Social Science & Medicine 52 (2001) 1151±1166

Comparative institutional response to economic policymanaged competition and governmentality$

Donald W. Light

Social and behavioral medicine, School of Osteopathic Medicine, University of Medicine & Dentistry of New Jersey,

40E Laurel Rd, Stratford, NJ 08084, USA

Abstract

This article provides a comparative conceptual framework for understanding why so many governments found

economic policies based on managed competition attractive and yet dangerous to implement. The frameworkconceptualizes governments as a kind of organizational complex and thus governments as an international populationof organizations, each embedded in a state that tries to harness and direct behaviour through what Foucault called

``governmentality''. This nascent concept is made more robust here and joined with Fligstein's historical research on theresponse of leading organizations when fundamental change threatens a population of organizations, by embracing anew conception of control that allows them to re-establish their control and pre-eminence. Fligstein studiedcorporations, but his model can be fruitfully extended to governments. Economic sociology has not to date been able to

do much comparative research on institutional responses to economic policy; but this set of case studies and conceptualframework provide such an opportunity. # 2001 Elsevier Science Ltd. All rights reserved.

Keywords: Comparative; Health economics; Medical sociology; Health care policy; Economic Sociology; Foucault

The project represented by this collection of papersstarted with a question: Given how economicallyunpromising and hazardous it is to subject clinical

medicine to commercial markets that reward less care(Glaser, 1993; Light, 1990,1991a,1994a), why did thecluster of policies known as ``managed competition''become the prevailing international model for health

care policy in the 1980s and 1990s? Why, further, didmany countries embrace it but then implement so little?Yet why does it seem that states and the nature of

governing health care have paradoxically changed any-way?Such questions call for rethinking the nature of the

state, the challenges to governance, and the proliferation

of competition policies as a social movement to rede®nethe legitimacy of health care as a welfare service. Theresistance to implementing managed competition in-

volved problems with health care being technically andsymbolically complicated, how embedded it is in asociety's institutional matrix (Ra�el, 1997) and inter-twined with cultural beliefs about illness and medicine

(Payer, 1980). In addition, when government o�cials,economists and health service researchers treatedmanaged competition uncritically as economic policy,

they failed to recognize it as a constructed reality and aset of solutions that re¯ect the interests of certain partiesthat will bene®t from it, or as an positivist ideology that

presents as scienti®c a political program that bene®tssome parties at the expense of others. As the costs topatients, their families, politicians or the state became

more tanglible, a number of governments became verycautious.These general, but unexpected features of response by

governments raise questions about how change di�uses

through populations of institutions. Most institutional

$This set of papers owes a debt to Neil Smelser, who

supported a special session at the 1997 annual meeting of

American Sociological Association. I am grateful to Steve

Birch, Ana Guillen, Kees van Rees, and Jonathan Mote for

helpful comments on earlier drafts of this paper.

E-mail address: dlight@Princeton.edu (D.W. Light).

0277-9536/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.

PII: S 0 2 7 7 - 9 5 3 6 ( 0 0 ) 0 0 2 3 6 - 7

studies assume and focus on fairly straightforwarddi�usion (DiMaggio & Powell, 1991); but this overlooks

the ways in which each institution in that population hasits own historically embedded structure, culture andpower relations. This comparative analysis of an

economic policy thus helps to explain, not only thedi�usion of an institutional change through an interna-tional population of governments, but also the culture,political and institutional forces that revised it, co-opted

it, and resisted it. Yet despite this resistance, theideology of managed competition has transformed thelanguage, values and institutional environment in which

services are provided, so that an ethos of professionalservice has been deeply altered by a commercial ethos.Governments and health care are in the process of

changing their values, their bases of legitimacy, theirclassi®cations and their practical actions to re¯ectcompetition theory, even as they resist its implementa-

tion in health care. These include rhetorics emphasizing``cooperation'' and the medical watchword of ``quality''to legitimate forms of cost accounting that devaluecaring and compassion. If the United States exempli®es

the full force of competition theory in health care, thenone sees its rise as causing deep loss of social capital(Portes, 1998), trust and caring (Gray, 1997; Mechanic,

1998).This essay develops a conceptual framework for the

case studies that follow it, by revising and integrating (1)

central ideas from institutional sociology, (2) Foucault's(1978, 1979) concept of ``governmentality'' as more thangovernance, (3) Pierson's (1996) revised concept of thewelfare state, (4) Portes (1998) synthetic conceptualiza-

tion of social capital and (5) Fligstein's (1990) dynamicmodel of how industries resolve periodic crises byadopting a ``conception of control'' which enables the

leading ®rms to reassert their legitimacy and control.Essentially, the need for governments to retrench afterthe oil crisis of the 1970s constituted an international

crisis of legitimacy for the population of governments asinstitutions that called for a new conception of control.Managed competition appeared to be a perfect solution

that would both legitimate cost containment and takethe heat of retrenchment o� governments, by restructur-ing the institutions of the state so that ``the invisiblehand'' of market forces would do the job. But even in its

ideal form managed competition has serious theoretical¯aws (Light, 1995a) and depletes the rich stock of socialcapital on which good clinical medicine depends. It also

threatens health care institutions and professions whichembody this social capital and which drew upon theirhistorically and culturally embedded legitimacy to resist

these new policies. Yet as the 21st century begins, itappears that managed competition has won a deepervictory by infusing the culture, language and measures

of ``health gain'' with its cost accounting and commer-cial ethos.

The need for conceptual robustness

Alongside its relevance to understanding the modernstate, these comparative case studies of competitionpolicy contribute to understanding social institutions.

They lead one to recognize current conceptual limita-tions in applying the new institutionalism to economicchange (Smelser & Swedberg, 1994; Brinton & Nee,1998). The new institutionalism in sociology arose to

correct certain limitations in institutional analysis(Brinton & Nee, 1998, Chapter 1) and weaknesses inthe development of economic sociology around network

ties. Such ties can be fragile, unstable, unpredictable andpersonal. ``The reliability of institutions provides analternative basis of trust or credible commitment that is

overlooked in the network embeddedness perspective''(Nee & Ingram, 1998, p. 23). Ironically, however, thede®nition of institution as ``a web of interrelated norms

} formal and informal } governing social re-lationships'' (p. 19) seems too broad to distinguishinstitutions from culture, though it may be meant as acounterfoil to the neoclassical economic concept of

rational actors maximizing self-interest. Nee and Ingramformulate six propositions about institutions. The ®rstproposition, that individuals uphold norms to capture

gains, assumes that people are little more than rationalactors working together to ``capture gains''. The secondproposition asserts that the main bene®t of frequent

interaction is to monitor each other's behavior moree�ectively and to lower costs, a social variant ofeconomic man. The desire to conform and its rewardscharacterize the third proposition. It seems like the

sociological counterpart to the neoclassical desire topro®t.1 There is not enough culture or history here orenduring bonds, too few commitments or fears, no

recognition of dedication or sacri®ce, no play or fun, nofaith or charity, no narcissistic pursuits of grandiosity.Yet all these are important in health care and in many

other social institutions. They certainly play importantroles in the economic sociology of managed competi-tion, particularly the history dedication and the grandi-

osity.Health care involves a deep commitment, institutio-

nalized in law, custom, and ritual, to helping the sickand vulnerable. Such help is often thankless, smelly,

repugnant, exhausting, even dangerous. If one is onsalary, then, why would clinicians as rational actorsbother, or bother very much to undertake such work? If

physicians are paid by capitation, then as rational actorsthey have all the less reason to bother, because the lessthey do the more they pro®t. Neither homo economicus

nor this version of the new institutionalism enables us to

1The last three propositions of Nee and Ingram tell us that if

informal norms are aligned with formal rules, things will go

well; and if they are not, they will not.

D.W. Light / Social Science & Medicine 52 (2001) 1151±11661152

explain the dedication of most clinicians. Social capitalneeds to be given a much more prominent role (Portes,

1998). While most of the research on social capital has(once again) bypassed health care, no other arenaexempli®es its importance. Portes (1998) identi®es ®ve

kinds of social capital, each characterized by a generaltime-spanning generosity: a sense of trust, extra-roleobligations, a sense of duty, a generalized ``currency''and debt that is invoked when someone needs to be

helped, and a bounded solidarity that provides cohesionin the fact of adversity. These have all been vital toclinical care for decades (e.g. Fox 1959).

At a still larger level, major acute hospitals such as theLeeds Royal In®rmary or the Swedish Royal Hospitalhave a mixture of historical, ®nancial and formal ties to

other institutions in their domain, as well as a distinctiveleadership and institutional culture, that institutionalanalysis must take into account if it is to understand

e�ectively their economic behavior and how the policiesof managed competition were variously resisted and co-opted. Social interactions in health care show that thenew institutionalism and economic sociology also need

to take cultural domains seriously and to include ananalysis of power relations, because Immergut (1992)and others have found that they account for signi®cant

di�erences between nations in how policies are formed,whether they can get passed, and their patterns ofimplementation.

Weighing in technological complexity, uncertainty and

risk

Another valuable contribution of research in thehealth care domain to the new institutionalism and

economic sociology is to identify the practical andtheoretical implications of how the technology, patternsof uncertainty and distribution or allocation of risk in

di�erent industries a�ect the organization of work,institutional structures, and economic behavior. Thesemay not be so important in some industries, but if one is

studying computer companies or health care, theirtechnological character and the distribution of riskstrongly in¯uence what kinds of economic exchange are

possible, what values, safeguards, and customs getinstitutionalized, and how institutions get formed andchange. Medicine poses the outer limits of thesevariables. Time and again, business executives who

con®dently enter health care come away concluding thatmedicine is far more complex and laced with moreuncertainties and risks than any industry they know. As

we shall see, this bears directly on both the economictheory of managed competition and the sociologicalrealities of its implementation.

Of course, there are sizable parts of health care thatare quite routine, not especially complex, and involve

little risk. But it is the uncommon and complex casesthat matter. Speci®cally, the patients that make up the

sickest 10 percent of a given population require 72percent of all medical resources to treat them, and thesickest 2 percent have disorders whose complexities and

risks are so great that it takes 41 percent of all costs totreat them (Berk & Monheit, 1992). The body is socomplex, its systems so interdependent, the centralnervous system and psyche so in¯uential on all other

systems, and individuals so variable, that there areconstant surprises, reversals, new unexpected develop-ments, side e�ects and interactions in treating the

seriously ill. Renee Fox (2000), in a new review ofmedical uncertainty, describes the reemergence ofinfectious diseases once thought conquered, the emer-

gence of new ones never imagined, and the specialists''dread of being asked to formulate a prognosis for theseriously ill because it exposes his or her ability to do so.

In addition, the more technologically sophisticatedmedicine becomes, the more complex interactions andunexpected side e�ects become, and these are com-pounded by large individual variations between one

person and another. The situation is akin to Perrow's(1985) analysis of complex systems whose very complex-ity will result in ``normal accidents''. But Perrow's

examples are systems like nuclear reactors that runsmoothly most of the time; complexity aside, the actionsand reactions are known and addressed in their design.

By contrast, serious medical care does not run smoothlymost of the time because it is often emergent andcontingent. One is not sure what the real problem is orhow well a treatment will work until after the fact. With

an aging population and its chronic disorders, thesefeatures will become even more prevalent.

The state and governmentality

Why have so many governments been attracted tomanaged competition as a policy for containing healthcare costs? To understand requires thinking of govern-

ments as a population of organizations or an ``industry''that faced a basic change in their environment andlooked desperately around for a new concept or modelthat would enable them to regain control (Fligstein,

1990). The nature and meaning of ``government'' and``control'' here pertain to Foucault's concept of``governmentality''. With this broadened framework in

place, we can review the case studies and then concludewith some thoughts about what will endure for thefuture.

The crisis of retrenchment

For 30 years after World War II, the economies ofWestern countries grew and with them grew both

D.W. Light / Social Science & Medicine 52 (2001) 1151±1166 1153

welfare and social programs (Wilensky, 1975). Expand-ing social welfare programs gained political credits, and

the process seemed natural (Flora & Heidenheimer,1982). Strong economies, experts claimed, producestrong welfare states (Wilensky, 1975). Or put more

critically, rapacious capitalism requires governments todevelop strong welfare programs as means to quell theire of those damaged by its exploitations. O'Connor(1973) concluded from observing economic growth in

the 1960s that legitimating pro®table capitalist accumu-lation requires the state to increase welfare programsand other expenses. Over time, however, the gap

widened between the private accumulation of wealthand the tax revenues of the state, putting the state into adouble bind. Its strategies for encouraging private

capital to maximize pro®ts underfunded its programsto discourage social unrest. The oil crises and stag¯ationin the 1970s threw states into a crisis of no growth,

decreasing purchasing power of its ¯at revenues, andincreasing unrest and demand for social programs.Among those programs, health care expenditures keptgrowing the most rapidly, even as the rest of the

economy did not. Governments faced a new set ofcircumstances in which their old habits of welfarepolitics did not work and in fact were bankrupting them.

Now, it seemed, strong welfare states weakened econom-ic performance. They needed to shrink, retrench in faceof the ``seemingly universal trade-o� between equality

and employment'' (Esping-Andersen, 1994, p. 24).But the politics of retrenchment are not the politics of

expansion in reverse. As Pierson (1994, 1996) shows, thepolitical landscape fundamentally changes. Health care

is probably the strongest example of the problems thatPierson identi®es for states that feel they need toretrench. Investments in human capital and programs

of social support are no longer seen to work. Moreover,the social and welfare programs themselves have createdstrong interest groups and dense networks among both

clients and sta� that make for concentrated opposition.By far the largest and most powerful group of recipients(reaching up through the working class and into middle

and managerial classes) are patients of a national healthcare system, be it largely private in its details or morepublicly run. And by far the most prestigious andpowerful of interest groups are the associations, societies

or colleges of physicians and nurses. Even } or perhapsespecially } in national health care systems, theyusually institutionalize themselves ®rmly into the rules,

regulations, pay structures, and budgets. Voters alsoreact more strongly to cuts than to new bene®ts, and thebroader the program, the more voters are a�ected. Most

government leaders, Pierson argues, want to avoid beingblamed and losing the next election rather than gainingcredit for ®scal frugality. But while ``blame avoidance''

characterizes Pierson's analysis, it alone would be only atactic within the old approach of governing through

social bene®ts and would face steep odds against interestgroups of clients and sta�. What one needs is a

reconceptualization of people's relationship to the statethat can legitimate retrenchment.

The professions as instruments of governmentality

A way of conceptualizing the dilemmas faced by the

state in a period of retrenchment is to sharpen andextend Michel Foucault's suggestive concept of``governmentality'' (Foucault, 1979; Burchell, Gordon

& Miller, 1991). From traditional concepts of sover-eignty which prevailed until the 18th century, he tracedthe gradual shift from ruling to governing, that is, to

steering and controlling a constellation of institutionsand parties that constitute the state. But the capacity ofgovernments to steer became limited by some essentialinstitutions and parties refusing to put public goals or

policies ahead of their own interests. Excluding or tryingto control those who resisted did not work well, createda democratic de®cit and eroded the legitimacy of the

state.In response, governance has evolved towards govern-

mentality, towards the shaping, guiding and directing of

people so that they are motivated to do what is best forthemselves and for society as a whole (Foucault, 1978).This has involved market-oriented strategies thatmobilize the interests of resistant parties to the goals

of the state, crafting a common discourse between thestate and di�erent parties, and creating commonfunding streams (Rhodes, 1997, p. 57). This more

process-oriented and market approach has graduallymoved the state toward being ``an ensemble of institu-tions, procedures, tactics, calculations, knowledges and

technologies, which together comprise the particularform that government has taken: the outcome ofgoverning'' (Johnson, 1995, p. 8). Thus, government

and the state2 are constantly rede®ning themselves asthey govern. This is the case in health care, wheregovernments everywhere are struggling with the ques-tion of how to get citizens and clinicians to behave with

restraint and forethought so that their health care costsare restrained as well.Central to the techniques of modern governmentality

is the harnessing and institutionalization of expertise,such as medicine and related ®elds. If set up prudently,experts in criminal or judicial matters, public health or

treatment of the ill, accountancy, civil engineering andconstruction, and other realms of expertise, can handlemyriad problems and dangers with dispatch, therebycontributing to the order, e�ectiveness and legitimacy of

2 In this essay I shall use ``state'' to denote the full range of

public institutions and ``government'' to denote the executive

and legislative branches that govern the rest of the state and

other aspects of society.

D.W. Light / Social Science & Medicine 52 (2001) 1151±11661154

the state. As Terrance Johnson implied in an importantessay (Johnson, 1995), although we commonly think of

the medical profession as having used the state toestablish its own autonomy and powers, the state usesprofessions as instruments of governmentality in a

number of realms. Real autonomy does not existanyway (Light, 1988; Johnson, 1995). Autonomyrequires state intervention to create and sustain it.In response to Larson's (1977) emphasis on how

the medical profession uses state powers to standardizeand monopolize health care, Johnson contendedthat Foucault would say the state also needs to

standardize and steer health care. The rise of theprofessions, then, can be seen as part of developingmodern governmentality and the state as ``the overcome

of governing''.Although Johnson carried his argument to its logical

conclusion that ``The duality, profession/state, is elim-

inated,'' this is unlikely, because governmentality, by itsvery nature, is made up of multiple parties, institutions,and professions, each with its own sense of value andway of doing things and bound to clash or at least grate.

Governmentality is far more complex to exercise thansovereignty and must deal with four persistent chal-lenges: how powerful vested interests are to be over-

come, how to co-ordinate them, how to enhance theircapacities to be partners of the state, and how to avoidinertia and log-jams (Ling, 2000). Market strategies

seemed to o�er a simple self-regulating solution to allfour.Among the professions, the medical profession in

most countries is the most important potential partner,

yet it negotiated the strongest measures of self-protec-tion, self-determination and dominance over its domains(Roemer, 1991). For decades, political leaders let it carry

out a number of social functions as is wished. Medicineadvanced rapidly and a growing number of socialproblems, such as unruly children, alcoholics, political

dissidents, and others, became ``medicalized'' andtreated with drugs, surgery, or other techniques. Butduring the 1970s, as states increased hospital beds and

physicians to keep up with growing demand, politiciansbegan to learn about studies showing there were largevariations in rates of surgery and hospitalization byphysician after controlling for clinical and non-clinical

variables. Why were some doctors spending so manymore resources treating the same problems than others?At the same time, these rates varied by the number of

surgeons and hospital beds per thousand. Was medicinea case where supply was creating its own demand, ratherthan the other way around? Politicians began to

question their longstanding trust in the medical profes-sion to carry out its vital functions as it thought best.High proportions of tests, procedures, hospital admis-

sions, and prescriptions were also found by specialtyteams to be unnecessary. Meantime, the oil crisis and

stag¯ation brought years of prosperity to a halt, putt-ing all government budgets under severe pressure.

No longer could politicians simply allow physiciansto treat illness and other forms of deviance as theywished. Furthermore, economic revival was seen to

depend on lower taxes and social contributions (Esping-Andersen, 1994). Once again, the question was how tolegitimate retrenchment, especially in the well-en-trenched yet highly charged area of medical and hospital

services?

The transformation of state control

The ``New Right'' proposed an answer, and it applied

to many of the social and governmental services thatmade up much of the apparatus of governance. Theconservative revolt against the welfare state in the 1970s

that gained dominance in the 1980s turned existingexpectations and relationships on their head:

* Welfare and social programs are not the fruits or

necessary accompaniments of economic growth; theyimpede it.

* Those who use them are not so much needy as lazy

and dependent.* Welfare programs are handouts, a sign of moral

weakness by the state and the recipient.* Patients should pay user fees; welfare recipients

should work; university students should pay theirown way.

* Government-run services are ine�cient, bureaucraticand unresponsive; they need the challenge ofcompetitors to wake them up and shed theirwastefulness.

* More broadly, the state is ine�cient, incompetent,and in¯exible; competition, deregulation, and priva-tization should replace state functions (Murray,

1984; Glazer, 1988; Saltman & von Otter, 1992,Chapter 1).

The new concept of governmentality advocated byThatcher, Reagan and other conservative leaders, inshort, went well beyond Pierson's tactic of blame

avoidance. It aimed to get people to change theirworldview not only of themselves, but also of others andof the state. This new governmentality transformedretrenchment into a moral crusade for self-su�ciency, a

smaller state, and economic vitality.The New Right thus provided what Neil Fligstein has

called a ``conception of control'' (Fligstein, 1990;

Fligstein & Freeman, 1995). Although his researchconcerns fundamental shifts in whole industries, it canbe instructively applied to governments as well. They

too are a population of organizations, very much inthe business of selling themselves, thriving, opening

D.W. Light / Social Science & Medicine 52 (2001) 1151±1166 1155

``new markets'' (broadly speaking), and being asentrepreneurial as are large businesses. Like an industry,

the ``governance industry'' experiences fundamentalchanges from time to time that render current ways ofdoing business ine�ective and that threaten the ability of

governments to control their fate. At such moments,political leaders, much like business leaders, search for anew vision, a new concept of what they are doing thatwill enable them to regain the control they had. The

New Right provided that vision by pushing the corenotion of governmentality farther than before. Ifgovernment is the ``conduct of conduct', then govern-

mentality should consist of techniques that put peopleon their own, put them at risk, and reward those whosucceed. (Those who fail should redouble their e�orts to

succeed.) Like all good conceptions of control, this oneprovided both a diagnosis of the crisis itself, how pastmistakes had led to it, and a vision of a bright future

with speci®c recommendations about how to assure it.During the 1980s, governments throughout the

industrialized world began to privatize entire govern-mental functions by selling them o� or exposing them to

competitive bids for the services they did. Nationalrailroads, airlines, gas, water and telephone serviceswere put up for sale and often brought windfall revenues

to governments. Opinions vary on how successful theseventures were, and the facts vary from case to case. Atthe same time ``the new managerialism'' brought similar

ideas to government programs by emphasizing recipientsas ``customers'' and performance-based results. Con-sumer-based results meant decentralizing decision-mak-ing and responsibility down as close to the customer as

possible. Mandatory competitive tendering and theestablishment of internal markets (among the providerswithin a public service domain) gave this new orienta-

tion a sharp edge.By the end of the 1980s, however, evidence indicated

that the new managerialism had been less transformative

than envisioned initially. Some competitive tenderingand devolved responsibility proved more cost-e�ective,but others seemed to have done little more than to

relocate the workers and the service (e.g. cleaning orfood preparation) down the street in companies. Forexample, in assessing the impact of the Thatcher reformsacross many sectors of the British government, Ling

(1998) concluded that internal resistance, organizationallimitations and di�culties in measuring either quality orperformance impeded implementation. Further, the

``new managerialism'' implied that government shouldshake itself free of both itself and the professionals onwhom it relies; yet specialized areas like health care are

inherently intertwined with the professionals essential toit. Were the new mandarins to shake themselves free,what would they do? In short, this new kind of

governmentality privileged new incentives and structureover strategy, means over ends.

Managed competition as a conception of state control

This problem of dependence on professionals for vitalsocial services was most acute in health care, wheretechnological complexity and diagnostic subtlety pre-

vail. Politicians and ministers in many states tried duringthe 1970s and 1980s to rein in health care costs, largelythrough budgetary measures, price controls and re-straints on supply. From an international point of view,

they largely succeeded, as evidenced by their halting thepercentage of GDP going to health care that had beenrising up to the mid-1970s.3 But within each country,

pressures to spend more on health care seemed relent-less, and a number of e�orts to confront the professionwith new evidence that it was wasting large sums of

money ran into the institutional protections the profes-sion had established decades earlier. Thus, when theAmerican economist, Alain Enthoven, declared that the

problem lay in the ine�ciencies of administrativeservices and the solution lay in reconceiving govern-mentality in terms of competition between comprehen-sive health care plans, the response was enthusiastic

(Enthoven, 1980,1985,1988).Enthoven's solution, managed competition, began

with his fully realizing how extensively health care fails

to meet the criteria for bene®cial competition to takeplace (Enthoven, 1988). Competition, as Adam Smithand those after him realized, is the opposite of solidarity.

It rewards people for maximizing their self-interest, andparadoxically it can bene®t everyone by increasingprosperity if markets are structured so that severalconditions are met. Among these, there must be many

buyers and sellers so that none can a�ect the marketoverall, and they must not have relations with each otherthat would a�ect their economic behavior. In health

care, there may be many buyers (i.e. patients) or a few(like insurers or other collective buyers), but often thereare not many sub-specialists or hospitals to choose from

as `sellers', especially if a country has done a reasonablejob planning its facilities and designing its trainingprograms. Another cluster of requirements is that buyers

and sellers can easily go into and out of the market,obtain full information cheaply, and have the market``clear'' their choices quickly. But in health care, it takesa long time and great expense for sellers (providers) to

be quali®ed or set up, and it is nearly impossible to makethem exit (close down) once they are established. Buyers(patients) ®nd entering most medical markets di�cult,

information scarce and costly to obtain, choices verycomplex and hard to interpret, and they usually cannotexit a treatment for any substantial problem easily once

3The United States, where vested interests had prevented the

availability of any of these mechanisms, had no e�ective

controls and watched the percent of GDP for health care

steadily climb.

D.W. Light / Social Science & Medicine 52 (2001) 1151±11661156

it has started. Patients do the last thing that a properbuyer should do; they turn to the `seller' (their trusted

physician of choice) to advise them or even to choose forthem what they should have. On top of all theseproblems, the services or ``products'' of serious medical

problems cannot be clearly de®ned (and therefore onecannot de®ne what is being bought or sold), becausethey are uncertain, emergent and contingent (Light,2000; Rice, 1998).

Although Enthoven has often presented variants ofhis model to suit the audience and politics of themoment, at its most complete (Enthoven, 1988) his

model calls for consumers to choose among compre-hensive, large-volume prepaid health care plans (likeAmerican HMOs)4 within a market that would be

protected from cherry-picking, cost-shifting and otheranti-social practices. Safeguards include requiring uni-versal access, a common bene®ts package, community

rating, open enrolment, a strong system of qualityassurance and excellent information on price andquality. Even then, Enthoven was so impressed by theproven ability of American hospitals and providers in

the 1970s and 1980s to exploit any set of regulations thathe also stipulated there be an oversight body to managethis highly regulated market actively. For ``. . .withoutactive collective management on the demand side, themedical plans would be free to pursue pro®ts or survivalusing numerous competitive strategies that would

destroy equity and e�ciency and that individualconsumers would be powerless to counteract'' (Entho-ven, 1998, p. 11).Enthoven's theory carries a strong message, namely

that competition requires very extensive regulation aswell as a watchdog organization with a smart, well-paidsta� to be sure that health care providers do not

circumvent or manipulate the regulations so that theycan win without really competing. Otherwise, the stakesare too high to attempt competition, because the

providers (sellers) have all the advantages. They haveintimate ties with and extensive control over patients;they decide the diagnoses and treatment plans; and they

control the clinical records. Thus price competition inhealth care entails not deregulation but more regulation.All markets are framed by the values, mores, rules andlaws of society; there is no ``free market'' as some people

customarily say, not even for street vendors sellingsausages.This model of managed competition; however; ap-

pears to surmount the obstacles to bene®cial markets in

health care; and it has been widely acclaimed as the keyto making health care services more e�cient and

responsive. Competition was a radical new weapon forthe government as a countervailing power (Light, 1994b)in its e�orts to provide a�ordable health care, a weapon

that would replace years of ®ghting well-entrenchedprofessional and institutional powers using administra-tive orders with market forces that would makeproviders compete to make health care more a�ordable.

The government could step aside and let competitors doits work.Managed competition seems an ideal model for

governments facing retrenchment. It shifts the blamefor escalating costs to the providers. Its advocatesemphasize the ways in which the existing system has

perverse incentives to raise costs and lock in budgets orformulas that perpetuate costly forms of care (Entho-ven, 1988). They note the variable quality, fragmented

services, and ine�ciencies that result from professionaldominance and autonomy (Zelman & Berenson, 1998).Managed competition also suits well the three strategiesthat Pierson (1996) identi®ed } obfuscation, division

and compensation. Some tactics of obfuscation identi-®ed by Pierson are particularly germane to market-based policies in health care:

(a) making negative e�ects harder to detect,(b) obscuring the links between the negative e�ect and

the policy that causes it,(c) increasing complexity,(d) shifting the burden,(e) dividing the opposition,

(f) changing measures so that reforms are di�cult toevaluate,

(g) delaying the implementation so that opposition

®zzles out,(h) di�using the targets of opposition groups and(i) blaming the victim (Ryan, 1971).

The other two strategies, division and compensation,complement each other. They involve bene®ting some

groups while cutting back bene®ts for others, andcompensating the losers in some way. Division andobfuscation work together when indirect changes in

eligibility are used, or cuts are made for future unknownpersons while current bene®ciaries are grandfathered in.These are the more negative aspects of governmen-

tality that Foucault does not discuss or even hint at; but

just as The Prince, with which he opens his analysis,impartially discusses defensive and covert tactics as wellas positive and overt ones, governmentality should be

understood to involve all kinds and shades of tactics.Besides shifting blame and responsibility to providers,managed competition decentralizes responsibility,

creates division and makes the system more complexand fragmented. These features suit the politics of

4HMOs stands for health maintenance organizations. These

organizations provide, directly or through contracts, all services

for their subscribers for a ®xed sum per annum per person. This

gives them the incentives to prevent as much illness and

disability as possible, to teach people how to take care of their

own health problems, and to hold down costs.

D.W. Light / Social Science & Medicine 52 (2001) 1151±1166 1157

retrenchment while still leaving the state in control ofthe overall budgets. The negative e�ects of meeting tight

budgets become di�used and harder to detect inthousands of local decisions. Its e�ects also take timeto work themselves out. Competition rewards some (the

winners) but not others (the losers, who are presumed tobe less competent). In these ways, managed competitionenables governments to use all nine of the tactics Piersonidenti®ed.

Hazards of governmentality through competition

A careful analysis, however, reveals that even in itsideal form, managed competition has ¯aws. This reviewwill highlight just four of ten ¯aws identi®ed (Light,

1995a). First, managed competition does not solve thefundamental sources of market failure } the uncer-tainty, contingency and information asymmetry of

medical practice, but instead tucks them inside its basicservice unit, the HMO. There, Enthoven assumes thatmanagers will resolve all these problems better thanprofessionals did in traditional health care. As research

on HMOs indicates, many of them get resolved indisturbing ways (Morgan et al., 1997; Himmelstein et al.,1999; MacPherson, 1997; Schelsinger, Dorwant, Hoover

& Epstein, 1997).Second, managed competition assumes that providers

cannot be trusted but managers can. This is the puzzling

premise on which the model rests. While doctors andeveryone else are assumed to maximize their interests ashomo economicus, managers seem to come from another

genetic strain, perhaps homo honorabili, sel¯essly dedi-cated to keeping the market balanced and fair (Light,1995a,b). Third, managed competition leads to oligopo-lies forming in each market (Sullivan, 1995). As hospital

and physician groups merge and payers consolidate,they minimize the stated objectives of competitionpolicy. Finally, managed competition undermines public

health and area-wide programs by dividing an areaamongst competing proprietary interests.When it comes to using competition as a new

conceptualization of governmentality, the stakes areeven higher. Managed competition requires a verysigni®cant reorganization of typical health services and

has a strong implicit agenda to reduce substantially theuse of hospitals as the citadels of most health caresystems. This constitutes a major change in balancebetween countervailing powers, and those two goals

alone step on so many powerful toes that most politicalleaders would hesitate to implement competition poli-cies.

As has been evident in the United States for sometime, the radical implications of the competitionparadigm, with or without Enthoven's safeguards in

place, are that purchasers will come to widely di�eringconclusions about how their budgets should be spent,

and clinicians or hospitals or other provider organiza-tions will take concerted actions to defend themselves or

to expand. Thus the issues of quality, priority and valueare intensi®ed as the state's ability to exercise controlgets dissipated. Provider groups may also use undesir-

able tactics. These can include professional and organi-zational forms of collusion, forming market niches,cornering markets, locking in vital supplies (includingsubspecialists), gaining market dominance, hiding true

costs and declaring false losses or charging high prices,using direct or indirect paybacks, shifting costs andcostly patients to someone else's budget (including the

patient's personal savings), and ®ring or burying insidecritics.5 Competition also threatens the organizedprofessions, the hierarchies of government departments,

and the executive teams of hospitals and other healthcare organizations.Competition involves not only decentralized imple-

mentation but with it mistakes, bad luck and poorjudgement that lead to failures and large sums lost. Justas the parent HMO in the United States, like ``Kaiser''or ``Health Net'' get blamed for such mistakes by one of

thousands of providers in the system, so in nationalsystems the government gets blamed for something amanager or doctor does in some part of the system.

Further, competition may increase (and often has)inequalities, transaction costs, and service dislocations.In short, far from taking the blame and the ``heat'' o�

the state, competition policies tend to put the state in thehot seat, in a decentralized competitive system overwhich they would have much less control than in theirtraditional health care system.

The irony of managed competition as the idealstrategy for the politics of retrenchment in theory andrhetoric, then, is how it can turn out to be the nightmare

strategy in fact. Far from depoliticizing health care, thestate needs to be more active and capable than before.Just as the theory of global markets predicted the

``eclipse of the state'' when actually such markets needcapable states (Evans, 1997a,b), so states that delegatesigni®cant responsibilities of governance to health care

markets ®nd themselves more involved than ever. Thetransformation of state control of health care costs to acompetition paradigm may lead to less control and morecosts. These are some of the reasons why many states,

after an initial burst of enthusiasm, shrank back fromchanging the institutional structure of health services toa competitive one.

The worldwide policy movement of managed compe-tition and its variants as the new way for governments toget providers to rein in costs took o�, with no small

proselytizing e�ort by Enthoven himself, when RonaldReagan and Margaret Thatcher, as world leaders of

5 I have witnessed these all, and there are probably others I

have not.

D.W. Light / Social Science & Medicine 52 (2001) 1151±11661158

state, embraced its explanation of health cost problemsand their solution. The World Bank (1987), chief funder

to developing nations, extolled the virtues of managedcompetition in 1987 and made it the centerpiece of a1993 world report (The World Bank, 1993). The US

AID (Agency for International Development) and theInternational Monetary Fund also converted to the newparadigm. All three pressed or required developingnations and those in the ex-Soviet block to restructure

their health care systems into managed competitionmodels as part of receiving large loans for economic ornational development. Yet competitive models are more

expensive to construct and run than any other organiza-tional model, and these countries have only $50±300 perperson per year to spend on health care (Laurell &

Lopez, 1996). Thus, by the early 1990s, managedcompetition had become the leading new conceptualiza-tion of state control over health care waste and costs,

not only in advanced wealthy nations but by ®at in poornations, where market structures alone could consumemuch of the health care budget for treating the seriouslyill. Most of the international consulting ®rms, seeing an

opportunity for costly, long projects in setting up healthcare markets and institutions, joined the managedcompetition movement and promoted it as champions

of modern scienti®c management. It became di�cult fornations struggling to make their health care systemsmore cost-e�ective to ®nd anyone to give them advice

outside the managed care paradigm.

Institutional responses to governmentality throughcompetition

Because it is much easier to make money by skimping

on equity, quality and service than to become moree�cient, managed competition threatens societal goals(Light, 1990). Yet these goals are heightened by the

transformation of governmentality from administrativeto competitive approaches. Managed competition alsothreatens the institutional arrangements of professional

control, the dominance of hospitals and the hegemonyof academic medicine. When governments start toimplement it, they ®nd themselves confronted by a

constellation of institutional realities such as theorganized medical and nursing professions, the hospitalas a technically and culturally unique institution, deep-seated lay convictions about what good medicine should

be like, and self-protective maneuvers by governmentaldepartments themselves. Thus government leaders ®ndthemselves adopting a concept of governmentality that

threatens existing institutional domains, transgressesboundaries and disrupts network relationships. Whathappens in each country depends on the type of welfare

state it has (Esping-Andersen, 1990), the kind ofgovernance structure it has (Immergut, 1992), the

professional culture of services, and the nature of itshealth care system (Roemer, 1991; Twaddle, 1996).

Devolving services, decisions and money to actors inpublic service markets, also have a paradoxical e�ect ongovernmentality. The further governmental services are

transformed or devolved to market-based actors, themore deeply involved the government must become inassuring that services are of good quality, good valueand equitable. The arms-length roles as purchaser and

manager of fair markets make these issues much morevisible than when services were just administered, yetmore di�cult to attain under the new mechanisms of

governmentality. The contractees are more motivated tocut corners, select easy cases, or ``cook the books''. Thegovernment is thus doubly pressured to become more

involved in setting standards and clinical guidelines, andin monitoring services. A new era of performance andoutcomes arises, with a rethinking of what services are

needed: a reconstruction of services, and the develop-ment of consumer rights, national standards andsystems of accountability. These are organizationale�ects of the economic theory.

Ruggie (1992) has called these e�ects ``the paradox ofliberal intervention'' though what she is describing isgovernmental delegation of its functions to the market,

which may or may not start with an intervention andoccurs under conservative rule as well. In a subsequentbook, Ruggie (1996) developed models of three liberal

welfare state regimes: the segmented regime based onmarket rationality, the interventionist regime based ontechnical rationality, and the integrative regime based onsubstantive rationality. The model for the segmented

regime implies that the ``paradox of market delegation''does not apply unless there are scandals of poor quality,bad service, waste, pro®teering, or corruption. Subse-

quent interventions over time may lead to the integrativeregime in which the government and its vendorsnegotiate goals and terms in a mutually supportive

relationship that leads to what might be called partner-ship rationality.

Managed competition as rhetoric and ideology

Other sociological e�ects of economic theory involve

institutions using the theoretical model as a rhetoric fortheir own ends. They take the features they like andleave the rest behind, even though at its ¯awed best, themodel requires that all its safeguards be put in place. For

some politicians and investors, it is a model they can useto capture parts of the public-sector economy. For somegovernments, market strategies keep all parties so busy

dealing with the complexities and mis®ts betweencompetition and medical services that parts of thehealth care system can be privatized so that the

government pays less, and patients and their familiespay more. Often, managed competition provides a

D.W. Light / Social Science & Medicine 52 (2001) 1151±1166 1159

rationale for reducing coverage of services throughincreased user fees, treatment limits, or through stricter

eligibility criteria. One must always look for suchchanges backstage cost-shifting or service reduction,when governments declare that frontstage competition

has succeeded in saving them money, and several caseshave been described by Hsiao (1994).Market forces are supposed to shake providers up and

make them accountable to a commodi®ed agenda. But

the shift from a service ethos to a commercial ethos, inwhich it is ``smart'' to get what you can and ``dumb'' tokeep working hard to help the needy, can unleash

thousands of private schemes that may disrupt servicesand raise costs. It is for these reasons that, paradoxi-cally, shrinking the state by letting the market take over

some of its services leads quickly to intense governmentinvolvement and a new raft of regulations. Under-standably, a number of governments have held back or

been very cautious about implementing competitionpolicies.A commercial ethos has a deeper e�ect; for as Ruth

Malone (1999, p. 19) observes in an eloquent essay, it

``leaves little space for the kinds of actions that embodydi�erent values } for example, generosity, mercy, orsolidarity''. Selling ``products'' is not a relationship and

creates divisions based on ownership, but ownership isill suited to public deliberation of policies. The market asmetaphor constructs moral silences. Even to de®ne the

problem as ``cost containment'' de®nes the issue in away that contains the answer. These deeper issues ofvalues and political philosophy are highlighted byproposals to implement competition policies and often

lead to deciding that competition will compromise thebasic goals of the health care system.

Comparative case studies of managed competition

The United Kingdom

The ®rst leading nation to embrace managed competi-tion, and to become an exemplar for other nations, wasthe United Kingdom (Light, 2001). With a swiftness

and completeness that was not possible in the UnitedStates, the Prime Minister assembled a small group andissued a White Paper (Secretary of State for Health,1989) that e�ectively commanded everyone in the entire

health care system to reorganize themselves into aversion of managed competition. She was the world'schampion of markets and the new tools of govern-

mentality to increase accountability, consumer respon-siveness, innovation, quality, e�ciency and productivity.She had held o� from addressing the health service as

she sold o� or reformed many other parts of the stateshe inherited, but pressures of costs, demands, and the

press spurred her to announce sweeping reforms at theend of the 1980s.

As adaptations go, British managed competition wasas close to the complete model as anyone is likely to see(Light, 1997). The NHS was already like a giant group-

model HMO, with universal, community-rated andcomprehensive coverage, capitated primary care foreveryone, referral specialty services at system-ownedhospitals and decentralized health authorities overseeing

services for their populations. Every health authoritywas transformed from being an administrative o�ce of apublic service to its purchaser, and every hospital and

community service unit became a seller who had tomake ends meet by obtaining contracts from purchasers.In fact, a trial year was mandated. Dangers of bank-

ruptcies and inequalities quickly led the government in1991 to limit open competition severely (Light, 1991b;Dawson, 1995). Even so, there were considerable

dislocations, especially as outer-London authoritiesdecided to use their new purchasing to shift theirinpatient cases to local hospitals by shifting theirbudgets from the great London hospitals.

Over the next few years, the number of managers rosesharply. The extra costs of creating a health care market(strangely missing from the economic theory of mana-

ged competition), were considerable, and the costs ofcontracting and managing increased greatly. Transac-tion costs were high, because health care is so complex,

and it is so di�cult to measure whether value for moneyhas increased or worsened, or whether providers havetaken a number of much easier routes to reducing theircosts. The most likely are avoiding or shedding costly

cases, substituting inferior service for good service, andcutting services short. The NHS budget jumped.Colleagues became competitors; morale declined (Wil-

liams, Michie & Pattani, 1998). Within a few years,upper-level managers concluded that the competitiveform of governmentality was keeping them from work-

ing together to provide good care at low cost. Soon thecompetition strategy was largely dropped by theconservatives in favor of managed cooperation (Coulter,

1997).The most interesting part of the reforms was an

afterthought that did not ®t the main model of apurchaser±provider split, namely GP fundholding.

Space does not allow a detailed account (Coulter,1997; Light, 1995b), but the idea of giving GPs thefunds and power to purchase selective specialty services

gained a great deal of attention, because GPs knew theirpatients and knew all about clinical medicine, when laypurchasers did not. The best proved clever and

successful in wresting discounts and better service fromspecialists, but these were the exception, and the AuditCommission study concluded that 90 percent of

GP fundholders lacked the time, interest or skill to domuch with their budgets (Audit Commission, 1996).

D.W. Light / Social Science & Medicine 52 (2001) 1151±11661160

Fundholding created two-tier access and of course acon¯ict of interest between GPs doing their best for their

patients and maximizing income. GPs who objecteddeveloped ``GP commissioning'', in which they usedtheir know-how to advise health authorities on how to

purchase more e�ectively for the entire population,rather than just for their own panel of patients. WhenTony Blair swept into o�ce on the promise to end two-tier competition in health care, he universalized com-

missioning or fundholding (depending on one's point ofview) and is now rapidly moving towards giving largeGP groups responsibility for purchasing all health care

services.

The Netherlands

In the Netherlands, an elite policy circle also perceivedduring the 1980s that health care costs were rising too

rapidly, that referrals and overuse of specialty care wasexcessive, that the services were ine�cient and in¯exible,and that there was little freedom of choice (Lieverdink,2001). In fact, the health care system already had a

successful system of primary care and gatekeeping, andit used budgetary controls at various levels e�ectively tohold down health care costs. Nevertheless, a commission

was formed and issued a major report in 1987 thatrecommended integrating current insurance arrange-ments into a single system that would provide coverage

for about 85 percent of services, so that funds couldcompete over the rates for the remaining 15 percent thatcitizens would purchase. This is a distinctly Enthoven

design for managed competition. The ensuing yearswitnessed sustained debate that articulated basic con-cerns. First, providers were probably more powerfulthan insurers, so that competition would create a sellers'

market rather than the envisioned buyers' market.Second, risk adjusters did not exist anywhere that wouldprevent adverse e�ects, such as providers avoiding

treatment of seriously ill patients. Third, competitionwould probably lead to consolidation on both sides ofthe market so that competition would be minimized and

the whole exercise would simply transform the healthcare system into a more costly bilateral monopoly.Nevertheless, the government is supporting competitionover new and extra services, which will naturally grow

over time, and more broadly the entire health careculture has shifted towards the language and rhetoric ofcommerce. There are grave concerns about the long-

term impact of both of these trends.

Sweden

In Sweden too there was a perceived crisis in costduring the 1980s, though by international standards

Swedish costs were reasonable and steady (Anderson,Smedby & Vagero, 2001). Managed competition was

seriously considered, and the debates led eventually toeight changes, ®ve of which are not competition-oriented

but do re¯ect the central role of the strong buyer in thecompetition model. Setting priorities for treatments,dismantling dental insurance, and integrating services

for older persons are examples of strong buyer actionsthat do not per se foster competition. Cautious, smallexperiments have been tried by some purchasers,followed by restrictions to safeguard the innocent.

Swedes found they were inexperienced in using ``marketbehavior'' in health care and found it di�cult toestablish realistic market conditions in health care.

Meantime, the Swedes have taken measures to reduceuse of acute inpatient services and supply of personnel,increased user fees for patients with a resulting reduction

of use by lower income patients, charged more fornursing home and home-care services, decentralizedresponsibilities to the county councils, and in general

reduced the cost of health care from 9.4 percent of GDPin 1980 to 7.5 percent by 1994.

Italy and Spain

In Italy and Spain, the governments embraced the``new managerialism'' and devolved both budgets and

responsibilities to regions (Cabiedes & Guillen, 2001).Hospitals too were made more responsible by givingthem prospective, risk-adjusted budgets. Managed

competition manifested itself largely by encouragingprivate providers to compete against public units forcontracts, and by giving patients more choice where to

go. User fees and co-payments have been increased,producing more inequality in the system. Since providerswere paid for treating them, there was competition formarket share but not for price. The Spanish government

seems bent on a commercial course. They tried toconvert public hospitals into private ones, but intenseopposition has made them back o�. But they are

persisting in less visible ways, like converting publichospitals into ``public health foundations'' instead.Inspired by GP fundholding, some GPs have been given

budgets to purchase specialty and hospital budgets.

Portugal and Greece

The e�ects of managed competition in Portugal andGreece are signi®cantly a�ected by both having a weakand underdeveloped public sector and a strong private

sector that has developed to ®ll in the gap (Cabiedes &Guillen, 2001). Policies to foster competition betweenthe two sectors for public contracts have led to pressures

to pay the high prices of private-sector providers.Managed competition has also led the governments toact more like purchasers than administrators.

For example, they are developing performance-basedsystems of pay, strengthening other measures of

D.W. Light / Social Science & Medicine 52 (2001) 1151±1166 1161

performance, experimenting with private managementgroups run public facilities, devolving budgets and

responsibilities to regional bodies, and allocating bud-gets on a risk-adjusted population basis. All such e�ortsare di�cult and elicit serious protest in a situation of

limited funds and a weak public sector.

Israel

Israel is one of the most revealing cases of howhistorical and institutional forces shaped behavior whencompetition policy was adopted for the entire system,

and in important ways it parallels the British case (Gross& Harrison, 2001). During the late 1980s, a perceivedcrisis in ®nancing and service delivery led to a national

commission being formed. In¯uenced by the British andDutch e�orts to create managed competition, thecommission proposed in 1990 regulated competition

between the nation's four non-pro®t sickness funds andbetween hospitals as well. After intense political debate,the National Health Insurance Law was passed in 1994and took e�ect in 1995. The funds began to compete by

use of aggressive marketing techniques that were alsoillegal to sign up healthier citizens and collaborated toreduce their costs by restricting services, the two easiest

ways to minimize costs and maximize surpluses. Thegovernment, like the British government, quicklyimposed restrictions to prevent such competitive beha-

viors, and in general the rhetoric of managed competi-tion enabled the central government to strengthen itscontrol and reduce expenditures.

New Zealand

New Zealand, which for historical and cultural

reasons follows development in Great Britain closely,elected a conservative government in 1991 that faced anacute ®scal crisis (Fougere, 2001). Over the previous

years, health care funding had tightened on the claimthat provider capture had made the health boards andservices ine�cient. Taking the British reforms of

managed competition a step further, it issued a radicalrestructuring of its universal health care system bycreating regional health authorities that would be at

arm's length from providers and by devolving a risk-adjusted population-based budget to them. It trans-formed public hospitals into independent units thatwould then compete for contracts from the authorities

and make a pro®t. This would make them de facto likeprivate hospitals. But the plan also provided for privateplans to compete with the regional authorities, and

anyone who did not like the public services couldtransfer their share to the private plan. Public responsewas immediate and hostile, suspecting that this was not

a plan to make health care services more e�cient, asdeclared, but to privatize both services and funding. The

New Zealand system would be ``Americanized'' into awasteful, two-tier, unjust system. The government's

popularity plummeted and it quickly withdrew theo�ensive features of its plan. The regional authoritiesstayed within the government's budget but also purchase

the services that their populations need, and thisrequired working closely with providers to ®nd innova-tive ways to serve patients more cost-e�ectively. Sincethe hospitals have a monopoly on specialty services in

their areas, true competition would put power and pricesin their control or produce a stando� between a singlepurchaser and a single provider. Acute specialty services

for the seriously ill have the political and moral upperhand, as ever. The result was a series of ad hoc, specialappropriations, and over-expenditures rose rapidly.

Increasing objectives to the reforms by patients andprofessionally alike helped elect a new governmentopposed to competition as a tool of governmentality.

Latin America

In Latin America, many nations also faced a ®scal

crisis in the 1980s, but one much more profound than inthe other countries studied here (Iriat, Emerson,Campinas & Waitzkin, 2001). The degree of indebted-

ness, of in¯ation, and of unemployment was muchgreater. Latin American states also tend to be quitedi�erent from European states, captured much more

deeply by a small self-perpetuating elite, sometimeshaving a long history of exploitation and corruption,and characterized by public institutions (such as

hospitals and public health facilities) that are sinecuresfor political appointees, ine�cient and unresponsive.Exposing them to market forces that reward respon-siveness to consumers and e�ciency seems just what

they need. But the motives of the change agents make agreat di�erence. Data are di�cult to obtain (which isdata in itself), but it appears that in several countries the

executive branch of government is bypassing thelegislative branch and open debate in order to makedeals with multinational corporations who want admin-

istrative contracts to run the public health care andsocial security markets in the name of managedcompetition. Such actions build on the tradition of

dictatorships. The corporations are also buying outdomestic HMOs and hospitals, which then receivepublic funds through contracts. Some medical associa-tions and unions have organized campaigns against

managed care organizations entering and running publichealth services. But the World Bank, the InternationalMonetary Fund and the Interamerican Development

Bank want these countries to reduce their indebtednessby decreasing expenditures for social services and byprivatizing them as conditions for receiving large loans

for economic development. The higher administrativeand marketing costs, plus pro®ts, reduce the funds

D.W. Light / Social Science & Medicine 52 (2001) 1151±11661162

available for clinical services, on the presumptionthat the latter will be made so much more e�cient that

full service will be possible. Limitations in health careand social security result in expanded markets forprivate policies, and the private companies direct their

marketing towards younger, healthy workers. In theseways, managed competition serves as a rhetoric andrationale for institutional changes that rede®ne public±private boundaries in favor of healthy and wealthy

clientele.

Other cases

Actual experiences with competition in health carehave also been disappointing. US experiences are

di�cult to use (though quite mixed), because di�erentforms of competition are used selectively in parts ofmarkets that have almost none of the conditions that

Enthoven identi®ed as necessary. Apparent successes areeasy to achieve in a system with costs, prices andutilization are 50±100 percent higher than advancedhealth care systems in other countries. A more

representative, whole-system and sophisticated applica-tion of managed competition is Singapore. A detailed setof mechanisms and rules of governmentality were set up

by the government, and the Singapore experience with acompetitive model featuring medical savings accountshas been celebrated by health reform consultants and

competition advocates. But when William Hsiao, aprofessor of health economics at Harvard, investigated,he found that the increasing role of private ®nancing

caused more rapid cost escalation, duplicate facilitiesthat were then underutilized and a rapid increase inphysicians'' incomes (Hsiao, 1995). Hospitals, he found,did not compete on price but on technology, to attract

the doctors with paying patients. By 1993, the Singaporegovernment concluded that ``the health care system is anexample of market failure. The government has to

intervene. . . (Hsiao, 1995, p. 263).Hsiao also investigated systemwide competition stra-

tegies in South Korea, Chile, and the Philippines (Hsiao,

1994). In South Korea, strong demand-side techniquesto control rising costs resulted in rapid expansion ofunregulated private care which drove up costs at a

compound rate in real terms of over 15 percent perannum. In Chile, market forces introduced to enhancee�ciency and contain costs led to a large private marketthat selects health risks and the well-paid workers, a

deprivation of the national public system, and worsecare for the seriously ill. The Philippines requiredemployers to o�er HMOs and o�ered loans to help

HMOs develop, so that they would reduce illnessthrough prevention and hold down the use and costsof hospital and specialty care. Most of the HMOs that

developed were for pro®t, and after a decade theFilipinos found that they charged higher premiums than

old-fashioned cost-reimbursement insurance plans, se-lected the healthy workers and subscribers, and spent

only 55 percent of total revenues on clinical services,using the rest for sales commissions, managers, executivecompensation, and pro®ts. (This compares to 75±85

percent for clinical services by large commercial insurers,and 90±95 percent by government-run services.) Theunderlying dynamic in these cases is analyzed by Evans(1997). He concludes that managed competition pro-

duces higher prices and incomes for providers, suppliers,and insurers, and they enable the wealthy to buyexcellent care without having to support it for everyone

else. One should add that they also handsomely rewardinternational consulting ®rms, who play a critical role inproselytizing managed competition worldwide.

The unanticipated e�ects of managed competition

From these comparative case studies, one can see theeconomic myths and sociological realities of managedcompetition. First, ``managed competition'' is not taken

up in its a full; coherent version that enables bene®cialcompetition to take place in health care. Rather, itsrhetorical themes are taken up as an ideology to

legitimate in varying degrees privatization, budget cuts,user fees, two-tier access, a commercial ethos, and therationalization of services. After trumpeting competition

in health care as the key to cost-e�ective services in 1993,the World Bank commissioned a series of papers byleading skeptics of competition in health care in 1995,

who detailed the dangers involved (Dunlop & Martins,1995). By 1998, two leaders of WHO-Europe reportedthat ``Several CEE/CIS (ex-Communist Bloc and SovietUnion) countries have belatedly sought to develop a

state regulatory apparatus to reduce the negativeconsequences of giving market forces free rein'' (Salt-man & Figueras, 1998). Nevertheless, a worldwide

search by large insurance and managed care companiescontinues for new pro®table markets, now that theinitial windfall pro®ts of conversion to managed

competition in the overpriced U.S. system have beentaken and little pro®t seems left (Stocker, Waitzkin &Iriart, 1999).

Second, all countries emphasize the ``purchaser±provider split'' as the radical change from administeredservices that is the key to implementing managedcompetition policies. But in fact, such a split is not

sustainable; because the purchaser (a health authority orother agency) must become intimately involved with the®ne points of provision in order to negotiate successfully

new organizational forms and successful instruments ofgovernmentality. The radical core is not a split butrather a shift from professional dominance to buyer

dominance that was identi®ed some time ago (Light,1989). This shift is not easy, because the health authority

D.W. Light / Social Science & Medicine 52 (2001) 1151±1166 1163

or other agency was embedded before and is embeddednow in networks of providers. Yet within these relations;

power shifts from the hospitals as the nodes in provider-based networks to purchasers as the nodes in emergentpayment-based networks (Light, 1998).

Third, not much competition is allowed because it istoo dangerous and disruptive in health care (for thereasons explained), and because neither the majorplayers nor the public really want it. Thus ``managed

competition'' becomes political rhetoric for what isactually the revolt of the payers and the transformationof institutional culture. But something deeper and more

permanent happens in the process of states debatingseriously the merits and dangers of the competitionparadigm. It leads them to question long-standing

organizational arrangements and learn new ways ofthinking. Most important, it has led them to think andact like purchasers, or commissioners. This transforma-

tion entails several axes of change:

* from administrating services, to purchasing healthoutcomes;

* from treating individual ill patients, to maximizingwell-being and health of populations;

* from trusting professional judgement, to evidence-

based protocols and evaluation;* from professional autonomy and variation, to

professional accountability and standards;* from historic provider-based budgets, to population-

based, risk-adjusted budgets;* from centralized administration, to devolved deci-

sion-making;* from comprehensive publicly funded services, to

service limits with private upgrades.

None of these, except the last, would be possiblewithout the computer and its capacity to gather andanalyze large data sets. For decades, the so-called health

care is not been about health but about treating the sick.Now health status and health gain are taking centerstage. Evidence-based medicine, guidelines, accountabil-

ity, and budgeting were not possible before. Now theyare (Kindig, 1997).The unanticipated consequence of managed competi-

tion is a reconceptualization of the state in its relations

with its residents, from providing certain services tomaximizing the health and functioning of a population.When I worked with the Oxford Health Authority, for

example, as managed competition was being implemen-ted in 1990, some o�cers talked seriously about thelarge number of diseases and physical problems brought

in during the winter by poorer people who lived inhomes with no central heat. They discussed whethersome of the budget for health care services should be

spent on putting central heating in those homes. Whynot, if it was more cost-e�ective and also enhanced the

lives of those people? Commissioning is di�cult to dowell and involves major issues of distributive justice

(Daniels, Light & Caplan, 1996), but the pre-requisitesfor e�ective commissioning and lessons from earlymistakes have been identi®ed (Light, 1998). In the

meantime, we are in a transitional period of dislocationas we pass from providing medical services to purchas-ing for health gain. Thus, the enduring contribution ofthe managed competition policy movement is likely to

be, not so much competition per se but the transforma-tion of state control from service administrator to strongpurchaser or commissioner for evidence-based quality

and health gain. That, it turns out, is what we have beenwanting all along.

References

Andersen, R., Smedby, B., & Vagero D. (2001). Cost contain-

ment, solidarity and cautious experimentation: Swedish

dilemmas. Social Science and Medicine, 52, 1195±1204.

Audit Commission (1996). What the doctor ordered: A study of

GP fundholding in England and Wales. London:HMSO.

Berk, M. L., & Monheit, A. C. (1992). The concentration of

health expenditures: An update. Health A�airs, 11, 145.

Brinton, M.C., & Nee, V. (Eds.). (1998). The new institution-

alism in sociology. New York:Russell Sage Foundation.

Burchell, G., Gordon, C., & Miller, P. (1991). The Foucault

e�ect: Studies in governmentality. London: Harvester

Wheatsheaf.

Cabiedes, L., & Guillen, A. (2001). Adopting and adapting

managed competition: health care reform in Southern

Europe. Social Science and Medicine, 52, 1205±1217.

Coulter, A. (1997). Evaluating general practice fundholding in

the United Kingdom. European Journal of Public Health, 5,

233±239.

Daniels, N., Light, D. W., & Caplan, R. (1996). Benchmarks of

Fairness for Health Care Reform. New York: Oxford

University Press.

Dawson, D. (1995). Regulating competition in the NHS. Centre

for Health Economics, York (Eng.), Discussion Paper 131.

DiMaggio, P., & Powell, W. W. (1991). Introduction. In P.

DiMaggio, & W. W. Powell, The New Institutionalism in

Organizational Analysis (pp. 7±38). Chicago: University of

Chicago Press.

Dunlop, D., & Martins, J. M. (Eds.). (1995). An international

assessment of health care ®nancing: Lessons for developing

countries. Washington D.C.: The World Bank.

Enthoven, A. C. (1980). Health plan: The only practical solution

to the soaring cost of medical care. Reading, MA: Addison-

Wesley.

Enthoven, A. C. (1985). Re¯ections on the management of the

national health service: An American looks at incentives to

e�ciency in health services management in the UK. London:

Nu�eld Provincial Hospitals Trust.

Enthoven, A. C. (1988). Theory and practice of managed

competition in health care ®nance. Amsterdam: North-

Holland.

D.W. Light / Social Science & Medicine 52 (2001) 1151±11661164

Esping-Andersen, G. (1990). The three worlds of welfare

capitalism. Cambridge, UK: Policy Press.

Esping-Andersen, G. (1994). After the golden age: The future of

the welfare state in the new global order. Geneva: United

Nations Research Institute for Social Development.

Evans, P. (1997a). The eclipse of the state?. Re¯ections on

stateness in an era of globalization. World Politics, 50, 62±87.

Evans, R. G. (1997b). Going for the gold: the redistributive

agenda behind market-based health care reform. Journal of

Health Politics, Policy, and Law, 22, 427±465.

Fligstein, N. (1990). The transformation of corporate control.

Cambridge, MA: Harvard University Press.

Fligstein, N., & Freeman, R. (1995). Theoretical and compara-

tive perspectives on corporate organization. Annual Review

of Sociology, 21, 21±43.

Flora, P. & Heidenheimer, A.J. (Eds.). (1982). The development

of welfare states in Europe and America. New Brunswick,

NJ:Transaction.

Foucault, M. (1979). On governmentality. Ideology and

Consciousness, 6, 5±22.

Foucault, M. (1978,1991). Governmentality. In: G. Burchell,

C. Gordon, & P. Miller (1991) The Foucault e�ect: Studies

in governmentality. London:Harvester Wheatsheaf

(Chapter 4).

Fougere, G. (2001). Transforming health sectors: new logics of

organizing in the New Zealand health system. Social Science

and Medicine, 52, 1233±1242.

Fox, R. C. (1959). Experiment perilous: Physicians and patients

face the unknown. Glencoe, IL, US: The Free Press.

Fox, R. C. (2000). Medical uncertainty revisited. ln G. L.

Albrecht, R. Fitzpatrick, & S. C. Scrimshaw, Handbook of

social studies in health and medicine (pp 409±425). London:

Sage.

Glaser, W. A. (1993). The competition vogue and its outcomes.

Lancet, 34, 805±812.

Glazer, N. (1988). The Limits of Social Policy. Cambridge, MA:

Harvard University Press.

Gray, B. (1997). Trust and trustworthy care in the managed

care era. Health A�airs, 16(1), 34±49.

Gross, R. and Harrison, M. (2001). Implementing managed

competition in Israel. Social Science and Medicine, 52,

1219±1231.

Himmelstein, D., Woolhandler, S., Hellander, I., & Wolfe,

S .M. (1999). Quality of care in investor-owned vs not-

for-pro®t HMOs. Journal of the American Medical Associa-

tion, 282, 159±163.

Hsiao, W. C. (1994). ``Marketization'' } the illusory magic pill.

Health Economics, 3, 351±357.

Hsiao, W. C. (1995). Medical savings accounts: lessons from

Singapore. Health A�airs, 14(2), 260±266.

Immergut, E. M. (1992). Health politics: Interests and institu-

tions in Western Europe. New York: Cambridge University

Press.

Iriat, C., Emerson, E. M., Campinas, B., Waitzkin, H. (2001).

Managed care in Latin America: the new common sense in

health policy reform. Social Science and Medicine, 52,

1243±1253.

Johnson, T. (1995). Governmentality and the institutionaliza-

tion of expertise. In T. Johnson, G. Larkin, & M. Saks,

Health Professions and the State in Europe (pp. 7±24).

London: Routledge.

Kindig, D. A. (1997). Purchasing population health: Paying for

results. Ann Arbor, MI, USA: University of Michigan

Press.

Larson, M. S. (1977). The rise of professionalism: A sociological

analysis. Berkeley: University of California Press.

Laurell, A. C., & Lopez Arcellano, O. (1996). Market

commodities and poor relief: the World Bank proposal for

health. International Journal of Health Services, 26, 1±18.

Lieverdink, H. (2001). The marginal success of regulated

competition policy in the Netherlands. Social Science and

Medicine, 52, 1183±1194.

Light, D. W. (1988). Turf battles and the theory of professional

dominance. Research in the Sociology of Health Care, 7,

203±225.

Light, D. W. (1989). Towards a new sociology of medical

education. Journal of Health and Social Behavior, 89, 307±

322.

Light, D. W. (1990). Bending the rules. Health Service Journal,

5222, 1513±1515.

Light, D. W. (1991a). The Restructuring of American Health

Care.. In T. J. Litman, & L. S. Robins, Health Politics and

Policy 2nd ed (pp. 53±65). New York: Wiley.

Light, D. W. (1991b). Observations on the NHS reforms: an

American perspective. British Medical Journal, 303, 568±570.

Light, D. W. (1994a). Escaping the traps of postwar western

medicine. European Journal of Public Health, 3, 223±231.

Light, D. W. (1994b). Countervailing powers: a framework for

professions in transition. In T. Johnson, G. Larkin, & M.

Saks,Health Professions and the State in Europe (pp. 69±79).

London: Routledge.

Light, D. W. (1995a). Homo economicus: escaping the traps of

managed competition. European Journal of Public Health, 5,

145±154.

Light, D. W. (1995b). The future of fundholding. London:

Institute of Health Service Management.

Light, D. W. (1997). From managed competition to managed

cooperation: Theory and lessons from the British experi-

ence. The Milbank Quarterly, 75, 297±341.

Light, D. W. (1998). E�ective commissioning. London: O�ce of

Health Economics.

Light, D. W. (2000). The sociological character of markets in

health care. ln G. L. Albrecht, R. Fitzpatrick, & S. C.

Scrimshaw.Handbook of social studies in health and medicine

(pp 394±408). London: Sage.

Light, D. W. (2001). Managed competition, governmentality

and institutional response in the United Kingdom. Social

Science and Medicine, 52, 1167±1181.

Ling, T. (1998). The British State Since 1945. London: Polity.

Ling, T. (2000). Unpacking partnership: The case of healthcare.

MacPherson, P. (1997). Is this where we want to go? Hastings

Center Report , 27(2), 17±22.

Malone, R. E. (1999). Policy as product: Morality and

metaphor in health policy discourse. Hastings Center

Report, 29(3), 16±22.

Mechanic, D. (1998). The functions and limitations of trust in

the provision of medical care. Journal of Health Politics,

Policy and Law, 23, 661±686.

Morgan, R. O., Virnig, R. A., Devito, C. A., & Persily, N. A.,

(1997). The Medicare-HMO revolving door } the healthy

go in and the sick go out. New England Journal of Medicine ,

337, 169±175.

D.W. Light / Social Science & Medicine 52 (2001) 1151±1166 1165

Murray, C. (1984). Losing ground: American social policy 1950-

60. New York: Basic Books.

Nee, V., & Ingram, P. (1998). Embeddedness and beyond:

institutions, exchange, and social structure. In M. C.

Brinton, & V. Nee, The new institutionalism in sociology

(pp. 19±45). New York: Russell Sage Foundation.

O'Connor, J. (1973). The ®scal crisis of the state. New York: St.

Martin's Press.

Payer, L. (1980). Medicine and culture. London: Penguin.

Perrow, C. (1985). Normal accidents: Living with high-risk

technologies. New York: Basic Books.

Pierson, P. (1994). Dismantling the welfare state? Reagan,

thatcher, and the politics of retrenchment. Cambridge,

England: Cambridge University Press.

Pierson, P. (1996). The new politics of the welfare state. World

Politics, 48, 143±179.

Portes, A. (1998). Social capital: its origins and applications in

modern sociology. Annual Review of Sociology, 24, 1±24.

Rhodes, R. A. W. (1997). Understanding governance: Policy

networks, governance, re¯exivity, and accountability. Buck-

ingham, UK: Open University Press.

Rice, T. (1998). The Economics of Health Reconsidered.

Chicago: Health Administration Press.

Roemer, M.I. (1991). National Health Systems of the World,

vol. I, II. New York:Oxford University Press.

Ruggie, M. (1992). The paradox of of liberal intervention:

health policy and the American welfare state. American

Journal of Sociology, 97, 919±943.

Ruggie, M. (1996). Realignments in the welfare state: Health

policy in the United States, Britain, and Canada. New York:

Columbia University Press.

Ryan, W. (1971). Blaming the Victim. New York: Random

House.

Saltman, R. B., & Figueras, J. (1998). Analyzing the evidence on

European health care reforms. Health A�airs, 17, 85±108.

Saltman, R. B., & von Otter, C. (1992). Planned markets and

public competition: Strategic reform in northern european

health systems. Buckingham, England: Open University

Press.

Schlesinger, M., Dorwant, R., Hoover, C., & Epstein, S. (1997).

The determinants of dumping: a national study of

economically motivated transfers involving mental health

care. Health Services Research, 32, 561±590.

Secretary of State for Health (1989). Working for patients.

London:HMSO.

Smelser, N., & Swedberg, R. (Eds.). (1994). The hand-

book of economic sociology. New York:Russell Sage

Foundation.

Sullivan, K. (1995). Strangled competition: A critique of

Minnesota's experiment with managed competition. St. Paul,

MN: COACT Educational Foundation.

Stocker, K., Waitzkin, H., & Iriart, C. (1999). The exportation

of managed care to Latin America. New England Journal of

Medicine, 340, 1131±1136.

The World Bank (1987). Financing health services in developing

countries: An agenda for reform. Washington DC:The

World Bank.

The World Bank (1993). World development report 1993:

Investing in health. Washington DC:The World Bank.

Twaddle, A. C. (1996). Health system reforms ± toward a

framework for international comparisons. Social Science

and Medicine, 43, 637±654.

Williams, S., Michie, S., & Pattani, S. (1998). Improving

the health of the NHS workforce. London: The Nu�eld

Trust.

Wilensky, H. (1975). The welfare state and equality. Berkeley:

University of California Press.

Zelman, W. A., & Berenson, R. A. (1998). The managed care

blues & how to cure them. Washington DC: Georgetown

University Press.

D.W. Light / Social Science & Medicine 52 (2001) 1151±11661166