Global Foundation of Public Administration

Managing Government, Governing Management

by Henry Mintzberg

Reprint 96306

Harvard Business Review

MAY-JUNE 1996

Reprint Number

Harvard Business Review HENRY MINTZBERG MANAGING GOVERNMENT, GOVERNING MANAGEMENT 96306

PAUL STREBEL WHY DO EMPLOYEES RESIST CHANGE? 96310

T.J. LARKIN, REACHING AND CHANGING FRONTLINE EMPLOYEES 96304 SANDAR LARKIN

CHARLES M. FARKAS, THE WAYS CHIEF EXECUTIVE OFFICERS LEAD 96303 SUZY WETLAUFER

ELECTRONIC COMMUNITIES AND THE NEW COMMERCE

DEBORA SPAR, JEFFREY J. BUSSGANG RULING THE NET 96309

ARTHUR ARMSTRONG, JOHN HAGEL III THE REAL VALUE OF ON-LINE COMMUNITIES 96301

KEITH CERNY HBR CASE STUDY MAKING LOCAL KNOWLEDGE GLOBAL 96302

V. KASTURI RANGAN, SOHEL KARIM, SOCIAL ENTERPRISE SHERYL K. SANDBERG DO BETTER AT DOING GOOD 96308

IAN C. MACMILLAN, MANAGER’S TOOL KIT RITA GUNTHER MCGRATH DISCOVER YOUR PRODUCTS’ HIDDEN POTENTIAL 96305

KARL WEICK BOOKS IN REVIEW PREPARE YOUR ORGANIZATION TO FIGHT FIRES 96311

PERSPECTIVES ELECTRIC UTILITY DEREGULATION SPARKS CONTROVERSY 96307

“Capitalism has triumphed.” That was the pat conclusion reached in the West as, one by one, the communist regimes of Eastern Europe began to fall. It has become such an article of faith that we have become blind to its effects. Those effects are highly negative – indeed, dangerous – because the conclu- sion itself is wrong. In my view, we have confound- ed the whole relationship between business and government, and we had best clear it up before we end up no better off than the Eastern Europeans once were.

The Triumph of Balance Capitalism did not triumph at all; balance did.

We in the West have been living in balanced soci- eties with strong private sectors, strong public sec- tors, and great strength in the sectors in between. The countries under communism were totally out of balance. In those countries, the state controlled an enormous proportion of all organized activity. There was little or no countervailing force. Indeed, the first crack in the Eastern bloc appeared in the one place (Poland) where such a force had survived (the Catholic Church).

The belief that capitalism has triumphed is now throwing the societies of the West out of balance, especially the United Kingdom and the United States. That the imbalance will favor private rather than state ownership will not help society. I take issue with Milton Friedman of the University of Chicago, who has been fond of comparing what he calls “free enterprise” with “subversive” socialism. The very notion that an institution, independent of the people who constitute it, can be free is itself a subversive notion in a democratic society. When the enterprises are really free, the people are not.

Indeed, there is a role in our society for different kinds of organizations and for the different contri- butions they make in such areas as research, educa- tion, and health care. The capitalism of privately owned corporations has certainly served us well for the distribution of goods and services that are ap-

HARVARD BUSINESS REVIEW May-June 1996 Copyright © 1996 by the President and Fellows of Harvard College. All rights reserved.

HBR M A Y - J U N E 1 9 9 6

by Henry Mintzberg

Henry Mintzberg is a professor of management at McGill University in Montreal, Canada, and at INSEAD in Fontainebleau, France. He is currently overseeing a ven- ture by five business schools in Canada, England, France, India, and Japan. The venture is seeking to create a next-generation master’s program for the development of managers.

MANAGINGGOVERNMENT

GOVERNINGMANAGEMENT

propriately controlled by open-market forces. The books published by Friedman and his colleagues are goods of that kind. But is their research? Or the health care received by poor people living near those professors’ offices?

Beyond Public and Private For as long as anyone cares to remember, we have

been mired in a debate over the allocation of re- sources between the so-called private and public sectors. Whether it is capitalism versus commu- nism, privatization versus nationalization, or the markets of business versus the controls of govern- ment, the arguments have always pitted private, in- dependent forces against public, collective ones. It is time we recognized how limited that dichotomy really is.

There are privately owned organizations, to be sure, whether closely held by individuals or widely held in the form of market-traded shares. And there are publicly owned organizations, although they should really be called state owned, because the state acts on behalf of the public. We as citizens no more control our public organizations directly than we as customers (or as small shareholders) control the private ones. But there are two other types of ownership that deserve equal attention.

First, there are cooperatively owned organiza- tions, whether controlled formally by their sup- pliers (as in agricultural cooperatives), by their customers (as in mutual insurance companies or cooperative retail chains), or by their employees (as in some commercial enterprises, such as Avis). Indeed, all countries in the West, including the United States, are to a large extent societies of co- operatively owned organizations. According to the National Cooperative Business Association, almost half of the U.S. population is directly served by some cooperative endeavor, and one in three people is a member of a cooperative. I did some work re- cently for a major U.S. mutual insurance company. The enterprise is vigorously competitive, yet it benefits from being cooperatively owned. Its execu- tives are quick to point out just how important the absence of stock market pressures is for their abili- ty to take a long-term perspective.

Second, we have what I call nonowned organiza- tions, controlled by self-selecting and often very diverse boards of directors. These not-for-profit organizations are often referred to as nongovern- ment organizations (NGOs), but they are also non- business and noncooperative organizations (NBOs and NCOs). Indeed, we are surrounded by non- owned organizations. Among them are many of our

universities (including Friedman’s University of Chicago), hospitals, charity organizations, and vol- unteer and activist organizations (the Red Cross and Greenpeace, for example).

From a conventional political perspective, the in- clination might be to lay out these four forms of ownership along a straight line from left (state own- ership) to right (private ownership), with coopera- tive ownership and nonownership in between. But I believe that would be a mistake because, here as elsewhere, extremes meet: It is the ends that are most alike. For example, from the point of view of structure, both private and state organizations are tightly and directly controlled through hierarchies – one emanating from the owners, the other from state authorities. In other words, we should fold that line over. What seems like a straight line is really more like a horseshoe.

As a horseshoe-shaped representation of the four forms of ownership would suggest, the leap be- tween state and private ownership can be made more easily than a shift to nonownership or cooper- ative ownership. That may be why so much of our attention has focused on nationalization versus pri- vatization. The leap is so simple: Just buy out the other side, change the directors, and keep going; the internal control systems remain intact. In Russia today, in many sectors, these leaps have been too simple: State control seems to have given way to equally devastating control by the private sector. A surer way of achieving balance – slower and more difficult but now being pursued successfully in some of the other Eastern European nations – is to make wider use of all four forms of ownership around the entire horseshoe.

Unfortunately, we in the West have not come to terms with the full range of possibilities. Because capitalism has supposedly triumphed, the private sector has become good, the public sector bad, and the cooperatively owned and nonowned sectors ir- relevant. Above all, say many experts, government must become more like business. It is especially this proposition that I wish to contest. If we are to manage government properly, then we must learn to govern management.

Customers, Clients, Citizens, and Subjects

“We have customers,” Vice President Al Gore an- nounced early in his term in office. “The American people.” But do you have to call people customers to treat them decently? We would do well to take a look at what customers, this now fashionable word, used to mean before the Japanese taught us a thing

MANAGING GOVERNMENT

76 PHOTOS BY CHRISTOPHER MAKOS/ALL HATS COURTESY OF BARNEYS NEW YORK

or two. The greatest of the U.S. corporations – those of the automobile industry – did not treat their cus- tomers very well. They long pursued deliberate strategies of planned obsolescence – a euphemism for building quality out. Moreover, at least one giant retail chain regularly used bait-and-switch tactics, luring in customers with low prices to sell them more expensive products. And in one well- known story, a famous consumer-products com- pany, in order to sell more toothpaste, first made the opening in its tubes bigger and then marketed toothbrushes with longer heads!

Business is in the business of selling us as much as it possibly can, maintaining an arm’s-length rela- tionship controlled by the forces of supply and demand. I have no trouble with that notion – for cars, washing machines, or toothpaste. But I do for health care. For cars, washing machines, and toothpaste, most in- telligent buyers can beware, as the expression goes; and we have protec- tive mechanisms in place for buyers who cannot beware. But caveat emp- tor is a dangerous philosophy for health care and other complex pro- fessional services. Sellers inevitably know a great deal more than buyers, who can find out what they need to know only with great difficulty. In other words, the private ownership model, much as it provides “cus- tomers” with a wonderfully eclectic marketplace, does have its limits.

I am not a mere customer of my government, thank you. I expect something more than arm’s- length trading and something less than the encour- agement to consume. When I receive a professional service from government – education, for example – the label client seems more appropriate to my role. (General Motors sells automobiles to its customers; Ernst and Young provides accounting services to its clients.) In fact, a great many of the services I re- ceive from government are professional in nature.

But, most important, I am a citizen, with rights that go far beyond those of customers or even clients. Most of the services provided by govern- ment, including highways, social security, and eco- nomic policy, involve complex trade-offs between competing interests. Tom Peters captures this idea perfectly with a story about getting a building per- mit to enlarge his house. I don’t want some bureau- crat at City Hall giving me a hard time, he said in one of his newsletters. I want proper, quick, busi- nesslike treatment. But what if my neighbor wants a permit to enlarge his house? Who’s City Hall’s customer then?

If I have rights as a citizen, then I also have obli- gations as a subject. The British, of course, retain official status as subjects of the crown – a throw- back to the days when individuals forfeited much of their autonomy over their “nasty, brutish, and short” lives, as Thomas Hobbes put it, in return for the protection of the state. But, British or not, in one way or another we all remain subjects of our governments – when we pay taxes, allow ourselves to be drafted into armies, or respect government regulations for the sake of collective order.

Customer, client, citizen, and subject: These are the four hats we all wear in society. As customers and citizens, we enjoy a reciprocal, give-and-take relationship with government. Government’s cus- tomers receive direct services at arm’s length; its citizens benefit more indirectly from the public

infrastructure it provides. But there is one major dif- ference between government’s customer-oriented activities and its citizen-oriented activities: fre- quency of occurrence. Review public sector activi- ties carefully – for example, go over a government telephone directory – and you will find relatively little that fits the pure customer category. (And some of what does fit is rather unfortunate, such as lottery tickets. Do we really want our governments, like that toothpaste company, hawking products? Couldn’t the current malaise about government really stem from its being too much like business rather than not enough?) In contrast, under the citi- zen category, you will find an enormous amount of activity in the form of public infrastructure: social infrastructure (such as museums), physical (such as roads and ports), economic (such as monetary poli- cy), mediative (such as civil courts), offshore (such as embassies), and the government’s own support infrastructure (such as election machinery).

As subjects and clients, we have relationships with government that are more one-sided. To para- phrase John F. Kennedy, the question for us as sub- jects is what we must do for our governments in the form of respecting state controls. In contrast, as clients who receive professional services, our ques- tion is about what the state provides to us. That

HARVARD BUSINESS REVIEW May-June 1996 77

Above all, say many experts, government must become more

like business. It is especially this notion that I wish to contest.

government phone book reveals all kinds of activi- ties under the subject category – policing, the mili- tary, regulatory agencies, and prisons. But more sur- prising is the prevalence of professional services that governments provide directly, or indirectly through public funding: all of the health care in some countries and much of it in others, much of education, plus other services such as meteorology.

Of course, not all government activities fit neatly into one of the four categories. Our national parks, for example, provide customer services (to tourists) and professional client services (to tourists strand- ed on mountain faces). Parks are also part of the public infrastructure we enjoy as citizens, and that fact requires us, as subjects, to respect the environ-

ment of the park. To take another example, the in- mates of prisons are most evidently subjects. But they remain citizens with certain rights and, inso- far as we believe in the role of rehabilitation, are clients as well. I introduce these four labels, there- fore, not so much for classification as for clarifica- tion – to further our appreciation of the varied pur- poses of government.

Let me link the roles of customer, citizen, client, and subject to our earlier discussion. Customers are appropriately served by privately owned organiza- tions, although cooperatively owned ones – such as mutual insurance companies – can often do the job effectively. Only in limited spheres is direct cus- tomer service a job for the state. When it comes to citizen and subject activities, we should stray be- yond the state-ownership model only with a great deal of prudence. The trade-offs among conflicting interests in citizen activities and the necessary use of authority in subject activities mandate a clear role for the state.

The client relationship is perhaps more compli- cated. It is not clear that those professional services widely accepted as public – certain minimum levels of education and of health care, for example – are particularly effective when offered directly by gov- ernment, let alone by private business. Neither one on its own can deliver all the nuanced requirements of professional services. Markets are crass; hierar- chies are crude. Nonowned organizations or, in cer- tain cases, cooperatively owned ones may serve us

better here, albeit with public funding to ensure some equity in distribution. Incidentally, relying on cooperatively owned organizations for profes- sional services is not unusual. Even such obviously commercial professions as accounting and consult- ing often deliver services through cooperatives – namely, professional partnerships.

The Myths of Management We have seen that a balanced society requires

various institutional forms of ownership and con- trol and that within the public sector there is a wide range of roles for government. How, then, should government activities be managed? To answer the

question, we first need to take a look at management itself – or at least at the popular myths about it.

Discussion of management is cur- rently all the rage. I should really say Management, following the lead of Albert Shapero of Ohio State Univer- sity, who years ago wrote an article titled “What MANAGEMENT Says

and What Managers Do” (Fortune, May 1976). We are talking about his capital-letter Management here – a narrow, stylized process that, according to my research, has surprisingly little connection with what effective managers actually do. Yet this is the kind of management that now inundates us – in bookstores, M.B.A. programs, and hyped training seminars, for example. But does it really apply to our roles as citizens, subjects, clients, and occasion- ally customers of government? Rarely, in my opin- ion. Let me explain.

Three assumptions underlie the Management view of management. M Particular activities can be isolated – both from one another and from direct authority. The princi- ple derives from the private sector, where many corporations are divided into autonomous busi- nesses, organized as divisions. Each unit has a clear mission: to deliver its own set of products or ser- vices. If it satisfies the goals set by the central head- quarters, it is more or less left alone. M Performance can be fully and properly evaluated by objective measures. The goals that each activity must achieve can be expressed in quantitative terms: Both costs and benefits can be measured. (In business, of course, the criteria are financial, and costs and benefits are combined to set standards for profit and for return on investment.) That way, there can be “objective” assessment, which is apolitical in nature. The system cannot afford a great deal of distracting ambiguity or nuance.

MANAGING GOVERNMENT

78 HARVARD BUSINESS REVIEW May-June 1996

I am not a mere customer of my government. I expect something more than arm’s-length trading.

M Activities can be entrusted to autonomous pro- fessional managers held responsible for perfor- mance. “Let the managers manage,” people say. Many have great faith in managers trained in the so-called profession of management. “Make them accountable. If they perform according to plan, as indicated by measurement, reward them. If they don’t, replace them.”

These assumptions, in my opinion, collapse in the face of what most government agencies do and how they have to work. To isolate government ac- tivities from direct hierarchical control in the man- ner that Management prescribes, there have to be clear, unambiguous policies formulated in the po- litical sphere for implementation in the adminis- trative sphere. In other words, policies have to be rather stable over time, and politicians (as well as managers of other agencies) have to stand clear of the execution of those policies. How common is that? How many government activities fit such a prescription?

Lotteries, to be sure, but what else? Less than you might think. Many government activities are inter- connected and cannot be isolated. Foreign policy, for example, cannot be identified with any one department, let alone any one agency. There are, of course, public sector activities that can be isolated horizontally from one another more or less, as in the case of police or prison services. But can they be isolated vertically – from the political process? Certainly, there has been no shortage of effort to isolate them. A few years ago, the United Kingdom made its prison service an ostensibly autonomous executive agency and appointed a high-flying busi- ness manager to run it. Recently, in a major scan- dal, the manager was fired – apparent- ly because he would not dismiss one of the wardens after a highly publi- cized escape of three inmates. On leaving, he complained to the press that there was more political con- trol over the service after it became “autonomous” than before.

How many politicians are pre- pared to relinquish control of how many of their policies? And how many policies in government today can simply be formulated in one place to be implemented in an- other, instead of being crafted in an iterative process involving both politics and administration? Learning is another of the current buzzwords of Management. Well, this process of crafting policies is learning – mindlessly applying them is not. The belief that politics and administration in govern- ment – like formulation and implementation in cor-

porate planning – can be separated is another old myth that should be allowed to die a quiet death.

Next consider the myth of measurement, an ide- ology embraced with almost religious fervor by the Management movement. What is its effect in gov- ernment? Things have to be measured, to be sure, especially costs. But how many of the real benefits of government activities lend themselves to such measurement? Some rather simple and directly de- livered ones do – especially at the municipal level – such as garbage collection. But what about the rest? Robert McNamara’s famous planning, program- ming, and budgeting systems in the U.S. federal government failed for this reason: Measurement often missed the point, sometimes causing awful distortions. (Remember the body counts of Viet- nam?) How many times do we have to come back to this one until we finally give up? Many activities are in the public sector precisely because of measure- ment problems: If everything was so crystal clear and every benefit so easily attributable, those activ- ities would have been in the private sector long ago.

Consider an example from England’s public- sector health care. A liver transplant surgeon in the National Health Service operated on ten patients. Two died. Of the eight who survived, one who had had cancer years earlier suffered a recurrence. An- other patient’s liver began to fail, and he needed a second transplant. Of the remaining six patients, only three were able to resume normal working lives. Asked about his performance, the surgeon claimed his success rate as 8 in 10. (Indeed, as soon as he replaced that failing liver, he was prepared to claim 9 in 11. He counted livers, not people.) An immunologist put it at 7 in 10, believing that the

surgeon should not have operated on the person who had had cancer. A cost-conscious hospital ad- ministrator put the figure at 6 in 10. The nurses claimed 3 in 10, taking into account postoperative quality of life.

Now, picture yourself having to make your own assessment. Where is the magic envelope with the one right answer? You won’t find it. The fact is that assessment of many of the most common activities

HARVARD BUSINESS REVIEW May-June 1996 79

Our faith in managers trained in the profession of management

collapses in the face of how government agencies must work.

in government requires soft judgment – something that hard measurement cannot provide. So when Management is allowed to take over, it drives everyone crazy. And no one more so than the “cus- tomer,” who ends up getting the worst of it.

Finally, there is the myth that the professional manager can solve everything: “Put someone prop-

erly trained in charge, and all will be well.” We are so enamored of this cult of heroic leadership that we fail to see its obvious contradictions. For exam- ple, in the name of empowering the workers, we ac- tually reinforce the hierarchy. So-called empower- ment becomes the empty gift of the bosses, who remain firmly in charge. And those bosses, if knowledgeable about nothing but Management it- self, sit in midair, all too often ignorant of the sub- ject of their management. Such a situation just breeds cynicism. In mortal fear of not meeting the holy numbers, managers run around reorganizing constantly, engendering more confusion than clari- fication. In other words, our obsession with Man- agement belies a good deal of the reality out there. Consequently, it distorts serious activities, as in the case of many public school systems that have been virtually destroyed by the power of the man- agerial hierarchy to direct classroom activities without ever having to teach anything.

Models for Managing Government How then should government be managed? Let’s

consider five models. Each is marked by its own way of organizing government’s controlling author- ity, or superstructure, and the activities of its agen- cies, or microstructure. (The budget authority would be part of the former, for example; an envi- ronmental protection agency, an example of the lat- ter.) Some of the models are older, some newer. Some we could do with less of, despite their current popularity; others we could use more of, despite their unfamiliarity.

The Government-as-Machine Model. Govern- ment here is viewed as a machine dominated by rules, regulations, and standards of all kinds. This

applies to the superstructure no less than to each of the microstructures. Each agency controls its peo- ple and its activities just as the agency itself is con- trolled by the central state apparatus. Government thus takes on the form of a hologram: Examine any one piece and it looks just like the rest.

This has been the dominant model in govern- ment, almost to the exclusion of everything else. As Frederick Tay- lor’s “one best way,” it was popular- ized in the 1930s in the public sector by Luther Gulick and Lyndall Ur- wick. Its motto might be Control, Control, Control. In fact, the term bureaucrat, for civil servant, comes from the influence of this model.

The machine model developed as the major countervailing force to

corruption and to the arbitrary use of political in- fluence. That is why it became so popular earlier in this century. It offered consistency in policy and re- liability in execution. But it lacked flexibility and responsiveness to individual initiative, so now it has fallen out of favor. In one form or another, how- ever, the machine model continues to dominate government.

The Government-as-Network Model. This is the opposite of the machine model: loose instead of tight, free-flowing instead of controlled, interactive instead of sharply segmented. Gover nment is viewed as one intertwined system, a complex net- work of temporary relationships fashioned to work out problems as they arise and linked by informal channels of communication. At the micro level, work is organized around projects – for example, a project to develop a new policy on welfare or to plan for construction of a new building. Connect, Com- municate, and Collaborate might be the motto of this model. Ironically, like the machine model, the network model is also holographic in that the parts function like the whole: Individual projects func- tion within a web of interrelated projects.

The Performance-Control Model. Capital-letter Management finds its full realization in the performance-control model, the motto of which could be Isolate, Assign, and Measure. This model aims above all to make government more like busi- ness. But we need to be specific here because the ideal is not just any business. There is an assump- tion, not often made explicit, that the ideal is the divisional structure that conglomerates in particu- lar have popularized. The overall organization is split into “businesses” that are assigned perfor- mance targets for which their managers are held accountable. So the superstructure plans and con-

MANAGING GOVERNMENT

80 HARVARD BUSINESS REVIEW May-June 1996

The belief that politics and administration in government can be separated is a myth that should die a quiet death.

trols while the microstructures execute. All very tidy. But not necessarily very effective.

For one thing, few people in business still believe in the conglomerate form of organizing. If the busi- nesses have so little to do with one another, why bother to have them in the same organization? What value is added by a remote headquarters that exercises control of financial performance alone? For another thing, a heavy emphasis on planning and measured performance reinforces conventional hierarchical control at the level of the microstruc- ture, where managers have personal responsibility for attaining impersonal targets. Thus the ultimate effect is to reinforce the old machine model. In other words, the performance model decentralizes in order to centralize; it loosens up in order to tighten up. And tightening up comes at the expense of flex- ibility, creativity, and individual initiative. Thus the brave new world of public management all too often comes down to nothing more than the same old machine management – new labels on the old bottles. It works fine where machine management worked – sometimes even slightly better – but not anywhere else.

The Virtual-Government Model. Carry the per- formance model to its natural limit and you end up with a model that can be called virtual government. Popular in places like the United Kingdom, the United States, and New Zealand, virtual govern- ment contains an assumption that the best govern- ment is no government. Shed it all, we are told, or at least all that it is remotely possi- ble to shed. In virtual government’s perfect world, the microstructures (the activities of agencies) would no longer exist within government. All that kind of work would take place in the private sector. And the super- structure would exist only to the extent needed to arrange for private organizations to provide public services. Thus the motto of this model might be Privatize, Contract, and Negotiate. The model represents the great ex- periment of economists who have never had to manage anything.

The Normative-Control Model. None of the above models has succeeded in structuring social authority adequately. Perhaps that is because social authority is hardly about structures. “It’s all so sim- ple, Anjin-San,” the confused British captain in Shogun, shipwrecked in a strange land, is told by his Japanese lover. “Just change your concept of the world.” Exemplifying a different concept of the world, the normative-control model is not about systems but about soul. Here it is attitudes that

count, not numbers. Control is normative – that is, rooted in values and beliefs.

The model is not well recognized in most West- ern governments these days, let alone in most Western businesses. It hasn’t exactly worked badly for the Japanese, but the more they have demon- strated its superiority in direct competition with the West, the more the West has retreated into its old machine model – or newer versions of it – which works in precisely the opposite way. Once upon a time, however, when there was still the concept of public service, it was really the normative model that managed to keep the machine model function- ing. In other words, service and dedication muted the negative effects of bureaucracy. But much of that attitude is now gone or going quickly.

There are five key elements that characterize the normative model: M Selection. People are chosen by values and atti- tudes rather than just credentials. M Socialization. This element ensures a member- ship dedicated to an integrated social system. M Guidance. Guidance is by accepted principles rather than by imposed plans, by visions rather than by targets. M Responsibility. All members share responsibil- ity. They feel trusted and supported by leaders who practice a craft style of management that is rooted in experience. Inspiration thus replaces so-called empowerment. M Judgment. Performance is judged by experienced

people, including recipients of the service, some of whom sit on representative oversight boards.

The motto of the normative model might be Select, Socialize, and Judge. But the key to all is dedication, which occurs in two directions: by and for the providers of the service. Providers are treat- ed decently and therefore respond in kind. The agencies can still be isolated horizontally, but verti- cal control by the superstructure is normative rather than technocratic. The model allows for rad- ically different microstructures: more missionary, egalitarian, and energized; less machinelike and less hierarchical.

There is no one best model. We currently func- tion with all of them. Tax collection would be in-

HARVARD BUSINESS REVIEW May-June 1996 81

We need to shift emphasis to the normative model, where control is rooted in values and beliefs.

conceivable without a healthy dose of the machine model, as would foreign policy without the net- work model. And no government can function ef- fectively without a significant overlay of normative controls, just as no government today can ignore the need to shed what no longer belongs in the pub- lic sector. Government, in other words, is an enor- mously eclectic system, as varied as life itself (be- cause it deals with almost every conceivable facet of life).

But some models are for the better and some for the worse. We might wish to favor the better. We all recognize the excessive attention given to the machine model. But we should be aware of its resurgence in the performance model. This is not to dismiss the performance model. The quasi- autonomous executive agency is fine for many of the apolitical, straightforward services of govern- ment – such as the passport office. Let’s just keep it there and not pretend it is some kind of new “best way.”

We need to be more appreciative of the network model, which is necessary for so many of the com- plex, unpredictable activities of today’s govern- ments – much of policy making, high-technology services, and research, for example. But reliance on this model can also be overdone. In France, both public and private sectors have long been dominat- ed by a powerful and interconnected élite who move around with a freedom and influence that is proving increasingly stifling to the nation. The net- work system in France could use a lot more agency autonomy to check the power of that élite.

It is my personal belief that we sorely need a major shift of emphasis to the normative model. As the

Japanese have made clear, there is no substitute for human dedication. And although much of Western business needs to take this message to heart, it has become especially important in government – with its vagaries, nuances, and difficult trade-offs among conflicting interests. An organization with- out human commitment is like a person without a soul: Skeleton, flesh, and blood may be able to con- sume and to excrete, but there is no life force. Gov- ernment desperately needs life force.

I believe this conclusion applies especially to client-oriented professional services, such as health care and education, which can never be better than the people who deliver them. We need to free pro- fessionals from both the direct controls of govern- ment bureaucracy and the narrow pressures of mar- ket competition. That is why nonownership and some cooperative ownership seem to work so well in those areas.

Governing Management If any of these ideas make sense, then we must

prove them feasible by beginning to temper the influence that business values and currently popu- lar Management thinking have on other sectors of society. In other words, government may need managing, but management could use a little gov- erning, too. Consider the following propositions: M Business is not all good; government is not all bad. Each has its place in a balanced society along- side cooperative and nonowned organizations. I do not wish to buy my cars from government any more than I wish to receive my policing services from General Motors. And I would like to see both pri-

82 HARVARD BUSINESS REVIEW May-June 1996

Virtual government contains a questionable

assumption that the best government is

no government.

vate and public sectors passed over, for the most part, in the direct delivery of health care in favor of nonowned and cooperatively owned organizations.

Societies get the public services they expect. If people believe that government is bumbling and bureaucratic, then that is what it will be. If, in con- trast, they recognize public service for the noble calling it is, then they will end up with strong gov- ernment. And no nation today can afford anything but strong government. Isn’t it time that all the knee-jerking condemnation of government in the United States stopped? As a Canadian who lives part of the year in France, I can testify that these negative attitudes are proving contagious, and they are doing none of us any good. M Business can learn from government no less than government can learn from business; and both have a great deal to learn from cooperative and nonowned organizations. People in the public sec- tor cope with their own kinds of problems: conflict- ing objectives, multiple stakeholders, and intense political pressure, for example. Yet their problems are becoming increasingly common in the private sector. Many of the most intelligent, articulate, and effective managers I have met work for govern- ment. Unfortunately, they are not very aggressive about letting their ideas be known. Businesspeople profit greatly when they listen to them.

Cooperatives have sophisticated ways of dealing with dispersed constituencies, as well as with spe- cial kinds of customers. And then there is the non- owned organization, wherein we find the West’s fullest realization of the normative model, about which we have much to learn. We can benefit great- ly from the experiences of both – as soon as we get beyond our narrow prejudices. M We need proud, not emasculated, government. Attacks on government are attacks on the fabric of society. We have individual needs, to be sure, but a society that allows them to undermine collective needs will soon destroy itself. We all value private goods, but they are worthless without public goods – such as policing and economic policies – to protect them.

Making numerous political appointments is now considered a natural part of the U.S. political process. (This was not always the case: Such ap- pointments are proportionately three times more common today than they were in the 1930s.)1 Each new administration simply replaces the top layers of the departmental hierarchies. I believe it is time that this was recognized for exactly what much of it is: political corruption; not technically illegal but nonetheless corrupting of a dedicated and experi- enced public service. It, too, stems from the mistak-

en belief that those who have managed something can manage anything (although many political ap- pointees have managed only a few lawyers or re- search assistants).

If political appointments are so wonderful, how come they are not used in the military? Imagine a U.S. president replacing all the one- and two-star and most of the three-star generals of the army with political appointees. There would be outrage. “You can’t run the army this way,” people would insist. “You have to have devoted, experienced people.” Well, why is it any different for the departments of commerce, education, or state? Other countries have found ways to achieve political control with- out resorting to political administration. M Above all, we need balance among the different sectors of society. This applies to attitudes no less than to institutions. Private sector values are now pervading all of society. But government and other sectors should be careful about what they take from business. Business has probably never been more influential than it is now. In the United States, through political action committees and lobbying activities, institutional interests (not only business interests) put enormous pressure on the political system, reducing the influence of individuals. The system is out of control. My argument here is not against business as business; rather, it is for balance in society. We need balance among our four sectors, and we need to balance our public concerns as indi- viduals with the private demands of institutions.

Today the prevailing mood supports the privati- zation of public services. Some of that thinking is probably useful. But a good deal of it is also just plain silly. And if we are so prone to scrutinizing what doesn’t belong in government, shouldn’t we be equally diligent in considering what doesn’t be- long in business? Take newspapers, for example. Can any democratic society afford to have all news- papers in the private sector, especially when they are concentrated in a few hands that can exercise great political influence should they choose? Other models of ownership can be found, indeed in some of the most prestigious newspapers in the world – for example, nonownership of The Guardian in England and multiple cooperative ownership (journal- ists and readers, alongside some institutions) of Le Monde in France. Let us not forget that the object of democracy is a free people, not free institutions. In short, we would do well to scrutinize carefully the balance in our societies now, before capitalism really does triumph. 1. Donald J. Savoie, Thatcher, Reagan, Mulroney: In Search of a New Bureaucracy (Pittsburgh: The University of Pittsburgh Press, 1994).

Reprint 96306 To place an order, call 1-800-545-7685.

MANAGING GOVERNMENT

HARVARD BUSINESS REVIEW May-June 1996 83

Harvard Business School Publishing Customer Service - Box 230-5 60 Harvard Way Boston, MA 02163 Telephone: U.S. and Canada (800) 545-7685 Outside U.S. and Canada: (617) 495-6117 or 495-6192 Fax: (617) 495-6985 Internet address: [email protected]

Harvard Business School Management Productions videos are produced by award winning documentary filmmakers. You’ll find them lively, engaging, and informative.

Please inquire about HBR’s custom service and quantity discounts. We will print your company’s logo on the cover of reprints, collec- tions, or books in black and white, two color, or four color. The pro- cess is easy, cost effective, and quick.

Telephone: (617) 495-6198 or Fax: (617) 496-8866

For permission to quote or reprint on a one-time basis: Telephone: (800) 545-7685 or Fax: (617) 495-6985

For permission to re-publish please write or call:

Permissions Editor Harvard Business School Publishing Box 230-5 60 Harvard Way Boston, MA 02163 Telephone: (617) 495-6849

HBR Subscriptions

HBR Article Reprints HBR Index and Other Catalogs HBS Cases HBS Press Books

HBS Management Productions Videos

HBR Custom Reprints

Permissions

Harvard Business Review Harvard Business Review U.S. and Canada Subscription Service P.O. Box 52623 Boulder, CO 80322-2623

Telephone: (800) 274-3214 Fax: (617) 496-8145

Outside U.S. and Canada Tower House Sovereign Park Lathkill Street Market Harborough Leicestershire LE16 9EF Telephone: 44-85-846-8888 Fax: 44-85-843-4958

American Express, MasterCard, VISA accepted. Billing available.