PUBLIC ECONOMICS AND PUBLIC ADMINISTRATION

Public administration has always overlapped economics. Efficiency was the holy grail of the progressive officials and academics who created the modern discipline of public administration. They sought to place public affairs "on a strict business basis," directed "not by partisans, either Republican or Democrat, but by men ... skilled in business management and economics." Consequently, they created a professional bureaucracy to manage "the increased importance of the public functions of the twentieth century city. Streets had to be paved for newly developed motor vehicles; harbors had to be deepened for big, new freighters. In addition, electric lighting systems, street railways, sewage disposal plants, water supplies, and fire departments had to be installed or drastically improved to meet the needs of inhabitants, human and commercial, of hundreds of rapidly growing industrial centers" [Weinstein, 1968: 93-95; see also Rubin, 1993]. Moreover, establishing a professional bureaucracy at the municipal level did lead directly to higher levels of investment in infrastructure and, thereby, to significant increases in economic growth [Rauch, 1994; see also Mauro, 1995].

Organizational efficiency once meant the Weberian bureaucratic paradigm, which was codified for the public sector in the Taft, Brownlow, and Hoover Commission reports [Barzelay, 1992; see, for example, Blau & Meyer, 1971]. In the years following publication of the first Hoover Commission report, public administration did not abandon the bureaucratic paradigm, but it drifted away from economics. Public administrationists discovered organizational psychology and behavior. Many in the field rejected the distinction between politics and administration, with its stress on neutral competence. Some were intimidated by the mathematics used by the rational choice disciplines. And a few rejected the traditional goals of economy and efficiency on ideological grounds

Public administration's drift away from economics was not entirely one-sided. Fifty years ago, most English speaking economists accepted Pigouvian welfare economics and Keynesian macroeconomics. They generally believed that government should set goals and objectives for the economy as a whole. Many admired the system of detailed centralized planning and control used by Gosplan in the Soviet Union to implement its long-term policies and strategic plans, an adaption of the Kriegwirtschaftsplan, the planning and control system used to mobilize Germany's resources during World War I -- i.e., Weberian bureaucracy carried to its ultimate conclusion. Indeed, Gosplan's approach was not unlike the planning and control mechanisms used in the United States and the United Kingdom to fight World War II. Gradually, however, most economists came to appreciate the dysfunctions produced by state allocation of productive assets and central planning and control and to recognize the impossibility of a Pigouvian social welfare function. For many this appreciation was reflected in a commitment to markets over almost any system of hierarchy or command. It is hardly surprising that this commitment was inimical to the idea and practice of bureaucracy -- or almost any other kind of organization or regulation, for that matter.

Public Administration's drift away from economics was interrupted temporarily by the flurry of excitement generated during the 1960s and early 1970s by program budgeting and systems analysis, the former grounded primarily in applied economics and the latter in operations research and management science, both rational choice disciplines. This period witnessed the publication of the work of Charles Hitch and Roland McKean [Hitch & McKean, 1960; Hitch, 1965; McKean, 1968], Jesse Burkhead and Jerry Miner's landmark text, Public Expenditure [1971], which focused on the supply and demand for governmentally provided goods and services, and other significant works concerned primarily with questions of resource allocation in the public sector -- mostly defense [Novick, 1965; Enke, 1967; Niskanen, 1967; Quade & Boucher, 1968; Fisher, 1970], but other areas as well [Schultze, 1968; Haveman & Margolis, 1970; Hirsch, 1970; Margolis, 1970; Williamson, 1970; Merewitz & Sosnick, 1971; Rivlin, 1971 & 1972; Quade, 1975]. This body of literature represents a mother lode of scholarship that still has not been adequately incorporated into public administration. At the same time, the topics raised by Hitch and McKean and their contemporaries have since been largely abandoned by economists and now find only occasional mention in standard textbooks in public finance/economics.

Arguably, the defining moment for the relationship between public administration and economics came at the Minnowbrook conference in 1970. The conference proceedings, which announced the New Public Administration [Marini, 1971], included a respectful comment on PPBS [Planning Programming Budgeting Systems -- an arrangement combining program budgeting and systems analysis (Parker, 1971)], but the rational choice theorists in attendance were roughly treated by the assembly and some claim virtually expelled from it. Their comments and Steven Bram's invited paper were subsequently deleted from the conference proceedings [Marini, 1971: 5].

Ultimately, the gulf between public administration and economics created a niche that two distinct intellectual hybrids vied to fill. The first of these hybrids involved the creation in 1965 of what has since come to be known as the Public Choice Society. The participants at its first two meetings included James Buchanan, Gordon Tullock, John Rawls, William Riker, Vincent Ostrom, Toby Davis, James Coleman, and Charles Plott. Public choice involves the application of economic logic -- methodological individualism and rational, self-interested decision making -- to questions and issues that had traditionally been the concern of political scientists and public administrationists; it has been one the great success stories of modern social and economic science.

The second of these hybrids involved the establishment of schools of public policy at some America most prestigious universities: Harvard, Chicago, UC Berkeley, Duke, Carnegie-Mellon, etc. These schools placed the rational choice disciplines of economics and operations research/management science at the center of their curricula. They continue to benefit from the halo effect produced by their location at elite universities, but most now acknowledge that their defining mission -- training policy analysts -- was fundamentally misconceived. On balance, they have not been intellectual successes and their faculty members have generally sought intellectual refuge in their home disciplines.

Given this history of continental drift, it is surprising, indeed, to find Zhiyong Lan and David Rosenbloom asserting in a recent editorial published in Public Administration Review[1992] that the discipline of public administration is undergoing a paradigm shift and that a rational-choice, economics-based paradigm has emerged preeminent.

PUBLIC ADMINISTRATION IS NOT ECONOMICS

Of course, Lan and Rosenbloom are wrong. The discipline of public administration may be undergoing a paradigm shift, but a rational-choice, economics-based paradigm has not emerged preeminent, at least not so far. There are at least three reasons why public administration cannot be economics [Zorn, 1989].

Public Administration is Prescriptive

First, there is the basic difference between engineering and science [Behn, 1996]. Public administration is concerned with prescription -- the identification of normative rules for decision makers that would lead them to make decisions that are optimal from the standpoint of the citizenry as a whole. Economics is concerned with prediction -- the identification of rules decision makers are likely to follow, given their incentives. Bluntly put, public administrators solve problems; economists explain choices.

Economic theory is useful to public administrators when it provides them with concepts they can use to diagnose problems accurately and to prescribe effective solutions to those problems -- i.e., concepts like opportunity costs, incentives, or capitalization that can be profitably applied to an array of problems frequently encountered by public administrators. But real-world problem solving also frequently raises questions of value and of right and wrong. Economic logic recognizes no good but efficiency, no evil aside from inefficiency [Box, 1992; Hood, 1995]. Morality ought to play an important role in the conduct of the public's business; economists often have trouble accounting for this simple fact.

Public Administration Is Realistic, Empirically Grounded, and Practical

Second, economics is an a priori, theoretical discipline; public administration is concerned with "pragmatic reform." Economists build elegant, logically consistent deductive models; public administrators deal with messy, real-world problems. Indeed, it can be argued that economists prefer rational choice theories to models that incorporate bounded rationality primarily because they are conclusive, not because they are right. Decision makers can be approximately rational in a nearly infinite number of ways; they can be rational in only one.

This difference between economists and public administrators is illustrated by the way they deal with the problem of voluntary provision of collective goods. Economists define a collective [or public] good in terms of two properties: jointness of supply and impossibility of exclusion. This means that once a collective good is supplied by some of the members of a group, it may be enjoyed by all. From this premise they deduce that the decision of some of the members of a group to provide the good or some quantity of it for themselves presents each of the other members with an opportunity for strategic behavior. Since the other members of the group can profitably engage in strategic behavior, economists conclude that they will. If the other members of the group can share in the good regardless of their contributions, economists predict that they will withhold or reduce their own contributions to its provision. Hence, the decision by some of the members of a group to supply a quantity of a collective good leads other members to "free-ride" on their contributions -- which is to say that, if contributions are voluntary, collective goods will be under provided or, in the extreme, not be provided at all [see Breton, 1989].

Crime control aptly illustrates this situation. Citizens can affect the level of crime in their community in two ways: by limiting their exposure to risk and protecting their property, and by helping the police fight crime. If their possessions or personal security matter enough to them, individuals may see a direct material benefit from investing in locks, guns, guard dogs, or security systems. They may even see the benefit from participating in volunteer citizen block-watches or banding together to patrol their own streets or financing a private security force to do so. The authorities can encourage these kinds of activities by providing guidance and technical assistance, by passing out police whistles, by urging people to mark their property so that it can be more easily identified when stolen, by helping to organize block watches, by setting up emergency call systems tied to rapid response, and by positioning themselves to provide back-up to private efforts.

However, self-defense alone will not control crime. Criminals must be identified, apprehended, and convicted. The police necessarily depend upon the citizenry to alert them to crime and to aid them in the conviction of criminals. Unfortunately, only where their safety or that of their loved ones is at risk or where their property is threatened will private citizens realize a direct benefit from intervening to stop in crimes in progress. And even where they have been personally victimized, citizens rarely individually benefit from helping the police identify, apprehend, and convict their assailants, since the harm has already been done and the criminal justice system seldom provides restitution. In these instances, the behavior expected of citizens, though of great value to their community, is not personally rewarding in an obvious way. Hence, individuals often shirk these onerous civic responsibilities, trying to free-ride on the efforts of their neighbors.

Many economists insist that, in the presence of collective goods, citizens must be coerced to perform their civic responsibilities [Breton, 1989]; otherwise jointly provided services will necessarily be undersupplied. This mind set reflects, in part, the propensity of economists to confuse a perfectly useful analytic construct, economic man, with living, breathing humans. Economic man is a rational fool. Given the opportunity to ride free, he will. Since economic man will not voluntarily cooperate with his neighbors to provide public or collective goods, he must be forced to do so. This mind set also reflects a propensity to overlook the vitality of human ingenuity in designing social arrangements and to ignore the availability of motivational alternatives to coercion.

In contrast, public administrators recognize that citizens often free-ride on the efforts of their neighbors, but they interpret this as problem to be solved rather than a necessary fact of life. Beekeeping provides the classic example of the failure to distinguish between economic theory and reality. Once upon a time, economists taught that beekeeping is a collective good and that, since fruit growers can rely on their neighbors' bees to pollinate their blossoms, most growers will. Hence, beekeeping must be undersupplied.

S.N.S. Cheung, then an assistant professor of economics at the University of Washington, did something that was rather extraordinary: he left his armchair to find out whether beekeeping is actually undersupplied [Cheung, 1983]. As the result of a careful study of beekeeping and apple-growing practices in Washington State, Cheung found a long history of contractual relationships between apple growers and beekeepers. These contracts provided for beekeepers to be compensated for their contribution to the growers' apple crops. He also found that apple growers implicitly covenanted with their neighbors to keep the same ratio of bees to trees. Apple growers who did not abide by the covenant were ostracized and treated to inconveniences by those who did. Consequently, Cheung concluded that free riding may not be a serious problem among real apple growers and beekeepers [see also Ostrom. 1990; Ostrom, et al., 1994, Breton, 1989].

The primary point that Cheung was trying to make is that real people are not rational fools. They often do contribute voluntarily to the provision of collective goods [Ostrom. 1990]. Furthermore, social conventions or group norms can discourage free-riding and reduce shirking. These conventions can take the form of ethical precepts, regularities imposed by institutions, or simply fixed rules of thumb for individual behavior. Moreover, group norms can be collectively enforced through the ostracism of those who fail to contribute and praise for those who do.

A second lesson to be drawn from the fable of the bees is that neither community norms nor the collective enforcement of those norms just happened. Knowledge of what to do and how to do it was provided by field agents of the United States Department of Agriculture's [USDA] Extension Service. This information gave apple growers a solid technical basis for group norms governing the behavior of individual growers. The field agents determined how many hives were needed and fairly apportioned responsibility for their provision. They also played a role in monitoring compliance with group norms and in passing that information along to growers. In so doing, the field agents could identify shirkers and the subsequent shortfall in the provision of bees that had to be made good by the rest of the community. Moreover, the USDA provided growers with a powerful collective sanction against free riders in the form of marketing orders and quota. Free riders were not merely subject to social ostracism; the other growers could have actually denied them access to the most lucrative markets.

The point is that voluntary contributions to the provision of public goods don't just spontaneously occur; opportunities for collectively beneficial action must be identified, individual contributions established, performance monitored, and defectors sanctioned [see Heckathorn and Maser, 1989; Maser, 1986]. This is, of course, the last and most important lesson of the fable of the bees: voluntary provision can be organized and must be managed. Because management implies a manager, it follows that someone, usually a public official, must be charged with mobilizing the community on the behalf of the public good, organizing provision of the good, creating incentives, and supervising enforcement of community norms [Powers & Thompson, 1994].

Public Administrators are Preoccupied with Technical Efficiency

There is a third reason why public administration is not economics. As a normative discipline, public administration is preoccupied with identifying decision rules that citizens would unanimously support. In practice this means that, just as economists don't like to make value judgments, public administrators are usually more comfortable condemning technical than allocative efficiency. Technical inefficiency means that managers fail to minimize cost or maximize output because they aren't using the best available technology. Technology means not only plant and equipment, but also the methods used to coordinate activities and to motivate performance. Best available, means in practice, not merely in theory.

A comparison of how Ford Motor Co. accounts for its purchases with how the Navy handles accounts payable illustrates technical inefficiency. Not so long ago, Ford cut the required number of manual accounting transactions to pay for goods from nine to three, permitting a 75 percent staff reduction in its accounts payable department [Hammer, 1990: 104, 107]. In contrast, it takes the Navy twenty-six manual accounting transactions and nine reconciliations -- thirty-five steps in all -- to process and pay for things [Hemingway, 1993: 8-12]. This system is not only time consuming, it often leads to bad service and excessive investment in inventories. According to the National Performance Review, it causes delays in obtaining repair parts that keep a high proportion of the Navy's cars and trucks out of commission and forces the taxpayer to fund ten percent more vehicles than the Navy really needs [Gore 1993: 12].

Computerization could eliminate more than half of the steps in the Navy's accounts payable process [Hemingway, 1993: 25]. But why are fourteen, let alone thirty-five, accounting records needed where Ford gets by with three? One answer is that Navy fails to capture information once and at the source. Instead, each step in the supply process -- requisition, receipt, certification of invoice, reconciliation, and revision -- is repeated at every level of the organization. Moreover, the people who produce information do not process it. Processing is handled by financial management specialists from the bottom of the organization at the top. Finally, the Navy does not build financial control into its job designs. Naval officers have little discretion as to the mix or quantity of resources used by their commands. Even in peacetime their effectiveness in managing resources often has little or no bearing on the evaluation of their performance.

Until recently, most economists assumed that technical efficiency was someone else's concern: engineers, maybe accountants, organization theorists, or even public administrationists. Nowadays they understand that technical inefficiency is often far more important than allocative inefficiency, but they tend to explain it in terms of structural or other factors [e.g., lack of competitive pressure] that are beyond the control of managers. While the efforts of economists to understand managerial failure has in recent years produced some powerful new theories and concepts that can that can help public administrators deal with a variety of problems [see below], economists still tend to overlook the most common cause of technical inefficiency: ignorance. Ignorance is Protean. As they say in the Navy, there are only a few ways to do right, there are an infinite number of ways to screw up. For example, the Navy's Byzantine system of accounting is arguably an unintended consequence of the Anti-Deficiency Act [33 U.S.C. §1214, 1257 (1905)]. The system was designed to insure that neither local commanders nor higher level authorities exceeded the obligational authority granted them by Congress. That authority is now divided into fifty separate accounts, 557 management codes, and 1,769 accounting lines.

At the same time it should, perhaps, be acknowledged that a preoccupation with technical efficiency leads public administration to slight allocative efficiency. Allocative efficiency has to do with matching supply to demand. It is of special concern to economists, who object to private monopolies, for example, not because their prices are too high, but because they produce less than they would under competition. Why is this is bad? It is bad, even where production is technically efficient, because consumers would willingly pay more for the things that aren't being produced than it would cost to make them. Hence, net benefits are forfeit [net benefits = willingness and ability to pay - cost > 0]. Economists refer to foregone net benefits as "deadweight loss."

It is not hard to find instances of allocative inefficiency in government or to identify some fairly common pathologies that induce it. Economics teaches, for example, that, in the presence of a capital market where funds can be obtained at a price, the welfare of the citizenry will be maximized by the implementation of all projects offering positive net present values. This means that the timing of benefit/cost flows usually does not matter so long as future benefits and costs are properly discounted. But governments often appear to be obsessed with the timing of outlays. They give too much weight to current costs and benefits and too little to future ones. Consequently, they often put off investment programs that would produce net benefits or they stretch out programs, thereby increasing their costs, in order to comply with arbitrary spending constraints.

For example, over the past two decades, the Department of Defense has reduced production rates for a variety of weapons systems below minimum optimal scale. In several instances this has had the result of increasing the present value costs of the total production run by more than 100 percent. Why? To reduce the annual deficit. In this case, the federal government was willing to trade huge future liabilities for small current reductions in the growth of the national debt, the interest upon which is now running about six percent per year.

Government's propensity to disregard questions of feasibility, especially administrative and economic feasibility, also often results in allocative inefficiency. For example, John Mendeloff [1988] has shown how this propensity leads OSHA to over-regulate and how overregulation leads to underregulation. Both over- and underregulation are examples of allocative inefficiency. According to Mendeloff, OSHA overregulates exposures to harmful chemicals in the workplace because it often sets standards without regard to the benefits and costs of reducing hazards to those levels [pp. 155-185]. John F. Morrall III [1986] estimates that as a consequence OSHA's proposed standards impose costs on industry that typically exceed benefits by a factor of ten. In the case of its proposed formaldehyde standard, costs were 25,000 times greater than benefits. Not surprisingly, these costs have inspired industry to embrace every legal, administrative, or political measure that might conceivably prevent, delay, or overturn promulgation of new standards. Consequently, nearly all of OSHA's discretionary resources have been absorbed defending standards that govern a mere handful of hazardous substances. The opportunity cost of overregulation can, therefore, be seen in the workers who are exposed to hundreds of underregulated substances because the resources needed to revise standards are simply not available.

Government also has a propensity to disregard the incentive effects of the prices it sets. This necessarily results in allocative inefficiency. Fortunately, this propensity seems to be waning. Nevertheless, some people still have trouble seeing a relationship between low grazing fees on BLM land and overgrazing, between low prices for agricultural water and wasteful farm irrigation practices or urban water scarcity, between low timber harvest fees and overlogging, between the low landing fees charged private planes and airport overcrowding, or between agricultural price supports and food surpluses.

Public administrators ignore allocative efficiency at their peril. A preoccupation with technical efficiency, like poverty, may be "no sin, but it is no great honor either."

THE RENEWAL OF INTEREST IN ECONOMICS

Of course, observers as perceptive as Lan and Rosenbloom could not have concluded that economics has emerged as public administration's preeminent paradigm unless their claim held a kernel of truth. A paradigm shift seems to be in the works; at least, many in the field now reject the traditional bureaucratic paradigm. Moreover, economics seems much more directly relevant to the concerns of public administration than in the recent past. There are three not entirely unrelated reasons for these changes: changing styles in political science, changes in the environment of public administration, and advances in economic science. I will deal with the first of these two issues briefly. The remainder of this essay will focus on the last.

Political Science

It is, perhaps, not too strong to say that a rational-choice, economics-based paradigm has emerged preeminent in American political science, including bureaucracy and public policy, subfields that are closely related to public administration. In my opinion this is a healthy turn of events. An unbiased observer would have to acknowledge, however, that political science, like most of the humanities and social sciences, is prone to academic fads. They come and they go, often leaving little or nothing behind in the way of accumulated knowledge. It is natural that we think of ourselves as the tip of progress's arrow, but intellectual history demands a more humble interpretation. Just as academics of past generations usually seem wrong-headed to us, so too are we likely to appear to the next. Nevertheless, for good or ill, when political science sneezes, public administration more often than not catches a cold. Political science has sneezed.

Changes in the Environment of Public Administration

Public administration in the United States has also been influenced by the "new public management" [Gore, 1993]. The new public management emphasizes "performance appraisal and efficiency; the disaggregation of public bureaucracies into agencies which deal with each other on a user-pay basis; the use of quasi-markets and contracting out to foster competition; cost-cutting; and a style of management which emphasizes amongst other things, output targets, limited term contracts, monetary targets and incentives, and freedom to manage" [Rhodes, 1991: 11; Dunleavy & Hood, 1994].

The new public management is a worldwide movement [Rhodes, 1991; see also Hood, 1991; Schedler, 1995; Osborne and Gaebler, 1992; Barzelay, 1992]. Arguably, it represents a paradigm shift in public administration. Herman Schwartz [1994], for example, claims that government is undergoing "a profound shift toward a new kind of regime .... not simply a shift towards less state, but also a shift to a different kind of state." He attributes this shift to international market pressures. He stresses that many of the governments that have embraced the new public management are or were dominated by social democrats. New Zealand, which under Labour governments has gone further than any other country in its embrace of the new public management, is a prominent example [see below].

The driving force behind the new public management is technological change. Reductions in information costs brought about by computers and computer networks and our increased capacity to use them have caused four major shifts in the comparative advantage of governance mechanisms and institutional arrangements. These are [see Reschenthaler and Thompson, forthcoming]:

1. The efficacy of the market has increased relative to government provision and control;
2. The efficacy of the market and other self-organizing systems has increased relative to hierarchically coordinated systems;
3. The efficacy of decentralized allocation of resources and after-the-fact control has increased relative to centralized allocation and before-the-fact control; and
4. The efficacy of process-oriented structures has increased relative to functional structures.

These changes are hardly surprising. As will be explained below, the comparative advantage of any institutional arrangement boils down to a question of information or transaction costs. Changes in information costs should and have dramatically altered the relative advantage of governance mechanisms and institutional arrangements.

Large organizations, for example, were once justified by economies of scale and scope. Economies of scale are produced by spreading fixed expenses, especially investments in plant and equipment and the organization of production lines, over larger volumes of output, thereby reducing unit costs. Economies of scope are produced by exploiting the division of labor -- sequentially combining highly specialized functional units in multifarious ways to produce a variety of products. Economies of scale and scope were made possible by hierarchy and bureaucracy, which broke tasks down into their simplest component parts and recombined them to produce complex goods and services, allocated scarce resources to administrative units, and established product-market strategies.

In turn, hierarchy and bureaucracy were made possible by innovations in organizational design, administrative controls, and operational engineering [Chandler, 1962]. As Nathan Rosenberg and L.E. Birdsall [1986: 235-37] explain, most of the entrepreneurs of the Industrial Revolution were merchants and financiers -- they knew little or nothing about production. Business did not learn how to organize and supervise large numbers of workers until the mid 19th century. As late as 1892, for example, Carnegie Steel avoided the problem of organizing and managing the work of its production employees. "The organization of the work of large numbers of employees was a new management function, and direct employment could not have become a general practice until recruiting, organizing, and supervising factory workers been sorted out and fitted into a hierarchical scheme." Of course, only very large organizations could take full advantage of the bureaucratic revolution. Only they could be completely vertically integrated or afford to devote substantial amounts of resources to gathering and processing quantities of data for top management to use to coordinate activities, allocate resources, and set strategy -- these are, after all, fixed costs; they contribute nothing directly to output.

The computer is rapidly eroding economies of scale in administration, production, and marketing and, thereby, the comparative advantage of hierarchy and bureaucracy. Today, any organization that can afford a computer workstation and software [about $20,000] can have first-class administrative systems, ranging from purchasing and inventory control to human resources management to financial planning and capital budgeting to marketing and logistics. Twenty years ago these systems were available only to giant organizations. Moreover, computerized production [which consists of machine tools or other equipment for fabrication, assembly or treatment, linked by a materials handling system to move parts from one work station to another, and operating as an integrated system under full programmable control] now permits organizations to produce customized services at mass-production prices.

In computerized production facilities, overheads are more important than production volume. In these facilities, direct manufacturing labor often accounts for less than 5 percent of costs; materials and purchased components typically account for thirty to forty percent more. This leaves at least fifty-five percent for overheads. Most overheads are transaction or information costs. They involve activities like purchasing, materials handling, marketing, accounting, and asset utilization. They are driven by an organization's policies, its operating and administrative procedures, and its customer relationships -- not output volume, rate, or even mix.

To control overheads, including the costs of holding materials, parts, and finished goods inventories, many organizations have adopted techniques like lean manufacturing and just-in-time delivery of parts and materials. They have also modified their managerial cost accounting systems to focus the attention of responsibility center managers, marketing and manufacturing teams, and especially product designers and engineers on controlling overheads. Two tools have been critical to this effort: cycle-time burdening and transaction cost accounting [activity-based costing]. Again, not surprisingly, this is feasible because technology generates information about underlying productive processes. Computer-assisted-design programs also produce cycle-time and transactions costs estimates. Universal product codes and optical scanning devices permit continuous monitoring and, therefore, real-time reprogramming of the production process.

As a result of the declining importance of economies of scale in production, the average size of the workplace has been falling throughout the industrialized world for the last twenty years [Economist, June 24, 1995: 4-6 Survey]. Large companies are "mimicking their smaller competitors by shrinking their head offices, removing layers of bureaucracy and breaking themselves up into constellations of profit centers, ... they are 'sticking to their knitting' -- concentrating on their core businesses and contracting everything else out, ... [and] they are putting a computer on every desk and giving power to front-line workers." As Shoshana Zuboff explains [1988: 204], efficient operations in the modern workplace call for a more equal distribution of knowledge, authority, and responsibility. To create value from information, members of the organization must be given the opportunity to know more and do more. This means "dismantling the very same managerial hierarchy that once brought greatness."

All of this looks like economics to many noneconomists, but it is really not. The intellectual justification for these changes comes primarily from management thinkers such as Peter Drucker, Theodore Levitt, Thomas Peters, Joseph Bower, Robert Anthony, Robert Kaplan and Robin Cooper, Henry Mintzberg, Alfred Chandler, Kenichi Ohmae, C.K. Prahaladad and Gary Hamel, and Peter M. Senge. The informational and organizational tools used in making these changes are described in the Harvard Business Review, Sloan Management Review, and Management Accounting, not the American Economic Review or even the Journal of Economic Behavior and Organization. Many public administrators now read these journals and have tried to use the new tools in their organizations -- hence the new public management.

Borrowing from the business-management literature is an old custom, as well as a recent phenomenon for the discipline of public administration. Business administration and public administration are both prescriptive, pragmatic disciplines [Hammond, 1990; Fitch, 1990]. Moreover, business schools and schools of public administration once shared the same proverbs of administration, just as we shared the Weberian bureaucratic paradigm and a common intellectual foundation in the works of Chester Barnard, Henri Fayol, Mary Parker Follett, Luther Gulick, Phillip Selznick, Frederick Taylor, and others. We both taught that bureaucracy is the solution to the "problem of maximizing organizational efficiency," that bigger is better, that organizations should be functionally differentiated and vertically integrated, and that top management always knows best. Besides, large organizations in both the public and the private sectors were fundamentally alike. Most were hierarchies; most distinguished between high-level management tasks [planning, organizing, staffing, and developing] and low level management tasks [controlling, operating, reporting, and budgeting]; and most centralized resource allocation and staff services.

Now changes are taking place in what managers do and how organizations are put together -- even in what they are. The new public managers are interested in the way business does things, not because it is better than government, but because they face similar problems and the business-management literature is full of ideas that seem relevant to meeting these new challenges.

Advances in Economics

While the business-management literature is central to the new public management, two bodies of economic literature have also profoundly influenced its reception and its implementation: public choice theory and the new economics of organization. Public choice theory has changed the way we think about government and how it works. Moreover, in explaining the rules that voters, elected officials, and bureaucrats are likely to follow given their incentives, public choice theory has given public administrators some useful new normative information. After all, in many cases, a good normative model is merely a good positive model run backwards. Nevertheless, when public administrators look to advances in economic science for help, it is not primarily to the public choice literature that they turn to, but the new economics of organization.

The new economics of organization focuses on incentive and control structures and on the allocation of property rights and the ownership so as to minimize intraorganizational externalities or spillovers. It comprehends concepts like the Coase Theorem, transaction costs, externality, asymmetric information -- including agency theory, moral hazard, and adverse selection, contract theory, incomplete contracts, implicit contracts, incentive contracts, search and signaling theory, team theory, and incentive compatibility that are directly relevant to managerial problems. It is the legacy of economists like Kenneth Arrow, William Baumol, Albert Breton, Ron Wintrobe, Harold Demsetz, Victor Goldberg, Michael Jensen, Paul Milgrom, William Niskanen, Gordon Tullock and especially Ronald Coase and Oliver Williamson. It provides the new public management with the solid analytical foundation needed to understand how, when, and where to delegate authority, replace rules and regulations with incentives, develop budgets based upon results, expose operations to competition, search for market rather than administrative solutions, or use quasi-markets and contracting out to foster competition.

The economics of organization has already influenced the design of a variety of institutional arrangements [ranging from emissions trading and "bubbles" to outright deregulation of the airlines and interstate trucking] in the United States and the privatization and securitization of an astonishing array of government-owned assets [and some liabilities] in Europe. Moreover, the evidence is accumulating that these arrangements work. It is partly because of this evidence that the ideas of the new public managers command the attention they do.

PUBLIC CHOICE THEORY

Public choice theory, like the older normative theory of public finance from which it evolved, starts with the demand for and the supply of collectively provided goods and services. With two exceptions, the theory of demand for a collectively provided good is identical to the theory of consumer demand for a private good. In both instances, demand reflects individual willingness and ability to pay to consume a good or service. Total demand for the service is, therefore, assumed to be a decreasing function of the price of the good, an increasing function of consumer income, and the size of the market for the good. The two differences between the theory of demand for a collectively provided good and the theory of consumer demand for a private good are that the quantity of service provided within a jurisdiction is determined by a political process, usually assumed to be some form of majority rule, and is necessarily uniform throughout the jurisdiction.

Hence, the quantity demanded of a collectively provided service will depend upon its price [P], the permanent income of the citizenry [Y], and population size [C], i.e.:

Di = f [Pe, Ya, Cb]

where:

e = the price elasticity of demand for good i;
a = the income elasticity of demand for good i; and
b = a value from 0 to 1, representing the degree of publicness of good i.

[This formulation is outlined in Borcherding, 1977; see also Bergstrom & Goodman, 1972]. The functional relationship that would obtain in any particular case would, of course, reflect existing political institutions as well as culturally mediated tastes and preferences [Breton, 1989]. There is a fair amount of evidence that this model of voter demand for collectively provided goods works reasonably well in practice. Literally hundreds of studies have demonstrated that family/per capita income, tax price, community size, and population served better explain cross-sectional variations in collectively supplied service levels that any other set of "determinants" [Oates, 1994; e.g., Duncombe, 1991; Romer, 1992]. Of course these variables work best where service levels are determined by direct referenda [Holcombe, 1977, 1986, & 1989; Inman, 1978; Romer & Rosenthal, 1982]. The one bug in the ointment is the "flypaper effect," so called because intergovernmental transfers tend to stick where they land [Oates, 1994; Rubinfeld, 1994; Ladd, 1994]. That is: intergovernmental grants appear to produce far larger increases in the output of government services than predicted by the income and price elasticities of the basic demand model.

The Median Voter and Bowen Equilibrium

In democracy, Di should reflect the tastes and elasticities of the median voter [Bowen, 1943] and would be in equilibrium where:

Vm = Tm

where:

Vm = the marginal benefit of good or service i to the median voter m
Tm = the marginal cost of good or service i to the median voter m

Then, where the cost of the good is equally shared by all voters [n], it follows that the marginal cost of the good to all voters is:

 

Since the Samuelsonian efficiency condition [1954, 1955], where the sum of the marginal costs of providing the collective good equals the sum of marginal benefits to all contributors, is:

 

it follows, that, for a Bowen equilibrium to be efficient, the marginal cost/benefit to the average voter would have to be equal to the marginal benefit/cost to the median voter, i.e., the following condition would have to be satisfied:

Consequently a Bowen equilibrium will be efficient if and only if marginal net benefits sum to zero at the output or supply level preferred by the median voter. This is of course something of a "knife-edged condition," except where the tastes, willingness, and ability to pay of all the voters in a jurisdiction are practically identical. One mechanism that might lead to the satisfaction of this condition is Tiebout sorting [named after the late Charles M. Tiebout, 1956] in which voters move to the communities whose governments best satisfy their preferences for collectively provided goods and services [Zax, 1989; Deller, 1990a, 1990b; Taylor, 1991; see, however, Hoyt, 1990]. While most economists grant that Tiebout sorting takes place, there is little consensus as to its significance. William Fishel [1992], for example, claims that the evidence from zoning and voting demonstrates that it is fairly complete; John Yinger, Howard S. Bloom, Axel Borsch-Supan, and Helen F. Ladd [1988], however, claim that the evidence from tax-capitalization is equivocal.

This is a significant claim, because, at the local level, to the extant that service and tax levels are capitalized in real property values, reliance on the property tax reduces differences between the mean and the median voter even in the absence of Tiebout sorting. Reliance on proportional or progressive income taxes probably has a similar effect at state and, perhaps, federal levels of government. The key point to be stressed here is, as Bowen explained [his conclusion was later popularized by Downs, 1959-60], the average voter's demand for collectively provided goods will normally exceed the preferred consumption level of the median voter. This means that in a democracy, where costs are equally shared or, perhaps, even where taxes are proportional to income, collectively provided goods will tend to be undersupplied.

There is evidence that Bowen/Downs undersupply does occur. For, example, Fabio Silva and Jon Sonstelie [1995] recently tested William Fischel's [1989, 1995] hypothesis that, by requiring equal spending per pupil across all school districts in the state and, thereby, reducing Tiebout sorting and widening the gap between the preferences of the average and the median voter, the California Supreme Court in Serrano v. Priest caused a reduction in public spending to support elementary and secondary education. Before Serrano, California ranked 11th among states in public school spending per pupil, 13 percent above the average of all other states. By 1990, California had fallen to 30th, ten percent below the average. Silva and Sonstelie found that one-half of this decline could be attributed to Serrano. They attributed the remainder to rapid enrollment growth during the 1980s.

Lindahl Equilibrium

There is a second condition whereby the knife-edged condition described earlier might obtain, at least in theory: where decisions about collective provision are made unanimously. Unanimity can be satisfied under Lindahl equilibrium [Feldman, l980], a special case of the Samuelsonian efficiency condition. Formally, Lindahl equilibrium is defined as a vector of expenditure shares [S1,S2,...,Sn] and a level of provision [D*] such that for all i, where i's expenditure share is Si, the quantity of the good desired is D* [i.e., because i's expenditure or Ti is SiD*, …Si = 1]. Furthermore:

Under this formulation, demand for all i is presumed to be normal, i.e., the higher i's share Si, the lower the D, i will want. Hence, if the Lindahl equilibrium were D1 and one i reduced their contribution to the provision of the good, the share of all other i would increase and D1 > D2.

This voting rule is seldom used, however, because of the transactions costs associated with finding Lindahl equilibrium -- search, bargaining, and monitoring costs [Heckathorn & Maser, 1989; Maser, 1986]. Because of this fact, James Buchanan and Gordon Tullock [1962] suggest that an optimal constitution would minimize the sum of the costs imposed upon dissatisfied minorities, a decreasing function of the size of required majorities, and transactions costs, an increasing of function. Indeed, Americans have tried to design a constitutional order with this goal, albeit not this formulation, in mind. The founding fathers designed the federal Constitution to protect lives, liberties, and property from the overweening ambitions of a reckless and extravagant executive, the passions of minorities, and the interests of narrow or temporary majorities. This structure was intended to insure that the United States would have a government of law and that the law would change slowly and incrementally and only when the direction of change was endorsed by a large majority of the citizenry. The founding fathers relied upon partisanship, jealousy between the two chambers of Congress, and the particularities of committee interests to make the federal government an agent of negotiation and compromise that would reach its decisions through the discovery of a lowest common denominator [Thompson, 1979].

Legislative Decision Making

Representation promotes efficiency in two other ways. First of all it minimizes the transactions costs of political participation [economists see participation as cost, not a benefit]. Second, it transforms some public goods, where each individual voter's preferences are known only to himself or herself and costs are shared, into private goods, where benefits and costs are denominated in dollars and knowledge of the net-benefit schedule of each voter/legislator is equally available to all [Shepsle & Weingast, 1984]. Once the public goods problem is laid to rest, efficiency should be easy to arrive at. According to the Coase Theorem [Coase, 1960], where we are dealing with private goods and decisions are made one good at a time, any decision rule will produce an efficient outcome -- except where prevented by restrictions on side payments, transactions costs, or just plain ignorance [Dahlman, 1979].

Consider, for example, the following problem: there are three legislators: X, M, and N. Each represents a smog producing district; however, X's constituents export all of their smog to M's; M's export all theirs to N's; and N's keep all the smog they produce, plus that which they import from X's and M's constituents. The problem that X, M, & N must deal with is underinvestment in maintenance of automotive smog control devices. This is a real market failure: smog control devices work properly only when they are properly maintained; the cost of maintaining this equipment is very low, at least compared to other equally effective means of abating pollution. Unfortunately, since individuals bear the cost of maintenance, but the benefit thereby produced accrues primarily to others; they have no incentive to do so. The result is underinvestment.

Compulsory inspection and maintenance [I&M;] is one way to deal with this market failure. It is an attractive example for our purposes for a number of reasons, one of which is that it can be assumed to impose roughly equal costs on voters in each constituency. Figure 1 illustrates the political calculus associated with a typical compulsory I&M; option. In this example the efficient solution is at xopt, where the vertical sum of Vx, Vm and Vn equals MC. But since each legislator prefers the I&M; option that equates T and V for the median voter in their constituencies: zero for X, xmed for M, and xhi for N. This example also clearly illustrates the typical gap between xmed and xopt -- although, in this case, xmed is closer in xopt than zero [see also Nelson, 1990; Bell, 1989].

[Figure 1 goes about here]

What happens when this problem is approached in a Coasian manner, searching at the margin for Pareto superior moves? For example, I&M; does not have to be compulsory to be effective. Automobile operators could be bribed to have their vehicles inspected and to achieve high maintenance scores at inspection. These bribes could be financed by a property tax levied on N's district. Under these circumstances, the situation outlined in Figure 2 would obtain. [Figure 3 shows the same circumstances in preference space and demonstrates that the efficient solution obtained in Figure 2 is also the Bowen Equilibrium.]

[Figures 2 and 3 about here]

Note that this solution has the attractive quality of making all three legislators better off than they would have been under the solution shown in Figure 1, but it is only one of possibly many Coasian solutions to the problem of underinvestment in the maintenance of smog control equipment. The Coase theorem merely states that, where decision makers are not constrained in their search for a solution, the solution they choose will be efficient.

Of course, real legislators do not always behave like Coasian paragons. Simple human fallibility goes a long way toward explaining this fact. Information on the incidence of benefits and costs of a proposal may be equally available to all, but it may also be available to none. It is easy to conceptualize a Coasian solution to a problem of market failure; it is frequently far harder to design a practical program to implement one [e.g., Hahn and Noll, 1982]. Hence, it is not surprising that legislators fail to behave like Coasian paragons. Like the rest of us, individually and, consequently, collectively, they make mistakes.

Explaining Oversupply

The basic model for a collectively provided good implies the likelihood of undersupply. To many public choice theorists this likelihood appears to fly in the face of reality. What they want to do is explain what they observe: an excess of government [Coughlin, 1991]. Two serious attempts to explain government excess have been mounted and, while neither entirely succeeds, both have taught us something about government. These are the interest-group theory of politics and William Niskanen's theory of spending coalitions, which introduced the notions of structurally induced equilibrium and agency theory to the study of politics and bureaucracy.

The economic theory of interest-group politics is largely the creation of the "Chicago school" of regulatory theorists: George Stigler [1971], Richard Posner [1971, 1974], and Sam Peltzman [1976]. They argued that a variety of government programs [agricultural subsidies, military procurement, tariffs and import quotas, most regulation of business, and the structure of the tax system] are the product of an exchange between elected officials, who receive votes and campaign contributions, and members of groups, who reap higher incomes from their political investment. The theory predicts that politicians will use their power to transfer income from those with less political power to those with more: from the rich and the poor to the middle class and from disorganized, diffuse interests to well-organized concentrated interests [e.g., geographically-targeted benefits and predominantly federally-financed costs (Ferejohn, 1974; Shepsle & Weingast, 1981; Hird, 1991)]. Who gets what depends upon the costs and benefits to individual group members of participating in political processes and the ability of groups to influence policy makers.

Of course, political scientists have known about interest groups for generations. What was new about the economic theory of interest-group politics? First, economists better understood the costs and benefits of government actions and their distributional consequences, which focused attention on the size of individual payoffs rather than the wealth of the player.

Second, economists brought Mancur Olson's theory of collective action to bear upon the question of interest group influence on the public policy process. Olson demonstrated [Olson, 1965; Olson & Zeckhauser, 1966] that political participation [interest-group activity, voting, etc.] imposed private costs upon the participant but tended to create benefits that were nonexclusive, which in turn led to free-riding and the "exploitation of great by the small." This focused attention on public policies that could deny benefits to nonparticipants, the concentration of benefits produced by those policies, and the support thresholds required to claim those benefits. Third, economists based their theory on an assumption that is often wrong, but still might be of considerable utility in partial equilibrium analysis, that elected official are exclusively concerned with maximizing the probability of their reelection, which is an increasing function of interest group support [since citizens won't even vote unless they are persuaded to do so by campaign advertising and workers].

The biggest problem with the economic theory of interest-group politics is that, while it often tells a plausible story about existing public policies, empirical tests of its basic assumptions and predictions seldom work very well, compared say to the presumption that legislators pursue their own notions of the common good [Peltzman, 1984 & 1990; Kalt & Zupan, 1990; Mayer, 1991; Graddy, 1991a&b;]. And, of course, one should be mindful that the economic theory of interest-group politics was created primarily to explain government regulation of prices and entry in industries like trucking and airlines. Both of which have been subsequently deregulated.

The one central claim of the economic theory of interest-group politics that has gone unchallenged is that many of the benefits of government action: tariffs, agricultural marketing orders, import quota, various types of price and entry control regulation, tariffs, pork barrel spending, and the like, do not accrue to their nominal recipients [Buchanan, Tollison, and Tullock, 198]. Instead, the rents created by government action are capitalized in asset prices, especially real property values, or competed away. Government created rents are often competed away because their existence leads to rent-seeking -- behavior aimed at getting or keeping rents. Most of these activities are directly unproductive and are, beyond some point, entirely wasteful.

Gordon Tullock [1967] was probably the first scholar to think systematically about the consequences of rent-seeking. He argued that, if everyone could freely participate in the rush for the spoils, each rent-seeker would expend the full amount of the potential transfer in its pursuit. In which case, the rents created by government would be dissipated by the directly unproductive activities incurred to capture them. Indeed, in an implicit analogy to the effect of product differentiation in the economics of imperfect competition, Tullock hypothesized that the waste involved in capturing rents could actually exceed the rent to be captured. In a related vein, Mancur Olson [1982] argued that, if rents are extensive and efforts to retain them pervasive, the inevitable result is a kind of policy gridlock, which devours ever more resources in defense of the economic status quo, stultify change, and, by diverting investment away from productive activities and inhibiting the process of creative destruction, reduces the rate of economic growth.

Indeed, an outsider cannot help but notice the amounts of money spent on campaigns for public office in this country or the resources employed to influence the legislative, administrative, and judicial processes. What seems remarkable about the American political system is not that it produces more rents than in other countries -- that doesn't seem to be the case [see Krueger 1974] -- but that the creation, maintenance, and distribution of rents attracts [or the gestation and implementation of any policy initiative, for that matter, requires] so much more effort here than elsewhere.

William Niskanen and the Budget Maximizing Bureaucrat

In a second attempt to explain government excess, William Niskanen, Chairman of the Cato Institute and former head of the President's Council of Economic Advisors, [1971; Blais & Dion, 1991] showed that a revenue-maximizing, single-product bureau with absolute monopoly and agenda-setting powers would be technically efficient but produce up to two times the optimal quantity of output. While there are, perhaps [see Carroll, 1989, 1990], too many monopoly bureaus, their agenda-setting powers are often limited, most use a variety of technologies to provide an array of services, and technical inefficiency is widespread. Hence, anyone who leaps from the presumed monopoly power of bureaus to the allocative efficiency of government is undoubtedly over-reaching.

Having said that, one must still recognize the remarkable scientific contribution made by Niskanen. First of all he, more than anyone else, demonstrated that the behavior of government officials, like corporate bureaucrats, could be deduced from their tastes and opportunities and that this approach is more effective than assuming that they are merely well trained robots. Second, Niskanen showed how the ability to control agendas presented to median voters could shift outcomes from their preferred positions. For example, confronted with a choice between nothing and a higher than preferred spending level [i.e. Vm < Tm ], the median voter should prefer the higher spending level as long as total benefits exceeded total costs [ ÚVm > ÚTm]. Of course, this same mechanism in the hands of a different agenda setter could lead to undersupply, but Niskanen argued that in the American congressional system, committees and subcommittees, the effective legislative and budget agenda setters, are likely to be dominated by program advocates [for evidence on this point, see Shepsle, 1976, 1979, 1986; Munger, 1984]. Finally, Niskanen adapted the structure-conduct-performance paradigm from the industrial organization theory to the behavior of government bureaus, demonstrating, among other things, that monopoly supply is a necessary, but insufficient condition for allocative inefficiency. Niskanen's strictures against bureaucratic monopoly have probably had more influence on the theory and practice of public administration than any other idea drawn from the public choice literature [Ezzamel, 1993; Ferris & Graddy, 1991; Mayston, 1993; Miranda, 1994 & 1995; Rogerson, 1990].

Political scientists have embraced Niskanen's notion of structurally induced equilibrium, primarily I think because it appeals to their concern with political institutions and their fascination with games of strategy. There is now an extensive literature on structurally induced equilibrium [see Bendor, 1990; McKelvey, 1976; Shepsle, 1979]. One conclusion that can be drawn from this literature is that, if referenda are not carefully restricted to a single issue dimension, the drafter can manipulate them to produce almost any outcome desired. Another widely-accepted conclusion is that an inability on the part of the legislature to achieve a stable collective choice could be a source of considerable bureaucratic discretion -- even were there no problems measuring bureaucratic performance or designing incentive mechanisms [Hill, 1985; Knott & Miller, 1987]. Of course, could is not the same as is [Langbein, 1996; McCubbins & Schwartz, 1984; McFadden, 1976, 1977].

Both economists and political scientists have also recognized the decisive role played by individually motivated agents in the determination of bureaucratic outcomes. Most, however, question Niskanen's assumption that bureaucrats are revenue maximizers [Blais & Dion, 1991]. In the meantime new theories have been developed that rely on a more careful or perhaps more imaginative description of bureaucratic tastes and opportunities. Contemporary models of bureaucracy stress the informational endowments of bureaucrats, the implicit and explicit contracts that link their actions to rewards, and their discretionary powers. The presumption that individuals within the State's administrative apparatus are single-mindedly driven to expand and protect existing programs and develop new programs of intervention has given way to the presumption that their utility functions might include some or more of the following arguments: effort and risk aversion, perquisite consumption, control benefits and other nonpecuniary benefits, and reputation [Laffont & Tirole, 1993; Gonzalez & Mehay, 1985; Mehay, 1986; Migue & Belanger, 1974; Schleifer, 1994; see also Antoci, 1995; De Fraja, 1993; Gemmell, 1990; Santerre, 1990; Whynes, 1993; McFadden, 1976, 1977]. Moreover, the dichotomy between competitors and monopolists proposed by the structure-conduct-performance paradigm: price takers versus price setters, has been largely superseded in the industrial-organization literature by a new technology of games under incomplete information [Tirole, 1988].

The newer models of bureaucratic behavior often seem inherently plausible. But like the economic theory of interest-group politics, they generally fail to yield successful empirical predictions beyond the ones for which they were custom tailored [see Conybeare, 1984]. For example, a while back L.R. Jones and I [Thompson & Jones, 1986] proposed a bilateral monopoly model of the budget process, with the central control office on one side and agencies on the other. We assumed that budgeters were primarily interested in cutting budgets -- that is after all what they are paid to do -- and that agency officials were motivated by a variety of considerations -- task accomplishment, perquisite consumption, control benefits and other nonpecuniary benefits, and reputation [see Thompson & Williams, 1979]. The typical outcomes of this model were less than optimal budgets and outputs and higher than minimum unit costs. If this model were generally valid, however, both budgeteers and agency heads would consistently oppose competition whenever it raised its ugly head. That does not always seem to be the case, however. In both New Zealand and Australia, for examples, officials in central control agencies have taken the lead in promoting competition within government.

Another example is due to Terry M. Moe [1990: 140, 1989; see also Tabellini & Alesina, 1990; Cooley & Smith, 1989], who argues that political authorities, especially legislators, favor administrative controls that are ineffective by design. He claims that legislators shun serious policy control and, instead, seek "particularized" control because they "want to be able to intervene quickly, inexpensively, and in ad hoc ways to protect or advance the interests of particular clients in particular matters." Detailed rules that impose rigid limits on an agency's discretion and its procedures help to satisfy this appetite. Consequently, detailed object-of-expenditure budgets are the norm, for example, not for historical reasons, but because they are suited to the needs of temporary governing coalitions, which are likely to be far more concerned with who gets public monies and where it goes, than with what it buys for the public at large. Furthermore, Moe argues that the rigidity characteristic of the American administrative process is largely the product of the efforts of temporary ruling coalitions to prevent future majorities from interfering with their handiwork [see also McCubbins, Noll, & Weingast, 1989].

Summing up

Public choice theorists are often cynical about politics and pessimistic about the workings of government. They disallow any role for what Steve Kelman [1987] calls "public" or "civic" spirit, except to the extent that self interest is defined to include an interest in the welfare of others [a tactic which has the effect of denying to public choice any Popperian bite whatsoever]. Moreover, as Michael Trebilcock [1994] explains, public choice theorists implicitly reject the notion that ideas have power [which, if true, would render public choice an exercise in futility], although the recent trend to privatization, deregulation, and to tax reform can hardly be explained any other way. As Trebilcock puts it, these policies show that ideas have force and that "... politics, to an important extent, is partly about what are thought to be good ideas as well as what are thought to be politically salient interests."

These problems aside, and clearly they are not small problems, what public choice theorists say about coalition formation, free riding, agenda setting, and bureaucracy is important, if for no other reason than because it has been useful in promoting a healthy skepticism [not cynicism] about government and interest group demands. It is not a bad thing to look beneath the packaging [see Stanbury, 1993]. Individuals and groups do often turn to government to obtain or preserve economic rents that would otherwise be unavailable to them. Government activities often are designed to interfere with efficient market solutions to resource allocation problems. Society has several common pools of wealth: [1] personal and business net assets; [2] government's net real and financial assets, not including natural resources; [3] publicly owned natural resources; [4] the stock human capital; [5] an environment pool that reflects the overall "quality" of the environment; and [6] the wealth of the future generations. The wealth of future generations is, of course, largely dependent upon the expansion of the first five pools. Politics, at its worst, is merely a means by which stakeholder groups use the collective, coercive power of government to tilt these pools in their direction. This produces a lot of sloshing about and considerable leakage. Moreover some groups are especially vulnerable to losses -- the young and a fortiori the unborn, for instance, must bear the consequences of the failure of current generations to expand these resource pools, but are largely excluded from political processes. Consumers and taxpayers, individually and collectively, are similarly although not so completely disadvantaged in political arenas. It seems likely that increased skepticism about existing institutional arrangements and governance mechanisms has encouraged experimentation with alternatives and increased receptivity to the lessons of the new economics of organizations and institutions.

THE NEW ECONOMICS OF ORGANIZATION

In a recent article, Robert D. Behn [1995] observed that public administration needs to focus on big questions, such as how public managers can break the micromanagement cycle of procedural rules and how they can measure the achievements of their agencies in ways that will help to increase those achievements. The new economics of organization focuses on these vital questions. Its basic idea is that the comparative advantage of any institutional mechanism or governance arrangement boils down to a question of information or transaction costs "and to the ability and willingness of those affected by information costs to recognize and bear them" [Arrow, 1969; see also Alchian & Demsetz, 1972; and Barzel, 1982]. Hence, the circumstances that create market failures: public goods, natural monopolies, externalities, moral hazard and adverse selection, etc., the problems that justify government action in a capitalist economy, are fundamentally information failures. Markets could deliver public goods, for example -- if information technology existed that would permit free-riders to be profitably excluded from enjoying them. Monopolies could be compensated to behave like competitors -- if information costs were lower. And, externalities could be eliminated by bargaining between self-interested individuals, without the intervention of government -- if transactions costs were zero. Much the same logic applies to the choice between organizations and markets and the kinds of governance mechanisms used within organizations.

The problem with the elegant, logically consistent deductive models created by the practitioners of the economics of organization is that many are neither empirically grounded nor very practical. Until very recently the economics of organization was a classic example of theory without data. Ten years ago, for example, when John Conybeare [1984] surveyed the literature on economic structure and performance in the public sector, he found a single empirical study to cite. This situation is changing [e.g., Frant, 1993; forthcoming], but not very rapidly. One reason for this state of affairs is that tinkering is required to make a satisfactory positive theory of public sector supply arrangements. One cannot, for example, simply presume that arrangements have been selected because they minimize the sum of operating and transactions costs [Moe, 1984; although it is best to be agnostic on this point, see Ferris & Graddy, 1991, 1994; De Groot, 1988] or that the arrangements one observes are optimal; careful, comparative, best-practice research is needed.

Second, the new economics of organization can be impractical, in part because its practitioners are often more interested in saying what cannot be done, than what should be, and in demonstrating that circumstances give rise to inefficiencies, than how one might go about minimizing them.

Consider the typical approach to asymmetric information -- agency, moral hazard, and adverse selection problems. Economists seem to fix on problems in which there is high degree of conflict between principal and agent, which usually leads to a substantial gap between what principals get and what they could get if they were better informed [Persky, 1994]. But intellectual interest ought not obscure reality. Many, perhaps most, asymmetric information problems have reasonably satisfactory solutions [Tirole, 1994]. Evidence that a potential problem is a real one requires information about magnitudes, and not just an existence proof. Agency, moral hazard, and adverse selection problems are common in corporate governance, since the interests of corporate managers are also not necessarily identical to those of the stockholders [see, for example, Williamson, 1988]. Yet many businesses survive and prosper nonetheless. Furthermore, these conflicts do not undermine completely the value of theories that ignore them. Moreover, when economists look closely at agency problems, they often shrink in size, when they cannot be made to disappear altogether [Baker, 1992; Baker, Gibbons, & Murphy, 1994; Campbell, Chan, & Marino, 1989; Hemmer, 1993; Itoh, 1993; Bates, 1993; De Groot, 1988; De Fraja, 1993; Good, 1992; Mayston, 1993; Wallis, 1991].

The difference between potential and real agency problem is demonstrated by James Buchanan's conjecture that risk-averse public officials will exploit their superior understanding of the production of government services to extract higher budgets from the public [1969:98-102]. Buchanan's argument goes something like this: from the public official's standpoint, it may be more costly to finance public spending with taxes than with debt. In which case, borrowing would lead to higher than optimal levels of government spending and services. Consequently, Buchanan concludes that pay-as-you go financing serves as a check on public officials, bringing their actions into line with the tastes of the citizenry. Pay-as-you go financing has much to recommend, but as it turns out, the risk-aversion problem that is at the heart of Buchanan's conjecture is strictly incentive compatible -- either it is not a problem or is easily soluble [Choate and Thompson, 1989, 1990; see also Wittman, 1989].

In the remainder of this essay I assume that agency problems are not completely intractable and can usually be mitigated if not completely overcome. Given this assumption, what can one take from new economics of organization that is both useful and valid?

The Control Problem

The propensity to devise inflexible and comprehensive rules is, perhaps, nowhere more irresistible than where government procurement is concerned. Consider the fruitcake, 250 tons of which are purchased by the Army each year. To preclude abuses on the part of unscrupulous bakers, to make sure there really are some candied fruits and nuts in the fruitcake, to guarantee adequate shelf-life and resistance to handling, and to insure palatability in all the far-flung places of the world where the American soldiers celebrate Christmas, the specifications for the MIL-F-1499 [Fruitcake] are eighteen pages long. Plastic whistles take sixteen pages, olives seventeen, chewing gum fifteen, condoms thirteen, and so on.

The late Frederick C. Mosher [1980: 545-47] not so long ago observed that government has experienced a sea change in its responsibilities and its tactics and concluded that these massive changes have rendered obsolete the traditional administrative controls we inherited from our forebears. In a similar vein, Allen Schick [1978: 518] has noted with concern that recent changes in the reach and scale of government have been accompanied by massive growth in the scope and content of rule-bound governance mechanisms: reporting requirements have multiplied; auditors scrutinize more closely the accounts of federal agencies and contractors; and direct controls in the form of rules and regulations have proliferated.

Schick concluded that we cannot afford to go on imposing direct controls over an ever widening sphere of activities -- that new solutions to the problem of administrative governance must be sought. He finished with a reminder that in many cases individuals can be more effectively influenced to serve the public interest "by inducements which allow them to pursue their own interests than by constraints which try to bar them from behaving as they want."

Perhaps the most useful contribution of the economics of organization is that it has made us cognizant of the alternative governance mechanisms that are available for influencing people to advance the policies and purposes of the institutions they serve: [1] outlay budgets, [2] responsibility budgets, [3] fixed-price contracts, and [4] flexible-price contracts, of the circumstances under which each has a comparative advantage, and the criteria to be used in choosing between them.

Governance in General

All governance system designers face the same key choices: what, where, when, and, whom to control. The choice of what and where to control is reasonably self-evident. Management control should be primarily addressed to the behavior of service suppliers [i.e., departments and agencies, contractors, etc.], the efficiency with which they produce goods and services, and ultimately the efficiency with which they use the assets at their disposal.

The choice of whom to subject to controls and when to execute those controls is less self-evident. In the abstract, a governance system designer has four sets of options, comprised of two choices of subject and two of timing.
[1] The subject may be either an organization or an individual; and
[2] Controls may be executed either before or after the subject acts.
Before-the-fact controls are intended to prevent subjects from doing undesirable things or to compel them to do desirable thing. These controls necessarily take the form of authoritative mandates, rules, or regulations that specify what the subject must do, may do, or must not do. The subjects of before-the-fact controls are held responsible for complying with these commands and the controller attempts to monitor and enforce compliance with them.
After-the-fact controls are executed after the subject, either an organization or an individual, decides on and carries out a course of action and, therefore, after some of the consequences of the subject's decisions are known. Because bad decisions cannot be undone after they are carried out, after-the-fact controls are intended to motivate subjects to make good decisions. Hence subjects are made responsible for the consequences of their decisions, and the controller attempts to monitor those consequences and to see that subjects are rewarded or sanctioned accordingly.
Combining the choice of subject with that of timing, we find that the governance system designer must choose among four distinct institutional alternatives: individual responsibility, before-the-fact or after-the-fact, and organizational responsibility, before-the-fact or after-the-fact.
Privatization
The significance of these alternative designs is partially reflected in debates over the merits of privatizing the delivery of public services. Proponents of privatization imply that the choice is between rule-governed, often over-regulated, monopolistic public bureaucracies and freely competing private firms [Hanke, 1987; Savas, 1982; see, however, Donahue, 1989]. Were that the choice, it is difficult to see how privatization could be wrong, since it resolves to a simple question of monopoly or competition. Other things equal, provision by competing private firms will usually be more efficient than provision by a public monopoly.
But other things are not equal. Consequently, the distinction often drawn by proponents of privatization between public and private production is often overly simplistic. It is simplistic because many proponents of privatization implicitly assume that governments are flawed but markets are perfect: that market participants are perfectly informed, that market transactions are costless, and that they involve the exchange of a discrete product for cash. However, when the choice of governance arrangements is at hand, that is precisely what we must not do. In the real world, where the production of a good or service may be characterized by increasing returns to scale or scope, a high degree of lumpiness in production [or consumption], asset specificities, or the absence of close substitutes, the purchaser must expend resources shopping and bargaining over prices and terms. If the item or service is to be made available at a later date, the purchaser must expend resources to insure that the supplier adheres to the terms of the bargain or contract. That is, to avoid costly mistakes, the purchaser must incur search, bargaining, monitoring, and enforcement costs. These transaction or control costs are part of the cost of acquiring a a good or service. It does not matter whether they occur inside of government or out; ultimately the public pays.
Consequently, the distinction between provision by a public agency and provision by a private entity fails to capture the full range of choices available to the governance system designer. It also fails to reflect all of the factors that are relevant to the choice.
First, although most goods and services are produced by organizations and not individuals, effective control ultimately presumes individual accountability. The distinction drawn by the proponents of privatization between public and private provision ignores the capacity of controllers to hold managers of public organizations under their jurisdictions individually responsible for their behavior and, thereby, their capacity to influence directly the rewards and sanctions that accrue to those individuals, such as salary and opportunities for advancement.
Controllers cannot possibly hold managers personally responsible where their relationship to the supplying organization is at arm's length and the structure of individual responsibility is veiled by the organizational form. The only way an organization can be rewarded [or punished] is by increasing [or reducing] its revenues. An organization's revenues can affect an individual manager's welfare, but only indirectly.
The difference between holding individuals and organizations accountable or between direct, personal influence and indirect influence is quite straightforward. Take the following example: if the quality of services supplied by an agency, the Defense Logistics Agency for example, is grossly unsatisfactory, the controller can recommend the dismissal of the agency director. Where the Department of Defense has an arm's length relationship with a service supplier, McDonnell-Douglas for example, and the relationship is unsatisfactory, all that the controller can do is recommend termination of the relationship. The controller can punish the supplying organization, but cannot punish the managers responsible for the failure -- although their actions might very well lead the organization's board of directors to to so. Unfortunately, punishing a monopoly [that is, any sole-source supplier] is like cutting off your nose to spite your face; rewarding one is like eating an eclair to celebrate staying on a diet. Consequently, where the supplying organization is a monopoly, the capacity to influence managers directly will have considerable utility, particularly where controllers can stimulate and exploit competition between alternative management teams.
This claim can be verified by reference to the private sector. In the private sector, most real natural monopolies make intermediate products, i.e., goods that are used to produce consumer goods or services. Natural monopoly [decreasing costs as output increases] can usually be attributed to spreading large, lumpy investments in specialized resources -- technological know-how, product specific research and development, equipment. -- over additional output. Investment in specialized resources often inspires a process called vertical integration ["backward" if initiated by the consumer goods producer, "forward" if initiated by the intermediate goods producer] [Joskow, 1988; Colbert & Spicer, 1995]. The new economics of organization tells us that vertical integration occurs because it permits transaction or control costs to be minimized, in part through the substitution of direct supervision for indirect influence [see Williamson, 1985; on the relevance of transaction cost economics to public administration, see Maser, 1986; Vining and Weimer, 1990; and Ferris and Graddy, 1991, 1994; see also Friedman, 1981 and Borcherding, 1983, 1988].
In the language of transaction-cost economics, investment in specialized resources is called asset specificity. An asset is said to be specific if it makes a necessary contribution to the provision of a good or service and has a much lower value in alternative uses. The corollary of asset specificity is bilateral monopoly, a circumstance that provides an ideal environment for opportunistic behavior on the part of both supplier and customer. For example, once intermediate product producers have acquired specialized assets, their customers may threaten to switch suppliers to extract discounts from them. In that case, suppliers may find it necessary to write off large parts of their specialized investment. Or, if demand for the final good increases greatly, the intermediate product supplier may be able to use its monopoly power to extort exorbitant prices from customers. Hence, where the relationship between the intermediate product supplier and his customer is at arm's length, the threat of opportunistic behavior may be sufficient to eliminate the incentive to make what would otherwise be cost-effective investments. Vertical integration can eliminate this threat. Indeed, where the intermediate product producer provides homogeneous goods or services [i.e., outputs that are easily monitored], total production volume is specified, and technologies are mature, vertical integration permits a bilateral monopoly to be governed satisfactorily by unbalanced or two-part transfer prices [Masten, 1993; Young, 1991; Schroeder, 1993; Prusa, 1990].
Vertical integration is, of course, only one way to deal with asset specificity [Walker & Poppo, 1991]. Some organizations invest in specialized resources and own design-specific assets, which they provide to their suppliers. This is called quasi-vertical integration. It is common in both the automobile and the aerospace industries, and, of course, it is standard procedure for the Department of Defense to provide and own the equipment, dies, and designs that defense firms use to supply it with weapons systems and the like [see Monteverde and Teece,1982]. Other organizations that rely on a small number of suppliers or a small number of distributors write contracts that constrain the opportunistic behavior of those with whom they deal. A well-executed contract can approximate the outcome from vertical integration [although such contracts are often very hard to write and, where one of the parties is inclined to exploit the other, prohibitively costly to enforce] without incurring the very real costs of vertical integration. In other cases, desired outcomes can be realized through alliances based on the exchange of hostages [e.g., surety bonds, exchange of debt or equity positions] or just plain old-fashioned trust based on long-term mutual dependence. In Japan, for example, buyer-seller relationships tend to be based on mutual confidence [Ravenscroft, 1993]. Toyota relies on a few suppliers that it nurtures and supports. It maintains tight working links between its manufacturing and engineering departments and its suppliers and explicitly eschews opportunistic behavior in the interest of maintaining long-term relationships.
Nevertheless, in one study of vertical integration in the U.S. aerospace industry, Scott Masten [1984] unambiguously demonstrated that asset specificity and, therefore, decreasing cost is basic to the make-or-buy decision. Where intermediate products were both complex and highly specialized [used only by the buyer], there was a 92 percent probability that it would be made internally; even 31 percent of all simple, specialized components were made internally. The probability dropped to less than 2 percent if the component was not specialized, regardless of its complexity.
The proponents of privatization also err in their implicit claim that responsibility can be vested in organizations if and only if the organization is private, and in individuals if and only if the organization is part of the public sector. The absurdity of this claim becomes clear as soon as it is explicitly stated; it is consistent with neither theory nor practice. For example, many state legislatures base their relationships with public entities such as universities or hospitals on arm's length relationships that are guaranteed by self-denying ordinances, which exempt the managers of these public entities from detailed oversight and direct control. The Department of Defense's decisions about base closure and realignment, for example, are exempted from direct congressional control by just such a self-denying ordinance [Twight, 1990]; public authorities are another good example [Frant, forthcoming]. Similarly, the recurring procurement fraud cases show that managers of private entities that supply services to the government can be held directly and personally responsible for their behavior when it violates federal law.
Finally, most of the proponents of privatization implicitly presume that the services provided to or for government are homogeneous or fungible, which implies that the problem of identifying the most efficient supplying organization or management team resolves to a simple question of price search, an elementary control mechanism that reveals information about the "customer's" demand for the service. In fact, many of the organizations supplying goods or services to or for government supply bundles of more or less heterogeneous products -- many of these products are hard to measure and costly to evaluate, some prohibitively so.
Choosing between Alternatives
Proponents of privatization do make one significant, unexceptionable claim: the choice of institutional design should depend on the cost and production behavior of the good or service in question. However, they frequently fail to carry this claim to its logical conclusion. This choice should depend upon minimizing the sum of production costs and transactions costs [Williamson, 1985].
At least two factors are relevant from a transactions-cost standpoint: the ease with which the consequences of operating decisions can be monitored and the desirability of inter-organizational competition. Most management control theorists believe that where consequences [that is, an organization or responsibility center's outputs] are easily monitored, control should focus on the consequences of the subject's decisions; where they are not, control should focus on their content [inputs]. Because consequences are easily monitored where entities produce homogeneous outputs or where a responsibility center within an entity performs fungible activities, it follows that controllers should rely on after-the-fact controls where homogeneous outputs are supplied. In contrast, it follows that they should rely on before-the-fact controls where each item supplied is, from the "customer's" perspective, intrinsically unique. Furthermore, this view has been reinforced by recent findings in organizational economics and agency theory [Mayston, 1993; Ezzamel, 1993; Macintosh, 1993; Whynes, 1993; Good, 1992; Brown, Thornton, Buede, & Miller, 1992].
At the same time, industrial organization theory tells us that inter-organizational competition is desirable only where costs are constant or increasing as quantity of output [rate or volume] increases [Tirole, 1988; Carlton & Perloff, 1990]. Where economies of scale or scope cause costs to decrease when output is increased, monopoly supply is usually appropriate [Barzelay (1992) refers to entities that have this characteristic as "utilities"; see also Kettl (1993)]. Because responsibility can be effectively vested in organizations only where customers or their agents are ultimately indifferent to the survival of one or more of the supplying organizations, the implication of this line of reasoning is that controllers should vest responsibility in organizations only where inter-organizational rivalry is practical and likely to be effective -- and in individuals, where it is not [Willig & Jatar, 1993; Aertsen, 1992]. These normative prescriptions are summarized in Figure 4.

FIGURE 4 GOES ABOUT HERE

Execution of Alternative Governance System Designs
These four basic sets of controls are all employed by government. But is each appropriately employed? Before this question can be answered, I must first show how these designs are used and explain the practical logic of their implementation. My discussion will concentrate on the use of before-the-fact controls. This does not mean that I particularly like them. On the contrary, I believe that controllers should resort to before-the-fact control designs only where production and transaction costs clearly renders their use the least objectionable alternative [Reid, 1990].
I concentrate on the use of before-the-fact controls because it seems to me that their implementation is not well understood, especially by those who most rely on them. Many participants in the policy process seem to believe that before-the-fact controls not only safeguard against abuse, but also, by reducing costs, improve performance. If failure occurs nevertheless, they tend to believe the solution lies in still more or better rules.
One possible explanation for the persistent faith in the efficacy of before-the-fact controls is that its devotees just don't understand how hard it is to execute them efficiently. For example, they appear to believe that the subjects of before-the-fact controls will comply with them simply because they are morally obligated to do so. Obviously, however, not everyone is inclined to respect moral authority, to respect the law, or to obey rules. It is necessary to monitor and enforce compliance with rules and to ferret out and punish noncompliance. It is also necessary to specify the content of before-the-fact controls -- to tell subjects what to do and what not to do in such a way as to find and enforce efficiency, which is no easy matter.
Moreover, many of those who believe in the potency of before-the-fact controls fail to understand that moral authority is all too easily eroded by an oversupply of rules. Moral authority, respect for the law, the inclination to obey rules are of critical importance to the stability and the efficacy of social arrangements. In reality, that they are far too important to be frittered away where other mechanisms of social control will suffice. Rather they ought to be carefully husbanded so that they will be available when and where they are really needed [see Tyler, 1990].
Before-the-fact controls are like after-the-fact controls in that they ultimately rely on incentives and sanctions for their effectiveness. The difference is that after-the-fact management controls are incentive or demand-revealing mechanisms, while before-the-fact management controls are incentive or demand-concealing mechanisms. This means that opacity is an essential characteristic of before-the-fact controls. The incentive aspects of before-the-fact controls are thus less clear than the incentive aspects of after-the-fact controls. This also means that their effectiveness is hostage to the skill with which they are executed, which explains why their incentive aspects are easily overlooked and, perhaps, why they are not better understood. In order to show how demand-concealing mechanisms work, I will first try to show how demand-revealing mechanisms work or, at least, what they are.
After-the-Fact Governance System Designs
By demand-revealing mechanisms, I mean those in which customers [or their agents] declare their willingness to pay for various quantities of goods, services, or activities. Customers transparently reveal a demand schedule that fully expresses their wants and preferences to their suppliers. Then they let suppliers figure out how best to satisfy those wants and preferences. The classic demand-revealing mechanism is the competitive spot market, where customers buy from any number of anonymous firms. When many suppliers are disposed to satisfy customer wants, customers simply choose the best price and quantity combinations offered -- those that moves them farthest down their demand schedules. In so doing customers reward the suppliers that are willing to do the most to satisfy their tastes and implicitly punish the rest. For example, they might order wheat from a broker, at the market price payable on delivery. In that case, there would be no formal contract. Customers would put no restrictions on producers. In fact, customers probably won't even know who grew their grain. But wheat farmers are nevertheless rewarded for their contributions. Government relies on these kinds of spot markets when, for example, it purchases electrical components off the shelf.
After-the-Fact Controls Transparently Reward Measured Performance or Results
The spot market is by no means the only demand-revealing mechanism used to govern transactions between buyers and sellers. There many variations on the basic theme of reliance on transparent rewards. But all of these variations have one common attribute: rewards are provided after operating decisions have been made by the producer, after assets have been acquired and used and outputs monitored. Because they are executed after asset acquisition and use decisions have been made, we refer to them as ex-post or after-the-fact controls.
A close analog to a spot market is seen where government uses prospective price mechanisms to reimburse free-standing service providers. The system used by the Health Care Financing Administration to pay hospitals for treating patients is an example. The enrollment-driven funding formulas used by some states to compensate post-secondary institutions for teaching students is another [Jones, Thompson, and Zumeta, 1986]. In both of these instances the subject is a freestanding organization, and the structure of authority and responsibility within the supplying organization is purely an internal matter. The government or its agent announces a price schedule and specifies minimum service quality standards [or a process whereby these standards are to be determined] and the time period in which the price schedule will be in effect.
Under prospective pricing, all qualified organizations will be paid a stipulated per-unit price each time they perform a specified service, such as enrolling a full-time equivalent student or treating a heart attack. This means, among other things, that the government's financial liability is somewhat open ended. It depends on the price offered and the quantity of service actually provided [Liability = Price Quantity], although not directly on the costs incurred by the suppliers.
Another close relative of the spot market is the fixed-price contract. Again, under fixed-price contracts, prices are fixed before hand and aren't affected by the supplier's subsequent costs. In contrast, under flexible-price contracts those costs are shared with the customer. The limit under flexible-price contracts is reached in the case of a cost-plus fixed-fee contract, where the customer assumes full responsibility for all legitimate, measured costs.
In a pure fixed-price contract government usually buys from numerous suppliers held at arm's length. Frequent bidding contests are held and orders are shifted among suppliers chosen simply on price. These price schedules can entail all sorts of complex arrangements, including rate, volume, and mix adjustments as well as inflation adjustments and sometimes default penalties. Fixed-price contracts also comprehend franchises to provide a specified services, perhaps at specified locations, for a fixed periods of time -- garbage collection at a military base, for example.
Under all of these demand-revealing mechanisms, government relies upon inter-organizational competition, along with the profit motive, to motivate service suppliers to make wise asset acquisition and use decisions and to produce efficiently. If inter-organizational competition is effective, organizations that don't make wise asset acquisition and use decisions fall by the wayside.
Demand-Revealing Mechanisms in Vertically Integrated Organizations
In some cases, even where the cost behavior of the service in question renders vertical integration and therefore monopoly supply appropriate, after-the-fact, demand-revealing mechanisms can still be effectively employed. This is done in businesses and businesslike public sector organizations by holding managers responsible for optimizing a single criterion value, usually a measure of financial performance, subject to a set of specified constraints. This control mechanism is, therefore, usually known as responsibility budgeting [Anthony and Young, 1988: 365-386; Thompson, 1991], sometimes as performance contracting [Islam, 1993]. The fundamental construct of responsibility budgeting is an account [or control] structure that is oriented toward administrative units that have purposes or objectives and use inputs [resources] to produce outputs [goods or services]. Ideally a responsibility budget contains a single number or financial performance target [e.g. a unit cost standard, or a profit or return-on-investment target] for each administrative unit.
Responsibility centers are classified according to two dimensions: [1] The integration dimension &endash; i.e., the relationship between the responsibility center's objectives and the overall purposes and policies of the organization; [2] The decentralization dimension &endash; i.e., the amount of authority delegated to the responsibility manager, measured in terms of the discretion to acquire and use assets. On the first dimension, a responsibility center can be either a mission center or a support center. The output of a mission center contributes directly to the organization's objectives. The output of a support center is an input to another responsibility center in the organization, either another support center or a mission center. On the decentralization dimension, discretionary expense centers, which mimic the governmental norm, are found at one extreme and profit and investment centers at the other. A support center may be either an expense center or a profit center. If the latter, its profit is the differences between its expenses and "revenue" from "selling" its services to other responsibility centers [Halal, 1994]. Both profit and investment centers are usually free to borrow, and investment centers are also free to make decisions about plant and equipment, new products, and other issues that are significant to the long-run performance of the organization.
Responsibility budgeting is largely concerned with setting targets, monitoring performance in terms of the targets specified, and rewarding managers accordingly. Responsibility centers coordinate their activities via a process of exchange and mutual accommodation. This is often possible now, where it wasn't really before, because the design of responsibility structures [the allocation of property rights and the ownership so as to minimize intraorganizational externalities or spillovers] and management information systems [performance measures, transfer prices, and costing procedures] are now more sophisticated. The increased sophistication of performance measures, transfer prices, and costing procedures is largely due to improvements in computer speed and software [Lapsey, 1994].
Where bilateral relationships are concerned, it is usually possible to set up some kind of transactional arrangement to eliminate or internalize spillovers [Ronan, 1992; Ronan & Balachandran, 1988; Balachandran & Ronan, 1989; Dorkey & Jarrell, 1991]. In most cases, these relationships can be governed via buyer-seller arrangements and, where they occur within the organization, by appropriate transfer prices. Where monopoly supply is appropriate, two-part tariffs [e.g., hook-up plus usage charges for telephone services, see Tirole, 1988; Cohen, Loeb, Stark, 1992] or unbalanced transfer prices can be used to manage relationships and to provide the supplier with an incentive to make long term investments in plant, equipment, or know-how. [Unbalanced transfer prices are often easier to use than cost-based two-part tariffs, since the latter must be very carefully calibrated to produce an efficient solution, but accountants don't like them very much] In other cases, support units can be made to compete within the context of a quasi-market dynamic with others supplying similar services inside and outside the organization and permitted to charge whatever the market will bear.
Such an account structure would look something like the following:
1. All administrative units are be classified as either mission or support centers.
2. All costs accrued by support centers -- including charges for the use of capital assets and inventory depletion -- are charged to the mission centers they serve.
3. Mission centers are funded to cover their expected expenses -- including support center charges.
4. A working capital fund provides short-term financing for support units.
5. A capital asset fund provides long-term financing of capital assets and encourages efficient management of their acquisition, use, and disposition.
The principal formal device by which transactions are currently accomplished within government is the revolving fund. These funds involve buyer-seller arrangements internal to federal government. The Navy had a revolving fund as early as 1878. Modern day revolving funds date to the 1947 National Security Act and, under the Defense Business Operating Fund, have grown to $40 billion a year. Two kinds of funds have been established under this authority: stock and industrial funds. Stock funds are used to purchase supplies in bulk from commercial sources and hold them in inventory until they are supplied to the customer -- usually a military unit or facility. Industrial funds are used to purchase industrial or commercial services [e.g., depot maintenance, transportation, etc.] from production units within the government. Both kinds of funds are supposed to be financed by reimbursements from customers' appropriations [Juola, 1993: 43].
Modern transfer pricing could expand the scope of this device and enhance its effectiveness by establishing rules for setting prices prospectively rather than retrospectively and my emphasizing the notational nature of the exchange. Transfer prices encourage efficient choice on the part of support centers and the units that use their services, only [1] if prices are set ahead of time, [2] if support centers nominally charge all of their expenses against revenues earned delivering services, and [3] if support centers are treated as responsibility centers -- their managers are authorized to incur expenses to deliver services and are held responsible for meeting the stated financial targets of their centers: the more highly capital intensive ones, as investment centers, most others either as cost or income centers [although in most cases their performance could be evaluated in terms of quasi-profitability -- i.e., benchmark cost/price less actual cost][Bailey, 1967: 343].
Under responsibility budgeting/performance contracting, the structure of authority and responsibility within the organization is of crucial interest to the controller. The effectiveness of responsibility budgets depends on the elaboration of well-defined objectives, accurate and timely reporting of performance in terms of those objectives, and careful matching of spending authority and responsibility. Their effectiveness also depends on the clarity and transparency with which individual reward schedules are communicated to responsibility-center managers and the degree of competition between alternative management teams [Govindarajan, 1988; Goold, 1991; Chia, 1995].
Examples of governments that have experimented with responsibility budgeting and accounting include New Zealand, Australia, Great Britain, Switzerland, Finland, the Province of Alberta, Canada, and many others [Schick, 1990; Cothran, 1993; Ezzamel. 1993; Schedler, 1994]. These governments have rapidly downsized their public bureaucracies to more efficient types of organizations and replaced reliance on monopolistic agencies with policies favoring competition and contracting out on the assumption that competition, both of advice and delivery systems, is essential to efficient service provision. As noted above, New Zealand has carried these reforms furthest. Most of the attention given to public administration in New Zealand has focused on its efforts to improve the quality of its financial reports and its performance measures. It is the first country to publish a rational set of general purpose financial statements: including a balance sheet showing assets and liabilities and an accrual-based statement of income and expenses. However, the changes made in the structure of its government are perhaps even more noteworthy [Scott, Bushnell, & Sallee, 1990; Ball, 1994].
First, Parliament privatized everything that was not part of the "core public sector." The residual "core" includes a mix of policy, regulatory and operational functions. Second, Parliament redefined its relationship with department heads. Departmental heads are appointed for fixed terms, with the possibility of reappointment. Each works to a specific contract, negotiated with the State Services Commission, which also monitors and assesses performance [hence, performance contacting, see World Bank, 1995; Islam, 1993]. Third, Parliament changed the way it appropriates funds.
The basis of appropriation now depends on the department's ability to supply adequate information about its performance. All departments started out in Mode A; the majority have now progressed either to Mode B or C. Under Mode A, departments are treated as discretionary cost centers and Parliament appropriates funds for the acquisition of inputs. Under Mode B, departments are treated like quasi-profit or expense centers. Under Mode C, departments are treated like investment centers. Appropriations pay for the outputs produced by the department and for any changes in the department's net assets. Under both Modes B and C, managers are free to make decisions [under C most] about investments in plant and equipment. Parliament relies on the fact that managers' financial performance is one of the main bases upon which their performance is assessed to insure that their decisions will be sound [Gul, 1994; Halal, 1994; see, however, Burke, 1990].
Before-the-Fact Governance System Designs
Before-the-fact management controls are demand-concealing mechanisms. Their distinguishing attribute is that they are executed before public money is spent. That is, they govern a service supplier's acquisition and use of both short-term and long-term assets, which means that the controller retains the authority to preview these decisions. Examples of before-the-fact management control include object-of-expenditure appropriations and encumbrance accounts -- these govern the kind of assets that can be acquired by governmental departments and agencies; apportionments, position controls, and the fund and account controls that regulate the rate, timing, and purpose of public spending [Pitsvada, 1983; Schick, 1964, 1978], and the similar rules and regulations that govern the behavior of private contractors [Goldberg, 1976; Kovacic, 1990].
Readers will recognize the combination of before-the-fact controls and individual responsibility in traditional government budgets. Most will also recognize the combination of before-the-fact controls and organizational responsibility in the so-called cost-plus contract -- the most notorious member of the administered or flexible-price contract family. I generally prefer the term "flexible-price contract" to "cost-plus contract," because I am concerned primarily with distinguishing these contracts from fixed-price contracts. Flexible-price contracts comprehend a variety of incentive and cost-sharing contract designs other than the classic cost-plus contract. In turn, flexible-price contracts are included in the broader category of administered contracts [see Goldberg, 1976].
Most governmental budgets are spending plans. To distinguish them from responsibility budgets, I use the term "outlay budgets." Under outlay budgets, supplying organizations are guaranteed an allotment of funds in return for providing a service for a stipulated period. They usually receive the allotment regardless of the actual quantity or quality of services provided [which has the effect of treating every government department, agency, or bureau as a discretionary cost center].
Flexible-price contracts are basically production plans. They fully specify product or service characteristics and a usually a delivery schedule. Under flexible-price contracts, supplying organizations are guaranteed reimbursement [complete or partial] for any legitimate expense incurred providing the service. Hence, the prices they are paid for providing services are determined retrospectively according to settled cost-accounting standards and the specifics of their contracts.
To say that controllers focus their attention primarily on a supplier's asset-acquisition decisions does not mean that they ignore performance in executing outlay budgets or price in executing flexible-contracts. Controllers usually take account of information about the future consequences of a supplier's decisions as well as information on its current and past behavior. Their attention to performance may be tacit, as in the execution of traditional line-item budgets, rather than express, as in the execution of performance, program, or zero-base budgets. But the consequences of asset acquisition decisions usually matter a great deal to controllers. What is crucial is that, under these control systems designs, attention to the performance consequences of spending decisions is necessarily prospective in nature. Controllers will not reveal a demand schedule that fully expresses customer wants and preferences to suppliers or leave it to suppliers to figure out how best to satisfy those wants and preferences.
Even under these control systems designs, the service provider, whether a department or an outside contractor, must assume some responsibility for managing output levels and delivery schedules, service quality, or price. Nevertheless, as I shall demonstrate, the logic of demand-concealing oversight requires supplier discretion to be carefully restricted. This means that suppliers must be subjected to fairly extensive, fairly detailed before-the-fact controls. An agency's outlay budget, for example, should identify all the asset acquisitions that it is to execute during the year, specify their magnitudes, and make it clear who is responsible for implementing each acquisition.
Of course, constraining managerial discretion is not the only function that before-the-fact controls perform. If it were, it would be hard to claim that they ever represented a least-objectionable alternative, let alone explain their widespread use. Rather, constraining managerial discretion is chiefly a means to an end, not an end in itself [Lewis & Sappington, 1989; Baron & Besanko, 1988; Reichelstein, 1992; Keren, 1989]. To show how subjects can be told what to do in such a way as to enforce efficiency, I will outline in detail the logic of employing the two basic before-the-fact governance system designs: flexible-price contracts and outlay budgets.
Flexible Price Contracts
There is a difference in the role that competition plays under fixed and flexible-price contracts. The difference is not that it takes place before the production of the service in question. [Economists refer to such a competitive regime as competition for the market, to distinguish it from competition in the market.] The recipients of fixed-price contracts often receive exclusive franchises prior to the delivery of services.
The difference between the role played by competition under fixed and flexible-price contracts can be summarized by the expression "moral hazard." Under flexible-price contracts, competition cannot be relied upon to keep prices low, let alone to enforce efficiency [Cohen & Loeb, 1990]. Once a flexible-price contract has been signed, the supplier is free to dip his hand into the customer's pocket. Because the supplier is spending somebody else's money, the normal incentives to cost effectiveness largely disappear. Decisions that affect cost, service quality, or price [i.e., asset acquisition and use decisions] must be made during performance of a contract, but once a contract is signed, the supplier can no longer be fully trusted to make them [Thomas & Tung, 1992]. This conclusion holds especially where the customer ignores information regarding the performance of suppliers on earlier contracts or cannot [will not] award future contracts based on good performance. Even where fixed-price contracts are concerned, the refusal to take past performance into account discourages supplier loyalty and eliminates incentives to improve the quality of the product delivered [see Kelman, 1990; von Ungern-Sternberg, 1994].
Why, then, would a customer ever sign a flexible-price contract? Why not simply write fixed price contracts? The answer is that a fixed-price contract is the mechanism of choice where controllers know precisely what their principals want and there are several potential service suppliers who know how to meet those preferences. Under those circumstances, service quality attributes offered, promised delivery schedules, and bid price allow the customer to evaluate proposals satisfactorily. Regrettably, all of these conditions are likely only where the service in question is fairly simple and relatively standard -- garbage collection, for example [von Ungern-Sternberg, 1994].
Technological Uncertainty and Financial Risk
In other cases, neither the controller nor the service supplier may have enough knowledge of the value of product attributes or production processes prior to performance of the contract to employ a fixed-price contract [Macintosh, 1993]. It is a simple fact of life that considerable experience is usually required to manage to a narrow range of outcomes; where specialized or unique services are involved, no organization is likely to have the required experience. Consequently, any organization that agreed to produce a unique service, according to a specified schedule, at a fixed price would incur a large financial risk. This risk can be shifted, but it cannot be eliminated.
Government, especially the federal government, can often bear financial risks better than supplying organizations. This is the case because of the size of the assets government commands and its ability to pool risk. Consequently, the cost to government will often be lower if it assumes a portion of the risk associated with acquisition of the service. Flexible or retrospective pricing is one way for the government to assume this risk. Moreover, the preferences of the government may change during performance of a contract. Under a fixed-price contract, it might not be possible to secure desired changes in service attributes if they involve increased costs for the vendor.
Of course, the indifference of the government to financial risk is easily exaggerated. Government is not immune to financial risk, otherwise it would never make economic sense for it to rely on outlay budgets [Carlton and Perloff, 1990: 503]. Moreover, while it may be true that doing business with government is risky, the risk is mostly unsystematic, and may, therefore, be diversified away.
Nevertheless. my point remains: customers should prefer flexible-price contracts to fixed-price contracts where it is cheaper for the customer to deal with uncertainty than it is for the contractor to do so or where the customer is more concerned with the ability of the contractor to provide a product that works than with price [see, however, Cohen & Loeb, 1989]. The question is: can before-the-fact controls be used to insure that the seller retains an interest in cost effectiveness?
Using Before-the-Fact Controls to Enforce Efficiency under Flexible-Price Contracts
Execution of a flexible-price contract must begin with a fully specified project spending plan detailing work to be performed, personnel, material, and equipment to be used, input quality standards, and scheduled milestones. This plan provides a basis for the enforcement of efficiency through bargaining and negotiations carried on during the performance of the contract [Reichelstein, 1992; Kathawala, 1990; Lewis & Sappington, 1989]. This process can be compared to a repeated prisoner's dilemma game, in which both parties have a common interest in reaching agreement but also have antagonistic interests with respect to the content of agreements. In this game, the customers try to get as much of what they want as they can at a given price, and the supplier tries to get the highest possible price for providing the service [Hofstede, 1967; see also Breton, 1989]. Bargaining power in a prisoner's dilemma game depends on the information available to each party. In particular, the customer's power is greatest where the customer [or his agent] knows the supplier's true cost schedule, but can withhold full information as to his preference or demand schedule [Morgan, 1949].
This is simply a more formal way of saying that strategic advantage accrues to the party that can best look ahead and reason back. To do so, one must be able to put oneself squarely in the other party's shoes [i.e., one must know the other party's costs under a variety of contingencies]. This is one purpose of "should-cost" models. It is also one of the purposes behind selecting agents who have walked in the other party's shoes [promoting trust is another] -- purchasing agents in manufacturing plants, for example, are usually recruited from the ranks of industrial salesmen and process engineers and vice versa. [The federal government's revolving-door laws enjoin this kind of personnel exchange, however. These laws probably increase the government's power to set an agenda but undoubtedly reduce its ability to understand or use the information that that power confers.]
In a repeated game the information available to the customer [or his agent] will depend upon his ability to control the sequence of moves and countermoves that comprise the game. Public choice theorists refer to this condition as agenda control [Hammond, 1986]. Given comprehensive before-the-fact controls, under which changes can be made only with the prior approval of the other party or his agent, the party suggesting or initiating a change must necessarily reveal valuable information to the other. This can work to the advantage of the customer or the supplier, or both.
For example, consider the following situation: ". . . contracts and specifications are drawn for . . . a ship and agreed to. . . . The contractor discovers he can do the welding of some plates less expensively by another means. About that time the client decides that some room on the ship should be larger.... The contractor can plead that he cannot easily change the room size: however, if the client will permit the altered welding maybe a deal can be struck" [Stark and Varley, 1983: 132]. But when flexible-price contracts are appropriately employed, there is every reason to believe that most change proposals will be initiated by the service supplier. Competition for the market provides an incentive to potential service suppliers to promise more than they can deliver, since contracts are usually awarded to the service suppliers who promise the most. Consequently, very few contract winners can make good on all their promises, especially where their managerial discretion is severely restricted by a full set of before-the-fact controls. This fact will usually become evident to the service supplier during performance of the contract. The service supplier will also learn of the tradeoffs between cost, service quality, and delivery schedule available to it and will eventually want to [or in some cases have to] change its promises or its plans.
Under a full set of before-the-fact controls, such changes are contingent upon prior approval. To secure approval service suppliers must reveal information about their capabilities and trade-off possibilities. As a result, power to enforce the preferences of the government passed to purchasing officers -- but only if they know what they are doing and how to make it happen [Brown, Thorton, Buede, & Miller, 1992; Reichelstein, 1992; Lewis & Sappington, 1989]. A practical example of this process can be seen in the development and production of the Army's multiple launch rocket system, where the relative expertise of the program manager permitted not only a confident source-selection decision but also highly effective contract management [Kennedy School, 1987].
Outlay Budgets
A similar logic [Wildavsky and Hammond, 1965] applies where outlay budgets have a comparative advantage -- under decreasing costs to scale over an array of specialized or unique services [the following is based on Thompson, 1993]. Outlay budgets can help to keep prices low and to encourage efficiency where large, lumpy investments in specialized resources are needed in order to provide services, where each problem, client, or task performed is in some sense unique, and where the most serious problems are supposed to be dealt with first.
Many organizational units in the government have these att