Ian Gough Economic institutions and the satisfaction of...

Ian Gough

Economic institutions and the satisfaction of human needs Article (Published version) (Refereed)

Original citation: Gough, Ian (1994) Economic institutions and the satisfaction of human needs. Journal of Economic Issues, 28 (1). pp. 25-66. ISSN 0021-3624 © 1994 Association for Evolutionary Economics This version available at: http://eprints.lse.ac.uk/60819/ Reprinted from the Journal of Economic Issues by special permission of the copyright holder, the Association for Evolutionary Economics. Available in LSE Research Online: February 2015 LSE has developed LSE Research Online so that users may access research output of the School. Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Users may download and/or print one copy of any article(s) in LSE Research Online to facilitate their private study or for non-commercial research. You may not engage in further distribution of the material or use it for any profit-making activities or any commercial gain. You may freely distribute the URL (http://eprints.lse.ac.uk) of the LSE Research Online website.

http://afee.net/?page=jeihttp://afee.net/?page=jeihttp://afee.net/http://eprints.lse.ac.uk/60819/

jei JOURNAL OF ECONOMIC ISSUESVol.XXVIll No. I March 1994

Economic Institutions and theSatisfaction of Human Needs

Ian Gough

The purpose of this paper is to evaluate different economic sys-tems using as a criterion their ability to satisfy human needs. Theconceptual basis is the theory of human need developed in Doyaland Gough [1991] and briefly summarized here. To assess thepotential of economic systems to satisfy human needs, thusdefined, I use a family of theoretical approaches from differentdisciplines broadly labelled "new institutionalist" or "new politicaleconomy." The economic systems to be investigated are distin-guished according to their dominant organizing principle: themarket, the state, and the community. Recognizing that "pure"models of each are historically and logically impossible, I evaluatecombinations of institutions that are as close as possible to thepure model: minimally regulated capitalism, state socialism, andvariants of communitarianism. Afler summarizing my conclusionsat that point, I then, in the next three sections, go on to considerthree variants of "mixed economy" capitalism: statist capitalism,corporatist capitalism, and neoliberal capitalism. Again I evaluateeach according to our criteria of need satisfaction before drawingsome general conclusions.

The author is Reader in Social Policy at Manchester University. The authorwishes to thank David Donaldson, Diane Elson. Andrew Gamble, Geoff Hodgson,Mick Moran, Peter Penz, David Purdy, and Paul Wilding for helpful comments onan earlier draft. The paper has originated out of, and is indebted to, years ofdiscussion and collaborative work with Len Doyal.

25

26 Ian Gough

Since this is an extremely ambitious project, it has necessarylimits that should be emphasized. First, the sole criterion accord-ing to which economic systems are compared is the optimum satis-faction of universal human needs, which will be defined shortly.Second, the focus is on need satisfaction within, not between, na-tion-states. It excludes global linkages between nation-states. Ef-fectively, this limits my focus to the developed world, though Ibelieve that some of the arguments are relevant for developing na-tions too. Third, it is concerned only with the ability of economicsystems to satisfy present levels of need satisfaction: issues ofeconomic sustainability and intragenerational redistribution areleft to one side. These are serious limitations, but they are madenecessary by the scope of the investigation remains. The paper isnecessarily broad and relies on secondary sources to buttress manyof its claims.

Need-Satiafaction as a Measure of Welfare Outcomes

This paper attempts to evaluate socioeconomic systems and in-stitutions according to the anticipated welfare outcomes enjoyed bytheir citizens. Welfare outcomes are conceived in terms of the levelof satisfaction of basic human needs. This approach thus differsfrom much contemporary research in both comparative socialpolicy and economics. The former has sought to explain variationsin "welfare states" by analyzing specific welfare inputs, such aslevels of state expenditure on social security, or more recently, wel-fare outputs, such as the specific social policies or the "welfarestate regimes" that characterize syndromes of social policies.^Much economics research, on the other hand, has concerned itselfwith the final outcomes of policies but has traditionally definedthese rather narrowly, such as, for example, rates of economicgrowth, monetary stability, rates of unemployment and employ-ment, and productivity growth [Strumpel and Scholz 1987; cf. Put-terman 1990]. Freeman [1989] undertakes a much broader andmore sophisticated evaluation of four "political economies," yet hestill restricts his evaluative criteria to two: growth rates and dis-tributional equity.

Both these approaches tend to ignore the final impact of allthese factors on the levels and distribution of well-being of thepopulations concerned (though this gap has been recognized bysome such as Alber et al. [1987]). The major reason for the lack of

Economic Institutions and the Satisfaction of Human Needs 27

progress here is an inability to agree on concepts and measures ofwell-being that have cross-cultural validity. The postwar periodhas witnessed a growth in research that utilizes concepts such asthe "level of living," "social indicators," "basic needs," and "humandevelopment" and that has informed comparative evaluation ofwelfare outcomes in the Third World. However, this work has hadlittle impact due in part to the changed political and economicclimate of the 1980s. It has also been criticized as lacking a unify-ing conceptual framework [Sen 1987] and more particularly for in-corporating Western cultural and political biases in the verynotions of universal need and social progress [Rist 1980; Doyaland Gough 1991, chap.8]. Though some of these issues have beenaddressed in some of the philosophical literature on need, therehas existed a barrier between this literature and the more applieddevelopment literature.

The absence of a theoretically grounded and operational con-cept of objective human need has inhibited the development of acommon calculus for evaluating human welfare. On the contrary,there is a widespread scepticism that human needs exist, or abelief that all needs are relative. Typical of the first view areneoliberals, such as Hayek and Flew, together with the dominantstrand in neoclassical economics. The second view, that needsexist but are relative, takes a variety of forms. For many Mar-xists, human needs are historically relative to capitalism; forvarious critics of cultural imperialism, needs are specific to, andcan only be known by, members of groups defined by gender, race,and so on; for phenomenologists and some social researchers,needs are socially constructed; for post-modernist critics and"radical democrats," needs are discursive and do not exist inde-pendently of the consciousness of human agents [Doyal andGough 1991, chap. 1]. Clearly, if any of these perspectives are cor-rect, then any common yardstick of welfare is unattainable andcannot be used to compare and evaluate different economic in-stitutions and systems.

Our theory attempts to overcome these limitations. The theoryis both substantive and procedural: substantive in defending, con-ceptualizing, and operationalizing the idea of universal humanneeds; procedural in recognizing the inevitable social determina-tion of products, policies, and processes that satisfy needs andthus in recognizing the necessity for procedures for resolving dis-

28 Ian Gough

putes in as rational and democratic a way as possible. I will merelylist the conclusions of our substantive theory here.

We argue that all persons have an objective interest in avoidingserious harm that in turn prevents them from pursuing theirvision of the good, whatever that is [Doyal and Goug^ 1991, chap.4]. This pursuit of the good entails, as others have argued, anability to participate in the form of life in which they find themsel-ves. Thus, objective basic needs consist, at the least, in thoseuniversal preconditions that enable sustained participation inone's form of life. At the most, they consist of those universalpreconditions for critical participation in one's form of life-thecapacity to situate it, to criticize it, and, if necessary, to act tochange it. Basic human needs, then, are the universal prereq-uisites for successful and, if possible, critical participation in one'ssocial form of life. We identify these universal prerequisites asphysical health and autonomy. In turn, autonomy of agency-thecapacity to make informed choices about what should be done andhow to go about doing it-is impaired when there is a deficit ofthree attributes: mental health, cognitive skills, and opportunitiesto engage in social participation.

Recognizing that these common human needs can be met in amultitude of different ways by an almost infinite variety of specificsatisfiers, we next go on to identify those characteristics of needsatisfiers that everywhere contribute to improved physical healthand autonomy [Doyal and Gough 1991, chap. 8]. These we label"universal satisfier characteristics," or "intermediate needs" forshort. We group these characteristics into 11 categories: adequatenutritional food and water, adequate protective housing, a non-hazardous work environment, a nonhazardous physical environ-ment, appropriate health care, security in childhood, significantprimary relationships, physical security, economic security, safebirth control and childbearing, and appropriate basic and cross-cultural education. Nine of these apply to all people, one refers tothe specific needs of children, and another to the specific needs ofwomen for safe child bearing. All 11 are essential to protect thehealth and autonomy of people and thus to enable them to par-ticipate to the maximum extent in their social form of life,whatever that is.

As developed thus far, our theory of needs is substantive, or "in-trinsic" [Hewitt 1992, chap. 10]. It identifies universal basic andintermediate needs and legitimizes the use of cross-cultural social


indicators with which to chart need satisfactions. It thus providesa means of empirically comparing the welfare performance of dif-ferent societies [Doyal and Gough 1991, chaps. 12-13]. However,my purpose here is to investigate theoretically the contribution ofdifferent social institutions to the satisfaction of these needs,which leads me on to the procedural dimension of our theory.Here, we indentify universal procedural and material precondi-tions for enhancing need satisfaction [Doyal and Gough 1991,chaps. 7 and 11]. These are attributes of social systems, and it isthese with which I am principally concerned in this paper.

Procedural preconditions relate to the ability of a group toidentify needs and appropriate need satisfiers in a rational wayand to prioritize need satisfiers and the need satisfactions of dif-ferent groups. In the face of radical disagreements over the per-ceived interests and needs of different groups, how can this bestbe achieved? To answer this we draw upon the works of Habermasand Rawls to sketch out certain communicational and constitu-tional preconditions for optimizing need satisfaction in practice.Habermas outlines a theory of communicational competence thatemphasizes the importance, for the rational resolution of debates-including debates about need satisfaction, of the best available un-derstanding and of truly democratic debate [Habermas 1970;Roderick 1986]. With modifications to his three principles, weargue that Rawls [1972] identifies the constitutional frameworkthat will enable citizens to engage in such debate.

In what follows, I will summarize our procedural preconditionsunder three headings.

PI. Rational identification of needs. Needs are defined and dis-tinguished from wants by appealing to an externally verifiablestock of codified knowledge, for example, knowledge about nutri-tion, child-rearing, or environmental control. The ability to tapand rationally to utilize this stock of codified knowledge—to engagein collectively identifying common human needs-is a first precon-dition for improving need satisfaction.

P2. Use of practical knowledge. At the next level, appropriateneed satisfiers have to be selected. Here we argue that thecodified knowledge needs to be complemented by the experiential-ly grounded understanding of people in their everyday lives. Forpresent purposes, we will assume that this knowledge can betapped in one of two basic ways. First, there is participation inmarket relations, where these are relatively unconstrained by con-

30 Ian Gough

tingencies of power or ignorance. Second, there are various formsof political participation and "claimsmaking" [Drover and Kerans1993]-the process whereby people collectively identify their dis-satisfactions, name their felt needs, and make claims against avariety of institutions.

P3. Democratic resolution. If a rational policy to identify andprioritize need satisfiers must draw on both codified and experien-tially grounded knowledge, then the inevitable disagreements thatresult must be confronted and resolved in a forum as open, asdemocratic, and as free of vested interests as possible. This is thethird procedural precondition by which different socioeconomic in-stitutions will be evaluated.

Next, "material preconditions" refer to the capacity of economicsystems to produce and deliver the necessary and appropriate needsatisfiers and to transform these into final need satisfactions. Weargue that there is a strong moral case for codifying the inter-mediate needs identified earlier in the form of state-guaranteedrights. However, the de jure codification of social or welfare rightsis no guarantee of their de facto delivery. To assess the latter, wedevelop a cross-cultural model of material production [Doyal andGough 1991, chap. 11] that yields four material preconditions forimproved need satisfaction. These include:

Ml. Production. The greater the total quantity and quality ofneed satisfiers produced, the greater the potential need satisfac-tion. The efficiency by which need satisfiers are produced is thusthe first of our material preconditions.

M2. Distribution. Next, need satisfaction is maximized if thesesatisfiers are distributed in line with the needs of individuals. Thisnormally entails individuals in households, though for certain col-lective satisfiers the unit of consumption is different and larger.

M3. Need transformation. These satisfiers are then transformedinto individual need satisfactions, a process that predominantlytakes place within (various sorts of) households. This, we argue,will reflect the distribution of satisfiers within the household, inparticular the degree of equality between men, women, andchildren. Final levels of need satisfaction will also be affected bythe direct effect of production processes and the quality of thenatural environment on human welfare.

M4. Material reproduction. The above processes take placethrough time, which requires that the stock of capital goods,natural resources, and human resources be at least maintained in


order to ensure further rounds of production and need satisfactionin the future. Though difficult issues of sustainability are raisedhere, a theory of human need must encompass material reproduc-tion and must extend beyond short-term horizons.

However, I have already indicated that to simplify theanalysis, this fourth material precondition is unfortunatelyomitted here. Thus, I am left with three procedural and threematerial criteria with which to evaluate different economic sys-tems.^

A Theoretical Framework for Macro-Social Analysis

Different economic arrangements are to be evaluated accordingto these criteria. To do this requires a set of theories and as-sociated knowledge with two major characteristics. First, theyshould be broadly applicable to a variety of socioeconomic sys-tems, yet be sensitive to the institutional variations betweenthem. Second, they should bridge the central fault-line in socialscience between the disciplines of economics and sociopoliticalscience. I will use for this purpose a body of works that can begrouped under the labels of "comparative political economy" and"the new institutionalism." This body of work has arisen at theconfluence of economics and sociopolitical science as a critique ofthe dominant paradigms in each: rational choice theory ineconomics and functionalism/behaviorism in sociology [Cammack1989]. It represents a return to the central concerns of classicalpolitical economy of Smith and his followers and to the critique ofthat political economy by Marx. Both were concerned with therelation between the economy and the state and with the effect ofsuch relations on human welfare [Esping-Andersen 1990, chap. 1;Gough 1979, chap. 1]. It also embraces the economic sociologypioneered by Weber [Holton 1992]. Let me briefly consider bothstrands separately.

On the one side, institutional economics emerged, initially inthe United States with the work of Commons and Veblen, whowere dissatisfied with neoclassical economics and desired to refor-mulate the discipline in at least three directions.^ First, technol-ogy and preferences are no longer conceived of as exogenous. Theeconomic environment is recognized as affecting access to infor-mation and the way that information is processed. This under-mines the view that individual agents are continuously

32 Ian Gough

maximizing or optimizing in any meaningful sense since theirpreferences are continually adapting in the light of their ex-perience. Therefore, second, the neoclassical assumption of equi-librium is replaced with the idea of agents learning and actingthrough real historical time. Economic life is characterized bystructural, not just "parametric" uncertainty, which imposes on ac-tors a reliance on routines and habits. These durable patterns ofbehavior define social institutions. The third characteristic of in-stitutional economics is thus a recognition of the role of institu-tions in economic life and a rejection of essentialist argumentsabout "the market." Self-seeking action and institutional struc-tures combine to generate a process through time characterized bylong periods of continuity punctuated by rapid breaks or institu-tional shifts. This paradigm also directs our attention to the in-stitutional contrasts between different economic systems.

Within social and political science, developments from a verydifferent starting point have resulted in a rather similar set ofpropositions. In explaining state activity within capitalist societies,the dominant paradigm was some form of structuralism, whetherframed by the requirements of industrial society and its economic,demographic, and bureaucratic correlates, or the requirements ofcapitalist society for the performance of accumulation andlegitimation functions. In both cases, the economy was conceptual-ized as isolated from social and political institutions, and the latterwere accorded no sources of autonomous development.

One attempt to overcome some of these problems can be tracedto central European scholars such as Polanyi [1957]° and Schum-peter [1976], for whom the interdependence of the market economywith the state and the community was a sine qua non. Anothersource of alternative thought has been the work of those scholarswho, in the tradition of Mill and deTocqueville, recognize the im-pact of democracy on state development. More recently, there hasbeen the project of "bringing the state back in" with its emphasison the state as an autonomous or independent actor, with certainspecific interests, that can act creatively to define problems anddevelop policy [Evans et al. 1985; Skocpol 1985]. All these ap-proaches attach little weight to the role of particular classes or so-cial agents in explaining state activity [Esping-Andersen 1990,chap. 1]. A common idea is that of institutional persistence and itscorollaries. Institutions are enduring, which means that at anytime any particular institution, including state structures, can be


"sub-optimal" or, to use a more explicit and loaded term, "dysfunc-tional" for the system as a whole [Cammack 1989].

Another strand in the reaction to structuralist perspectives insocial and political theory has identified social classes as a keypolitical agent. Developing from social democratic theorists of theAustrian school, this strand has emerged as the class mobilizationthesis associated with various Scandinavian writers [Korpi 1983;Esping-Andersen 1985]. Another source has been a "contradictory"Marxist analysis that stresses the role of class conflict in shapingsocial and state development [Gough 1979]. Alongside and partlycritical of these, a broader, more diffuse institutionalism hasdeveloped in recent years that recognizes the role of other institu-tions, including firms, other economic organizations, and bodiesrepresenting class interests. This sociological institutionalismvaries according to whether or not it countenances an explicit rolefor structural or environmental forces alongside institutional be-havior in explaining policy developments.

One prominent characteristic of all these sociopolitical or "his-torical-structural" schools of new institutionalism is a view of in-stitutional change as discontinuous, contested, and problematic.Another is the situation of societal and state-centered variableswithin a more systemic framework. For example, according toHall [1986], the major components in explaining changes in publicpolicy are the organization of capital, labor, the state and thepolitical system, and the position of the nation within the interna-tional political economy. Within this field, however, institutionsresist change and develop in a path-dependent manner.

Taken all together, these "new institutionalist" theories mark aconvergence between economic and sociopolitical analysis, whichprovides a fertile framework for a macro-social analysis ofeconomic institutions. In particular, they enable comparisons tobe made of different socioeconomic systems and different stages ofdevelopment. It is this framework of institutionalist thinking thatI will use to derive hypotheses concerning the impact of differenteconomic systems on levels of need satisfaction.

A Taxonomy of Economic Institutions

Economic debate and policy prescription today is dominated bythe respective merits of markets and public planning, so much sothat it is tempting to focus on free market capitalism, centrally

34 Ian Gough

planned economies, and various sorts of market-planning mix.However, this would neglect a third set of economic relationshipscurrently being rediscovered in the economics literature, which canbe gathered under the label "community." The list of writers thusdistinguishing three fundamental forms of economic organizationis long. It includes economic historians [Polanyi 1957, chap. 4; Bos-well 1990], political scientists [Streeck and Schmitter 1985],sociologists [Bradach and Eccles 1989], organizational theorists[Powell 1990], and institutional economists [Thompson 1991].Table 1 illustrates the key concepts identified in some of thesetaxonomies.

Table 1. Taxonomies of Economic Systems

AuthorPolanyi

Streeck/Schmitter

Bradach/Eccles

Powell

PrincipleForms ofintegrationPrinciples ofcoordinationand allocationEconomiccontrolmechanismsForms ofeconomiccoordination

MarketMarketexchangeDispersedcompetition

Price

Markets

StateRedistribution

Hierarchialcontrol

Authority

Hierarchy

CommunityReciprocity

Spontaneoussolidarity

Trust

Networks

Drawing on Polanyi [1957] and Putterman [1990, chap. 1], wecan substantively define the economy as the sphere of social ac-tivity in which people produce, distribute, and consume thematerial requirements to meet their wants and needs. Thisgenerates recurring interactions among elements and agents inthe system. According to all the major representatives of classicalpolitical economy, including Smith and Marx, a major feature ofsuch interaction is a division of labor, both within "enterprises"and between them. This division of labor raises productivity but inturn requires some mechanism or mechanisms for coordinating theactions of the numerous interacting agents. It is to this fundamen-tal question that the three solutions identified above have emergedover the course of human history.^

First, I look at markets. Here private agents exchange entitle-ments to goods and services with each other. Where a large num-


her of such commodity exchanges regularly take place we canidentify the social institutions of a market. This fonn of coordina-tion entails private rights in the use, consumption, disposition,and fruits of economic resources and goods and the rights to trans-fer these rights, except the ownership of labor [Putterman 1990,59-60]. The prices or terms at which these exchanges take place isdetermined solely by the free negotiation of the parties concerned.Thus, economic coordination is decentralized, ex post, and uncon-scious.

The second form of coordination is by authoritative regulationsissued in hierarchical organizations. Where these organizationsare themselves coordinated by authoritative regulation backed bycoercion, we may speak of a state system of coordination. Such asystem normally entails state ownership of the means of produc-tion, apart from labor. Coordination here is thus centralized, exante, and conscious.

The third ideal-type form of economic coordination is more dif-ficult to specify. Nowhere in the modem world does it provide ageneral mode of economic coordination, though it exists withincertain sectors such as some voluntary organizations and socialmovements. When "community" is advocated as a normativemodel of a desirable economic system, it appears in differentguises. Excluding those who explicitly identify community withpre-modem, hierarchically organized, status-bound societies, weare still lefl with a great variety of views. First, there is the ideaof communism held by radical socialist thinkers such as Marx,Morris, and Kropotkin [Miller 1989]. This idea has been revived inthe last three decades in response partly to the belief thatdevelopments within capitalism are laying the foundations forcommunitarian economic relationships (in, for example, the workof Gorz [1982] and Van der Veen and van Parijs [1987]). Second,at the opposite pole, there is the libertarian view of communityespoused by Nozick [1974]. Here membership of a community isvoluntary and self-chosen. Third, there are new attempts to con-ceptualize a "democratic communitarianism" drawing upon thecurrents of decentralized socialism, personalist Christian democ-racy, ideas of corporatism, and civic humanism [Boswell 1990].This strand tends to equate community in the modem world withnational citizenship [Miller 1989]. The last two conceptions havebeen explicitly concerned to augment those of market and state,not to replace them.

36 Ian Gough

However, underlying these differences are some commonthemes distinct from the other two modes. Economic coordinationwithin communities is by democratic negotiation. Solidaristic sen-timents of loyalty and reciprocity within social groups facilitatesuch consensus-building. The opposition between separatist in-dividualism and state collectivism is overcome by a new focus onthe quality of human relations. Coordination may thus be con-ceived as decentralized, ex ante, and conscious.

In the next three sections, I will evaluate the potential con-tribution of these three forms of economic coordination to the satis-faction of human needs. The intention is to try, so far as ispossible, to abstract from real-world complexities by consideringthese three economic systems as "ideal types." However, this is notstrictly possible. According to the "impurity principle," any actualsocioeconomic system will contain, alongside its dominant prin-ciple, at least one other economic structure based on different prin-ciples for the whole to function [Hodgson 1984, 85-9 and 104-9].Thus, market economies must incorporate a system of authorityand operate within a set of specific social relationships. A puremarket society is a logical contradiction. Similarly, a centrallyplanned economy encounters contradictions that can only beresolved via decentralized market and civil relationships. In thesetwo cases, then, I consider models that incorporate the minimumdegree of "impurity" or contamination by other principles, drawingon empirical and historical evidence where appropriate. The thirdform of economic coordination, via community networks, poses dif-ferent problems since it has not existed as an even modestly self-sustaining form in the modem age. I will consider brieflyconceptions of community as an overarching principle of economiccoordination before again pointing out the dependence of such aprinciple on the other two modes of economic coordination.

Free Market Capitalism

The defining characteristic of a free market economy is thateconomic coordination is decentralized, emerges as a result ofvarious individual actions, and is not consciously controlled. Freemarket capitalism is used here to refer to a combination of thisform of coordination and private property ownership. It is this thatI will now briefly evaluate according to the societal preconditionsfor optimizing need satisfaction sketched earlier.


Our first procedural precondition is excluded by definition.There is no collective sphere of economic decision making. Freemarket capitalism is an economic system that responds to con-sumer wants backed by money-a system in which literally nobodymakes decisions on the composition of output and its relation tohuman needs. Against this it can be argued that capitalism ex-tends, to the maximum feasible extent, people's freedom tonegotiate about human needs. If individuals are sufficiently wellinformed and have the freedom to act on that information, then itcan be claimed that subjectively defined wants will closely ap-proximate generaiizable human needs.

The second procedural precondition-that the practical know-ledge of people be effectively tapped in identifying improved needsatisfiers-constitutes a strong claim for markets. Markets makeuse of the dispersed knowledge of millions of actors, and the con-tinual process of discovery that they are free and able to make en-genders the restless innovation and productivity of capitalism.

The third condition-a democratic forum within which debatesover how best to meet needs can take place—is also claimed forcapitalism. The conjoint development of capitalism and of certainforms of representative democracy in much of the world stemsfrom the decentralization of decision making and power in marketsociety. Furthermore, decentralized "claimsmaking" is facilitatedif citizens have the rights and capacities to form associations topress their own perceptions of needs and satisfiers within thepolitical arena.

Against these claims must be set much recent analysis of un-regulated markets and their political implications, which isrelevant to our first procedural precondition. Wants can divergefrom needs in significant ways, abetted by market society.Markets are an inefficient source of knowledge and can interferewith the communication processes necessary for human needs tobe identified and agreed upon. For example, it can be in the inter-ests of individual producers to supply distorted information if thiswill maximize profits and if they are able to do so. The sheer num-ber of commodities produced in developed market societies meansthat consumers have inadequate knowledge of their charac-teristics and insufficient time to find out. Furthermore, it hasbeen argued that unregulated market society undermines com-munal ties around which less individualistic conceptions of needcan form. If wants are endogenous to the economic system, this

38 Ian Gough

undermines any simple view of the market as a responsemechanism to autonomous consumer desires.^^

Unregulated markets can also distort the nature of democraticdebate within the political sphere. It is rational for actors to at-tempt to pursue their interests within the political sphere, result-ing in democratic abuses such as clientelism and worse. On theother hand, markets offer commodified need satisfiers to those whocan afford them, which undermines their incentive to participatein political debate over alternatives. Opportunities for exit reducevoice. Last, the reliance of all existing market societies on agendered division of labor constrains the ability of women to par-ticipate in democratic fora [Bowles and Gintis 1986, chaps. 4-5].The implication of these critiques is that either authoritativeregulation or collective sentiments or both are necessary correc-tives to the unconstrained pursuit of individual self-interest inmarket settings if human needs are to be recognized andprioritized.

Turning to material preconditions, the claims of marketcapitalism are strong. Markets not only utilize the dispersedknowledge of millions of separate actors, but they also providethem with incentives to act on that information in such a way as tomaximize efficiency at a moment of time [Hayek 1948; Gray 1992,chap. 2]. Though the strong claims of Pareto efficiency require un-realistic assumptions, the argument that markets enhance produc-tive or "x-efficiency" remains. The ability of capitalism to producegoods in prodigious quantities and to innovate totally novel kindsof products is of considerable relevance in assessing its contribu-tion to the satisfaction of human needs. However, it is not real-world capitalism that is investigated here, but a model of aminimally regulated market economy.

This model has several major weaknesses, according to a longhistory of economic analysis. The failures of unregulated economicmarkets to satisfy consumer wants are so well known that I willnot detail them here. They include tendencies to monopoly, the in-ability of markets to supply public goods, the self-defeating produc-tion of positional goods, and the inefficiency or diswelfares causedby the tendency of markets to meet wants in commodified forms[Penz 1986]. Laissez-faire capitalism may be an efficient systemfor satisfying certain wants by means of commodities, but that isall. There are further limitations stemming from the untram-melled pursuit of individual self-interest. This engenders profound


uncertainties, results in "prisoner's dilemma" situations where alllose out in the absence of cooperation, and encourages oppor-tunism and short-termism, which harms longer-term conceptionsof self-interest.

Second, the distribution resulting from the operation of un-regulated markets cannot, even in rich societies, offer entitle-ments to basic need satisfiers to all citizens [Sen 1981]. Regardingour third material precondition, need transformation, marketsociety will tend to dissolve non-capitalist relations includingthose between the sexes. However, recent scholarship contendsthat gender inequalities are perpetuated via the conjunction ofpaid and unpaid work and the institution of monogamous mar-riage. A market economy requires altruistic, collective behavior onthe part of women in the household because their unpaid laborprovides a flexible cushion that permits men to respond to marketsignals [Elson 1988]. Thus, formal equality coexists with genderedinequality, that in turn affects the levels and distribution of needsatisfactions in society [Pateman 1988; Bowles and Gintis 1986,chap. 4]. Furthermore, capitalism can harm those needs directlymet in relations either at work or in the community. Theautonomy of workers may be undermined once firms and factoriesbecome established institutions in market societies and the tech-nical division of labor is extended [Wood 1982]. At the same time,the erosion of community bonds creates new diswelfares for whichmore and more commodities cannot necessarily compensate[Hirsch 1976].

To conclude, then, a minimally regulated, free marketcapitalist society suffers from many drawbacks as an institutionalsetting within which human needs can be satisfied. On both proce-dural and material grounds, it is found wanting. As Polanyi hasargued, a strict market economy (even with the concessions to theexistence of other institutions made here) is neither desirable norlogically possible. The implication of much institutional economicanalysis, as well as of political science and sociology, is that torealize their procedural and material potential, market relationsneed complementary regulation by public authorities and by net-works of more solidaristic relations in civil society-what Etzionicalls the "social capsule" [Etzioni 1988; cf. from differentstandpoints Wolfe 1991 and Gray 1992]. It is much more interest-ing to investigate various forms of mixed economic systems. But

40 Ian Gough

before I do that, I should turn to another relatively homogenouseconomic system-^ centrally planned economy.

State Sociali8m

Here the dominant form of economic coordination is planningby a central authoritative apparatus. Though this rules out privateproperty ownership, it does not entail any single form of collectiveownership. However, in historical practice (except in what wasYugoslavia), de jure ownership of the bulk of capital and land hasbeen vested in the central state. Again, the extent to which thiseconomic system presupposes a specific political form of the stateis disputed. However, in all real-world cases since 1917, and beforethe revolutionary reforms of 1989, representative democracy wasdenied and official communist parties exercised a pervasive andpowerful role in the state apparatus. It is these forms of propertyownership and state that will be assumed in what follows.

State socialism, inaugurated in 1917, had as its conscious goalto replace market-determined production for profit by plannedproduction for human needs. Of course, such a system can have,and has had, different goals, such as victory in war or crash in-dustrialization. Moreover, Marx and his followers drew a distinc-tion between communism and socialism, between the terminus anda station along the way. But let us accept, for the purposes of thisargument, that Marx's slogan "To each according to his needs" isindeed the final goal of socialist society. How well is the modelstate socialist economy sketched above able to realize this goal?

It forms a stark contrast with the previous model of unregu-lated capitalism. Centralized planning to meet needs takes centerstage, whereas citizen participation, whether as economic or politi-cal actors, is marginalized. In light of our procedural precondi-tions, there are benefits and disadvantages. On the one hand,codified knowledge can in theory be utilized effectively to identifyneeds and to marshall resources to meet them, especially in condi-tions of underdevelopment and scarcity. On the other hand, thereare few sites where the experientially grounded knowledge ofpeople can be utilized. They are denied a creative role in theeconomic sphere. They are also prevented from making claims incivil society and within the workplace. Finally the absence of civiland political rights undermines the capacity of the political process


rationally and democratically to adjudicate on different ways ofimproving need satisfaction.

All these are clear procedural defects. What is more, the onepositive feature, a planning apparatus committed to improvinghuman welfare, has in practice severe limits. The political elite isrelatively insulated from other points of view, which results in adistortion of the codified knowledge upon which planning is based.The bureaucratic planning apparatus acquires considerable powerto pursue its own interests, establishing in the process what hasbeen called a "dictatorship over needs" [Feher et al. 1983].

As regards our material preconditions, state socialism exhibitsseveral advantages in principle over unregulated capitalism, atleast at low levels of development. The planning apparatus canprioritize the production of basic need satisfiers such as basiceducation, primary health care, basic foodstuffs, or family plan-ning services. Entitlements to these can be ensured via suchmeasures as a radical redistribution of land assets, full employ-ment policies, and direct public provision of services (thoughregional variations in distribution are often harder to combat).Need transformation can also be enhanced via policies to educateand improve the status of women, to control births, and to providealternative forms of child care.

Against this must be set the case that central planning en-counters growing problems of coordination, and notably so aseconomic development proceeds. Centralized planning, even aidedwith the most modern computers, cannot effectively coordinateeconomic transactions in a modern economy where the number ofdifferent products runs into the millions. The result is that large-scale projects and homogenous products are given priority at theexpense of many essential yet disparate need satisfiers. Com-partmentalization of interests within the planning apparatus in-terferes v̂ rith the adjudication between projects. More problematic,at the enterprise level it is extremely difficult to set targets tomotivate managers and workers to produce what the plan re-quires. Where targets are expressed in physical terms, factorieshave an incentive to distort products in order to achieve targetspecifications. This results in shortages and poor quality, whichembraces many key need satisfiers such as housing [Nove 1983].

At the distribution level too, the commitment to work-relatedrewards can discriminate against those, especially women, with amore tenuous link with paid labor, while special nonmonetary

42 Ian Gough

benefits for the elite undermine overall equality. There is nocountervailing system of distribution to that determined by the of-ficial economy [Szelenyi 1978]. Despite a formal commitment topolitical and economic equality for women, the structural proper-ties of state socialism deny this in practice. Shortages and the at-tendant queues add to the double burden stemming from agendered division of domestic labor. Not only does this harm theneed satisfaction of women, it can interfere with the effectivenessof the need transformation process and thus the overall levels ofneed satisfaction.

As with market economies, analysis and evidence suggest thata relatively pure command economy is neither desirable norfeasible, according to our need-related yardstick. This is perhapsmore surprising since both the intent and ideology of statesocialism have proclaimed the meeting of human needs as an ex-plicit and high-priority goal. Yet the conclusion is clear: marketsand networks in civil society are necessary to overcome thedeficiencies of a pure central planning model. And indeed, this iswhat was found in all state socialist societies between 1917 and1989, albeit in distorted forms. All exhibited, alongside the officialeconomy, what Markus [1981] calls "second" and "third"economies. The second economy was comprised of self-employedand private production units together with "moonlighting" andother unlawful enterprises. The third economy embraced the sys-tem of "tolkachi"-networks of informal relations between andwithin the bureaucracy and state enterprises formed to overcomethe mutual problems that they faced.

Community, Communitarianism, Communism

If community as a generalized system harks back to a mythicalpast, communism reaches forward to a Utopian future. Asdeveloped by the Utopian socialists and even their critic, Marx, it isa society of absences: without markets and money, without state,without hierarchical, horizontal, and sexual divisions of labor,without inequality and scarcity. The tension between individualself-interest and collective interest is overcome through the trans-formation of social relations and human identity. For many critics,this vision is not logically realizable: it is "evasive, confused andproblematic" [Soper 1981, chap. 9]. In particular, it overlooks theconstraints stemming from human psychology, human biology, and


the limits to the biosphere. Yet the last two decades have wit-nessed a renewed interest in communitarian alternatives, partlyin response to a view that technology and other trends make pos-sible a post-industrial society of one form or another. These takeon board some of the above critiques, but hold onto some of thebasic tenets of a communitarian position: principles of spon-taneous solidarity, relations of reciprocity, and small-scale com-muntities with participatory democracy. An example is Gorz's[1982] vision of a dual society, which combines a "heteronoous"domain of work and authority alongside an autonomous domain ofself-determined activity, where the latter is dominant.

In terms of our procedural preconditions, such a model has onemajor advantage. It permits what Miller calls "dialogicdemocracy," a form of negotiation in which genuine learning takesplace including learning about basic needs and how to meet them.People's experiential knowledge can be tapped, but sectional inter-ests can be negotiated in a forum that would encourage the reach-ing of a democratic and rational consensus [Miller 1989, chap. 9;Doyal and Gough 1991, chap. 7]. Moreover, such democraticnegotiation would extend beyond the formal political arena toembrace work relations and certain other relations within society.

It is likely that such an arrangement would also permit theutilization of both codified and experiential knowledge in thedomain of production. The deficiencies of markets and state plan-ning can be mollified if networks and negotiation generate alter-native sources of information and motivation. This is moreprobable if they are based on relations of trust, reciprocity, andmoral obligation [Boswell 1990, chap. 2]. Such a moral solidaristiccommunity could prioritize the production of need satisfiers, dis-tribute them according to urgency of need, and reorder interper-sonal relations to develop gender equality and more effective needtransformation. In this way, collective needs can be asserted overindividual wants as the dominant goal of a communitarianeconomy.

Against this must be set several fundamental problems. First,if such communities are "all-embracing" like medieval monas-teries, they risk coercing their members into agreement about theends of life and the goals that individuals ought to value and pur-sue. Individuals are "engulfed" by the community-in other words,one of their basic needs, autonomy, is severely restricted [Plant etal. 1980, chap. 10; Miller 1989, chap. 9]. This can be overcome if

44 Ian Gough

membership of communities is voluntary, as Nozick [1974] recom-mends. But then another problem is faced: some individuals-"mis-fits" and outsiders-would not be accepted by any community.Excluded from the only social systems that offer participation andsustenance, it is almost certain that their need satisfaction wouldbe threatened. A more general procedural problem arises becausesolidaristic communities create distinctions between insiders andoutsiders, which inhibits the fostering of universalizable interestsand thus the identification of true human needs. To overcome thisrequires some higher level of authority that is separate from andsuperordinate to the separate communities.

In terms of our material preconditions, communitarianism ap-pears at first sight superior to the other two alternatives. Butagain, this is to simplify the relations of modem economies (or im-plies turning one's back on the whole process of modernization andthe progress in meeting needs that this has sustained). Most com-munitarian solutions pay insufficient attention to the problem ofcoordination. For Gorz, local exchange of the products of small-scale enterprise would be via the medium of "labor-time vouchers."But as Nove [1983] has argued, either this requires very small-scale production, in which many of the productive advantages ofcontemporary capitalism are lost, or the value of the voucherswould need to fluctuate according to supply and demand, in whichcase they would be indistinguishable from money. Intercommunityrelations on a broader scale are still more intractable.^^ Moreover,communitarian advocates like Gorz tend to evince a romantic viewof unpaid, communal, and household labor, ignoring much recentfeminist scholarship and its argument that "community is fun-damentally a gendered concept" [Finch 1984,12].

For these and other reasons not adequately covered here, "com-munities," even democratic and need-prioritizing ones, cannot bythemselves mobilize the resources necessary to optimize the needsatisfaction of their members. I have spent a short time on thisthird set of economic institutions to disabuse any lingering beliefthat "community," "reciprocity," "networking," or "negotiation" canby themselves provide a third alternative to economic organizationand a surer way to meet human needs. It also sets the scene for anintegration of community with market and state, as proposed bysome recent writers.


Summary So Far

Table 2 summarizes the pros and cons of the three "pure" orparadigmatic systems of economic coordination as frameworks forthe optimization of need satisfaction. We may summarize theirdeficiencies another way by returning to the ideas of need thatthey each embody. Free market capitalism essentially equatesneeds with wants, an equation that is logically flawed and morallyuntenable [Doyal and Gough 1991, chaps. 2 and 61. Statesocialism by contrast operates with an idea of universal and objec-

Table 2. Evaluation ofThree 'Pure" Economic Systems

CriteriaPI. Rationalidentificationofneed8

P2. Use ofpracticalknowledge

P3.Democraticresolution

Ml.Productionof needsatis flers

M2.Distributionaccording toneed

M3.Effectiveneedtransfor-mation

Market: UnregulatedCapitalismAbsent. Unregulatedmarkets weaken "socialcapsule'VcoUective ethic

Markets tap but distortdispersed knowledge

Representativedemocracy weakenedby market exit andinequality

EfHciency in commodityproduction but marketfailures and absenceof noncommodity forms

No entitlements toneed satisliers

Potential for autonomouslearning harmed byinequality in work andunpaid household labor

State:StateSocialismPrioritization of,but dictatorshipover, needs

Absent anddiscouraged

Certain socialrights butabsence of civiland politicalrightsPrioritization ofneed satisfiers.but informationand motivationfailures

Entitlementsdistorted byabuse and labormarket links

Autonomouslearning restrictedat work, inconsumption, andvia unpaid house-hold labor

Community:CommunismRational use ofcodifiedknowledge, butincorporation ofindividualsRational use ofdispersedknowledge within.not between.communitiesWidespreaddialogicdemocracy.Absence ofcodified rightsPrioritization ofneed satisfiers.but problemsof coordinationbetweencommunitiesEntitlements toneed satisfierswithin, notbetween.communitiesGreater freetime plus auto-nomous domain.but gendered/householdinequalities?

46 Ian Gough

tive need but equates this with the views of the party and statefunctionaries. Need is identified with one particular form ofcodified knowledge, which reflects constellations of power incom-patible with the pursuit of truth. Communitarian models interpretneed as those interests defined by particular cultural groups orcommunities. They thus make relative the idea of universalhuman need and denude it of an evaluative or moral role. None ofthe three systems embody a notion of human need that is univer-sal and objective, yet open-ended and cumulative.

I now tum to see how far this ideal can be realized withinvarious forms of mixed or "impure" economic systems. I focus heresolely on mixed capitalist systems, that is, where markets have adominant role in economic coordination and where private owner-ship of the means of production is the dominant form of propertyownership. According to the tripartite model developed above, thisgenerates two fundamental forms of capitalist mixed economy. Thefirst is statist capitalism, where market coordination is accom-panied by a substantial degree of state steering of the economy.The second is corporatist capitalism, where the market is accom-panied by coordination via networks of negotiation between keyeconomic actors. Where both of these are absent, or weaklydeveloped, or deliberately undermined, we may identify a thirdvariant: neoliberal capitalism.

I now look at each of these in turn. Though we are here movingfrom "as pure as possible" economic systems to "impure" or mixedsystems, I continue to abstract from the complexities of the realworld and to analyze models of idealized mixed systems. A real-world economy, such as Germany's, will in practice exhibit fea-tures drawn from all these in a bewildering array.

Statist Capitalism

Statist capitalism may be interpreted as a return to seven-teenth and eighteenth century mercantilism. But it can claim atleast three intellectual and historical sources since the emergenceof industrial capitalism in Britain in the later eighteenth century.First, and most important, is the continental European perspectiveof statism associated with the writings of Weber, Hintze, List, andWagner, among others [Skocpol 1985]. This stresses the existenceof "states" (as distinct from "governments") that develop extensivecapacities and a wide range of roles. These states engage in rela-


tions with other states, promote economic development deemedessential for the competitiveness of second-order industrializingcountries, and develop social policies to enhance social integra-tion. This case for statist capitalism is essentially integrative anddevelopmental. Second is the "socialized liberalism" of Anglo-Saxon thought, beginning with J. S. Mill and continuing in thewritings of other "reluctant collectivists" such as Keynes andBeveridge. Here the case for state intervention may be typified aspragmatic and reactive. Third is the strategy of welfare statismdeveloped by Fabian social thought and social democratic politicsin the twentieth century. Here the state's role is proactive andegalitarian. These three strands-conservative, liberal, andsocialist-have thus generated different conceptions of theeconomic and social role of the state. There has been no singleroute to statist capitalism.

To help define the concept of statist capitalism further, weneed to consider in more detail the distinctive roles of the state ineconomic and social intervention. These can be classified invarious ways. According to Putterman [1990, chap. 2.5],capitalism can be modified by means of four types of intervention.First, and, he argues, least contradictory to the essence ofcapitalism, is to correct for market failures such as monopoly, ex-ternalities, and the inability to provide public goods. Second is tomodify the distributive results of market mechanisms combinedwith private ownership via an assortment of redistributivepolicies. Third, there are a set of reactive macroeconomic interven-tions intended to correct for systemic market failures in the factormarkets for capital and labor, of which Keynsianism is the best-known example. This third form indicates that the market cannotbe self-regulating in important areas of activity. Fourth, there areproactive interventions to steer the economy in a desired direc-tion. These last, including indicative planning and specific or"parametric" industrial policies, attempt to provide direction toeconomic activities at the industry or enterprise level. I will definestatist capitalism, as an ideal type, as a system where all fourlevels of state intervention are practiced. ̂ ^

Let us now evaluate statist capitalism according to our proce-dural and material preconditions for optimizing welfare. Intheory, it can overcome the deficits of laissez-faire capitalism. Interms of our procedural preconditions, collective interests can bedefined in a nonutilitarian way and asserted over powerful sec-

48 Ian Gough

tional interests. State planning can provide the means to prioritizecertain basic needs as goals of policy and can act to modify or steerthe market where it prevents their achievement. Democratic,educational, and administrative processes can supplement, orwhere necessary override, self-interested action in the market toimpose universal, need-oriented values over the pursuit of privatewants and sectional interests.

In the material domain, market failures can be compensated orregulated to improve the composition of output in a welfare-oriented direction. Thus, monopoly and externalities can be taxedor regulated by public authority. At the same time, the strengthsof markets as mechanisms for identifying need satisfiers, notablythose that take the form of commodities and are congruent withwants, are retained. At the distribution stage, the lack of entitle-ments of the poor and the maldistribution of resources according toneed can be corrected by using the familiar instruments of the wel-fare state. These can include not only taxation, social securitybenefits, and public services, but also wage and price and trainingpolicies designed to alter the distribution of primary incomes. Last,the effectiveness of the need transformation process can be im-proved in at least two directions. Equal opportunities legislation,public support for child care, and other family-support policies candiminish gender inequalities, while employment programs ofvarious kinds can reduce unemployment and thus directly enhanceeconomic participation with benefits for individual autonomy.

These potential benefits of a "mixed economy welfare state"have come under attack in recent years from proponents of theNew Right, who contend that govemment failure is always andeverywhere more likely than and more pernicious than marketfailure. In practice, they contend, none of our procedural precondi-tions are met in a mixed economy, even if they were desirable.State intervention is not rational because it cannot concentrate thedispersed knowledge of actors in a single body; to imagine other-wise is to suffer a "synoptic delusion." Nor is such interventionuniversal in intent, since governments are susceptible to numerouspressures from organized interest groups seeking to advance theirown interests, and these pressures are self-reinforcing. At thesame time, state action weakens the effectiveness of markets andthus their ability to identify those needs that are congruent withwants. State intervention also generates inefficiency and"sclerosis" in production to meet needs, both directly within the


public sphere and indirectly by harming the efficiency of theprivate sector in a variety of ways. Thus, democratically managedcapitalism negates its own goals and undermines the materialbasis for satisfying its citizens' needs.

Many of these criticisms have in turn been criticized by writerswithin the institutionalist school. The public choice models ofDowns, Olson, and others do not in fact predict human behaviorvery welL People often act unselfishly or in "unproductive" waysby voting in elections, by not cheating when nobody is looking, andso on. Furthermore, these neoliberal critics adumbrate an essen-tialist view of the market and the state. Neither are situated inrelation to each other, or in the context of the moral and socialorder of which they are a part.'^

Nevertheless, a new institutionalist analysis would recognizecertain elements of truth in the neoliberal critique and add somemore defects of its own. An interventionist state entails a dangerof clientelist politics, wherein special interest groups can lobby or"capture" state agencies to pursue their specific interests. Thisdanger is especially pronounced when state intervention expandsfrom the first to the fourth of the categories above-fromparametric regulation to enterprise-specific regulation. Ratherthan the state representing the public interest and imposinggeneralizable goals over sectional interests, sectional interestsmay extend the pursuit of their goals through political as well asmarket means [Skocpol 1985; Rueschemeyer and Evans 1985]. Bydefinition, this will impede the identification of universal needs.At the same time, state intervention may lack legitimacy andstability. Neither bureaucratic nor technocratic rationality is ade-quate once state intervention shifts from allocative to productiveactivity [Offe 1975; Mayntz 1983].

Turning from procedural to material preconditions for im-proved need satisfaction, statist capitalism is still vulnerable, al-though it exhibits several advances over minimally regulatedcapitalism. States may lack either the willingness or the capacityto intervene in the appropriate ways (or both). The former re-quires a minimum degree of autonomy from dominant forces incivil society, thus making it vulnerable to the problems discussedabove. It also requires bureaucratic capacities, including materialresources, a bureaucratic "esprit de corps," and access to relevantexpertise, which are not always forthcoming. Moreover, wherethey are forthcoming, they may well generate further limits. The

50 Ian Gough

limits of bureaucratic state regulation are by now well rehearsed.The lack of detailed, "thick" information, or experientialknowledge, leads to the formal application of rules, which cangenerate inefficiencies [Rueschemeyer and Evans 1985]. Where thestate is directly delivering need satisfiers in the form of public ser-vices, this can lead to the abuse of clients and the provision of in-appropriate satisfiers [Doyal and Gough 1991, chap. 14]. Togetherthese procedural and material defects can generate anauthoritarian, corrupt, and (what is referred to in Brazil and else-where as) an "anti-welfare state," which acts to protect the inter-ests of powerful groups at the expense of the needs of the mass ofthe people.

At best, a proactive state is no more than a means for theachievement of a needs-oriented policy: it may be a necessary con-dition, but it cannot be sufficient. Statist capitalism may be moreconducive to meeting human needs than unregulated capitalism,but the answer is indeterminate in the absence of further informa-tion on the direction and nature of state policy. To answer this, wemust turn to the nature of civil society and the case for a thirdmode of economic coordination.

Corporatist Capitalism

Institutionalist economics argues that successful market rela-tions need to be "embedded" within not only a system of publicauthority, but also a network of relations in civil society. Markettransactions in conditions of uncertainty require a degree of trustbetween the parties that they will behave according to the agree-ment [Bardach and Eccles 1991]. On this basis, networks ofrelationships that sustain trust are featured as a third form ofeconomic coordination. In contrast to market or hierarchy, thesecoordinate through less formal, more cooperative and negotiatedmeans. These in turn enhance a longer-term perspective and abroadened conception of self-interest, which help reproduce thenetworks over time. ̂ ^

The other major contributor to a renewed interest in such a"third way" has been the emergence of democratic "corporatism"notably in Western Europe in the 1960s and 1970s, which has beentheorized by Schmitter, Lehmbruch, Streeck, and others [see Wil-liamson 1989 for a survey]. There are two basic components: first,the centralized organization and representation of major interest


groups in society and their mutual bargaining, and second, theregular incorporation of these groups into the policjrmakingprocess via bargaining with the state and political parties (some-times called "concertion"). Katzenstein [1985, 32] adds a third fea-ture of democratic corporatism: an ideology of social partnershipthat integrates differing conceptions of group interest with vaguebut firmly held notions of the public interest.

This third form of economy has been conceptualized as a dis-tinct "associative" logic of social order by Streeck and Schmitter[1985] and as a "democratic communitarian" third way by Boswell[1990]. Both recognize various historical antecedents, includingDurkheim's writings on solidarity and corporations, personalistChristian democratic thought, the doctrines of the RomanCatholic church (in particular the Papal encyclicals of 1891 and1931), and the associationalism of Saint Simon and earlysocialists. Boswell has done most to theorize this third way as aderivative of communitarian thought. Rejecting all-inclusive com-munities, for reasons similar to those advanced above, he arguesin favor of fostering "fraternity" and participation in largergroups. He claims that the nation-state is still the prime site ofsuch community identification today [Boswell 1990, chap. 3]. Thisis close to Miller's [1989, chap. 9] argument that nations are theonly possible form in which an overall community can be realizedin modern societies, so long as this community is sited within apolitical organization of citizenship.

Economic forms of such a national community can be fosteredin various ways, including corporate public responsiveness andcollaborative industrial relations. In all these examples, "externalcolloquy" is the crucial element that prevents organizations frompursuing their own narrow goals and from defying the public in-terest. Perhaps the most notable modem-day example of this"public cooperation" is corporatist participation in certainEuropean countries. This parallels Streeck and Schmitter's [1985]characterization of "associationalism" as a distinct model of socialorder in the modern world. ̂ ^ Here collective actors of functionallydefined interest associations are constrained and enabled to relateand negotiate with each other. "The central principle is that ofconcertation, or negotiation within and among a fixed set of inter-est organizations that mutually recognize each other's status andentitlements and are capable of reaching and implementing rela-

52 Ian Gough

tively stable compromises (pacts) in the pursuit of their interests"[Streeck and Schmitter 1985,10].

To explain the emergence and persistence of these structures of"public cooperation" or "responsible associative governance," twodistinct approaches have typically been adopted drawing onDurkheim and Marx, respectively. The first looks for features inthe social structure that enhance solidarity, such as the continuityof organizations, their numbers in relation to the size of the nation,the background proximities between decision makers, and thesalience of communitarian beliefs [Boswell 1990, chaps. 5-9]. Thesecond, however, explains them in terms of class structure, power,and conflict. Workers have an incentive to unite and pursue collec-tive action to overcome their individual powerlessness in the labormarket. The two dominant power resources that they can con-struct are trade unions and class-based political parties. Accordingto Przeworski [1986], it would be rational for a workers' move-ment, under plausible assumptions about the behavior ofcapitalists and workers, to pursue a strategy of accommodationwith capital. Thus, corporatism is another label, and a confusingone, for societal interclass conflict and bargaining [Korpi 1983;Esping-Andersen 1985,1990].

On the basis of this second perspective, Katzenstein [1985,chap. 3] distinguishes two fundamental forms: liberal corporatismand social corporatism. The former is found where powerful andcentrally organized business communities confront relativelydecentralized and weak labor movements. The latter is foundwhere there exist strong, centralized, and politically powerful laborunions, with or without an equivalent business community.^^ Thework of Esping-Andersen [1990] suggests that liberal corporatismis often combined with the influence of Christian democratic ideol-ogy, whereas social corporatism is the associate of socialdemocratic ideology. Thus, the two explanations may be combinedto explain in different ways the persistence of two distinct forms ofcorporatism, concertation and public cooperation. Both, however,envisage corporatist arrangements as a complement to the role ofmarket and state. Most analyses also assume a substantial pro-ac-tive role for the state. Associationalism is thus, in practice, com-bined with statism to form a hybrid third form of capitalism.

What, then, are the pros and cons of corporatist capitalism as aprocedural and material framework for the improvement and op-timization of need satisfactions? At a procedural level, corporatist


capitalism offers several advantages. By encouraging reciprocity,shared norms, and trust, it nourishes a rational yet democraticprocess of identifying collective interests and thus, potentially,universal human needs. By retaining the informationalmechanisms of the market, it enables practical knowledge to betapped. Yet by fostering dialogic democracy, it discourages ashort-term view of economic self-interest and the incentive forthose with money to exit while enhancing the mechanisms ofvoice. Futhermore, Offe and Wiesenthal [1980] argue that in theprocess of class struggle and bargaining, workers' organizationscan only achieve their interests by partially redefining them. This"dual logic of collective action" means that the labor movement in-terprets material well-being broadly, moving beyond sectionaleconomic interests toward something approaching a conception ofbroader human needs. This suggests that institutions of social cor-poratism will tend to pursue need-related goals to a greater extentthan those of liberal corporatism.

Against this must be set several risks. In the absence of auniversal framework for public cooperation, corporatist practicescan degenerate into cartels, particularism, and clientelism. Both arelatively autonomous state and a shared normative frameworkare necessary to counteract these threats. Bargaining between or-ganized groups by definition excludes unorganized groups, whichare likely to comprise those whose needs are most clearly ignoredand whose "claimsmaking" needs to be most encouraged. This alsocontributes to an imbalance of power, which undermines the effec-tiveness of democratic practices. It is probable, however, that so-cial corporatism promises a more inclusive and equal system ofinterest representation than liberal corporatism and is thereforeless open to these criticisms. Last, the national basis of assoc-iationalism today threatens to exclude outsiders, such as migrantworkers, from the benefits of citizenship and participation.^*

Our material preconditions for need satisfaction are more like-ly to be met in several respects than under the previous economicsystems considered here. By supplementing market and statemechanisms with networks of interest intermediation, cor-poratism offers several gains in the production of need satisfiers.Information passed through networks is "thicker" than informa-tion obtained in the market and "freer" than that communicatedin a hierarchy [Kaneko and Imai, as quoted in Powell 1991; cfElson 1988]. Longer-term perspectives will also foster the produc-

54 Ian Gough

tion of more efficient services and programs to meet needs. Accord-ing to Streeck and Schmitter [1985], "private interest govenment"more effectively combines policy formation with policy implemen-tation and thus enhances the delivery of specific need satisfiers. ̂ ^

As regards distribution, democratic communitarianism is likelyto prioritize policies to eliminate poverty, defined as a degree ofdeprivation that seriously impairs participation in one's society[Boswell 1990, chap. 3]. Insofar as this is so, it will aid the dis-tribution of satisfiers according to need. Social corporatism is like-ly to go further and add a more radical redistribution to its agenda.The emphasis on worker participation is likely to promote the pur-suit of need-related policies within the production process. How-ever, the impact of corporatist capitalism on the needtransformation process is at best neutral or indeterminate. Insofaras it prioritizes production-based interest groups, it could act tomarginalize women and the household sphere. The historic in-fluence of Catholicism in European variants of liberal corporatismhas imparted a bias against gender-equality policies that is absentin social corporatism.

Neoliberal Capitalism

The 1980s have witnessed a reaction against both statist andcorporatist capitalism on the part of those arguing for a restorationof minimally regulated capitalism. This combines elements ofliberal and conservative thought in a novel combination, dubbed"the free economy and the strong state" by Gamble [1988]. ThisNew Right program is of course associated with the Thatcher andReagan administrations in the United Kingdom and the UnitedStates. In many Third World countries, it has been imposed fromthe outside via IMF-led structural adjustment programs.

I will deal with this third form of contemporary capitalismbriefly, since it seeks to reestablish a system of minimally regu-lated capitalism that has already been surveyed. However, it doesintroduce a new element: the paradoxical development of thepowers of the state in order to "roll back the state." This stems inparticular from the argument of public choice theory, that interestgroup, bureaucratic, and electoral pressures generate a continuallyexpanding but inefficient set of state interventions in the economyand society. Thus, the modem democratic state subverts thefreedom of the market order [Dunleavy and O'Leary 1987, chap.3].


To overcome this requires a strategy to reduce the powers of boththe state apparatus and organizations in civil society. Thus, twocharacteristic policies flow from this: on the one hand, deregula-tion, privatization, and tax cuts and on the other, a restatement ofthe rule of law and a reduction of the powers of trade unions andother institutions that lie between the state and the individual[Gamble 1988, chap. 2].

It is likely that such a combination of policies will prove to beharmful to our procedural and material preconditions for needsatisfaction. The advantages and deficiencies of minimally regu-lated capitalism have already been rehearsed. To the negativeoverall balance must here be added, however, the deliberate use ofstate power, not to further collective, generalizable goals, but tobuttress the pursuit of individual interests and the market order.Furthermore, state power is also used to disperse the networks ofcorporatist negotiation that might form the alternative basis forthe emergence of generalizable interests. Bereft of the counter-vailing power of public authority and of networks of publiccooperation, we would predict that this form of capitalism willserve less well as a societal framework for improving human needsatisfaction than corporatist capitalism and many forms of statistcapitalism.

Conclusion

Table 3 summarizes the pros and cons of the three mixed formsof capitalism. Neoliberal capitalism, I predict, would be no moreconducive to human flourishing than minimally regulatedcapitalism. Indeed, according to Gamble [1988], its defining fea-ture-a combination of "free market and strong state'-promises apoorer performance. It is bereft of both the countervailing powerof public authority and the networks of public cooperation. Thisform of capitalism has a poor chance of realizing the proceduraland material framework for improving human need satisfactionidentified earlier.

The potential impact of statist capitalism on human well-beingis, I conclude, indeterminate. While it has a potential to correct forthe tunnel vision and market failures of minimally regulatedcapitalism, it also contains a potential for authoritarian, clien-telist, and bureaucratic features that distort both procedural andmaterial effectiveness. At best, a proactive state is no more than a

56 Ian Gough

Table 3. Evaluation of Three Mixed Economic Systems

CriteriaPI. Rationalidentificationofneed8

P2. Use ofpracticalknowledge

P3.Democraticresolution

Ml. Efiidentproduction ofneed satisliers

M2.Distributionaccording toneedM3.Effectiveneed trans-formation

StatistCapitalismIdentification ofcertain collectiveinterests, but elitedominationIndeterminatepotential toimprove marketeffectiveness

Wider domain ofpublic sphere,but bureaucratismand/or clientelismPotential to overcomemarket failures butbureaucratic failures

Indeterminatepotential toredistribute accord-ing to needIndeterminatepotential to improvelabor and genderequality

CorporatistCapitalismSocial capsule andcollective ethicfavor identificationofneedsPotential to com-bine market andnetwork knowl-edge, but exclusionof unorganizedNurtures dialogicdemocracy, butexclusion ofunorganizedPotential to over-come market andbureaucraticfailures

Social entitlementsto basic needsatisflers likely

Social corporatism:potential toimprove labor/gender equality

NeoliberalCapitalismAbsent. Both marketand state weaken"social capsuIeVcollective ethicMarket-basedknowledge fostered;claimsmakingdiscouraged

Market and state usedto restrict democraticpublic sphere

Efllciency in commodityproduction butmarket failures andabsence ofnoncommodity formsNo socialentitlements toneed satisfiers

Market and genderrelated inequalitiesin labor andhousehold

means for the achievement of a needs-oriented policy: it may be anecessary condition, but it cannot be sufficient. Statist capitalismmay be more conducive to meeting human needs than unregulatedcapitalism, but the answer is indeterminate in the absence of fur-ther information on the direction and nature of state policy.

In principle, corporatist capitalism permits the dominantmarket mechanism to be regulated by both public action and socialconstraints collectively negotiated by key economic actors. Thus, ithas the potential to overcome market and state failures in thematerial realm and to foster some form of dialogic democracy inthe procedural realm. Against this must be set the danger that un-organized groups will remain excluded from the corporatistdecision-making bodies, and thus that their needs will be over-looked or overridden. Though this danger is greater under liberal


corporatism, it is still present under social corporatism, par-ticularly for groups identified according to extra-economic criteriasuch as women and ethnic groups.

To arrive at some definitive ranking of these different sets ofeconomic institutions is not possible in the absence of explicittrade-offs between our six preconditions. While we argue thatRawls [1972] and the work of some of his followers, such as Pogge[1989], provide some important signposts to help in answeringthis question, we do not pretend to advance a comprehensive solu-tion [Doyal and Gough 1991, chaps. 7 and 11]. My own view isthat the weight of argument emanating from institutional orpolitical economy theory favors corporatist capitalism on both pro-cedural and material grounds, and within this category it favorssocial over liberal corporatism. Neoliberal capitalism appears tooffer the poorest framework for optimally satisfying universalhuman needs, while statist capitalism is indeterminate.

Let me conclude by noting the two ways in which this analysiscould be advanced. One is normative and entails enquiringwhether feasible alternative socioeconomic arrangements couldperform better than social corporatist capitalism in meetinghuman needs. It is important to repeat that only mixed capitalistsystems are considered here. The claims of market socialism orthe economics of partnership, for example, are not investigated.̂ *^The second route is empirical. It entails constructing operationalindicators of these idealized economic systems that can be appliedto real-world national economies. These can then be correlatedwith the historical record of substantive need satisfaction of dif-ferent nation-states. In this way, the conclusions reached in thispaper can hopefully be tested against real-world evidence [Goughand Thomas 1993].

Note8

1. However, we have argued in Doyal and Gough [1991,chaps. 6-7] that this goal is closely associated withother moral goals including the Rawlsian interpreta-tion of social justice.

2. For a survey of research on the former, see Wilenskyet al. [1987]. For a critique and the case for a focus onpolicy outputs, see Alber et al. [1987] and the work ofEsping-Andersen [1990].

58 Ian Gough

3. For the moment, too, we leave open the question ofwhether all six of these societal preconditions are com-patible, or whether there are conflicts between any ofthem.

4. See Hodgson [1988] and Etzioni [1988]. The labels areconfusing here. "The new institutionalism" usuallyrefers to modifications of neoclassical economics,which take into account such factors as the dynamicnature of all economic life as an adjustment to uncer-tainty [Hayek 1948, and the Austrian school] or theproblems stemming from information and transactioncosts and the incentives these give for the estab-lishment of durable economic institutions [Coase1937, 1960; Williamson 1985]. These all, however,retain a commitment to rational, maximizing in-dividuals as the basic units of analysis, a feature ex-plicitly rejected by the American institutionalistschool of Commons, Veblen, and others described inMayhew [1987].

5. However, Polanyi can be criticized for failing to"embed" the concept of the market itself within socialrelations and for thus retaining an essentialist idea ofmarkets. See Lie [1991].

6. Examples include Katzenstein [1985], Hall [1986], andWeir et al. [1988]. For a general analysis, see Marchand Olsen [1984].

7. As Cammack [1989] notes, this second strand is closeto an "integrated Marxist account" that combines classorganizations and a relatively autonomous state ac-ting within a field or general logic of internationalcapitalism. Nor does the first strand necessarily entailthe second, or vice versa. Yet, as Cammack points out,some notion of systemic prerequisites, or "environ-mental incentives," is necessary if one is to assess theextent to which institutions are functional or dysfunc-tional. For this reason, and because the two are sooften intertwined, I will take the two strands togetheras constituting the structural-historical strand of newinstitutionalism.

8. For a related but different taxonomy, see Sjostrand[1992]. As I will argue below, no real-world economy


or mode of production relies solely on only one ofthese mechanisms; and this applies also to most real-world institutions such as the family.

9. As do Streeck and Schmitter [1985]. As a result ofthis identification, they posit a fourth "associative"model of social order distinct from that of "com-munity." However, I will argue below, following Bos-well [1990], that their associational order can beconsidered as a subset of a broadly communitarianmode of economic coordination.

10. This paragraph draws in particular on Hodgson[1988, chaps. 7-9], Penz [1986], Liess [1976], Hirsch[1976], and the essays in Ellis and Kumar [1983],notably those by Crouch, Ellis and Heath, andKumar.

11. This section draws on Westoby [1983], Nove [1983],Feher et al. [1983], Nove and Nuti [1972], Komai[1980], and Hodgson [1984, chap. 11].

12. One person to address these issues is Devine [Cohenand Devine 1988], who proposes a comprehensive sys-tem of interest representation at national, regional,industry, and enterprise levels, coupled with an in-stitutionalized form of "negotiated coordination" todetermine all investments and capacity changes inproduction units. He also specifies the roles of ademocratic state in regulating the economic system.However, if this system is designed to supplantmarket-determined prices in many parts of theeconomic system, it is extremely likely that familiarproblems of interest group behavior would be en-countered. For example, the demands such a commit-tee system would make on citizens' time wouldencourage many, especially the least organized, to optout. If it is advocated as a third form of coordinationto supplement state and market relations, then it hasmuch to offer in developing a mixed form of economicsystem discussed below.

13. On this basis, Katzenstein [1985, 20] regards Japanand France as exemplars of statist capitalism in thedeveloped capitalist world, though his conclusion isdependent on the existence of the third category of

60 Ian Gough

corporatist capitalism discussed below. See also Shon-field [1965, chaps. 5 and 7] on France.

14. At times, this generates a marked inconsistency be-tween the analysis of interest group formation and therequirement for restraint and virtue in the publicsphere, an inconsistency that Hayek overcomes by ad-vocating traditional values and a "strong state" torestrain the rationalist pursuit of self-interest. Thisrecognizes and reinstates the interdependence ofmarkets on state and community, but in a way thatthreatens our third procedural precondition.

15. Complementarity, accommodation, and reciprocity aresaid to characterize successful network relations ineconomic production such as those in Japan. See read-ings in Thompson et al. [1991] and Hodgson [1988]. Itis interesting that the genesis of reciprocity is ex-plained in two different ways, corresponding to thedivision between economic and sociological/anthropo-logical paradigms discussed above. On the one hand,game theory shows how cooperative behavior can en-hance individual interest-satisfaction. On the otherhand, anthropologists emphasize the normative stand-ards and obligations that sustain exchange relations.The centrality of cooperation and networks is agreedbut for very different reasons [Powell 1991].

16. Streeck and Schmitter [1985] are explicit that associa-tions signal a fourth order of economic coordinationdistinct from market, state, and community. However,elsewhere they see them as a series of pragmatic ad-justments within capitalist society [1985, 23], withhistorical antecedents in late medieval cities [1985,10]. They also share several features in common withthe community order, for example, a logic of inter-dependence between actors, compared with one of in-dependence in markets and dependence in hierarchies[1985, 11] and a central role for negotiation betweenroughly equal entities-the difference being that theentities are organizations rather than individuals. Forthese reasons, and in the light of Boswell's [1990] ar-guments, I consider that they are better conceived of


as modern forms of community order within mixedcapitalist economies.

17. Kohli [1987] argues that the b

Ian Gough Economic institutions and the satisfaction of...

Documents

Transcript of Ian Gough Economic institutions and the satisfaction of...