureaucracy, we find a conspicuous absence of the counter-accusations characteristic of the debates over corporate power. Instead, we find models based single-mindedly on the monopoly assumption. All the theories of public bureaucracy put forward to date assume that bureaucrats are monopolists. Sometimes the monopoly assumption is explicitly made, as in Niskanen's work or in that of Migué and Bélanger.

Suppose for the moment that capitalists compartmentalize labor markets as described in the previous chapter. Why do workers put up with the possibilities for work, often appallingly limited from a middle-class point of view, that capital offers?

To many the question will seem as ridiculous as asking why men put up with injustice and oppression. An obvious answer is that they have to: Rebels are crushed by those who want the world as it is; those who refuse bad jobs find none better. But even if it were proved that fear suffocates revolutionary thoughts, brute necessity is not an exhaustive explanation of why workers accept certain jobs.

There is much evidence, we will see, that workers choosing jobs often pick one that most of us would find unappealing. Immigrants, for example, frequently prefer relatively high-paying, unskilled, insecure, dead-end jobs to jobs that pay less but offer the chance to learn skills–jobs, in other words, with a future. There are also cases of unskilled workers who more or less accidentally reveal an extraordinary aptitude for some line of work, and then refuse promotion. Nor is it uncommon for very skilled craftsmen to refuse to compete for promotions for which they are and know themselves to be superbly qualified. And there are, to take a final and more pointed example, many instances of workers bearing the economic and psychological hardships of unemployment rather than accepting just the kinds of jobs that immigrants seem most willing to take.