Why a New Public Management? Why Now?

Fred Thompson

Paul Krugman calls 1913 the high-water mark of the First Global Economy. He notes, that over the preceding century, the world economy had been transformed by technology and the widespread acceptance of the belief that free markets, with secure property rights, were the best way to achieve economic progress.

After 1913 the market atrophied -- long-distance trade shrunk, private international movements of capital virtually disappeared, and a third of the world rejected private property.

How does one explain this reversal? Perhaps, more importantly, how does one explain the even more astonishing reversal of fortune that, at the start of the 21st century, the world has returned to more or less the same ideology of free markets, small governments, and sound money that prevailed at the beginning of the 20th.

The answer to the first question must be that bureaucracies replaced alternative institutional arrangements, primarily markets in the first half of the 20th century because they outperformed them. How? Presumably, or so Alfred Chandler argues, because of technological innovations that led to massive economies of scale and/or scope.

What were the changes in technology that caused bureaucracies to out-perform markets? Here the surprising answer is changes in organizational arrangements themselves. That is: changes in organizational design, personnel systems, operational engineering, accounting systems, and control technologies. This answer reflects the currently fashionable view among economists that the comparative advantage of institutional arrangements boils down to a question of information costs and that actual arrangements are solutions to information problems &emdash; the costs associated with search, bargaining, monitoring, and enforcement. Hence, transformations in organizational arrangements must be largely driven by changes in information costs.

Elsewhere, Gil Reschenthaler and I (1994) have argued that changes in organizational arrangements produced four major shifts in the comparative advantage of alternative institutions in the late 19th century. These are:

1. The efficacy of centralized allocation and ex-ante control increased relative to decentralized allocation of resources and ex-post control, which had the effect of increasing the payoff to scale.

2. The efficacy of functional structures increased relative to process-oriented structures, which had the effect of increasing the payoff to scope.

3. The efficacy of hierarchically coordinated systems increased relative to self-organizing systems, which had the effect of increasing the payoff to vertically integrated systems of command and control.

4. The relative efficacy of government provision and control increased, which had the effect of decreasing the payoffs to free markets, secure property rights, and minimal government intervention.

Of course, if these shifts explain the rise of bureaucracy might not recent innovations in organizational design, operational engineering, accounting systems, and control technologies, by reversing these shifts, also suffice to explain its fall?

Prussians perfected the bureaucratic model: Heinrich von Stein, Gerhard von Scharnhorst, August von Gneisenau, and Helmuth von Moltke during the 19th century. Their administrative innovations included detailed centralized materials requirements and logistical planning, control by rules, standard operating procedures, and the merit principle, functional administrative design, decomposition of tasks to their simplest components and narrow job descriptions, and sequential processing.

The American contribution to this system lay primarily in activity and cost measurement, in process engineering: standardization of components, processes, and products, and in the use of electric motors to reconfigure workflow. These innovations made possible the moving, or continuous, assembly line, in which each assembler performed a single, repetitive task. The moving assembly line was first implemented at Henry Ford's Model-T Plant at Highland Park, Michigan, in 1914, increasing labor productivity tenfold and permitting stunning price cuts -- from $780 in 1910 to $360 in 1914. Ford made everything he needed for his cars from the raw materials on up. Of course, total vertical integration required the organization of huge numbers of activities and employees. Workers, staff specialists, and middle managers had to be recruited, sorted out, and fitted into a merit-driven hierarchical scheme -- that is, bureaucracy.

Not only did bureaucracy make large, complex organizations efficient, it also made them inevitable. Only very large organizations could take full advantage of bureaucracy. Only they could afford to devote substantial amounts of resources to gathering and processing quantities of data for top management to use to coordinate activities and allocate resources. Hence, it seemed that bigger organizations were necessarily better. And, there seemed to be no natural limits to this conclusion.

In the US the progressive movement created modern public administration. To a remarkable degree the progressive reforms -- an executive, input-oriented budget, a professional civil service and merit-based public personnel administration, control by rules, standardization of procedures, task specialization, and a strict administrative hierarchy, with clearly delineated staff and line functions -- were based on the Prussian model. In a few instances -- the War Department under Elihu Root and the USDA Forest Service under Gifford Pinchot, the NYC Dept. of Sanitation under Col. George E. Waring, for examples -- progressives proudly acknowledged the source of inspiration for their administrative reforms. Elsewhere, they expressed some discomfort at copying the governance institutions of an undemocratic, militaristic regime. One of the best-known apologies for this practice was Woodrow Wilson’s argument that politics and administration are different functions, making it possible to borrow administrative practices from an authoritarian state without thereby threatening democratic politics -- “If I see a murderous fellow cleverly sharpening his knife....”

Regardless of their source, progressive reforms led to dramatic improvements in the delivery of government services and in the productivity of public employees. Anecdotes to this effect abound: significant reductions in disease following Col. Waring’s reforms, the forest rangers’ erstwhile reputation for efficiency, widespread replacement of government contracting out by in-house production, and, perhaps most telling, the early 20th Century enthusiasm for postalization (i.e., running businesses like the US Post Office). Not all of the evidence is anecdotal, however. Cross national comparisons show, for example, that total factor productivity growth in surface transport once tended to be higher in nationalized systems than where government regulated price and entry and higher in regulated systems than in competitive ones. Similar evidence exists with respect to most so-called public utilities. Empirical evidence also exists as to the consequences of the wave of reform that transformed the governments of many U.S. cities in the last century. Controlling for city and time effects, bureaucratic reform led to significantly increased rates of infrastructure investment and economic development (Rauch, 1995). Overall, by the middle of the last century, despite far higher employment growth, value added per worker remained 40 percent greater in the public sector than in the private.

My point is that bureaucratic arrangements once successfully provided security, jobs and economic stability, ensured fairness and equity, and delivered the “one size fits all” services needed during the era that lasted from the turn of the last century to the mid-1960s. In the meantime, however, the organizational arrangements invented at the dawn of the industrial era have become increasingly anachronistic.

What Goes Around, Comes Around

Centralization, executive, input-oriented budgets, standardization, and direct supervision of the flow of raw materials and components through the production process were eventually rendered obsolete by innovations in organization pioneered by General Motors under Alfred P. Sloan. The best known of these is the multi-product, or M-form, organizational structure, in which each major operating division serves a distinct market segment, retains considerable autonomy, and keeps its own books, and is evaluated using the DuPont system of financial measurement. Short run coordination between GM's consumer goods divisions and the divisions making components was achieved via buyer-seller relationships -- quasi- arm's length transfer pricing arrangements. Longer run coordination was achieved via the first modern capital budgeting system used in the US. GM's organizational innovations were widely emulated by American businesses during the 1950s and sixties.

Improvements in educational levels and advances in automation have reduced the relative efficacy of bureaucratic personnel systems: control by rules and standard operating procedures, task specialization, and sequential processing. Indeed, these have been superceded in many industries by modern, people-based human resources management practices: self-managed teams, control built into job design, and decentralization of decision-making as basic principles of organization, highly selective hiring of new personnel and employment security, extensive training, comparatively high compensation based on organizational performance, reduced status distinctions and barriers across levels, and extensive sharing of financial and performance information throughout the organization (Pfeffer, 1998). The consequences of these high performance HR practices include faster organizational learning and innovation, greater flexibility, skill acquisition, and productivity, and ultimately improved customer service.

More recently American businesses have abandoned functional compartmentalization along with vertical integration. Arguably, these trends are being driven by reductions in communications, logistics, and information processing costs -- reductions stimulated if not caused by the introduction of computers and by our increasing ability to use them. These reductions are breaking down economies of scale and scope built upon functional specialization and vertical integration. As a result, even large companies are mimicking their smaller competitors: shrinking head offices, removing layers of bureaucracy, and concentrating on core businesses.

This has led to flatter as well as smaller organizations, organized around a set of generic value-creating processes and specific competencies. Some single-mission organizations are now organized as virtual networks, some multi-mission organizations as alliances of networks. Philip Evans and Thomas Wurster (1997) refer to both of these kinds of organizational arrangements as hyperarchies, after the hyperlinks of the World Wide Web. Evans and Wurster assert that these kinds of organizations, like the Internet itself, the architectures of object-oriented software programming, and packet switching in telecommunications, have eliminated the need to channel information, thereby eliminating the tradeoff between information bandwidth (richness) and connectivity (reach). How far hyperarchy will go is an open question. Evans and Wuster claim that it will destroy all hierarchies, whether of logic or power, “with the possibility (or the threat) of random access and information symmetry.”

These changes have already influenced business to a greater or lesser degree. They have had almost no effect on the production and delivery of public services. As a result, productivity growth in the public sector has lagged productivity growth in the private sector by a remarkable degree. Indeed, low government productivity almost wholly explains the gap between value added in manufacturing and in services. Evidently, value-added per worker is only 5 percent lower in private services than in manufacturing, but government productivity lags manufacturing productivity by a third. The 20 percent of the American workforce employed by government generates less than 15 percent of total output. This means that, if government workers were as productive as nongovernmental workers, GDP would be five percent higher.  More dramatically, had value-added per government worker increased at the same rate as in the goods sector from mid-century on, GDP would have been thirteen percent higher than in 2003 ($1.4 trillion, about three-fourths of total federal, state, and local expenditures).

Moreover, we now live in an economy where workers demand autonomy and citizens/customers demand superior service and more choice. Old-fashioned business bureaucracies cannot meet these demands; neither can old-fashioned government bureaucracies.

What the new public management calls for is the adoption of the organizational designs and practices that are transforming business: decentralized, flatter, perhaps smaller, organizations, organized around sets of generic value-creating processes and specific competencies, high performance HRM practices, modern information technology, balanced responsibility budgeting and control systems, and loose alliances of networks (Jones & Thompson, 1999). An example of what I am talking about is the New Zealand Post, which under its CEO, Elmar Toime, transformed itself from a typical bureaucracy to a profitable state-owned enterprise and the most efficient postal service in the world. This entailed a 30 percent reduction in workforce, but because of changes in organizational design and HRM practices, “these reductions were accomplished without leaving the organization weakened by a distrustful and unmotivated workforce” (Pfeffer, 1998: 186).

Ultimately, we favor these things not only because we want to make the public sector more productive, but also because we want it to be more democratic. Old style bureaucracy is authoritarian and hierarchical, those attributes never comported well with democratic values. Moreover, the requirements of directing giant, vertically integrated, functional organizations has tended to overwhelm the capacity of the public and its elected representatives to attend to the general welfare. Limiting the scope of the public sector to the provision of services that truly are infused with the common interest cannot but enhance the efficacy of democratic governance mechanisms.

Business Values and Bureaucracy

Thompson’s interpretation of organizational history and the buildup of bureaucracy convey an impressive amount of intelligence in a very short space. He is indeed the master of efficiency. His descriptions are vivid and illuminating, even though his announcement of the fall of bureaucracy is premature. His prescriptions are well-intentioned, though overly imbued with business sector zeal.

Thompson writes beautifully, but what is he prescribing? Two things we already have enough of: Old clichés about the superiority of business practices over public administration practices, and gizmo-sounding slogans offered as management reforms. He says we need “high performance HRM practices,” along with “generic value-creating processes” and “balanced responsibility budgeting.”

Thompson presents service sector productivity data as if they were established facts rather than controversial assertions. Economists openly acknowledge the trouble with measuring service-sector productivity. Moreover, rapid productivity increases may be more difficult to accomplish in services than in goods. Ignoring these controversies, Thompson implies that there is a public employee problem behind the productivity data.

By drawing a circle around a bullet hole, Thompson identifies the New Zealand postal service as a bulls-eye, an exemplar of what the remaining hundreds of thousands of public sector organizations could do if they would just transform themselves into “decentralized, flatter, perhaps smaller, organizations, organized around sets of generic value-creating processes and specific competencies, high performance HRM practices, modern information technology, balanced responsibility budgeting and control systems, and loose alliances of networks.”

The predictable upshot of Thompson’s essay is his claim that productivity growth in government is less than productivity growth in the private sector, and, moreover, it is up to the public sector to remake itself in the image of business. Never mind that 4th quarter 2002 productivity improvement (the most recent available at this writing) in the U.S. non-farm business sector showed a productivity decline of .2 percent.  (And yet no one views this as an employee issue.)

Two things are ignored amidst the praise for whiz-bang new organizational designs that have supposedly displaced old-time bureaucracy in the private sector but not in the public sector: a) The displacement of old-time bureaucracy with new forms has not occurred in the private sector; b) There have been tremendous changes in service delivery instruments that have taken place in the public sector, including contracting out, policy networks, interagency task forces, and special district governments.

A. Bureaucracy as rhetorical foil

While the term “bureaucracy” always makes good cannon fodder, the reports of its demise are premature. On one hand, bureaucracy is not as totalizing as Thompson makes it out to be. Henry Ford’s company never did, in fact, make “everything he needed for his cars from the raw materials on up.” The “total vertical integration” that bureaucracy supposedly achieved is a chimera. On the other hand, the reforms that Thompson offers are an extension of the same instrumental rationality that also sustains bureaucracy. Precision, speed, clarity in communication, reduction of friction, reduction of personnel costs -- these are the technical advantages of bureaucracy, according to Max Weber. It is no coincidence that Thompson’s management reforms are intended to achieve these same sorts of effects.

Meanwhile, the characteristics that make bureaucracy what it is remain intact: “The principles of office hierarchy,” “levels of graded authority,” and “a firmly ordered system of super- and subordination” (Weber 1946: 197). The suppositions of expert training remain. Duties of the job continue to demand the full working capacity of the employee. General office rules, to this day, must be learned and followed.

Moving beyond bureaucracy in a serious way involves a critical assessment of the instrumental rationality that informs modern organizations of all kinds. Max Weber (1946) said that the decisive reason for the advance of bureaucratic organization “has always been its technical superiority over any other form of organization” (p. 214). It is worth remembering the source of such advance:

Today, it is primarily the capitalist market economy which demands that the official business of the administration be discharged precisely, unambiguously, continuously, and with as much speed as possible. Normally, the very large modern capitalist enterprises are themselves unequalled models of strict bureaucratic organization. Business management throughout rests on increasing precision, steadiness, and, above all, the speed of operations (Weber 1946: 215).

The ethic of instrumental rationality that capitalism and bureaucracy have brought about is now so thorough-going that bureaucracy -- the primary vehicle for spreading instrumental rationality into the culture -- is ridiculed for no longer being efficient enough. Yet bureaucracy has set the terms, and the game is still being played according to the rules of bureaucratic rationality. Efficiency and rapid, unambiguous communication remain the foremost criteria that are used to assess the “technical superiority” of organizational form. Hence the “fall” of bureaucracy in the corporate sector is surely an over-statement. Instead, performance management, outcome measurement, and results-oriented managerialism are tributes to the triumph of bureaucracy’s core value, instrumental rationality.

B. Democratic innovations

Thompson wants public sector bureaucracies to operate more like a business, and thinks that democratic values run in precisely the same direction. We can agree that democratic structural innovations that increase information flow (bottom-up as well as top-down) should be developed in the public sector -- and, indeed, this is happening in a vibrant way. However, Thompson ignores the rapidly expanding body of research on policy networks, decentered governance, and deliberative policy implementation in Denmark, England, USA, and the Netherlands. Public administration theorists concerned about democratic governance watch these experiments and innovations closely.

Amidst increasing specialization (which began as a bureaucratic phenomenon), it seems as though the literature on organizational efficiency and the literature on democratic policy implementation exist in separate spheres of specialization.

Even when management reforms can be justified on efficiency grounds, it would be claiming too much to assert that one has, thereby, justified the reforms on democratic grounds as well. Can efficiency and democracy be discussed on the same page? Some say yes. As Thompson puts it “Ultimately, we favor these things not only because we want to make the public sector more productive, but also because we want it to be more democratic.”

However, when the discussion shifts from efficiency to democracy, the politics are exposed. Conservative ideology equates democracy with limited government, rather than with expanded participation in it. Thompson says, “Limiting the scope of the public sector to the provision of services that truly are infused with the common interest cannot but enhance the efficacy of democratic governance mechanisms.” One begins to suspect that, by the end of the essay, Thompson’s critique of public sector productivity, on the supposedly politically neutral terrain of efficiency, has transformed itself into an essay on limited government. An economic definition of democracy, like the others, has its upside and its downside.

C. The business ethic

Thompson ignores the downside of the business ethic, dubbed “infectious greed” by Alan Greenspan. Avarice would also be a good term to use. In the wake of the scandals at Global Crossing, Adelphia, Enron, Arthur Anderson, WorldCom, and so on, it takes chutzpah to hold the business model up as one worth emulating in the public sector. Instead, the indulgent and dangerously righteous business model generates a public sector ethics that should give one pause.

Consider this example from my home area in South Florida. The Pembroke Pines city manager Charles F. Dodge and his assistant Martin J. Gayeski were paid $306,000 per year between them, and they now will be entitled to pension benefits because they are retiring.  This story (from Miami Herald reporter Scott Andron) would be ho-hum news, except for one catch.

Now Dodge and his subordinate will be paid $3,345,000 over five years -- as consultants for Charles F. Dodge LLC the brand new consulting firm that just won a non-competitive contract to manage the city. I wonder if this is the sort of high-performance HRM practice that Thompson has in mind.

There are several features of the deal worth considering:

  1. The city will now pay $363,000 per year additional for city management services.
  2. Before the last city council meeting the city was paying $306,000 per year total. The new total cost is $669,000 per year, more than double.
  3. The former city manager Dodge (now consultant Dodge) will begin immediately to collect pension benefits (of unknown amount).
  4. There was no competitive bidding process. The city’s attorney said there was no need for it.

According to city council members, Dodge and Gayeski were doing an outstanding job managing the city. Dodge justified his contract amount by touting several personnel inactions. He saved the city money by not hiring an education director to oversee the city’s charter schools; by not hiring a project manger to oversee development of a park; by not hiring an additional assistant city manager. The city council was persuaded by this argument, and I would not be surprised if advocates of “new public management” would likewise be persuaded.

What is my objection? In a culture dominated by corporate practices and business ethics, limited government has slightly different connotations than it did for the American founders when the business corporation had no legal status. Unlike classic liberal political theory that extolled civil society and the autonomous individual, Thompson extols corporate practices.

Thompson’s reforms are presented as good for democracy. They might save taxpayers money and they might make governmental organizations more effective. These would be good things. But are they democratic in any robust sense of the term? I doubt it. Something more profound is happening, instead. The instrumentally rational predisposition, honed and improved though corporate practices, stands ready to colonize culture and its politics.


Government does not have an employee problem; it has a public-management policy problem. According to the Volker Commission:

• The federal government is not performing nearly as well as it can or should. The difficulties federal workers encounter in just getting their jobs done has led to discouragement and low morale.

• A clear sense of policy direction and clarity of mission is too often lacking, under-cutting efficiency and public confidence. As a result, there is real danger of healthy public skepticism giving way to corrosive cynicism.

To me democracy means self-determination -- the power to influence decisions that are central to one’s life. It’s true: I like the idea of limited government. Government works best where participation in public-policy making is widespread. Unfortunately, because people treat participation in decision-making as a cost and because they do not see government as central to their lives, many opt out. One does not have to be a right winger to believe that government performance might be improved or that the citizenry's abilities to control its elected agents might be enhanced if their attentions and responsibilities were suitably focused on fundamental issues.

More importantly, work is central to our lives. But, at work, we are often silenced. “The principles of office hierarchy,” “levels of graded authority,” and “a firmly ordered system of super- and subordination” are inimical to democracy.  They are also increasingly inimical to high performance. Nowadays, high performing entities are more likely to be designed around team-based collaborations that successfully spread authority and responsibility throughout the organization and thereby mobilize the collective intelligences of their members.

IBM’s Dallas TX facility is an example. It mimics a market. Everyone is either a customer or provider, depending on the transaction, thereby transforming the facility into a network of exchanges. Each exchange is a closed loop involving four steps: request from a customer and offer from a provider, negotiation of the task to be performed and definition of success, performance, and customer acceptance. Until the last step is completed, the task is unfinished. IBM uses powerful computers to track these loops and monitor the progress of each transaction. The result has been to empower workers, eliminate boundaries and bottlenecks, and boost productivity through the ceiling.

As Hugh Miller notes, some government organizations (e.g., Oregon’s housing authority under the leadership of Rey Ramsey [Hecht & Ramsey, 2001]) have copied well-managed businesses by organizing themselves into similar alliances of networks, sharing their top management, core competencies, and a common culture, and using computers to chart activities and operational flows. Their control systems are like those of centralized bureaucracies in that they collect a lot of real-time information on operations. Unlike the control systems of stove-piped centralized bureaucracies, however, which passed the exercise of judgment up the managerial ranks, this information is used to push it down into the organization, to wherever it is most needed, at delivery, in production, or to the client.

Anyone committed to democratic governance should be truly excited by these innovations.


Beniger, James R., The Control Revolution: Technological and Economic Origins of the Information Society (1986).

Chandler, A.D. The Visible Hand: The Managerial Revolution in American Business (1977).

Evans, P.B., and T.S. Wurster “Strategy and the New Economics of Information,” Harvard Business Review, September-October, 1997: 71-82.

Hecht, Ben, and Rey Ramsey, ManagingNonprofits.org: Dynamic Management for the Digital Age (2001).

Jones, L.R., Fred Thompson, Public Management: Institutional Renewal for the 21st Century (1999).

Pfeffer, Jeffrey, The Human Equation: Building Profits by Putting People First (1998).

Rauch, James E.  “Bureaucracy, Infrastructure, and Economic Growth: Evidence from U.S. Cities During the Progressive Era,” American Economic Review.  85/4 (Sept. 1995), 968-979.

Reschenthaler, G.B., and Thompson, Fred  “The Information Revolution and the New Public Management.” Journal of Public Administration Research and Theory.  6/1 (1996) 125-144.

Thompson. Fred. “Fordism and PostFordism,” Encyclopedia of Political Economy. Routledge: London, 1998: 404-407.

Weber, Max. 1946. “Bureaucracy.” Chapter VIII in H.H. Gerth and C. Wright Mills, eds. and trans., From Max Weber: Essays in Sociology. New York: Oxford University Press, pp. 196-252.

Womack, James P., Daniel T. Jones, Daniel Roos, The Machine that Changed the World (1990).

The basic texts on information costs include Kenneth Arrow, The Organization of Economic Activity (1969); Yoram Barzel, Measurement Costs and the Organization of Markets. Journal of Law and Economics. 25/1: 27-48 (1982); Paul Milgrom, John Roberts, Economics, Organization, and Management (1992); and Oliver E Williamson, The Economic Institutions of Capitalism (1985) and Organization Theory: From Chester Barnard to the Present and Beyond (1990).

Comparative analysis of the industrial and bureaucratic revolutions are found in Karl Polanyi, The Great Transformation (1944, reprinted 1985) and Nathan Rosenberg, L.E. Birdsall, How the West Grew Rich: The Economic Transformation of the Industrial World (1986). It is reasonable to conclude that the changes that led to mass production also promoted mass politics and that product standardization encouraged egalitarianism. Mass production required the organization of vast numbers of unskilled laborers to work assembly lines. Like circumstances imply like interests. Widespread common interests led to powerful industrial unions and many instances popular political mobilization as well. Mass marketing and product standardization had a significant leveling effect with respect to consumption, reducing the benefits of relative wealth and, as a result perhaps, organized opposition to relatively egalitarian social policies. Consequently, from the standpoint of the argument made here and with the benefit of 20/20 hindsight, the bureaucratic (authoritarian) populism (socialism) of the first half of 20th century now appears to be an almost irresistible consequence of industrialization.

The Prussian military's role in the development of both the factory system and bureaucratic organizations is described in Andre Corvisier, Armies and Societies in Europe, 1492-1789, translated by A.T. Siddall (1979); Martin van Creveld, Command in War (1985) and Supplying War: Logistics from Wallenstein to Patton (1977); T.N. Dupuy, A Genius for War: The German Army and the General Staff, 1807-1945 (1977); Archer Jones, Civil War Command & Strategy (1992); and Walter Millis, Arms and Men: A Study of American Military History (1956). The adoption of Prussian practices in the public sector in the US is explicitly described in Ben W. Twight, Organizational Values and Political Power (1983); see also James Weinstein, The Corporate Ideal in the Liberal State: 1900-1918 (1968).

The new paradigm in government is best described by Michael Barzelay, Breaking through Bureaucracy (1992).





Hugh T. Miller is Professor and Director of the School of Public Administration at Florida Atlantic University. He is the author of Postmodern Public Policy (Albany: State University of New York Press, 2002).

Fred Thompson is Grace and Elmer Goudy Professor of Public Management & Policy, Atkinson Graduate School of Management, Willamette University. He edits the International Public Management Journal and is a CARR fellow, London School of Economics, and confesses to being a managerialist and an instrumental rationalist (is there any other kind?).