Fundamental Institutions of Capitalism: A Radical View (Routledge Frontiers of Political Economy)

The Fundamental Institutions of Capitalism

‘This is a very important and well-researched book. It is a major attempt to address some serious analytical problems concerning the nature of the world in which we live. Its major contribution lies in its conceptual analysis of fundamental economic structures and relations of the modern world. Few other authors have addressed these issues in such depth as Screpanti. In this respect and more, it is a very welcome contribution. Its particular value lies in its detailed analysis of legal contracts, human agency and power relationships. Screpanti’s contribution is to open up much of this territory. Appropriately, a wide range of material, from the work of Marxists, institutional economists, behavioural economists and other social theorists, is drawn upon. The book penetrates into areas where few other authors have ventured to explore.’ Geoffrey M. Hodgson Research Professor at the University of Hertfordshire ‘I enjoyed reading this work. It raises questions to which we do not seem to have answers.’ Professor Frank H. Hahn University of Siena The Fundamental Institutions of Capitalism presents a radical institutional approach to the analysis of capitalism. Ernesto Screpanti discusses a wide range of topics and puts forward a number of provocative arguments that expose common ground in both neoclassical and Marxist orthodoxies. It will appeal to a broad audience of social scientists including advanced students and professionals with an interest in politics and economics. Ernesto Screpanti is Professor of Political Economy and Professor of the History of Political Economy at the University of Siena. He has published extensively on the economics of labour, institutional economics, the history and methodology of economics and previous publications include An Outline of the History of Economic Thought, with S. Zamagni.

Routledge Frontiers of Political Economy

1

Equilibrium Versus Understanding Towards the rehumanization of economics within social theory Mark Addleson

2

Evolution, Order and Complexity Edited by Elias L. Khalil and Kenneth E. Boulding

3

Interactions in Political Economy Malvern after ten years Edited by Steven Pressman

4

The End of Economics Michael Perelman

5

Probability in Economics Omar F. Hamouda and Robin Rowley

6

Capital Controversy, Post Keynesian Economics and the History of Economics Essays in honour of Geoff Harcourt, volume one Edited by Philip Arestis, Gabriel Palma and Malcolm Sawyer

7

Markets, Unemployment and Economic Policy Essays in honour of Geoff Harcourt, volume two Edited by Philip Arestis, Gabriel Palma and Malcolm Sawyer

8

Social Economy The logic of capitalist development Clark Everling

9

New Keynesian Economics/Post Keynesian Alternatives Edited by Roy J. Rotheim

10 The Representative Agent in Macroeconomics James E. Hartley 11 Borderlands of Economics Essays in honour of Daniel R. Fusfeld Edited by Nahid Aslanbeigui and Young Back Choi

12 Value, Distribution and Capital Essays in honour of Pierangelo Garegnani Edited by Gary Mongiovi and Fabio Petri 13 The Economics of Science Methodology and epistemology as if economics really mattered James R. Wible 14 Competitiveness, Localised Learning and Regional Development Specialisation and prosperity in small open economies Peter Maskell, Heikki Eskelinen, Ingjaldur Hannibalsson, Anders Malmberg and Eirik Vatne 15 Labour Market Theory A constructive reassessment Ben J. Fine 16 Women and European Employment Jill Rubery, Mark Smith, Colette Fagan, Damian Grimshaw 17 Explorations in Economic Methodology From Lakatos to empirical philosophy of science Roger Backhouse 18 Subjectivity in Political Economy Essays on wanting and choosing David P. Levine 19 The Political Economy of Middle East Peace The impact of competing trade agendas Edited by J.W. Wright, Jnr 20 The Active Consumer Novelty and surprise in consumer choice Edited by Marina Bianchi 21 Subjectivism and Economic Analysis Essays in memory of Ludwig Lachmann Edited by Roger Koppl and Gary Mongiovi 22 Themes in Post-Keynesian Economics Essays in honour of Geoff Harcourt, volume three Edited by Peter Kriesler and Claudio Sardoni 23 The Dynamics of Technological Knowledge Cristiano Antonelli

24 The Political Economy of Diet, Health and Food Policy Ben J. Fine 25 The End of Finance Capital market inflation, financial derivatives and pension fund capitalism Jan Toporowski 26 Political Economy and the New Capitalism Edited by Jan Toporowski 27 Growth Theory A philosophical perspective Patricia Northover 28 The Political Economy of the Small Firm Edited by Charlie Dannreuther 29 Hahn and Economic Methodology Edited by Thomas Boylan and Paschal F. O’Gorman 30 Gender, Growth and Trade The miracle economies of the postwar years David Kucera 31 Normative Political Economy Subjective freedom, the market and the state David Levine 32 Economist with a Public Purpose Essays in honour of John Kenneth Galbraith Edited by Michael Keaney 33 The Theory of Unemployment Michel De Vroey 34 The Fundamental Institutions of Capitalism Ernesto Screpanti 35 Transcending Transaction Alan Shipman 36 Power in Business and the State An historical analysis of its concentration Frank Bealey

The Fundamental Institutions of Capitalism

Ernesto Screpanti

London and New York

First published 2001 by Routledge 11 New Fetter Lane, London EC4P 4EE Simultaneously published in the USA and Canada by Routledge 29 West 35th Street, New York, NY 10001 Routledge is an imprint of the Taylor & Francis Group

This edition published in the Taylor & Francis e-Library, 2002. © 2001 Ernesto Screpanti All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Screpanti, Ernesto, 1948– The fundamental institutions of capitalism / Ernesto Screpanti. p. cm. – (Routledge frontiers of political economy) Includes bibliographical references and index. 1. Capitalism. I. Title. II. Series. HB501 .S475 2001 330.12’2–dc21 ISBN 0–415–24767–5 (Print Edition) ISBN 0-203-45338-7 Master e-book ISBN

ISBN 0-203-76162-6 (Adobe eReader Format)

00–067373

Contents

vii

Contents

1

List of tables

ix

Introduction

1

The employment contract, transaction institutions and capitalism

6

Profit making and labour mobilisation 6 The employment contract 7 Transaction institutions and social reproduction 23 2

Individuals, culture and behavioural institutions

44

Premisses 44 Bounded rationality and intentional action 54 The multiple self and false consciousness 64 The ideological conditions of social reproduction 89 Conclusions 99 3

The state and normative institutions

112

The incompleteness of institutions 112 Where the state comes from 114 The state action 126 4

Forms of co-operation and power The struggle for survival and the struggle for power 145 Co-operation 147 Power 157

145

viii Contents 5

Production governance structures

185

A taxonomy of governance structures 185 The capitalist firm 186 The market 215 The governance of knowledge production 230 6

Different forms of capitalism

255

The need for a general theory of capitalism 255 Institutional systems 256 Forms of capitalism 269 Conclusions

287

Towards the autonomy of capital 287 References Index of names Subject index

291 303 307

List of tables ix

Tables

1.1 1.2 2.1 2.2 4.1 4.2 6.1

Types of contracts that can be used to mobilise work A classification of social actors Mental feelings A classification of human actions Elementary forms of co-operation Types of governance relations Institutional forms of capitalism

8 27 46 89 150 150 267

x List of Tables

Introduction 1

Introduction

This book originates from a certain dissatisfaction with some kinds of orthodoxy prevailing in contemporary heterodox thought, especially Marxism. Recent research has brought to light a few problems and paradoxes that impair the critique of political economy at its most abstract theoretical levels. The first is the problem of exploitation in production. To foreground the production process as the privileged sphere of capitalist exploitation, Marx put forward a series of hypotheses that enabled him to isolate the subject for investigation. For instance, he assumed the ‘law of value’: all commodities are exchanged at a price that coincides with their cost of production, so that the possibility of exploitation occurring in the market rather than in production can be excluded. Moreover Marx maintained that all markets are competitive, to exclude the possibility of exploitation arising from monopolistic power. A principle of equivalence is assumed whereby a commodity is traded at a price that equalises the present value of future returns to the cost of production. Under these conditions there are no unequal exchanges in the market. Finally, to isolate the analysis of exploitation from the conditions of monetary circulation and natural scarcity, he assumed that all surplus value coincides with profit, that land rent and money interest are nil. Then he tried to demonstrate the following thesis: although the remuneration of labour is equal to its cost of production – that is, to ‘the value of labour power’ – the capitalists succeed in extracting a positive residual income from the production process. An objection could be raised. If wage coincides with the cost of production of labour power and with the present value of its future returns, where does the surplus value come from? Why does competition in the labour market not push the wage rate to a value that eliminates profits altogether? Marx would answer that if profit is reduced below a normal rate, investment will shrink, the industrial reserve army will rise, wages will decrease and the capitalists will eventually earn what they want. But one could rebut: This is precisely what any neoclassical microeconomic textbook teaches. Profit tends to stabilise at a rate that remunerates the productive contribution of managers without leaving any residue. Marx would insist: the working of the labour market is such that wages are pushed to settle at a subsistence level which has nothing to do with a Marshallian demand price for labour. This reply would be a trap since it implies that the law of value, which was assumed to hold in all markets, just does not hold in the case of the labour market.

2 Introduction Exploitation would be caused by the market forces that push wages toward subsistence and not by the production powers that compel workers to produce a surplus value. Labour would not be paid under conditions of equal exchange, with the consequence that exploitation would appear to take place in the market and not in production, quite the contrary of what it was intended to demonstrate. A second problem is the definition of social classes. According to a widespread interpretation of Marxian theory, the two fundamental classes in a capitalist system are the ‘capitalists’, defined as the private owners of means of production, and the ‘workers’, defined as propertyless wage earners. This distinction, however, crucially depends on the classical saving hypothesis: the workers receive a subsistence income and have zero propensity to save. If they received a wage high enough to enable them to save, they would also earn some capital income, for example in the form of dividends on shares administered by a pension fund. The wealth accumulated with workers’ savings is a form of capital owned by the workers. It follows that the latter should be classified in the same social class as the managers who earn both salaries and dividends. Obviously the managers would be much richer than the workers. But this observation would only enable us to define a class of rich and a class of poor people – an unacceptable definition for a Marxist – instead of a class of exploiter capitalists and another of exploited proletarians. Third problem: the paradox of capitalism without capitalists. According to Lenin the property regime established in the Soviet Union immediately after the October revolution was one of ‘state capitalism’. He expected worker control of the state to grant socialist use of public wealth. Yet such control has never come about. Thus it was plain state capitalism. But a system without private ownership of the means of production cannot be considered capitalist if a capitalist is defined as a private owner. So what was the social nature of the former Soviet Union? It was neither socialism, nor did it seem to be capitalism. If one does not like the theoretical tricks invented ad hoc because of the incapacity to give up some categories of orthodox Marxism (degenerate worker state, bureaucratic collectivism etc.) one cannot but stick to Lenin: it was state capitalism. In this case, though, one must be able to draw all the theoretical conclusions: capitalists cannot simply be defined as private owners of capital. Note that the same problem arises with the forms of public wealth existing in market capitalism systems. Thus the paradox does not just pertain to the social nature of the Soviet Union, and actually bears on the problem of identification of the capitalist class in all systems in which the traditional bourgeoisie has faded out. Indeed this is the question of the definition of capitalism in general. The fourth problem is perhaps the most serious of all. It concerns the workers’ ‘human capital’ and has been brought to light by G. A. Cohen in Self-ownership, Freedom, and Equality. According to a certain interpretation of Marx’s thought, the workers are exploited by the capitalists because they have to work more than is necessary to reproduce the value of their labour power. And it is real exploitation, a form of injustice, for the workers’ productive contribution comes from the use of labour power – a commodity that belongs to them. But then the proletarians are not strictly propertyless. They are, in Marx’s words, the ‘owners of labour power’.

Introduction 3 This commodity accrues as a flow of labour time. Therefore it presupposes the existence of a stock from which it flows. Whatever the stock is called, ‘labour capacity’, ‘workers’ abilities’, ‘nerves, brain and muscles’, ‘human capital’ and so forth, the workers have ownership of it because they have ownership of themselves. Thence one might ask: if all people have self-ownership, how is it possible to overcome exploitation? Marx admits that in the first phase of communism certain principles of bourgeois right remain in force, including one that implies selfownership: ‘to each according to his own ability’. Let’s suppose that, to trigger a process of socialism construction, all the starting conditions for the ownership of non-human capital are zeroed. What is to prevent the birth of a new kind of capitalism, in the presence of an unequal distribution of natural endowments (talents, time preferences, risk aversion)? Who could legitimately forbid me, if I have selfownership, to use my poetic talent to accumulate wealth with copyright royalties? And who could forbid me to use this wealth to hire and exploit a secretary who is poetically less endowed than me? Things are no simpler in the second phase of communism. Why should I use my ‘human capital’ to enable those less endowed to freely satisfy their needs? Obviously, if communism is an Earthly Paradise in which everybody has become a saint, there would be no problem. But what would be the need for communist institutions in an Earthly Paradise? If it were a land of Cokaigne, in which you only have to reach out a hand toward the tree of abundance to get what you want, all problems would again be solved. Actually Marx had something like a land of Cokaigne in mind when he held that communism would be made possible by such widespread productive development that almost every form of scarcity would be eliminated. Only in this case he was able to reconcile the preservation of individual freedom, intended as the individual’s ownership of his own person, with the constitution of universal freedom from needs. However, in a communist world in which not all men are saints and not all goods are abundant, only an institutional system that limits individual freedom in some way (for instance through progressive taxes) can enforce the principle ‘from each according to his ability, to each according to his needs’. This system must be based on the general principle that individuals do not have self-ownership. Justification of the principle of non self-ownership must be ontological rather than political: it is conceived as a universal principle of justice. If you accept it for some institutional systems you cannot exclude it for others. However, if selfownership is assumed to ‘prove’ labour exploitation when the capitalist system is investigated, it cannot be excluded to avoid the possibility of exploitation of the more talented when a communist system is designed. Rejection of the principle of self-ownership implies that the workers do not own any ‘human capital’, even if capitalist ideologies induce them to believe they do. But, if there is no such thing as a stock of labour capacities in the workers’ balance sheet, where does the flow of goods they sell as labour power come from? And if they can sell only what they own, what do they sell in exchange for a wage? All these problems have a common origin: an excessive emphasis on property

4 Introduction rights in the definition of capitalism. It is possible that the institutions which regulate property rights and the mechanisms which govern their distribution are in reality historically contingent rather than essential, and that they vary considerably without any change in the capitalist nature of the process of surplus value extraction, in other words, that ‘juridical private property in means of production is not an essential part of capitalism’ (Zarembka 1992: 129). If this is so, then what are the fundamental institutions of capitalism? Which ones are present in any form of capitalism regardless of its historical particularities? There are other theoretical difficulties besides the four referred to above, from the problem of value to that of the laws of movement, but I shall not dwell on them here, first because I consider them less important than the four already mentioned and second because they are marginal with respect to the doubts which have given rise to this work. However there is one very general and important difficulty that I cannot overlook. In his earlier philosophical works, Marx, in criticising Feuerbach and some young Hegelians, developed a conception of human agency which could well be defined in terms of ‘institutional individualism’. In his ontology of the social being, the subjects of history are concrete individuals rather than generic humankind; and they are human subjects, that is, limited in their capacities and powers, moved by material interests and strongly conditioned by social relations and the technological, institutional and cultural context in which they act; and yet they act and make history. In his economic works, on the other hand, Marx normally resorted to a methodology of aggregates that undoubtedly justifies those who attribute him a holistic attitude. I think both positions are valid – ontological individualism, by virtue of sound philosophising before the problem of human liberation; methodological holism, to come to grips with the ineluctable complexity of social systems in the presence of bounded scientific rationality. But one would have expected some effort to be made to link up the two levels of discourse. In Capital, Marx often appeals to the representative agent device or to the law of great numbers. The modern – or rather the postmodern – reader would have liked to see something more, as, for example, an analysis of the institutions that condition behaviour, govern the aggregation of collective agents, select and discipline individual and social choices. Obviously, it is easy to raise such demands in our times of institutional and evolutionary revival. Yet the present-day reader too must be satisfied by Marx if his theory is not to be reduced to an antique. Actually I nourished the hope of finding a satisfactory answer to the structureagency problem by delving into new institutional economics. My disappointment over this gave rise to dissatisfaction with another kind of orthodoxy – one which is rapidly catching on in contemporary institutional thought. In particular I was unable to find a solution to the problem of singling out the fundamental institutions of capitalism – the institutions that create the conditions for labour exploitation in a free society and under the rule of law, those that give birth to social classes and push people to aggregate in social relations of conflict and co-operation. Worse still, in spite of the enormous amount of energy spent by modern economists in analysing contracts and especially the employment contract, I have not found in

Introduction 5 new institutional economics a satisfying account of the bases of power relations that regulate the production process. Instead I have found a subtle confusion between the employment contract and the contract for services which leads to the former being interpreted as an incomplete form of the latter. This confusion prevents us from seeing the difference between what happens in a factory and what happens in a market. I have also found an even greater confusion between the employment and the agency relationship, by virtue of which we are induced to believe that the bond linking a worker to his employer is not unlike the one linking a doctor to his patient. Not to mention those who try to pass off the employment contract as an insurance arrangement. And I shall refrain from delving into the ideological content of all this confusion. Nor have I found any satisfaction in wiser streams of thought, such as those originating from the Coasian insight into the authority relations regulating cooperation in a firm. Here too I have discovered an excessive emphasis on property rights, and an error which is similar to that made by the many orthodox Marxists who believe the capitalist’s power in the factory originates from the ownership of means of production. But are we sure that the owner’s authority derives from the residual right of control on his wealth? Ownership attributes a right on the residual income produced with the means of production, but these alone do not produce anything. Ownership does not attribute any residual right of control on the actors who handle the means of production, if these actors have not explicitly undertaken an obligation to obedience. And it is the employment contract, not the ownership of capital, that constitutes this kind of obligation. Thus my view is that the fundamental institution of capitalism, the institution that makes exploitation in a capitalist production process possible, has nothing to do with the distribution of property rights. My thesis is that a fundamental institution of capitalism is the one which constitutes the basic power relations in the labour process, the employment contract. This is what I argue in Chapter 1. The first chapter can be read as a premiss to the rest of the book. Once the very foundation of the analysis of capitalism has been recast, as I try to do in Chapter 1, many other categories of the critique of political economy have to be rethought, which is what is attempted in the other chapters. Let me conclude by thanking the friends and colleagues who have helped me with their criticism and suggestions: Sam Bowles, Guglielmo Carchedi, Massimo De Francesco, Nicola Dimitri, Luca Fiorito, Frank Hahn, Geoff Hodgson, Bruno Jossa, Kathy Khang, Francisco Louça, Bruno Miconi, Mario Morroni, Edward Nell, Ugo Pagano, Angelo Reati, Ermanno Tortia, Andrew Tylecote, Stefano Vannucci. I discussed some of the arguments developed in this book in the OPEL newsgroup, where I received many encouraging comments, especially from Riccardo Bellofiore, Rakesh Bhandari, Francisco Paulo Cipolla, Paul Cockshott, Gerald Levy, Gilbert Skillman and Paul Zarembka. Obviously, in thanking my advisers I accept all responsibility for any errors. Finally, a very special word of thanks to Lynn Kirby, who has intrepidly striven to make my English readable.

6 The employment contract, transaction institutions and capitalism

1

The employment contract, transaction institutions and capitalism

Profit making and labour mobilisation Mr Boss decides to set up a business to earn profits. To be precise, he decides to use workers to produce pins. The perfect world in which Mr Boss lives has certain peculiar characteristics. There are many workers who, to avoid unemployment, are eager to work for a wage. All individuals are self-interested but loyal, and nobody conceals information. Both Mr Boss and his workers are risk neutral. It is not difficult for anyone to collect the information necessary to take a personal decision, nor does anyone find the economic calculations complicated. All agreements are simple and there are no transaction costs. Furthermore, whatever the technique used to produce pins, the means of production and working skills are not tied to the production of pins and can be disinvested at any time without loss of value, apart from normal depreciation. Finally a criterion of valuation is in force, called ‘the principle of equivalence’, according to which the price of any commodity coincides with its production cost and the present value of its prospective yields. Mr Boss has a major problem: what kind of agreement should he propose to the workers to make them produce profits for him? He considers several options. The first is to propose that they become partners in his firm. All the partners, including Mr Boss himself, are equal, all of them do the same work and have the same decision-making power. At the end of the production process the partners democratically share the value added (net of interest on capital),1 and therefore they all earn an income proportional to their productivity. Mr Boss cannot earn any profit. The second possibility is to give the workers a mandate: to work autonomously in Mr Boss’s factory, i.e. to work by preserving their decision-making power as far as labour activity is concerned. However, their decisions must pursue Mr Boss’s interests. Since the latter knows the workers are loyal, he has no doubt about their will to supply the efforts and intelligence necessary to reach his own ends. The workers will be paid fairly, so that their incomes will be proportional to the value of their services. Mr Boss will again be disillusioned: no profit will remain for him. The third possibility is to ask the workers to sell him the specific labour services necessary for the production of pins. Mr Boss in this case knows all details of the labour process before stipulating the agreement and is able to define the exact

The employment contract, transaction institutions and capitalism 7 labour services required of the workers. The latter will readily accept the agreement, since their services will be remunerated fairly, i.e. in accordance with the ‘human capital’ from which they originate. Also in this case Mr Boss will discover he can earn no profit, since the value added in production, net of interest and depreciation allowances, will coincide with the value of labour services. The fourth possibility is to propose a sharecropping contract to the workers: they will take all production decisions and implement labour activity autonomously; at the end of the production process income will be divided into two parts, one for Mr Boss and one for the workers. It would appear that the former could earn some net profit with an agreement of this type, and this will in fact occur if his income share is greater than the interest on capital. But then the agreement will be considered unfair by the workers, since they will have supplied a production and decision-making activity and produced a value, of which only a part returns to themselves. However, in the special case in which the principle of equivalence is applied in defining income shares, the remuneration of Mr Boss will again be nil. There is, finally, a fifth option: to propose to the workers an agreement by which they accept to give up their decision-making autonomy for a certain number of hours a day, during which they will perform labour activity under the command of Mr Boss. With this agreement there is no need for the workers to be informed beforehand about the labour activity they will be ordered to carry out in the production process. Nor will they be entitled to know it. Their activity during labour time will no longer be their action, since it will be a manifestation of Mr Boss’s will and a means for the realisation of his ends. Whatever income they are paid, there will be no way of linking it to the labour services, since these services belong to Mr Boss. What he will do during their labour time and with their labour activity is his own business and prerogative. Thus the workers’ remuneration does not take the form of the price of a commodity, nor does the value of the product of their activity take the form of a value produced by them. Wage is not the price of labour services, but a compensation for obedience. And there is no reason or possibility of appealing to the principle of equivalence to ask for the present value of labour pay to be equal to the ‘human capital’ embodied in the workers’ capacities. It is possible therefore that Mr Boss succeeds in using labour activity to produce a net income, i.e. a surplus value, which is over and above the wage paid to the workers for their obedience. Note that surplus value includes interests and rents, besides profits. If Mr Boss aims at maximising his profits, he in fact works at maximising surplus value. Interests and rents are a cost to him, but are not determined by him. Therefore the only way for Mr. Boss to increase profits is to increase surplus value. The employer who controls the labour process is an exploiter who works in the interests of the whole capitalist class, including the rentiers.

The employment contract Some kinds of contracts that can be used to mobilise labour To bring out the moral of the story, let me now reformulate its message in a more rigorous form. The scene takes place in a very unreal world. The reader should

8 The employment contract, transaction institutions and capitalism have understood that the hypotheses describing this world have no descriptive pretensions. In fact they have two objectives. First of all they sustain a counterfactual reasoning. All the assumptions used in recent neoinstitutionalist literature to account for the employment contract have been removed: bounded rationality, class differences in risk aversion, incompleteness and asymmetry of information, opportunism, costs of control, transaction costs, asset specificity. I want to show that even in this strange world there is good reason for resorting to the employment contract; none of those special assumptions is necessary for the existence of this kind of contract. The second objective is more ambitious: to bring to light the rationale of the employment contract by showing its real nature. This is why the force of abstraction has been driven to such sidereal heights. The principle of equivalence, according to which the price of a commodity coincides with its cost of production and the present value of its future yields, is not in force in the capitalist world, even if it is in force in most economic theory.2 Yet it has been assumed here. It is precisely at this level of abstraction that the real nature of the employment contract can be understood. In fact, since it is not a contract for the sale of a commodity, there is no possibility of applying the principle of equivalence to it. It is easy to grasp the nature of the employment contract by comparing it to other kinds of contracts that can be used to mobilise work. Four of them are considered: partnership, agency, services, and sharecropping.3 There are two categories of subjects, B and L, who stipulate a contract at time t0 to regulate the performance of a production activity to be completed at time t1. The contracts establish that L will be paid an income w, the amount of which can be defined either at time t0 or at time t1, depending on the kind of contract. L will perform an operation, Oi, during the period of the production process, t1–t0. The various forms of contract differ in several characteristics, in particular: the object exchanged, the time at which w is decided, the time at which Oi is decided, the subject who decides Oi. Table 1.1 shows schematically the main characteristics of the five kinds of contract. In the Partnership contract two or more subjects confer goods or services for the common exercise of an economic activity, with the aim of sharing its profits. There are many kinds of partnership, but here, for the sake of simplicity, I shall assume that capital is rented by all the partners in equal shares, and that labour services of the same value are contributed by all the partners. Thus profits are shared in equal parts. There is no difference between B and L. The object of Table 1.1 Types of contracts that can be used to mobilise work Kinds of contract Partnership Services Agency Sharecropping Employment

Object of transaction

Time at which w is decided

Time at which Oi is decided

Decisor of Oi

Labour services Labour services Labour services Labour services Obedience

t1 t0 t0 t1 t0

t1–t0 t0 t1–t0 t1–t0 t1–t0

L L L L B

The employment contract, transaction institutions and capitalism 9 transaction in this contractual form is the set of labour services contributed by the partners. The amount of labour contributions is decided ex ante, i.e. at the moment the contract is signed, t0. The specific labour services, however, are decided during the production process, i.e. in the period t1–t0. All the partners have equal decisionmaking power. Finally remuneration is decided in time t1, the moment of closing of the production process, when profits are distributed. This kind of arrangement cannot generate surplus value for anyone if the income shares coincide with the values of labour services contributed. With the contract for services subject L undertakes to supply a service or a set of services for subject B without any bond of subordination. L takes an obligation to furnish a specific job and not to carry out a generic labour effort. Both the nature of the service and the remuneration are fixed at the time of stipulating the contract, t0. The way the service is produced, though, is L’s business, and will be decided during the production process, in the period t1–t0. What matters is that, at the established term, L is able to deliver the result of labour activity. Also in this case neither of the subjects will earn a profit if the value of the service is equal to the income paid by B to L and coincides with the cost borne by L. With the contract of agency the agent, L, takes a commitment to perform actions in the interest of the principal B. The agent preserves his personal autonomy, so that he himself decides the labour services during the production process. His compensation can be fixed at the moment of contract stipulation, but a part of it can be linked to economic performance and decided at the end of the production process. If the principle of equivalence is in force, compensation coincides with the value of the services. The residual income belongs to B, but it will coincide with the interest on capital.4 The sharecropping contract consists of an agreement between the owner, B, and the sharecropper, L, by which the latter performs the labour activity and the former supplies the means of production. Although the commodities to be produced can be decided by B at time t0, the labour services are decided autonomously by L in the period t1–t0. Profits are shared by the two parties. Therefore, if the principle of equivalence is applied, the value of the labour services must coincide with the sharecropper’s share, in which case no net profits will accrue to B. Shares need not be fixed at 50 per cent, and in general the sharecropping scheme can be considered a prototype of a vast series of pay schemes in which the compensation of L is linked to the outcomes of labour activity: performance pay, group bonuses, payment by piece-work, sales commissions etc. In all these cases B will get no profits from L’s labour activity if the workers’ compensation coincide with their productivity. The contract of employment and the capitalist firm In the employment contract w is decided at time t0, whilst Oi is decided in the period t1–t0. It is the employer who determines Oi, since the contract assigns him the prerogative of deciding autonomously on everything concerning the use of labour. This prerogative derives from the employees’ obligation to carry out labour

10 The employment contract, transaction institutions and capitalism activities under the command of the employer. In the contract of employment no labour services are exchanged, so that the employees do not need any information about them. The labour services may be known ex ante to the employer, but this is not a necessary condition. In reality, due to the constant drive to innovate the labour process, employers do not normally know ex ante the labour services they will order their employees to produce. The contract of employment differs from all the other contracts as it does not entail the sale of a commodity. Even the partnership contract, insofar as the partners contribute labour services, takes the form of an agreement for the exchange of commodities, namely a multilateral agreement. Furthermore, the income earned by the workers with the other contracts is compensation for the production contribution brought by their services. This is not so with the contract of employment, in which the workers sell no services. What they actually sell is an obligation to obedience. With this obligation they give up their decision-making autonomy for a certain number of hours a day, and the employer acquires the power of command over them. The labour activity carried out after the contract stipulation is not an action of the workers, but takes the form of an action of the employer. As a consequence, the product obtained by this means belongs to the employer, so that there is no reason to link the workers’ remuneration to their productivity. Of course there is always a production contribution of labour activity, but this is not a worker’s contribution. Finally, precisely because the contract of employment is not a contract for the sale of a commodity, the workers’ remuneration does not take the form of the price of a commodity. Wage is not the price of a ‘labour commodity’ or ‘labour power’ or ‘labour services’. It is compensation for the renunciation of freedom. The fact that the wage is not the price of labour services is a first condition for the exploitation of labour. Since the production contribution of labour activity does not relate to the worker, there is no way of reducing the wage to the value of labour services; and, since labour is not a commodity, there is no way of referring its value to a production cost. This implies that it is not possible to appeal to the principle of equivalence to determine the wage rate. And, if justification of the principle is sought via an appeal to perfect competition, it could be observed that the wage rate has never been strictly determined by the law of demand and supply – not even at the beginning of modern capitalism, as Adam Smith so well understood. A second condition for the exploitation of labour occurs in the production process. Since employers have command over the workers, it is their right to direct and control labour activity in pursuing their own ends. And if their ends are reduced to the profit motive, the contract of employment gives them the social instrument for extracting surplus value from the use of labour. Labour services are produced by the employers’ actions, and their product belongs to them. Therefore, by directing labour activity the capitalists can produce a value which is over and above the compensation paid to the workers for their obedience. Note that all the other forms of contract can also be used to generate exploitation, but only insofar as market conditions enable the exploiter (who could be either B or L) to infringe

The employment contract, transaction institutions and capitalism 11 the principle of equivalence. None of the other forms of contract generates the conditions for exploitation in production. The capitalist firm is founded on the workers’ commitment to obedience and is an organisation based on a hierarchy of power whose purpose is to produce profits. The capitalist firm is a nexus of contracts of employment, the exact opposite of what the philosophers of the ‘nexus of contracts’ believe. To understand its rationale there is no need to know anything about the market, let alone its imperfections. It is not born out of any imperfection in mercantile exchange. The conditions for the existence of the capitalist firm are not generated by transaction costs, information asymmetries, or any other kind of market failure. The theories that sustain the opposite view are in fact based on a misunderstanding about the forms of contract used to mobilise labour. To understand the rationale of the capitalist firm one must avoid confusing the employment contract with that of agency. With the latter, the agents preserve their decision-making autonomy: labour services are exchanged together with a mandate to define them during the production process. This kind of agreement is perfectly defined in the market sphere. But, if it works, in an economic context in which the agent is risk neutral like the principal, there is no way for either of them to extract profits from the labour process. It is also necessary to avoid confusing the employment contract with the contract for services. Specific services are exchanged under the latter, and, if the market does not fail, they are paid their real value. When we ask a plumber to repair our wash basin we know the commodity we require and, if the market mechanism enforces the principle of equivalence, there is no way of extracting surplus value from the plumber’s work. The contract of employment is another thing, and must not be considered as an incomplete contract for services.5 Since it is not a contract for the sale of a commodity, it does not contribute to the market allocation of labour. The allocation of labour is defined in the labour process within a hierarchical structure. The rationale for this structure has nothing to do with market failures, transaction costs, uncertainty, complexity, technical change and so on: it is simply to be found in the profit motive. Note that the employment contract does not presuppose employees know ex ante the nature of the labour process. The contract for services, on the contrary, requires that presupposition, the objects of transaction in the latter contract being labour services: these are known to B ex ante, even if they are decided by L in the period t1–t0, but must be known to L if he is to be in possession of all the information required to enter the contract. This theory of organisation is not a theory of a generic firm. It is a theory of the capitalist firm. Its capitalist character is essential for understanding its nature. And it is precisely this character that reveals the essence of its hierarchical structure. At this level of abstraction, it is only an elementary hierarchy. It has no control functions, for example, since there are no information asymmetries. In Chapters 4 and 5 I will show that the closer we come to the real world, the more complex the firm’s hierarchy becomes. But only at the present level of abstraction is it possible to grasp the basic function of capitalist hierarchy. It remains true that the legitimacy of the capitalist command, whatever its particular function, derives from the

12 The employment contract, transaction institutions and capitalism workers’ commitment to obedience. The hierarchy created with the contract of employment is a hierarchy of power, and this is fundamental. It is a simple structure, implying the existence of only two layers: those who command and those who work. This simplicity brings to light the essence of every other capitalist social structure: all have roots in the dual relation of power. To sum up, the thesis argued in this section is not that the contract of employment is the only form of transaction that allows exploitation. All the other kinds of contract can be used to this end if market conditions allow a party to infringe the principle of equivalence, which is normally the rule when markets are not perfectly regulated. With the contract for services, for example, B can exploit L if w is lower than the value of labour services, or L can exploit B if it is higher. On the other hand, it is even possible that the wage rate fixed in a contract of employment is so high and the employer so inefficient in directing the labour process, that, ex post, exploitation in production can turn out to be nil. What I have tried to demonstrate is that the employment contract is the form of transaction on which the capitalist firm is founded. The capitalist firm is an economic organisation that uses capitalist control in the labour process with the object of extracting a surplus value for the accumulation of capital. The social classes’ bargaining strength determines wages, while the employers’ managerial ability determines labour productivity. The magnitude of surplus value depends on the interaction between these two factors. With other kinds of contracts, on the contrary, the magnitude of exploitation depends only on the strength of the parties in the bargaining process, and not on capitalist control of the labour process. Thus the first fundamental difference between the employment contract and the other forms of contract resides in the capitalist’s prerogative of controlling the labour process through command over labour. There is a second fundamental difference. If the functioning of markets guaranteed the principle of equivalence, no exploitation would be obtained with the kinds of contracts used to exchange labour services. However, application of the principle of equivalence in fixing the remuneration of labour services in the case of the employment contract is quite compatible with the workers’ exploitation, since with this kind of contract the labour services belong to the employers, not to the employees. Thus, whilst exploitation with the other kinds of contract can be interpreted as the consequence of an infringement6 of that distributive principle, this interpretation is no longer possible with the employment contract. Labour capacity and the boundaries of authority Since the employment contract constitutes a power relation, it requires a definition of the boundaries of authority. The usual employment contracts, which say nothing about the specific activities the workers will be required to carry out in the labour process, are instead rather eloquent about the range of authority. Nevertheless, the contractual and legal definition of the boundaries of authority are doomed to remain substantially incomplete. The reason is that they also depend on the workers’ capacities. No law can validate an order to do something which is beyond the

The employment contract, transaction institutions and capitalism 13 reach of an employee. In practice the authority relation implies that bosses have the prerogative to tell the subordinates to do whatever is not ruled out by norms, agreements and the labour capacities. Thus there are two kinds of limitations of authority in the employment contract: institutional and technical. Institutional boundaries, for example, can establish that a fine and not a whip can be used to discipline workers. Technical boundaries can establish that a bricklayer cannot be required to accomplish accountancy tasks or to lift a two ton block of marble. The former kind of boundaries refer to the rules defining the prerogatives and the obligations of the parties, in particular the rules regulating the way employers can exert command, control, punishment and enforcement. Most of these limitations are defined by the law, others by custom and still others by the contract itself. Many such rules are ingrained in the common values and expectations of the individuals belonging to a given social setting and are agreed upon implicitly. The other kind of boundaries refer to the workers’ technical capacities and limit the range of activities the employees can be assumed to be able to perform. Most of these too are defined by the contract itself but also in this case there are many which are agreed upon implicitly. The workers’ capacities have two dimensions, which can be specified in terms of competence and ability. Competence defines a worker’s technical knowledge, whereas ability defines the effectiveness with which any knowledge can be applied. Thus the ex ante technical boundaries of authority can be represented in the form of an area of potential technical power, as surface l in Figure 1.1, where the horizontal axis represents competencies and the orthogonal axis abilities. Competence consists of bits of knowledge, it is a quanta set. And any bit of knowledge can be applied with a specific degree of ability. Therefore the area of potential technical power need not be compact. The sides of rectangle l in Figure 1.1 represent the technical boundaries of authority. The competencies axis has been drawn as a dotted line to convey the idea that competencies are not a compact set. The limits of authority are also defined by contractually determined labour time, that is, by the number of hours workers are committed to spend in labour activity. Contractual labour time, Lc, is represented on the vertical axis of Figure 1.1. Thus the sides of the parallelepiped define the limits of power as potential power. Once the employment contract has been signed, the employer can exert command in the labour process by ordering the workers to apply their capacities in the production of some specific services or outputs. Any labour service or product results from a combination of certain competencies, abilities and labour time. In this way the exercise of command contributes to transform technical capacities into outputs. From an ex post point of view the volume of potential power becomes a set of scattered lines, like Li–Oi in Figure 1.1, any one of which represents a peculiar combination of competencies, abilities and labour time required to produce a specific service or product. In this case the boundaries of authority delimit the technical efficacy of command. Power, now, is no longer potential: it is exerted

14 The employment contract, transaction institutions and capitalism

X x

Labour time

Abilities

Ly

Lc Le

L3 L1 O1

L2 O2

O3

λ

Competencies

Figure 1.1

power. Note that the overall labour time effectively spent in production, Le, is given by the summation of the labour times spent in the different labour activities, L1, L2 and L3. It might well be lower than contractual labour time, Lc, for instance because workers shirk. Workers’ capacities are not perfectly known by the employers. Furthermore some of them take the form of tacit knowledge, that is, bits of skills that cannot be expressed, even by the workers themselves, as a list of instructions. Thus the employment contract, which may be neither complete nor incomplete in describing the technical implications of labour services, can well be utterly incomplete on the definition of authority boundaries. In other words, there can be both ex ante and ex post information asymmetries in the employment relation. Ex ante, because the workers’ capacity cannot be better known to the employer than to the employee. Ex post, because the transformation of potential ability into actual effort cannot be perfectly controlled. There is no employment relation which is not infected with the basic problems of hidden action and hidden information, for power and exploitation are conflict relations. This is a fundamental reason why the labour process in a capitalist economy is a contested terrain. It is also a fundamental reason for the existence of a more or less complex hierarchy of monitoring and control within the capitalist firm, as well as many kinds of incentive pay schemes. Workers, in their pursuit of autonomy and better working conditions, continually try to play down labour

The employment contract, transaction institutions and capitalism 15 activity by hiding their labour capacities. Employers, in their pursuit of profits, continually try to overcome the workers’ resistance and to cope with the contractual and workplace information asymmetries. The level of labour productivity and the rate of exploitation depend dramatically on the success of these policies and the result of that struggle. All such problems will be dealt with in greater detail in later chapters. What I want to argue in the present one is that the very existence of capitalist exploitation in the production process does not depend on the upshot of that struggle. If the employment contract guarantees the possibility of capitalist exploitation even in the case of asymmetric information in the workplace, this is all the more so in the case of complete information. This is why I have previously assumed, for the sake of abstraction, that no information asymmetries exist. Complete information means that each agent possesses all the information required to enter a transaction. Since the employment contract constitutes a power relation, all the information required to define the authority boundaries (including those concerning the workers’ capacities) is assumed, in this chapter, to be common to all parties. Since this form of contract does not define the labour services ex ante, there is no need to assume anything about the workers’ possession of any information concerning those very labour services. This assumption has a counterfactual implication and aims at demonstrating that the rationale of the employment relation cannot be found in the impossibility of complete contracts: it must be found in power and exploitation. What do you sell when you sell labour? Seemingly a trivial question. And one could give it a trivial answer by complying with the commonplace that labour is a commodity: you sell a commodity. Yet a doubt remains raised by Karl Polanyi (1957) that the description of labour as a commodity is a misrepresentation of reality and that the labour market is a system of coarse fictions. One could ask: What commodity do we buy from the plumber when we ask him to repair our wash-basin? And is it the same thing he bought from his assistant when he hired him with a contract of employment? And suppose the repair has been done by the assistant under the direction of his employer. Whose work do we pay, that of the plumber or that of the assistant? Talking of ‘work’ one usually refers to a series of operations of object transformation. This is ‘labour activity’. Obviously it is not a commodity. We do not ask the plumber to sell us the operation of unscrewing a few metal rings, taking a old pipe away, cutting another pipe etc. We simply ask him to repair our wash basin; to sell us a service. To supply this service he has to perform some actions, i.e. take decisions about some particular labour activity, carry out those decisions or make somebody else carry them out, so as to produce some effects of object transformation in accordance with a goal – in this case, supplying a functioning pipe. A ‘labour action’ is a set of decisions about a labour activity which produce effects in accordance with an objective. Labour actions are different

16 The employment contract, transaction institutions and capitalism from labour activities: first, because they consist of decision-making and, second, because they produce effects in accordance with an objective. Yet, labour actions are not commodities. We do not buy the plumber’s decisions or decision-making capacity. We buy the services produced by actions, the effects of labour actions. These are ‘labour services’. It is labour services that are bought with a contract for services. What is paid in exchange is the value of services. The subject who is paid is the decisor of the action that produced the services, not the one who carried out his decisions. The plumber is paid for his services, regardless of whether his decisions have been carried out by him or his assistant. The principle of equivalence is complied with if the value of the service coincides with the costs of the action that produced it. These costs may include the wage paid to the assistant and the yield of the plumber’s ‘human capital’, besides those of the raw material and the means of production used in the action. If the principle of equivalence is complied with, there is no exploitation of the plumber. But what about the assistant? What does he sell to the plumber? He is hired with a contract of employment. He sells neither labour activity, labour actions nor labour services. He does not sell labour activity, because, although the assistant will be asked to carry out some labour activities, the contract does not specify the operations he will be required to perform; in fact he can be employed in many different tasks and operations and, in the event of innovations, even in activities which were not known at the time of stipulating the contract. He does not sell labour actions, because he is paid to execute the employer’s decisions, not to take decisions, nor to determine labour actions. He does not sell labour services, because these are the results of labour actions and therefore are produced by the decisor of the actions, not by their executor. The employment contract generates the employer’s capacity to decide about labour activity. This is generated by virtue of the worker’s obligation to carry out labour activities under the command of the employer. The employment contract generates the employer’s ability to carry out labour actions without performing labour activities, but simply by directing the worker’s labour activity. Thus workers do not sell a commodity. That is why their wage is not the price of a commodity. It is not the price of labour activities, because these may be multifarious, heterogeneous, variable and unknown at the time the wage is fixed; the wage, in fact, will not vary when labour activities vary. Nor is it the price of labour actions, because these are effected by the employer, not by the employee. Nor, finally is it the price of the labour services, because these are produced by labour actions and belong to the employer. As already observed, wage is the compensation for loss of freedom, i.e. for obedience; and obedience is a social relation, not a commodity. So, to say that the principle of equivalence is infringed in the fixing of wages has no sense. This principle cannot be applied, for wage fixing has nothing to do with determining the value of a commodity. Yet the principle remains in force in determining the value of the ‘product of labour’, which is the value of the labour services. Since labour services belong to the employers, their value is paid to

The employment contract, transaction institutions and capitalism 17 them. Thus, as already observed, compliance with the principle of equivalence is perfectly compatible with the exploitation of labour. The exploitation of workers has nothing to do with the legitimacy of the property rights claimed on the ‘products of labour’, nor with the ‘fairness’ of exchange relations, as Marx so well understood. Exploitation can exist because the contract of employment severs the remuneration of labour from the principle of equivalence, and severs the ‘product of labour’ from labour activity. To understand the nature of the social relation established by the employment contract, it may be useful to go back to its origin: the locatio operarum, whose existence at the end of the first century BC is known. The locatio operarum, despite the similarity of locution, should not be confused with the locatio operis, which corresponds to the contract for services. The latter defines the renting of a specific work or service, whilst the former defines the renting of generic labour. The concreteness and materialism of Roman law provides us with some concepts of transaction institutions which are free from the ideological and moralistic encrustations brought about by modern mentality, and may help us to grasp the real nature of those institutions. In Roman law, the problem of what is actually sold when labour is sold is expressed as the problem of what is the obligation of the lessor (locator) in the locatio operarum: is it an obligation to give or to do? Locare means ‘to place’, ‘to set’, ‘to put’, and generally refers to the obligation to deliver a thing to the lessee, who undertakes to pay a rent and give the thing back in due time. The most common type is precisely the renting of a thing (locatio rei), a plot of land, a means of production etc., in which the obligation is an obligation to give. Not so with the locatio operis, in which, having sold a service, the worker is obliged to do some specific labour actions. The locatio operarum, however, the renting of generic labour, is different, since it is impossible to define ex ante which specific actions must be ‘done’. According to a certain interpretation, the obligation taken by the worker in the locatio operarum is an obligation to give (as in the locatio rei) and not to do (as in the locatio operis). The exchange form entailed by the employment relation belongs to the do ut des (I give, you give) kind, and not to the do ut facias (I give, you make) kind. One of the strongest arguments in support of this interpretation, besides the one already mentioned (impossibility of defining ex ante the works to be done), is based on the observation that the concept of locatio operarum is also used for the renting of slaves (locatio operarum servi, or, more simply, locatio servi), in which the lessor, the owner of the slave, hires out a thing (the slave) and therefore is obliged to give the use of a man: the obligation to give a thing, and not to do a work. Similar to this is the obligation undertaken with the locatio operarum of a free man: again it is hiring out a man’s use; the main difference being that in this case the lessor hires out the use of himself. In fact the locatio operarum of a free man is also called locati sui, the letting of himself. This is not a substantial difference, however, because, although three individuals appear in the locatio servi, it remains true that the third individual, the one who is the object of transaction, is not a legal subject: the slave or, rather, his use, is the exchanged thing. In the same way, the

18 The employment contract, transaction institutions and capitalism object of transaction in the employment contract of a free man, is the use of the employee, the use of a man as a thing. Only two legal subjects appear in both cases, the object of the transaction being a thing, the thing the lessor undertakes to give. But precisely what is this thing? Clearly it is not the ownership of the ‘human capital’ embodied in the slave or the free employee, which legally remains an endowment of the slave owner or the free man. It is in reality the potestas, the power, the faculty of command the owner can exert on the slave’s activity or the free man on his own, that is, the prerogative of deciding the labour activity of the worker who is the object of transaction. To hire out the use of a worker means precisely to hand over the prerogative of determining his activity with an action of command. In the case of both the slave and the free employee, the seller gives up his potestas. In both cases the hired worker is obliged to obey the buyer during renting time. Now it is perhaps easier to answer the question ‘what do you sell when you sell labour’. Just reflect on the acuteness of Cicero’s observation: ‘wage is a compensation for the [journeymen’s] servitude’ (De officiis, I, 42). You sell your freedom. The expression ‘labour market’ is a metonymy which shades into euphemism or irony, depending on the point of view. That market is in reality a ‘freedom market’. Alienation, capital fetishism and the ideology of commodity labour7 ‘Political economy’ – writes Marx (1978b: 73) – ‘conceals the estrangement inherent in the nature of labour by not considering the direct relationship between the worker (labour) and production’. There is an ideology which has always dominated economic thought, including most of socialist thought: the ideology of commodity labour. It is based on the idea that workers are persons endowed with a ‘human capital’, i.e. certain specific labour capacities, of which they sell some flows in exchange for a price, called wage. This ideology is perpetrated by virtue of a theoretical misinterpretation – the mistaking of the employment contract for a sale contract or, rather, the interpretation of the former as an incomplete form of the contract for services. To understand the ideology of commodity labour it is necessary to grasp the meaning of labour alienation. And to do this one must start from the legal meaning of the concept: alienation as the conveyance of a thing to another. But only if one reflects that, with a contract of employment the worker transfers not a commodity, but freedom, it is possible to grasp the meaning the concept takes in the social and political sciences: alienation of labour means ‘loss of freedom’. It means, first, that the product of labour activity rises up before the workers as an independent power. It does not mean that the product of labour becomes an object, but that it appears to the workers as a thing which exists outside them, which is independent of them, which belongs not to them but to others. This occurs because, as the employers have acquired command over labour activity, the latter becomes, in the labour process, a manifestation of the employers’ own will. Labour activity, which is a physical and mental activity of the workers, is transformed

The employment contract, transaction institutions and capitalism 19 into an economic action of the employer. Therefore the products created through it belong to the employer not to the worker. But how can the workers become estranged from the products of their own activity, if they were not estranged from themselves in that very activity? Alienation also means estrangement from labour activity. During labour time the workers do not express their own free will, do not expend a free energy and do not take any free decision. Their activity is not voluntary, it is forced labour. In this sense alienation means that workers perceive their own work as an activity carried out under someone else’s command. Wage labour just means alienated labour. And if you do not like Marx, let me insist on Cicero: ‘journeymen work in a vile employment, for nothing which is really free can exist in a factory’ (De officiis, I, 42). The mystery of the employment contract is that it induces the workers to feel estranged from themselves. This mystery is created by a sort of existential contradiction, since, formally, workers enter into the contract as autonomous subjects, but, by doing so, they renounce their freedom: subjects who freely give up freedom. How do they bear this shame? The simplest way is to conceal to their conscience one of the two facets of the dilemma. And it is perhaps an inalienable aspiration to human dignity that induces them to conceal the more shameful: since they consider themselves human beings, they like to feel free. Hence their labour obligations must be considered as a manifestation of freedom. They freely accepted the employment contract and, in order not to feel slaves, even if only part-time, they convince themselves that they did not sell the other party a certain number of hours of their life, but only a commodity. Obviously the other parties, who do not consider themselves slave-owners – indeed, who consider themselves benefactors of humankind – have nothing to object. And this is how the commonplace, thence the ideology, of commodity labour is originated. It must be clear, however, that the roots of this ideology are dug deeply into the workers’ false consciousness. Economists build on the products of this false consciousness. In the elaboration of the ideology of commodity labour, the economic subjects are helped by the product of labour taking the form of a commodity. Thus, if commodities are produced by means of commodities, then the very activity of producing commodities can naturally appear as a commodity. It is the commodity nature of the labour product that enables the conscience to do the somersault by which the workers avert all the awareness of their alienation. It helps to transform real alienation into an estrangement of consciousness. There is nothing strange and nothing contradictory in being formally free at the bargaining table and substantially enslaved in the factory. The situation appears paradoxical only because often the human conscience has difficulty in perceiving its own division as normal. Thus it is only to avert the unhappiness of a divided conscience that workers are induced to seek refuge in the false consciousness of commodity labour. And political economy, in transforming that product of false consciousness into a theoretical concept, does just its job: it gives the dignity of science to the forms of commonplaces. Now, it is precisely the notion of a commodity called ‘labour’ or ‘labour power’ that gives rise to commodity fetishism, whereby social relations are perceived as

20 The employment contract, transaction institutions and capitalism relations between commodities. There is nothing particularly fetishistic in an exchange of a beaver for a deer or of a dollar for a cigar. The parties perceive and understand immediately that they are exchanging parts of their incomes and therefore some products of their work. But when they exchange a wage against an obligation to obedience and tell themselves the story of an exchange of commodities they are cheating themselves, they are deforming their perception of the world. It is here that a social relation is concealed under the shell of a commodity relation. It is here that the very idea of a commodity called ‘labour power’ is born. But work is a flow of labour activities in time. Since the obligation to obey is delimited in a time period, labour power must be defined as a flow commodity. Then it seems to presuppose the existence of a stock commodity, call it ‘labour capacity’ or ‘human capital’. If a worker is able to sell labour power, it is because he has the ownership of it. Which implies he has the ownership of himself, of his capacity to work. The very person of the worker is perceived as an object that can be owned. This is why it is called ‘human capital’. The idea of self-ownership is grounded on commodity fetishism or on what could properly be defined ‘capital fetishism’. In a capitalist system all economic relations between persons tend to take the form of relations between capital stocks. The idea of a free person endowed with inalienable rights is given the meaning of a free ownership of the capital embodied in the person’s economic capacities. And all persons’ capacities are valued as wealth, as abilities to produce value. However, self-ownership is a juridical impossibility in a system in which individuals are endowed with inalienable personal rights, in one that forbids slavery. In fact if nobody can own another person, no person can be bought and sold. Therefore nobody can sell themselves or their human ‘capital’. But a property right which is limited by the prohibition to sell is not a property right proper. Imagine a situation where the state allots plots of land to farmers entrusting them with all residual rights of control, including the right on residual income, but not with the right to sell the plot. Are these farmers entrusted with real property rights? Certainly not. They are indeed entrusted with a full right of use, i.e. with possession, not with proper ownership. Similar to this is the situation of a person’s ownership of himself. The worker has a full right to use himself as he likes. He has even the right to sell this use, e.g. by undertaking an obligation to obey during labour time, but he has not the right to sell himself: his person is inalienable because his freedom is inalienable. There is nothing that can be called ‘human capital’ in the same sense you call a machine a capital asset. Therefore the ideas of ‘self-ownership’ and ‘human capital’, as the products of that cognitive deformation process I call ‘capital fetishism’, are parts, indeed fundamental parts, of the capitalist ideology. Capital fetishism relates to ‘humans whose skills are being objectified’ (Ashton and Green, 1996: 18) or, rather, humans whose skills are considered a wealth asset. This does not mean those products of cognitive deformation are not real things. ‘The categories of bourgeois economics’ – says Marx – ‘consist precisely of forms of this kind. They are forms of thought which are socially valid, and therefore objective, for the

The employment contract, transaction institutions and capitalism 21 relations of production belonging to this historically determined mode of social production’ (1990, I: 169). Clearly it is the objectivity of forms of thought. And, as objective forms of thought, they have objective consequences, for they enter the sphere of motivations and justifications and produce real behavioural effects. They also serve to give an ideological justification to the power relations governing the production of capital. These relations are accepted in the minds of individuals as exchange relations among free and equal contracting parties. Free, because they do not sell freedom but just a commodity. Equal, because they are all capitalists, owners of some kind of capital asset, real or personal. The workers are induced to think that, in selling their labour power to a company, they are capitalists who make an investment, for they just sell a flow of commodity springing from the capital stock they own. They believe they are making the same transaction made by a manager who offers his capacities to the company; and a transaction similar to that made by the company’s creditors – the selling of a flow of commodity, of the use of a money stock. If they cannot be considered stockholders properly, the workers can be treated anyway as stake-holders of ‘their’ company on the same foot as all the other people who invest some of their wealth in it. This is capital fetishism at its best. Power, exploitation and value Marx states (1990, I: 167n): ‘Value is a relation between persons … concealed beneath a material shell’. Which persons and which shells? Let me stick to the abstraction of an economy where commodities are exchanged at production costs. Let me further assume that, in this economy, only two kinds of subject act in the labour process, employers and employees. The relation between persons established by the employment contract is a power relation. The workers’ alienation of freedom is countered by the employer’s acquisition of power. The true social exchange generated by the contract of employment is the ‘exchange’ between power and freedom. It is the worker’s renunciation of freedom that creates the employer’s power. As argued above, this kind of power takes the form of ‘potential power’ (i.e. the prerogative of command over labour), ex ante, at the contract stipulation; and that of ‘exerted power’ (i.e. the exercise of command over labour), ex post, during the production process. But a production process is not just a labour process, i.e. a process of technical transformation of commodities. It is in reality a combination of the labour process and the valorisation process. The valorisation process consists of the imputation of exchange values in the inputs and the outputs of production whereby power generates surplus value. Thus the analysis of the power relation developed so far can now be completed. The ex post definition of power in the valorisation process is represented again in Figure 1.1 (see page 14), where the mappings from the potential power space, LAC (Labour time, Abilities, Competencies), into the commodity set, X, represent the labour process, while the mapping from the commodity set into the labour space, L, represents the valorisation process. Ly is the net product measured in wage units.

22 The employment contract, transaction institutions and capitalism Wage, as a compensation for the commitment to obedience, is a lever and a measure of the employer’s power. The lower the wage, the higher the amount of working time spent by the workers for earning a certain income; and the greater is the amount of labour time that can be purchased by employers with a given income. The shell of material relation beneath which social relations are concealed is made of a very abstract substance indeed: the money value of commodities. It is difficult to see in the dollar price of a kilo of pins the amount of labour time that can be purchased by the subjects who own those pins. Nor is it easy to see the amount of value that can be produced by the labourers paid with those dollars. The money prices of commodities appear on the surface to be intrinsic properties of the commodities themselves, not what they really are, namely the expression of power relations. To measure them as real prices, by taking a commodity as numeraire, does nothing to improve things. Worse still, it gives strength to the impression that value is an intrinsic property of commodities – a property linked, for example, to their technical production conditions, scarcity or utility. Moreover, to express the wage in monetary terms or in terms of a real numeraire contributes to consolidate the impression that it is the price of a commodity which is itself dependent on an intrinsic property of a thing, the ‘value of labour power’ or the ‘marginal productivity of labour’. But labour is not a commodity and wage is not a commodity price. If wage is the price of the command over labour, then it can become an ideal numeraire, a measure of value which, albeit still conventional, is endowed with some interesting properties. By expressing the value of commodities in terms of this numeraire it is possible to break the shell of the appearance of material relations. The measure in purchasable labour,8 i.e. in terms of wages, immediately reveals the power relations giving rise to that value. Let W be the wage bill, w the wage rate, R the surplus value, assumed as identical to profits, Y = W + R the net income and L the level of employment. Assume, for the sake of simplicity, that L = Lc = Le, i.e. that contractual labour time coincides with effective labour time. The wage bill is the income expended by the employers to pay the workers. By using a wage as numeraire, the wage bill is expressed in purchasable labour. This is a measure of the employers’ potential power, that is, the quantity of labour time (of a particular kind)9 which can be set in motion through that fund: W =L w

By expressing the net income in terms of purchasable labour, one obtains a measure of a source of power, a measure of the labour time, Ly, that could be set in motion through the product of labour: Y = Ly w

Furthermore, if the wage share is measured as W/ Y=L/Ly, an interesting result is obtained. Its reciprocal, Ly / L, is the factor of exploitation, i.e. a measure of the potency of exploitation as an instrument of power accumulation. It is a ratio between the labour that can be purchased by using the product of labour as a wage bill and

The employment contract, transaction institutions and capitalism 23 that which has actually been commanded to obtain that product, i.e. a ratio between the labour purchasable by, and the labour contractually10 embodied in, the net product. The lower the wage share, the faster the command over labour can accumulate. In other words, the factor of exploitation, measured in purchasable labour, is a sort of power multiplier: it indicates the potential growth rate of power, the rhythm of power accumulation if all profits are invested. Finally, by expressing the ratio between the surplus value and the wage bill in purchasable labour one obtains a measure of the rate of exploitation in terms of power: R Y − W Ly − L Ly + = = −1 W W L L

The rate of exploitation is not just a ratio between surplus value and wages. The valuation in purchasable labour shows that exploitation is the basis of a mechanism of power accumulation: the higher the rate of exploitation, the greater the ability to accumulate power generated by command over labour. This formula brings a fundamental fact to light. If it is true that, by working in exchange for a wage, the workers work at producing their own fetters, then it is also true that, whenever they accept a wage share cut, they agree to work at strengthening their fetters. In the capitalist mode of production, exploitation and domination are two sides of the same coin. One is not possible without the other, since one sustains the other. Power serves to exploit, exploitation to accumulate power. But one cannot see this in the money prices of commodities. Everybody knows that money is power. Few people, though, know the reason, because one of the powers of money is its ability to conceal its nature. The valuation in purchasable labour reveals it.

Transaction institutions and social reproduction Transaction institutions and the civil society Transaction institutions are typified manifestations of the social actors’ will through which social relations are set up. All kinds of contracts, from sale to marriage, as well as many unilateral legal acts, such as deeds of gift, death-bed wills, promises and commitments, are transaction institutions. Their institutional nature is linked to their repeatability and their social diffusion. They are forms of transaction used continually by many social actors. In other words, transaction institutions have four fundamental characteristics: 1 2 3 4

they are forms of manifestation of the social actors’ wills; they constitute obligations among social actors; they are applied repeatedly in the same forms; they are socially accepted and used by many actors.

Transaction institutions are produced by the social actors and, in principle, their application should not require the intervention of any subject other than the parties themselves who are directly involved in the transactions. In Chapter 3 I

24 The employment contract, transaction institutions and capitalism argue that in reality external intervention is necessary. But I am provisionally assuming that any transaction institution is always capable of fully defining the obligations of the parties involved. I refer here to a hypothetical world in which civil society, that is, the set of interacting social actors, is able to produce all the social relations that bind the actors together. Social relations are nothing but the obligations constituted by transaction institutions. A contract, even a simple sale contract, defines not a relation between a person and a thing, but a relation among social actors. The contract for the sale of a commodity regulates the transfer of rights claimed over that commodity, thus it constitutes reciprocal obligations among the parties involved. Civil society is formed through the social relations created by the transaction institutions. The rationale of transaction institutions is to be found in their capacity to establish stable social relations. They are transactions, i.e. the undertaking of obligations. Through these obligations social relations are constituted. They must be institutions, i.e. typified, permanent and socially diffused forms of transaction. Through this characteristic social relations acquire stability. Social actors are multifarious, heterogeneous and mutable. If their relations were to be reinvented each time they are constituted, society would continually undergo disruptive and destabilising thrusts, and would be governed by chaos. Transaction institutions help to solve this problem. Institutionally determined social relations acquire structure and regularity. Moreover, institutions tend to legitimise transactions. Since they are socially diffused, they are considered socially accepted. This undoubtedly contributes to their stabilisation – an effect which is particularly important in a society dominated by antagonistic relations. The implicit legitimisation brought about by the social diffusion of transaction institutions induces the social actors to consider these normal and often even just. The ultimate foundations of property rights also stem from transaction institutions. The right to use things exclusively derive from obligations undertaken by the social actors. The right of an individual to dispose exclusively of a thing implies the exclusion of other individuals from the use of that thing. Therefore it can be founded on the obligation undertaken by other individuals not to interfere with that private use. An obligation of this kind, for example, is explicitly undertaken on by the seller in a sale contract or by the donor in a deed of gift. It is true that property rights are formally only transferred in a sale or a gift. But it is also true that the transferred right has been acquired through another transaction. This means that the set of transaction institutions forms a self-reference system in which each one refers to another, so that a civil society and the rights enjoyed therein are selfconstituted. Since a social actor’s property right over a thing consists in the exclusion of other social actors from its enjoyment, it can be constituted through the other actors’ acceptance of this exclusion. Even in the case of property of goods produced by the owner, it is impossible to recur to natural right or any other metaphysical hypothesis to justify the obligation of non-owners not to interfere. In reality the sole reason why social actors can claim rights over a good they have produced is that they can claim

The employment contract, transaction institutions and capitalism 25 rights over other goods, the means of production, the raw materials and the ‘human capital’, used to produce them. The owner’s right over a self-produced good derives from the obligation undertaken by other social actors toward his means of production. Labour activity in itself cannot generate property rights on its product, either by natural law or for any other transcendental reason. So much so that, for example, the workers employed in a capitalist firm cannot claim any right over the firm’s products, in spite of the fact these have been produced through their activity. Indeed the transaction institutions that constitute the firm define the workers’ obligations in such a way that it is impossible for them to claim any right on the products. However, the property right claimed by employers on the commodities produced in their firms derives from the fact that they have been produced by means of tools and labour activities which are considered legitimately under their control. It must be made clear that only the social origin of property rights stems from transaction institutions, not necessarily their complete definition, nor, as will be seen in Chapter 3, their legal sources. The problem is that transaction institutions generate the obligations of the parties directly involved in them, but are unable to impose on everybody the claims that derive from those obligations. In Chapter 3 I argue that, in order to define property rights completely, it is necessary to refer to normative institutions, i.e. higher level institutions. But I will also show that those higher level institutions originate, albeit indirectly, from the actions of social actors, and that they do not create transaction institutions, but merely complete them. The notion of transaction institution may be useful in clarifying a question recently raised by some neoinstitutional economists – that of the priority of the ‘market’ over ‘institutions’. According to certain economists, organisations, interpreted as forms of institutions, are created to cope with market failures, especially in relation to problems of high transaction costs and incomplete and asymmetric information. Obviously this kind of problem does not arise in the world of Mr Boss. However, this chapter’s level of abstraction is sufficient to make it clear that ‘natural’ markets do not exist, i.e. no exchange system can do without institutions, and therefore there is no sense in postulating a market priority over institutions. It is impossible to conceive a market transaction by ignoring transaction institutions. Even the most simple commodity sale necessitates the stipulation of a sale contract, and it is unreasonable to assume that every time two actors decide to exchange something they have to reinvent the contract form. In reality only a few typified and diffused forms are used to deal with all transactions. The markets themselves, as a system of recurrent and widespread exchanges, are constituted on the basis of transaction institutions. In a very general sense, therefore, it is the institutions that have priority over the markets. One might think this conclusion is valid only because it is based on a particular concept of transaction institution rather than on that of an organisation or governance structure. However, institutions have priority also because it is not possible to define the contracting parties in an exchange by ignoring institutions. First of all, even if this argument is confined to individual parties, one must take into account the fact that the latter can constitute themselves only insofar as they are endowed with some rights as, for example, freedom of contract and right of

26 The employment contract, transaction institutions and capitalism ownership over the exchanged commodities. These rights are originated by institutions. Second and more important, when the contracting parties are collective actors, as in the case of organisations, unions etc., they are created through institutions. Therefore these have priority over the market also because they are necessary to the constitution of those very subjects who meet in the market. Transaction institutions and the social actors The reader will have noticed that a new term has been introduced. In the preceding sections the contracting parties were called ‘individuals’ or ‘subjects’. In the last section they have been called ‘social actors’. Actors because they are considered as subjects of actions. Social because it must be a social action, i.e. an action which generates relations among individuals. A social actor is not necessarily an individual subject. There are collective actors, i.e. groups of individuals who act together. Transaction institutions play a constitutional function with collective actors. In fact these stem from the societary obligations undertaken by the individuals who belong to them, and societary obligations are the basis on which the group’s rules and working procedures are determined. There are two kinds of collective actors: movements and organisations, which are distinguished mainly by two aspects. In organisations, transaction institutions are mostly applied explicitly and formally, so that the members’ obligations are pretty well defined and the internal working rules are fairly clear and known to all members. In movements, on the contrary, there is no formal definition of individual obligations and the working rules are badly defined and accepted by the members only implicitly. The other important aspect of the distinction concerns the division of tasks among the members, at least as far as the decision-making function is concerned. In organisations there is always a subset of members to whom the decision-making function is formally assigned. Not so in movements, where anyone can take decisions and no one is obliged to execute them. Nevertheless a movement can be a collective actor so long as the individuals who belong to it act together. Finally it is necessary to distinguish between two forms of organisation: representative and corporate organisations. The former are set up to pursue the objectives of all the members, and there is no organisation goal other than those of the individual members. The latter is a special kind of organisation which has developed a life of its own, i.e. a tendency to corporate action which is independent of its members’ objectives. There are some particular kinds of organisation in which the interests and goals of the members are bent or compelled to serve a propensity to action which belongs to the organisation itself. These are the ‘corporate organisations’ or ‘corporate actors’. Note the use of the expression ‘propensity to corporate action’, rather than ‘goal of a corporate action’. A goal is a disposition to intentional action and as such refers only to individuals. An organisation cannot have a goal that does not fall within the ends of at least some of its members. But there can be an internal motion, a tension within the organisation, which emerges from members’ behaviour

The employment contract, transaction institutions and capitalism 27 Table 1.2 A classification of social actors

ìIndividual actors Social actors í ì Movements îCollective actors í ì Representative organisations îOrganisations í îCorporate organisations

or from interaction with external actors, and which is not reducible to the members’ goals. This emerging tension is the ‘propensity to corporate action’. A typical corporate actor in capitalist systems is the capitalist firm, which tends to develop a propensity to growth quite independently of its members’ ends. I will deal with this issue in Chapters 5 and 6. For the moment suffice it to have postulated the existence of corporate actors. The various kinds of social actors are schematically presented in Table 1.2. Institutions and social actors come into existence through an evolutionary process. This is now common knowledge among economists. The basic idea is that institutions and collective actors establish themselves when they satisfy the needs of at least one set of individuals and die away when they become useless. I will go further into this in Chapters 2 and 4, after the concept of the institution itself has been widened. In this chapter, however, I shall tackle an argument that is anything but a forgone conclusion in current literature. The evolutionary approach presupposes the notion of a struggle for life and, especially in complex animal species like homo sapiens, a struggle for dominance and subjugation. It is not so much a problem of adaptation to the environment, as a question of struggle among individuals for the control of resources. In humankind this type of struggle does not necessarily imply destruction of the competitors, but it has always implied domination and exploitation of man by man. Many institutions, insofar as they are used as instruments of conquest, take the form of weapons of subjugation, which says a lot about the ideological content of those theories of institutional evolution that try to account for institutions in terms of collective welfare. This thesis must be enunciated quite explicitly: the winning institutions in the evolutionary process are not necessarily those that maximise or prompt collective welfare; even less so those that maximise adaptation of humankind to the environment. Humankind as such is not a subject of action. Only individuals and specific collective actors are subjects of action. The following observation ensues: some institutions establish themselves even if they are not conducive to the interests and aspirations of all the individuals who nevertheless comply with them. Since institutions are tools for building social relations, in a society in which individuals are self-interested and interests clash they may serve to establish exploitation relations and consolidate or modify power relations among the social actors. Most individuals aim at pursuing their personal interests. Social exchange and co-operation are instruments that improve the conditions of life of human beings, and certainly allow some individuals to achieve better economic results than they

28 The employment contract, transaction institutions and capitalism can obtain by working individually. This is so independently of the economies of scale derived by the division of labour. Through exchange and co-operation, some individuals can take advantage of the labour activity of other individuals. Exploitation does not have to be demonstrated: it is a fact proved by history. It has always been one of the functions of co-operation and exchange. This is an undeniable fact, even though, from a theoretical point of view, it may come as a surprise in the light of an hypothesis of perfect rationality. But the surprise should not conspire against the fact or the hypothesis. It should instead bring us to understand that there can be no exploitation without oppression. Only the existence of asymmetric power relations, in other words, a concentration of power in the hands of restricted social groups, can account for the reason why many social actors voluntarily take part in co-operative activities and exchanges through which they become the subject of exploitation. In the struggle for existence it may be perfectly rational to accept exploitation if refusal to co-operate within the restrictions of the holders of power results in a deterioration of life conditions. A slave may rationally decide not to escape. Slaves need not necessarily remain in permanent doubt, nor day after day do they need to take a rational decision not to escape. They might even accept, as legitimate or normal the obligations that stem from exploitation relations. The fact that the transactions which constitute and regulate those obligations are institutionalised contributes to this attitude of compliance and therefore to social stabilisation. Through transaction institutions, civil society, whatever its social structure, also strives to achieve self-legitimisation. Before closing this section I have to jump to a lower level of abstraction and touch on some arguments which will be treated in more detail in later chapters. This is necessary to immediately come to grips with an important aspect of real capitalism. Individuals are not only self-interested actors, they are also heterogeneous. They have different knowledge, culture and aspirations and different interests. Since they enter into social relations through transaction institutions, it is inevitable that different kinds of institutions develop. Thus there are many heterogeneous institutions in a given society at a given moment – heterogeneous, not only because they satisfy different needs, but also because the same kind of need can be satisfied in different ways. For example, as already shown, labour mobilisation can be pursued by using at least five different types of contract. Some of them can be used as complements. A capitalist firm can use all five kinds of contract simultaneously: a partnership contract to set up the firm and limit the partners’ responsibility; an agency relation to engage a lawyer; employment and sharecropping contracts to hire workers; and finally a contract for services to contract out works. However, the same kinds of contract can also be used as substitutes, as alternative instruments to activate co-operation. Thus a worker, for instance, could find himself having to choose between alternative ways of earning his living: entering an employment contract with a capitalist firm; organising a workers’ co-operative; setting up a personal firm in order to sell his services as an independent contractor. In conclusion, it is possible to formulate a general principle, which will be called the ‘principle of institutional plurality’: because of the heterogeneity of

The employment contract, transaction institutions and capitalism 29 social actors, many different institutions, able to satisfy the same kind of social need in different ways, can exist in a given society. Obviously nothing prevents a restricted set of institutions from predominating over all the others which satisfy the same need. Such institutions give tone to the social setting. The notion of ‘mode of production’ is based on them. A mode of production is a model of society in which production activity is regulated by certain core institutions that are considered fundamental or typical. Social theory often resorts to model building, that is, the construction of pure institutional models in which only core institutions are taken into account and investigated. Model building is a useful instrument, and can be aimed at abstraction and simplification. Nevertheless it is always better to preserve a mental reserve when dealing with pure institutional models, and not forget the principle of institutional plurality. This principle, when referred to a specific mode of production, can be redefined as the ‘principle of impurity’. It states: in every mode of production, beside the core institutions, there are always others that are somehow extraneous or promiscuous. Thus, for example, the employment contract, which is typical of the capitalist mode of production, was also widely used in Roman times. Slavery, especially in the form of exploitation of minors or prostitutes, is still practised even in the most advanced capitalist societies. The principle of impurity is important to maintain a certain degree or realism in investigating any particular mode of production. But it is especially relevant from a theoretical point of view. As will be seen in Chapter 2, this principle is useful in solving two major problems of the theory of institutional evolution. It serves to justify the existence of institutional variety, which is an essential condition for evolutionary change, and it helps to account for the social actors as agents endowed with some freedom of choice even in situations where their aspirations are institutionally conditioned. Capitalism in general and the social classes Capitalism is a mode of production in which: all economic subjects are formally free from any personal bondage; wage labour is used to extract surplus value from the production process; surplus value is spent to accumulate capital. Thus this mode of production is characterised by three fundamental social and economic conditions. It is necessary to identify the institutions that make them possible. The first condition implies in reality a negative characterisation of the institutional setting. It excludes all the institutions which were used to extract surplus value in pre-capitalist systems: slavery, bondage, vassalage etc. If the level of abstraction of this chapter allows me to abstract from historical times too, I would say that capitalism came into being with individual freedom. Therefore its very advent opens up a theoretical problem: how to extract surplus value from the production process if workers are not bound by any personal obligation toward capitalists? That advent itself resolves the problem: capitalism came into being with the diffusion of the employment contract. This is the fundamental institution of capitalism, the institutional condition for the existence of capitalism in general.

30 The employment contract, transaction institutions and capitalism As I will show in Chapter 6, there can be different forms of capitalism, depending on the institutional systems that regulate property rights and sustain capital accumulation. But in common they must all have the use of the employment contract. The employment relationship is sufficient for identifying the working class as a ‘class in itself’ and as one of the key figures of capitalism. There is no need to know anything else in general about the workers to understand that they belong to a class undergoing exploitation in the production process and subjugation in the labour process: it is sufficient to know they have taken on the obligation to carry out subordinate work for a capitalist employer. This is capitalist exploitation in general because that kind of contract represents an institutional condition that holds in any form of capitalism. It is, however, still only a class in itself. The notion of ‘class’ is used here in the mathematical sense, to define a partition of civil society in terms of social actors who share a common property. The property defining the partition is the obligation to undertake subordinate work; in other words, a legal characteristic. In the definition of ‘class in itself’ workers are only juridical figures, social entities still deprived of body and soul. Nothing is known of their class interests, their political relations with other social actors, their ability to organise themselves as a collective actor, in other words of their being a ‘class for itself’. But nothing more can be expected at this level of abstraction. As far as the class of employers is concerned, it is for the moment impossible to identify it precisely even as a class in itself. They are undoubtedly capitalists, yet nothing is known of how and why they are. The employment contract, which is a necessary condition for the existence of capitalism, is not however a sufficient one. It does not guarantee the use of surplus value to accumulate capital. Thus while it makes it possible to identify with precision one of the two contracting parties, the worker, as a typical figure of the capitalist mode of production, the other party remains undefined as ‘capitalist class’, so much so that I have had to resort to the generic term of ‘employer’. The contract of employment defines the employer as only a potential capitalist. It reveals nothing of the way surplus value is spent. Yet, in order to be able to speak of capitalism, one must know the conditions that ensure the use of surplus value for the accumulation of capital. Therefore to identify the capitalist class as a subject of capitalist exploitation, it is necessary to identify other institutional conditions besides that regulating the employment relationship. Since history has produced different institutional systems that can be defined as capitalist, it is necessary to define different capitalist forms, each of which is characterised by specific institutions that guarantee capital accumulation. In each of these forms the capitalist class acquires different characteristics, and therefore only after descent to a lower level of abstraction is it possible to identify the capitalist class. A general definition of the capitalist class will be proposed only in the last chapter of this book. There is a fundamental theoretical difference between the concept of ‘working class’ and that of ‘capitalist class’. The former is perfectly defined in general on the grounds of the analysis of the employment contract, while the latter is not. As

The employment contract, transaction institutions and capitalism 31 particular institutional forms of capitalism change, so may the capitalists’ faces: they may be private owners of the means of production, managers, state functionaries etc.11 A change in the form of capitalism may bring about a change in the nature of the capitalist class, in the social figure confronting the working class in the production process and in the stipulation of the employment contract. What cannot change is the fact that the exploited class is that of the wage workers. The working class has a sort of theoretical primogeniture: first comes labour, in the analysis of capitalism, then the capitalists. Social reproduction: the financial conditions Strong objection is often raised against the very idea of exploitation as an essential characteristic of capitalism: if the workers are rational agents and free subjects, what prevents them from refusing the exploitation and oppression imposed by the employment contract? Actually there appears to be nothing to prevent them. There are alternative institutions which are often used by the workers to regulate their interactions with other social actors. Every day thousands and thousands of self-employed concerns are set up in the capitalist world, by artisans, shopkeepers and professionals, who sell their services through contracts for services or agency relations, or the products of their work by means of sales contracts. And worker co-operatives are set up with which the workers stipulate no employment contract. There is nothing strange in this in the light of the principle of impurity. The notion of a capitalist mode of production does not imply the exclusive existence of capitalist institutions. It only implies that a particular set of institutions, among which the employment contract, is predominant; that the great mass of workers is able to find a job only by signing an employment contract. Thus the question should be reformulated: what prevents most rational agents from choosing institutions that are alternative to the employment contract? Now the problem can be posed as follows: assuming all individuals are rational, only a limitation of choices can induce workers to agree to earn their living by submitting to an exploitation and oppression relation. In other words, it is necessary to identify the economic conditions that induce many workers to accept rationally a contract of employment. On this theme the socialist tradition has developed a theory which is now consolidated. Self-employment requires control of the means of production. Without this, workers cannot control their own work. If economic conditions are such as to prevent workers from controlling the means of production required to perform a labour activity, the choice of transaction institutions which enable selfemployment is precluded, so that the acceptance of an employment contract may become a forced ‘choice’. This, however, is still not an exhaustive answer to the crucial question, for one could ask: what prevents the workers from controlling the means of production? It is not the problem of the birth of capitalism. Granted the existence of the initial conditions for the emergence of capitalist institutions, the stability and robustness

32 The employment contract, transaction institutions and capitalism of those institutions still have to be accounted for. The really interesting problem, therefore, is what socialist literature defines as ‘the problem of social reproduction’. It is a matter of identifying the economic conditions and social mechanisms ensuring reproduction of the capitalist social structure, i.e. those conditions and mechanisms which, on the one hand, limit workers’ choices to unemployment and the employment relation and, on the other, enable the capitalists to preserve their power with the passing of time. The classical solution in socialist literature focuses on the institution of private ownership of the means of production. In the Marxian theory, for instance, the concentration of wealth in the capitalists’ hands ensures them ownership and, as a consequence, control of the means of production, whilst total deprivation of wealth compels the proletarians to accept wage work. The economic mechanism that stabilises the social structure in this model consists of a distributive process which reproduces in income distribution the same inequality existing in wealth distribution. The assumption that workers earn a subsistence wage and have a zero propensity to save accounts for their impossibility to accumulate wealth and therefore to escape exploitation through class mobility. The limitations of this theory are well known. The assumption of a zero propensity to save could be justified if referred to the type of capitalism that prevailed in the first half of the nineteenth century, but today is no longer acceptable. If a consistent mass of workers earn incomes over and above the subsistence level, they can accumulate wealth. Therefore, if control of the means of production depends on the possibility of acquiring property rights over them, then the mass of proletarians could in time have become capitalist. But this did not occur historically, even in the most advanced capitalist countries. In reality ownership of the means of production is neither a necessary nor a sufficient condition for production control. It is not necessary, as is proved by the fact that today both private and state companies are largely controlled by managers who do not own the means of production. The diffusion of shareholding and profitsharing among employees not leading to a diffusion of workers’ control of the means of production proves that it is neither a sufficient condition. Thus what has to be identified is a set of general conditions of social reproduction which are valid for all forms of capitalism. Some of these conditions will be dealt with in Chapters 2 and 3, but two of them can be identified straight away. The first is finance. Capital investment requires availability of purchasing power in advance of the production of profits. This is true for the set-up of a firm. But it is also true as regards the growth of firms. Since capitalism implies capital accumulation, it calls for continual growth in investments with a consequent ever-increasing demand for finance. Therefore, although self-finance is an important accumulation instrument, external credit is an essential element. It is a known fact that capitalist firms have privileged access to finance facilities through numerous channels, bank, state and commercial credit, new share and bond issues, etc., which are not so easy for workers to achieve. There is a general mechanism that limits the institutional choices open to the workers and therefore contributes to reproduction of the capitalist social structure. The gist of this mechanism comes under what could be called the theorem of

The employment contract, transaction institutions and capitalism 33 ‘class-biased credit rationing’, which states that in the capitalist mode of production capitalist firms have privileged access to finance for investment, while workers have a more limited and costly access. The most important reason why workers are strongly credit rationed is that creditors tend to grant loans only to reliable and solvable debtors; to subjects: who possess some capital to offer as collateral; who present profitable and low risk projects, and who have managerial ability and a reputation of solvency. Only rarely is a worker or a group of workers adequately endowed with these qualities. Because of their low incomes, workers are unable to accumulate consistent wealth resources to offer as collateral. Generally, the only door open to a worker aiming to jump outside his class is that of self-finance – a door that is narrower than the eye of a needle. Some people do succeed in making that jump, mainly by setting up small co-operatives or individual self-employed enterprises; but too few of them to cause extinction of the working class. The self-financing process also functions as a class selection mechanism. Firms and capitalists normally earn higher incomes than workers and have a higher propensity to save. Therefore, insofar as accumulation is limited by internal finance, a production sector managed by workers cannot, in the long run, grow faster than the capitalist sector. Even if workers’ enterprises succeed in obtaining external finance, their low level of capital will produce a higher debt ratio than that of capitalist firms. This raises their bankruptcy risk and consequently the cost of finance: not only are workers’ enterprises heavily credit-rationed, but external finance costs them more than it would a capitalist firm. This contributes to reducing their net profitability and therefore to hindering their birth and growth. An important instrument for implementing selective credit rationing is the determination of the risk structure of interest rates. It is not the only one, of course, and even more important are the banks’ policies in directly rationing the quantity of loans and in demanding differential collaterals. However, without loss of generality, we can ignore the latter two instruments and focus on the interest rates structure. For the sake of simplicity let me assume that only two classes of potential borrowers exist, workers and capitalists. Let me assume, moreover, that their respective investment projects yield the same prospective rates of profit, but also that, because of their poorer managerial ability, the workers’ projects are considered riskier. Under the hypothesis that there is no pure quantity rationing, credit supply is endogenous. But, under Kalecki’s increasing risk hypothesis, we have a rising credit supply curve. Class biased credit selection will be implemented through the fixing of two sets of interest rates, which I will call ‘privileged rates’, rp, and ‘usury rates’, ru. The minimum privileged rate, or prime rate, is rp* , while ru* is the minimum usury rate, with rp* < ru* . The supply curves for worker credit, SL, and capitalist credit, SB, are represented in Figure 1.2, where the workers’ and capitalists’ demand curves, DL and DB, are also traced. The workers receive an amount of credit equal to NL, the capitalists an amount equal to NB. The latter will obtain almost all the finance they ask for, whilst the former will be strongly rationed.

34 The employment contract, transaction institutions and capitalism r SL

SB

ru ru* rp DL

rp*

0

NL

DB

NB

Credit

Figure 1.2

The risk structure of interest rates is represented by the spread ru – rp, which depends on the banks’ preference for money and their propensity to risk bearing. The levels of the rates are determined politically by the central bank, which regulates access to base money by using the discount rate, rd. Credit becomes more selective when the difference ru – rp rises, given rd; or when rd rises, given the spread ru – rp. The role of the state in regulating the financial conditions of social reproduction will be dealt with in Chapter 3. Social reproduction: the production conditions The second condition of choice limitation relates to the labour process. Imagine a situation in which the owners of a firm decide to sell it. The buyers could be other financial capitalists, the firm’s manager or the workers. If the firm were bought by external shareholders, the latter could substitute the management or renew the internal managers’ mandate. In either case the firm would remain under the control of entrepreneurs. Little would change for the workers if control were exercised as a function of capital accumulation. If the firm were bought by the workers, would these be able to offer the banks any guarantee of capitalist efficiency or guarantees which are considered at least as strong as those offered by an autonomous management? If the answer to this question is negative, the theorem of class-biased credit rationing is well justified. The concept of ‘capitalist efficiency’ defines the ability to use production to further capitalist accumulation; the ability to extract surplus value and to invest it to sustain company growth. Capitalist efficiency is not, like Pareto efficiency, a thing which either is or is not there. It can be there to a greater or lesser extent. It rather resembles X-efficiency, as it refers to the effectiveness of the managers’

The employment contract, transaction institutions and capitalism 35 control of the production process besides that shown in mastering the external competitive environment. The managers’ ability to control a firm efficiently implies their ability to control the workers. In the same way, the workers’ ability to control a firm efficiently implies their ability to control the management. The problem then is: who is better equipped to control the other party? There is now vast literature sustaining the view that the control of a firm is optimal when it is assigned to the social group which is better equipped with managerial abilities, and that, since companies are actually governed by the entrepreneurs and not by the workers, this proves that the former are more capable to control the latter rather than the other way round. Let me call this the ‘orthodox view’. Many reasons are offered to justify this asymmetry, the most interesting of which are: asset specificity, asymmetric information, asymmetric knowledge and heterogeneity of motivations. If the entrepreneurs’ skills are specifically linked to the firm they manage, while the workers’ skills are not, the latter can count on finding alternative employment opportunities whilst the former cannot. Therefore if the workers have the right to fire the managers, they could use this power to compel the latter to accept incomes which would be inadequate to induce them to invest in their skills. Efficiency would then be impaired. This problem would not arise if control were assigned to the managers. Therefore managerial control is more efficient than workers’ control. Asymmetric information, in this context, means that one of the two groups possesses information or can perform actions which cannot be controlled or which are too costly to be controlled by the other group. If the managers are difficult to control, they could obtain advantages that would remain hidden from the workers. If the workers do not possess hidden information, they could be easily controlled and would be unable to gain any rent on their activity. Therefore if the workers controlled the company they would be unable to compel the managers to pursue the workers’ goals, whilst, if the managers had control they would be able to pursue their own goals by directing the workers’ activity. The managers’ control would be efficient from the managers’ point of view, whilst the workers’ control would be inefficient from the workers’ point of view. If managers are the ‘functionaries of capital’, their control would enhance capitalist efficiency. Asymmetry in knowledge endowment occurs when only one of the two groups controls the other group’s technical or organisational knowledge. This is a pervasive kind of asymmetry in contemporary capitalism. It has been fuelled by the growth of big business and mechanisation, by the tendency of companies to incorporate scientific research in their activity, by the birth of scientific management and, more recently, by the development of automation. Most of all, it is sustained by a general phenomenon which lies outside the companies structure, i.e. the tendency of contemporary culture to separate scientific from practical knowledge. Workers are educated in schools that cultivate manual dexterity and the ability to carry out subordinate work. The dominating classes, on the contrary, are educated in abstract and strategic thinking, in decision-making ability, in risk and responsibility bearing, in independent activity and the passion for power. Entrepreneurs are therefore

36 The employment contract, transaction institutions and capitalism endowed with the knowledge required to manage companies, to take risky decisions, to exploit the workers’ knowledge. The latter, on their side, are not even capable of reading a firm’s balance sheet. So how could they control its management? It is obvious that managerial control is more efficient than the workers’ control. As to the heterogeneity of motivations, it seems sensible to expect that, if the workers had control of their firms, they would tend to use it to improve their working and life conditions. Probably they would raise their incomes to the detriment of profits and accumulation, modify the labour process to make it less boring, tiresome and frustrating, reduce labour time, improve safety in the workplace and stabilise employment. The entrepreneurs, on the other hand, are not subjected to hetero-directed activity, and enjoy the exercise of power and a level of income which grow with the size of the firm. It is more than natural for them to use power to expand production and feed accumulation. No wonder therefore that the banks privilege them rather than factory councils.12 Thus the ‘orthodox view’ correctly describes the reality of capitalism: the technical, informational and cultural conditions of production in a capitalist economy justify the assignment of control to those who have it. Yet it does not tell the whole truth. Another part of this truth is brought to light by a ‘radical view’ which has grown up in opposition to the orthodox. The radical view maintains that it is the social structure of control that determines its technical and cultural bases. More precisely, it maintains that, once control has been taken away from the workers and taken on by a class of people who use it to extract surplus value and feed their own power, it will be handled to mould the technical conditions of production and the knowledge and information endowments of workers so as to sustain and enhance capitalist control itself. Technology and culture are not neutral, but are shaped by those who control them to favour the pursuit of their goals. And since control is in the hands of capitalists, the evolution of technology and culture reinforces the efficiency of capitalist control. The orthodox and the radical view are opposed but not alternative. They appear to be complements rather than substitutes. The orthodox view, when taken on its own, errs on technological determinism and does not account for the reasons why the mode of production has evolved in the direction that privileges capitalist control. If technical and cultural change is exogenous and neutral, one cannot understand why it has always privileged the dominating classes. The radical view, taken on its own, reveals a sort of conspiratorial vision of history and gives the impression of deliberately ignoring an indisputable historical fact: the superiority of capitalist control over worker control. However, when taken together, the two views contribute to account for the robustness and institutional stability of the capitalist mode of production and therefore to resolve the problem of social reproduction. The argument can be formulated using the Marxian notions of ‘production relations’ and ‘forces of production’. Social production relations, that is, the social conditions for capitalist control mould the evolution of the ‘forces of production’, that is the technical and cultural conditions of production. Then the evolution in the latter justifies and sustains those social relations. The system is robust in that, once triggered off, it can expand by invading a given institutional system to the

The employment contract, transaction institutions and capitalism 37 point of becoming dominant in it. Moreover it is stable as it is able to absorb or eliminate any technological, institutional and cultural innovation which is otherwise not compatible with its own dynamics and structure. Both robustness and stability are ensured by the positive feedback interaction between production relations and the forces of production. This thesis will be argued in greater detail in Chapter 6. Here I have at least to sketch some of its implications, for the factors sustaining evolutionary robustness and stability include those brought to light by the theorem of class-biased credit rationing. The conditions of capitalist control over production can lead to bankruptcy worker-managed co-operatives that do not conform to the rules of competition; or they can induce them to conform, e.g. through transformation into managercontrolled firms. A company controlled by its workers cannot grow if it does not produce profits for accumulation. If it does not grow it cannot innovate. If it does not innovate it is expelled from the market. Moreover, if it is financed externally, lack of profits will prevent it from meeting its financial commitments. But, if the banks and the financial markets foresee this possibility, they will not grant loans or will grant them only at a high cost, so that a co-operative firm would not even be able to set up. If the firm aims to produce profits at least at the same level as capitalist firm competitors, it must permit an efficient use and control of labour; and an independent management is better able to guarantee this control than a factory council. The workers will therefore have to elect a board of directors and assign it wide decision-making powers. The directors, if they want the firm to survive and prosper, will have to control and exploit the workers in the same way as their capitalist colleagues do. The financial conditions of social reproduction, as brought to light by the theorem of class-biased credit rationing, work well insofar as they are sustained by another set of conditions. These represent the production conditions of social reproduction, and are highlighted by both the orthodox and radical views. They too can be expressed in the form of a theorem: in the capitalist mode of production the technological and cultural conditions of production are produced by the capitalists’ actions, which mould them in such a way that capitalist control turns out to be justified as the most efficient. I will call this the ‘theorem of class-biased control’. Since the conditions for production control sustain those of credit rationing, one could wrongly infer that the theorem of class-biased credit rationing is not necessary to account for social reproduction. To understand why it is indeed necessary, one can envisage a hypothetical institutional system dominated by capitalist relations, in which, however, some sort of euthanasia of the rentiers has taken place. The state, for example, could control the banking system making it play the role of risk insurer and enabling it to endogenously produce all the money needed to sustain accumulation. The risk and liquidity premiums would be cancelled, and all the companies would obtain free access to credit. In this fantastic land of Proudhonia the theorem of class-biased credit rationing would not apply. However, capitalist control of production might remain more efficient than worker control, at least initially. Let me assume that a Proudhonian worker co-

38 The employment contract, transaction institutions and capitalism operative is able to produce, besides the income required to repay loans, a value added sufficient to pay the workers the same income they would earn by working as employees of a capitalist firm, but not sufficient to accumulate profits for growth to the same extent as the capitalist firm. The availability of credit would enable the co-operative to start up, survive and even grow, but not at the same pace as capitalist firms. The members of the co-operative could use their control over their firm to modify the techniques and production process so as to make work less frustrating and tiresome. Moreover they could utilise part of the company profits to launch an on-the-job training programme that would allow them to acquire the information, knowledge and skills required to augment their control capacity. In this world of free credit, the innovation represented by co-operative firms could acquire a certain robustness. These firms are assumed to be, at least initially, less efficient than capitalist firms if they do not produce as much profit as them. They would not be able to grow as fast as the latter. However, they would be more attractive to the workers, since they could offer them higher rewards in work significance, recognition, decision-making autonomy and cultural enrichment. As time passes, an increasing number of workers might be tempted to abandon capitalist firms and set themselves up as co-operatives. So that, despite the capitalists firms’ greater capacity to growth in size, the co-operatives sector may expand more rapidly than the capitalist firm sector, simply because it would take workers away from it. It is also possible that the co-operative investments in the workers’ skills, besides providing the workers with a greater stimulus to social commitment and self control, raise their productivity to the point where their firms would even become more efficient than the capitalist firms. In Proudhonia workers cooperatives could enjoy evolutionary robustness. This is a simplified counterfactual example which aims at sustaining the argument that the theorem of class-biased control alone is not sufficient to account for the social reproduction of a capitalist system, and that the theorem of class-biased credit rationing is also necessary. Are the two theorems together sufficient? The answer is again negative. Neither industrial nor financial firms operate in an institutional, political and cultural vacuum. First, the four conditions supporting class-biased control, namely asset specificity, information and knowledge asymmetries and the heterogeneity of motivations, are fed by a cultural and educational system which is largely outside the firms’ sphere. Second, the transaction institutions that are used for the constitution of firms presuppose a normative and political setting which is itself outside the firms’ sphere. The structure of the financial systems that guarantees selective rationing of credit is strongly conditioned by the political apparatus responsible for normative and government actions at macro level. Finally the very working of markets, i.e. the relational fabrics that guarantee competitive selection, depends on a normative and institutional setting which is again external to the firms. In other words, Proudhon’s dream, to set off a revolution by using free credit, is unfeasible because it is simplistic: it does not take into account the fact that the conditions for social reproduction concern the structure and working of the institutional system in its entirety. An analysis of them cannot therefore be restricted

The employment contract, transaction institutions and capitalism 39 to investigating what happens in factories and markets. Cultural and political conditions too have to be investigated, which will be done in the next two chapters. On sources Karl Marx well understood the role played by the employment relationship in creating the social conditions for exploitation. Although he went further into the problem in Capital (1990), in his first works in economics had shown he understood that, with the contract of employment, ‘the free labourer sells himself and, indeed, sells himself piecemeal. He sells at auction eight, ten, twelve, fifteen hours of his life, day after day’ (1978a: 205). This is the idea on which the principal thesis of this book is based. Marx’s theory, however, is lacking as regards the institutional aspects of the employment relationship, especially its ‘constitutional’ function. Indeed Marx himself was trapped in the ideology of commodity-labour, being unable to grasp the difference between a contract of employment and a contract for the sale of a commodity. And, although he was very clear on the fact that wage is fixed through a hostile struggle between workers and capitalists, he always followed Smith’s and Ricardo’s teaching in using the market metaphor to account for wage determination. He understood that ‘labour’ is not a commodity. However, since he based his analysis of capital on the theory of commodity-form and that of the valorisation process on the theory of labour values, he was compelled to introduce the notion of ‘labour power’ to account for the employment contract as a commodity sale contract. Moreover Marx thought that the fundamental institution of capitalism is private ownership of the means of production and that wage labour is a consequence of this institution. I have changed the order of logical priority: the contract of employment is the fundamental institution of capitalism in general; it creates wage labour independently of the way property rights are allocated. I began working on this idea in the late 1980s but, although I discussed it in three seminars between 1989 and 1994, I published it for the first time in Screpanti (1994) and, in an English version, in (1998). When the first chapter of this book had been written I came to know of the important work of D. P. Ellerman (1992), in which a similar idea is developed. Ellerman’s theory, however, does not precisely coincide with mine. Thus it might be worthwhile pointing out at least two of his arguments which I find perplexing. Ellerman, following Samuelson, reduces the wage to the rental of a flow of services and likens the employment contract to a contract for renting the services of a capital stock. He does not grasp the substantial difference between a contract of employment and a contract for services. As a consequence he remains trapped in the ideology of commodity labour and does not understand the fundamental fact that what the worker sells in exchange for a wage is an obligation to obedience and not a flow of goods. He maintains that the activity of a worker in the labour process is a labour action, in other words that the workers, in deciding to obey a command to perform a certain operation in fact decide that operation. Ellerman fails to understand that the workers’ obligation to obedience deprives them of

40 The employment contract, transaction institutions and capitalism decision-making capacity and creates the conditions for labour services becoming factually and legally products of the employers’ decisions. G. M. Hodgson (1999) also recognises that the employment contract is a fundamental institution of capitalism. He maintains, however, that this institution is not disjoined from those that define property rights. This seems a fairly sound idea. After all, the employment contract can be considered as a transaction institution that also constitutes property rights, namely the employers’ ownership of the goods produced by their employees. Furthermore there is no doubt that the distribution of property rights contributes to determining the distribution of incomes and surplus values and, as I will show in chapters 4 and 5, even to producing exploitation relations outside the labour process. However, one principle must hold firm: the employment contract is the fundamental institution of capitalism as it is necessary to guarantee labour exploitation in the production process independently of how property rights are allocated. The idea that the general conditions for capitalist class relation are posed in the labour process and not in the sphere of legal institutions defining property rights has been investigated by G. La Grassa (1993, 1996) and G. La Grassa and M. Bonzio (1990). This idea is full of philosophical implications for the theory of socialism. As has been brought to light by D. P. Ellerman (1992), B. Jossa and G. Cuomo (1997) and B. Jossa (1998), the overcoming of capitalism through the building of socialism is not so much a problem of changing the structure of property rights as one of abolishing wage labour. In contrast to Marx’s view, I believe that not all social relations take the form of commodity exchange in the capitalist mode of production and that the one which is truly fundamental is not based on commodity exchange. For this reason it is necessary to ground the analysis of capitalism on a more general concept than that of ‘commodity exchange’. I have tried to do this by introducing the notions of ‘transaction’ and ‘transaction institutions’. The attempt to account for the constitution of civil society on the ground of transaction institutions is inspired by an argument put forward by John R. Commons (1924), namely that transaction is the fundamental measure in economy and law, a sort of social electrolysis which guarantees economic choices, the exercise of power and the aggregation of men. The difference between the employment contract and the sale contract has been grasped by the economists, although not yet by all of them, only recently and mainly thanks to Herbert A. Simon (1951). The inspiration I drew from this essay does not however extend to the point of espousing its conclusions. In fact I explicitly abstracted from social differences in risk aversion in attempting to bring out the rationale of the employment contract. It seems that jurists are better equipped than economists to understand the nature of this kind of institution. After all, their investigations have a bimillenary tradition. My interpretation of the locatio operarum as a case of the do ut des (I give, you give) form of exchange, is drawn from R. Martini (1958). Marxist jurists have been misled by Marx’s idea that labour power is a commodity. K. Renner (1976), for instance, while recognising that ‘wage labour

The employment contract, transaction institutions and capitalism 41 is a relation of autocracy with all the legal characteristics of despotism’ (p. 114), insists in treating the exchange of labour as a case of do ut facias (I give, you make), thus likening it to the contract for services. G. Kay and J. Mott (1982), who provide an indepth investigation of the role of the capitalist state and legal system in codifying, regulating and legitimising the employment relationship, follow Karl Renner in interpreting it as a do ut facias form of exchange. The idea that the employment contract performs a ‘constitutional’ function in the organisation called a ‘capitalist firm’ would locate this theory in the ‘constitutional paradigm’, for which I refer to J. S. Coleman (1986) and V. J. Vanberg (1994), who also developed the notion of ‘corporate actor’ and applied it to the study of the firm. However, the nature of the internal hierarchy of the capitalist firm as an exploitation-oriented structure has been captured with greater acumen by S. A. Marglin (1974). The analysis of the conditions for the social reproduction of capitalism draws on many contributions of recent radical and institutional thought. We owe to G. M. Hodgson (1988) the formulation of the ‘principle of impurity’, which I have tried to justify with the ‘principle of institutional plurality’. The theorem of classbiased credit rationing comes from post-Marxist approaches to the monetary circuit. It has now become a sort of folk-theorem to which it is difficult to attribute a precise paternity. A clear enunciation can be found in A. Graziani (1989), a close investigation in S. Bowles and H. Gintis (1996). A treatment of endogenous money supply under the hypothesis of increasing risk can be found in Screpanti (1997). As far as class-biased control is concerned, the ‘orthodox view’ has been developed in many researches following a neoclassical institutional approach. I will limit myself to citing just a few classical works: R. H. Coase (1937, 1960), A. Alchian and H. Demsetz (1972), D. C. North and R. P. Thomas (1973), O. E. Williamson (1985) and S. Grossman and O. Hart (1986). Some fundamental works that develop the ‘radical view’ are: H. Braverman (1974), S. A. Marglin (1974, 1991), R. Edwards (1979) and S. Bowles and H. Gintis (1983). Finally an integration of the two views in a model of the organisational equilibrium has been accomplished by U. Pagano (1991a; 1991b) and U. Pagano and R. Rowthorn (1994). To return to Marx, I have drawn extensively on what I consider one of the deepest of his philosophical achievements, namely the theory of alienation he expounded in Economic and Philosophical Manuscripts. The theory of alienation is more enlightening than that of commodity fetishism developed in Capital. Alienation is linked to the expropriation of labour activity generated by the employment relationship, whilst fetishism depends on commodity exchange, i.e. on the opacity with which the commodity values, as determined in the market, reflect the social division of labour. Finally, the idea of measuring value in purchasable or commanded labour has been put forward by Adam Smith (1970) as a basis for a suitable theory of value in a capitalist economy. Marx appreciated this idea, but not to the point of giving up a labour theory of value that appeared to him stronger from an ontological point of view, in spite of the fact that Smith had correctly relegated its validity to a precapitalist economy.

42 The employment contract, transaction institutions and capitalism

Notes 1 The rate of interest charged to Mr Boss is determined institutionally. In a sense it is arbitrary and, for the sake of abstraction, it could even be assumed nil – an assumption that would ease understanding of the argument developed in the following sections. Marx does this assumption throughout volumes I and II of Capital and the first three parts of volume III. Interest rate determination will be dealt with in more detail at the end of the present chapter and in Chapter 3. 2 The principle of equivalence is implicit in the assumption of perfect competition and is present, not only in neoclassical pure economic theory, but also in classical and even Marxian theories. Marx and Engels called it the ‘law of exchange’ or ‘law of value’. 3 In using these concepts, I try to abstract from any particular legal system of contemporary world, mainly because there is a great deal of variation in the definition of concepts in present forms of legislation. 4 Some confusion sometimes arises because of a too generic use of the ‘principal-agent’ model, especially among the economists – a use that tends to subsume in it all the contractual forms suitable for the mobilisation of work. To avoid this confusion, I shall ignore the fact that the agency relation is not considered a contract in common law systems, and define the ‘contract of agency’ in the Roman sense of mandatum: a contractual form according to which a principal (mandator) gives a mandate to an agent (procurator) to take decisions and undertake actions autonomously in the interest of the principal himself. This kind of contract does not imply the acquisition of a faculty of command by the principal, nor the obligation to obedience of the agent. 5 The tendency of many neoinstitutionalist economists to treat the employment contract as an incomplete contract for services dates back to Commons (1924), in whose work it goes hand in hand with the inclination to reduce capitalism to the market system. 6 Among the conditions ensuring this infringement one must include the existence of monopolistic and monopsonistic practices in the markets for goods, labour services and credit. The psychological, institutional and political conditions contributing to the determination of interests rates in financial and money markets are also very important. To the extent that self-employed workers do not own their means of production, they may be exploited through the market by financial capitalists, and the more so the higher the interest rate. For instance, the fact that the sharecropping contract does not enable the owner to earn a profit in case his product share coincides with interest payments means that the worker is not exploited in the production process by the owner himself; however, if a surplus value has to be produced to repay the interest on capital, the worker is exploited by virtue of the financial market conditions which make for the existence of a positive interest rate. A similar argument applies to the payment of rents for the use of land. 7 This section is partly inspired by the section of Marx’s Economic and Philosophical Manuscripts dedicated to ‘alienated labour’, and is actually written by using ample paraphrases from it. It is also inspired by the theory of commodity fetishism developed in Capital. The reader will observe, though, that my interpretations of the notions of ‘alienation’ and ‘fetishism’ are not so standard. 8 The value of a commodity ‘to those who possess it’ – says Smith (1970: 133) – ‘is precisely equal to the quantity of labour which it can enable to purchase or comand’. Following Smith, ‘labour commanded’ is the term normally used by economists to define this measure of value. I am compelled, though, not to conform to this usage. To avoid confusion with the notion of ‘command over labour’ as defining the exercise of power in the labour process, I will use the expression ‘purchasable labour’, instead of ‘labour commanded’, as a synonym of ‘value measured in wage units’. 9 If the numeraire is an average wage, L is the purchasable time of a labour force whose skill composition corresponds to that of the employed workers. Here, though, for the sake of simplicity, it is convenient to assume that labour is homogeneous and the wage rate is unique. 10 Note that, if contractual labour time differs from effective labour time, Lc ≠ Le, the rate of exploitation depends explicitly on the former. In fact wage is normally paid on contractual

The employment contract, transaction institutions and capitalism 43 labour time. Effective labour time, obviously, is not inconsequential in determining the rate of exploitation, for labour productivity depends on it. As can be observed in figure 1.1, Ly can be taken to be a function of Le. The case of piece work and other forms of incentive pay, in which wage is partly commensurate to effective labour time, is investigated in chapter 5. 11 This is why the terms ‘capitalists’, ‘managers’ and ‘entrepreneurs’ are used here as synonyms: they generically define the social actors who perform the entrepreneurial function, those who have control over production activity and take investment decisions. 12 Note that, to account for the first three conditions of the efficiency of capitalist control, I have removed some of the hypotheses defining this chapter’s level of abstraction. The loyalty of all the individuals living in Mr Boss’s world, as well as the availability of all information, would not permit the emergence of any form of opportunism from asset specificity or asymmetry in information and knowledge. The fourth condition, though, is compatible with that level of abstraction: even in such a simplified world social actors could be moved by heterogeneous goals.

44 Individuals, culture and behavioural institutions

2

Individuals, culture and behavioural institutions

Premisses The problem of ‘human nature’ Man is the result of his fortunes, the term ‘fortune’ meaning both ‘wealth’ and ‘a hypothetical force unpredictably influencing events’. This proposition becomes especially meaningful when referred to men living in a capitalist society. There is no special insight in the idea that human nature cannot be defined by abstracting it from history and that rather it is the result of an unpredictable evolutionary process. More incisive, though, is the observation that history has led to a cultural and institutional setting in which human behaviour is conditioned by the economic dimension of life to the point where human substance is reduced to the accumulation of wealth and power. Certainly this observation is useful in an investigation focusing on the institutions of capitalism and their influences on individual behaviour. In this chapter I do a step back and two forward with respect to the preceding one. Back, because I focus on the social actors as individuals rather than social classes. The idea that collective action is the outcome of the actions of the individuals who take part in it is little more than self-evident. It is compatible with the proposition according to which ‘the whole is more than the summation of the parts’, since the relations among the parts in a social aggregate may be so complex that individual actions do not combine ‘additively’, but rather ‘multiplicatively’, ‘exponentially’ or even worse, so that the most diverse forms of non-intentional aggregate outcomes may ensue from intentional behaviour. The move forward consists in removing two simplifying and counterfactual hypotheses, namely complete information and unbounded rationality. Dominated by the terror of indeterminacy, the fathers of traditional neoclassical theory had to assume those two hypotheses to account for the behaviour of homo economicus, with the result that they transformed individuals into social atoms and behavioural automatons and deprived them of any possibility of real choice and error, that is, of any humanity. Removal of these two hypotheses is necessary to understand real human action. Understanding human action requires three problems to be resolved. First, the nature of the sentiments and moods which move individuals to action must be grasped. This is not just a question of egoism or altruism or accounting

Individuals, culture and behavioural institutions 45 for the propensity to innovation, entrepreneurship and change. The real problem is to avoid the hypotheses on human motives being referred to a fictitious ‘state of nature’. Second, there is the question of the sense in which human action can be defined as rational, so that it succeeds somehow in adjusting the means to the ends. This is also the problem of bringing to light the social conditions that make for rationality of actions and those that hinder it. Finally, the sense in which individuals act autonomously must be highlighted, and this is the question of identifying the limits within which the objectives of human actions are defined freely. Egoism, altruism and mental feelings In identifying the basic motivations behind economic actions, economists have always oscillated between two contrasting hypotheses: altruism (or moral sentiments, sympathy and the like) on the one side, and egoism on the other. Adam Smith too underwent this oscillation. However, after his abandonment of the philosophy of moral sentiments, the overwhelming majority of economists, from the classicals to Marx, from the neoclassicals to the Keynesians, have accepted the egocentric orientation, so much so that individualist ontology has now become part and parcel of economic science. Only recently, particularly because of the many theoretical failures of neoclassical economics and partly due to the contemporary explosion of neo-institutional, behavioural and evolutionary approaches, a partial re-evaluation of altruism as a basic bond of social life has begun to take place. Taken alone, however, neither of the two hypotheses holds water. The idea of absolute altruism, i.e. that all individuals are always altruistic, clashes with the common experience of the evils of the world and their human origin. It is also too simple an hypothesis: rather than working out the problem – the problem of what binds society – it serves to evade it. But, more than that, it is an absurd hypothesis; for, if altruism consists of the desire to benefit other people, it implies that there are egotistic people, that is, people who enjoy being benefited. This, however, does not validate the reverse idea that no individual is an altruist or the other similar and more widespread one according to which most of them are just egoists. The assumption of overriding egoism is often justified in terms of realism – a justification that cannot be accepted, however, without qualifying it with the observation that the world is full of examples of systematic and diffused unselfish forms of behaviour. Just think of the wide diffusion of voluntary work even in the greediest societies, or of the many kinds of social relations that are based on love, solidarity and a sense of duty or justice: families, charitable associations, some political parties etc. More often, the hypothesis of predominant egoism is justified in euristic terms. Some economists believe it is the most economical, in the sense of Occam’s razor. If you can account for social life even in the case of general egoism, you need not resort to easier assumptions. Other economists maintain that, even if it is not true that all human actions are motivated egoistically, it is certainly true that at least the economic ones are. The weakness of both these arguments is the same as for

46 Individuals, culture and behavioural institutions neoclassical economics – a science whose scope is defined by its method rather than its object. But it is possible to raise two more radical criticisms to the traditional economic theory of human motivations. The first is based on the observation that egoism and altruism are often treated as criteria of rational choice in an approach which remains basically utilitarian. One tries to maximise the utility of somebody, either his own or his neighbour’s. In reality human actions are often moved by quite different forces from profit calculation: love and hate, and envy, friendship, contempt; and anger, fear, hope, audacity and so on. The second criticism is based on the observation that both the hypotheses of egoism and altruism, as used by most economists, are metaphysical assumptions: they define human nature by abstracting it from the historical and social context and referring it to a fictitious state of nature. The point is that humans have never been observed in their natural state and indeed it is impossible to do so. These criticisms open up two paths of research which promise to thoroughly redress economic theory. The first concerns the role played by sentiments, moods and emotions in determining human actions and motivations. The second, the way in which mental feelings are affected by the cultural, social and institutional context. Mental feelings, intended as drives to action not reducible to a rational calculation, are multifarious and heterogeneous, but can be classified, for the purpose of economic applications, into a few general categories. They can be ordered by means of two simple criteria, one pertaining to the nature of the object, the other to the disposition of the subject. From the point of view of the object, it is possible to distinguish mental feelings oriented towards persons from those oriented towards situations. I will call the former ‘sentiments’ and the latter ‘moods’. From the point of view of the subject it is possible to distinguish the mental feelings produced by an attitude of open-mindedness from those produced by an attitude of narrow-mindedness. Thus the sentiments produced by an individual’s open-mindedness towards his neighbour are called ‘altruistic sentiments’, whilst those produced when the individual’s horizon is restricted to the self are called ‘egoistic sentiments’. The moods characterised by open-mindedness to change are called ‘innovative moods’, whilst those characterised by narrowmindedness to a given situation are called ‘conservative moods’. Table 2.1 illustrates schematically the classification. Love, solidarity and a sense of justice, for instance, belong to the N-W square; envy, greed and avarice to the S-W one; hope, audacity and love of risk to the N-E one; fear, doubt and risk aversion to the S-E one. Table 2.1 Mental feelings Nature of the object

Attitude of the subject

Openmindedness Narrowmindedness

Persons

Situations

Altruist sentiments Egoist sentiments

Innovative moods Conservative moods

Individuals, culture and behavioural institutions 47 Mental feelings so defined must not be confused with emotions. The former refer to the states of mind that affect the formation of human motives and aspirations and determine actions and decisions. They belong to the realm of causes. Emotions, by contrast, belong to the realm of effects: they consist in psycho-physiological experiences caused by the outcomes of actions. For example, joy, rejoicing, enjoyment, elation, bliss, boredom, regret, disappointment, grief, shame and guilt are emotions, not sentiments. Emotions have a valence, which can be positive or negative and ranked on a satisfaction-dissatisfaction scale. Once this ranking has been defined, the temptation may arise to reduce all emotions to a single variable (and why not call it ‘utility’?) Now, although it might be legitimate to give in to this temptation in certain circumstances, the following qualifications must always be kept in mind:

• • • •

emotions are heterogeneous and cannot easily be reduced to one another; since they are caused by the outcomes of actions and since these may be triggered by sentiments, emotions are normally associated with sentiments;1 the intensity of emotions is related to the intensity of the sentiments associated with them; even if actions bring about emotional satisfaction, they are not necessarily undertaken with the intention of obtaining satisfaction; often they are directly triggered by sentiments regardless of their possible emotional effects.

Coming to the second research path, recall that it is opened up by the following questions: if men live and act in specific historical and institutional contexts, is it correct to believe that the original and exclusive sources of their actions developed from ‘natural’ aspirations, sentiments and moods? Or is it more reasonable to maintain that culture and institutions contribute to mould human nature? One could be even more radical in this attempt at concreteness. After all we are not interested in knowing the nature of Villanovian humans, but rather of those living in the capitalist society. Thus the real problem is: how does the spirit of capitalism affect the formation of human character? Capitalism is the most revolutionary of all historically known social systems. This is so, not just because accumulation, innovation and change are essential determinants of its dynamic, but also and especially because it exalts the role of the individual in the struggle for existence whilst it tends to debase the traditional social aggregates – family, kinship, guilds, ethnic groups etc. – from which people obtain a sense of belonging and protection. Furthermore capitalism fosters the search for rationality and efficiency as a means of self-assertion. In capitalism, more than in any other known social system, the selection of the fittest requires the use of instrumental reason and nurtures egoistic sentiments and innovative moods. In other words, human nature, as moulded by capitalism, is that of an egoistic, rational and dynamic individual. In this system all individuals, in varying degrees, are made of such an amalgam. But not all of them come out as winners in the struggle for life. The winners will be those who are better endowed with that kind of sentiment, mood and rationality.

48 Individuals, culture and behavioural institutions The prototype of the fittest animal in this world is the innovator entrepreneur, a greedy and sharp individual, eager to get the better of his competitors, avid for power and wealth, always desirous of doing his best, audacious, optimistic, open to the future, and a lover of risk and innovation. In the mind of this person there is not too much room for solidarity sentiments, nor for moods that predispose to peace and leisure. Capitalism nurtures the mental feelings located in the S-W and N-E squares of Table 2.1; for it is these sentiments and moods that open up the road to success, and only the individuals who drive fast along this road are selected as winners. What about the mental feelings prevailing among the lower classes? They are contradictory. The struggle for existence in capitalist civilisation does not lead to the elimination of the losers. Human societies transform the losers into exploited and dominated subjects, their existence being as necessary to the reproduction of the system as that of the winners. Thus there is a tendency for the mental feelings located in N-W and S-E squares to find refuge among the dominated classes. These classes in fact take in the individuals who have proved unable to develop, in the right way and the right extent, the sentiments and moods that pave the way to the accumulation of wealth and power. Moreover it is precisely among these individuals that the ‘sour grapes’ mentality tends to form. Proletarian cultures thrive on the idea that money and power do not make for happiness. Many philosophies of peace spring from the experience of defeat in the struggle for life. However, the fascination of a winner’s sentiments is strong. Since egocentric and innovative feelings open the road to success, everybody tends to attach positive values to them, even people who are not well endowed with them. Social order is perceived as sustained by the individuals’ efforts at improvement, and the ideology by which ‘the weak are food for the strong and the poor are servants of the rich’ – as Geacomo Leopardi says – is perceived as intrinsically right and worth investing some sentiments. This contradiction in the feelings prevailing among the lower classes is a basic cause of the instability of their collective actions. During normal times, when capitalist growth shows up strength and welfare and when the perspective of radical social changes fades away, the losers’ aspirations to deliverance and self-assertion bend toward individualistic feelings. However, since power is a positional good, the majority of those who make bids for power are doomed to failure. Thus the spreading of egocentric sentiments among the lower classes tends to generate rancour, envy and hate. But it may also happen that, in some historical circumstances, the values of mutual aid are perceived among these classes as instruments of social revenge and liberation; for the experience of failure on the ground of individualistic values may cherish the vision of a better world and the hope of conquering it through collective action. I am not trying to describe Arcadia, nor to say that the poor are good and the rich bad. What I am saying is that, although the spirit of capitalism permeates the life and action of all individuals, it shows up at its best in the dominating classes; and that, if social relations exist which rest on altruistic sentiments, they are less likely to carry away the lives of exploiters. It remains true that, since almost all

Individuals, culture and behavioural institutions 49 the economic relations in the capitalist system are based on antagonism, the economic actions of all individuals are more or less dominated by egoistic sentiments. And even when the struggle to improve living conditions opens to the hope of conquering a better world, it is probable that sentiments of class hate and envy play an important role. Summing up, altruism must not be ignored, but it seems sensible to assume that egocentric sentiments predominate in determining economic actions. The justification for this assumption, though, must be sought in the nature of capitalism rather than in human ‘nature’. The desires for power, recognition and self-realisation There are some peculiar kinds of egocentric sentiments that deserve a deeper investigation. Entrapped in Adam’s original sin – the sin of assuming man to be a homo economicus – few economists have succeeded in grasping some of the most important egoistic motivations for social action. Thus we must get away from economic science, and look for enlightenment in the work of psychologists and sociologists, especially organisational sociologists, if we want to discern those motivations. We would then discover that there are at least three particular mental feelings which are not reducible to utilitarian anguish and which account for social action better than the Smithian parable of the butcher, the brewer and the baker: the desire for power, the desire for recognition and the desire for self-realisation. To tell the truth, the importance of the first kind of desire was acknowledged by Adam Smith himself, before he became an economist, when he observed that human action is often driven by the love of domination and authority over others, […] a certain desire of having others below one, and the pleasure it gives one to have some persons whom he can order to do his work. (A. Smith, 1978: 192) The desire for power can be defined, in the words of David McClelland, as the aspiration to have ‘impact, control, or influence over another person, group or the world at large’. As I will show in Chapters 3 and 4, this motivation is the only one that satisfactorily explains the behaviour of politicians and capitalists. It can be located in the S-W box of Table 2.1. It is a sentiment with which society normally associates negative value judgements, notwithstanding its essential role in keeping together and animating class societies. Common sense often values egocentric sentiments and actions negatively. Not always is this condemnation justified and there are cases in which it is reverted into praise by conventional wisdom itself. In particular there is a group of mental feelings which, although classified in the S-W and N-E box of Table 2.1 and notwithstanding their egocentric orientation, are judged as good and highly conducive to social co-operation. They may take the most varied forms, from vanity to self-respect, from pride to the sense of honour, from dignity to the

50 Individuals, culture and behavioural institutions aspiration to equality, deliverance and citizenship. In general they can be defined by grouping them under two category called the ‘desire for recognition’ and the ‘desire for self-realisation’. The desire for recognition consists of the aspiration to see one’s own merits and identity recognised by others. Everyone is motivated by this desire to a greater or lesser extent. Says Mrs Folie on behalf of Miss Selfloue: Finally in any maner trade of life, what can a man goe about to dooe semely, and with a good grace, either by hym selfe, ore afore others […] if this my damoisell Selfloue be not his aduauncer, whom woorthly I holde therfore in steede of my syster […] and what can be more foly, than one to lyke hym selfe, and stand in his owne conceite? but than againe, what thyng can be propre, or becommyng, or well dooen, in case the doer of the same mislyketh him selfe in it? So that take awaie this saulce of Selflikyng, which is euin the verie relesse of mans life and doynges, and by and by ye shal see the Oratour cold in his mattier, the Musicien mislyked with all his discant, the Plaier hissed out of the place, the Poete and his muses laught to skorne, the Peincter and his art naught set by, the Phisicien for all his medicines walke an hungred. […] Here nowe I recke not much, to passe ouer vntouched, how no maner acte, or noble deede was euer attempted, nor any arte or science inuented, other, than of whiche I might fully be holden first author. (Erasmus, 1965: 29–30) The desire for recognition can take two different forms, which can be accounted for by recalling a conceptualisation proposed by Fukuyama and O’Neill, namely the distinction between ‘merits ‘ and ‘standing’. Recognition of merits refers to individual talents and abilities and is important in knowledge and culture production, whilst that of standing refers to social position and mainly affects political and social struggle. The desire to see one’s merits recognised must not be confused with utilitarian egoism. Smith’s butcher offers a good meal in exchange for money or another good which is of use to him. Erasmus’ poet, on the contrary, publishes a poem not so much to make profits as to see his creativity recognised. The desire for merits recognition prompts social co-operation more than utilitarian angst because in the presence of asymmetric information, the latter kind of sentiment might generate opportunistic behaviour, whilst the former normally induces spontaneous revelation of personal information. The scientist who elaborates a new theorem tries to publicise his knowledge as soon as possible. The profitseeking entrepreneur who discovers a new production method tries instead to keep it secret for as long as he can. In Chapter 5 I show some consequences that the two different kinds of egocentrism may determine in the sphere of knowledge production. It is important to note that the desire for merits recognition is a sentiment that predisposes people to creative activity, that is, to innovative moods and knowledge production.

Individuals, culture and behavioural institutions 51 As to standing, it gives rise to two contrasting sentiments which could be called the ‘desire for status’ – the wish to be recognised by others as superior to them – and the ‘desire for deliverance’ – the wish to be recognised as their equal. Now, the desire for status, as in the case of merits recognition, should not be confused with utilitarian egoism, for status positions, although usually associated with income bracket positions, may give satisfaction independently of the income they might bring about. Nor should it be mistaken for the lust for power. First, power seeking may account for status seeking but cannot be reduced simply to that, for power gives satisfaction in itself, at least to some persons. Second, although the desire for status is usually associated with the aspiration to power, that for deliverance is instead associated with the aspiration to freedom. The lust for power and social prestige and the aspiration to deliverance play a fundamental role in social conflict. This is shown in Chapter 4, where I argue that, since power and status are positional goods, the struggle over their distribution is ineluctable in a capitalist society – a struggle which is one of the motors of social and economic change. Finally, the ‘desire for self-realisation’ is quite a different thing. It is the need to express one’s personality in one’s own work, to see one’s activity as an act of creation, to discover the manifestation of one’s talents in the products of labour actions. Recognition by others is not necessary to satisfy this desire, for satisfaction comes from the reflection of the creator’s genius in the act and product of his creation. In Chapter 5 I argue that the desires for self-realisation and merits recognition provide the most important ‘non-economic’ motivations to inventive activity and can help to overcome some of the limitations that capitalist appropriation poses to the efficient use and production of knowledge. The scheme of rational choice Traditional neoclassical theory accounts for human action with a model known as the ‘scheme of rational choice’. Individuals face constraints defining the opportunities within which they can make their choices and are endowed with preferences or tastes defining the possible goals of choices. Constraints may be physical, such as natural endowments and the availability of goods; institutional, such as customs and laws; economic, such as prices and incomes. Within their opportunity sets individuals make the choices suggested by their preferences. If the choices are rational they produce outcomes that yield the maximum utility compatible with individual possibilities. There is no need to dwell on all the criticism that has been raised against this theory. It is, however, worthwhile recalling some of the most important reasons for dissatisfaction. First, individuals must be endowed with unbounded rationality, or, to be precise, extensive computational ability, something which even today’s most advanced computers do not possess. Second, the individuals must be endowed with a complete knowledge of opportunities or their probabilities; but the world is

52 Individuals, culture and behavioural institutions too varied and complex for this hypothesis to be reasonable, and experience shows that many economic choices are made in conditions of radical uncertainty. Third, individuals must have an accurate knowledge of their tastes and must be able to order their preferences by complying with some criteria of logical consistency, such as completeness, transitivity, reflexivity, convexity; but such rigorous individuals are rare to come by. Completeness, in particular, is not so obvious as it might seem at first sight. Maybe you are unable to make a consistent choice between buying a CD-player and a freezer or buying a new washing machine and a television set. This kind of failure could well make the very scheme of choice fail, even if all the other assumptions are accepted. Incompleteness can render choice undetermined. It could be caused by cognitive limits of the deciding subject or by qualitative differences of the object of decisions. But often there is a deeper and more fundamental reason at the basis of the inability to choose: that pleasure is a multidimensional experience. Some critics focus on the observation that different actions can provide different kinds of satisfaction. Can you compare the intensity of pleasure you obtain from a poem with that you obtain from a beefsteak, or decide to remain indifferent in the choice between a marginal poem and a marginal beefsteak? Even worse, satisfaction is usually brought about by the consumption of composite goods and consists in reality of a combination of different kinds of pleasure. How do you compare, at the margin, the pleasure caused by reading a poem while listening to an overture with that produced by consuming a beefsteak and a glass of wine? Supporters try to cope with this problem by asking: even if it is difficult for an external observer to aggregate the different kinds of emotional satisfaction experienced by an individual, why should it be difficult for the individual himself? Any of us seems to be able to make that aggregation in his personal fruition of pleasure. However complex and multifarious are our emotions, we perceive them as a unitary experience. As Pareto (1989: 348) says, ‘observation shows that people compare pleasures every day, that they judge that two pleasures are equal’. It seems that we are able continually to transform our complex set of heterogeneous emotions into a scalar variable. This process of reduction to unity is a psycho-physiological operation; we do it unconsciously, but nevertheless we do it. So why can scientists not assume we do it? And why can the experience resulting from it not be called ‘satisfaction’ or ‘pleasure’ or even ‘utility’? A more serious problem is whether the experience of satisfaction or of the emotions underlying it are a correct guide to rational choice. Are people always able to foresee the intensity of emotions caused by a certain course of action? If it is only a question of making a decision on the basis of an ex ante estimation of the outcomes of an action, the problem seems to be surmountable, and the expected utility approach does well. Yet a probability distribution is intrinsically incomplete when events are only partially known or unrepeatable. Moreover, when decisions are made not on the grounds of expected emotions, but under the urge of passions, no concept of utility is able to help our understanding. A man who courts a woman he loves (a sentiment) does not make decisions in anticipation of the joy (an emotion) he expects from her response. He acts under the pressure of a mental

Individuals, culture and behavioural institutions 53 feeling independently of the subjective probability of ending up in bliss or distress. His preferences may appear to the economist as lexicographic or incomplete. The loophole of treating sentiment-induced actions as irrational is not a good scientific strategy, since more often than not sentiments contribute to maximise utility in a way that economic calculation cannot do. You can apply cost-benefit analysis in dealing with a prostitute. But with a lover you play with sentiments, and often obtain higher emotional satisfaction without any need for calculation. Nor is it reasonable to try to avoid this difficulty by assuming that, since sentiments are associated with emotions, one can reason as though choosing on the grounds of sentiments is equivalent to choosing on the grounds of their emotional outcomes. Apart from the fact that this kind of association is rarely reducible to a one-to-one mapping, it remains true that, normally, sentiments cannot be chosen. But the most serious difficulty, in my opinion, is another. A given action may be moved simultaneously by several sentiments and motives: you can give a dollar to a beggar out of pity and to earn a place in heaven; are you always in the position to decide whether it is better to spend it on buying a cigar? Whatever the course of action you choose, you might end up by experiencing contrasting emotions. If you give the dollar, you might feel pleased with your altruism and with the step you have taken toward heaven; but, at the same time, you might regret your lost dollar and, if you are a complicated man, you might even feel distressed because the decision to give the dollar to earn a place in heaven in fact turns you away from it; you might also regret the cigar you could have bought with that dollar and, at the same time, rejoice for the health of your lungs; finally you might meditate morbidly on the fact that you have encouraged the practice of begging. The multiplicity of emotions caused by an action is the consequence of a plurality of motives and sentiments. This may be the reflection of a multiple self – a division of personality which goes hand in hand with the multiplicity of roles, positions and relations that an individual undergoes in such a complex individualistic system as a capitalist society. And the multi-motivational aspect of decision-making does not just imply heterogeneity of goals. Often it implies their inconsistency. Is it reasonable to assume that an individual who is unable to define consistently the motivations for his actions is clever enough to order them completely? In fact an individual might remain entrapped in his indecision and decide not to decide or to postpone choice. An overdetermination of choice motivations may give rise to indeterminacy of behaviour. The existence of multiple selves seems a plausible explanation of why ‘a person sometimes prefers one thing, sometimes another, with no apparent reason for the switch’ (Sudgen, 1997: 12). I will return to this conundrum in the third part of the present chapter. Meanwhile let me observe that the solution the great utilitarian economists found to the problem was in reality a trick. Bentham, for instance, who was aware of the multidimensional aspects of pleasure, avoided the difficulty with an act of force, that is, by forcibly reducing the various kinds of emotions to a homogenous magnitude called ‘utility’. This strategy would work well enough if the problem were only one of aggregation of emotions. But it does not work at all when there are contrasting emotions. The ordinalist economists, in a more subtle way, supplemented their refusal of the

54 Individuals, culture and behavioural institutions notion of ‘utility’ with that of the logical inconsistency of behaviour and tried to pass off indecision as indifference. But to be indifferent about two courses of action is not the same thing as being uncertain about which one to choose. Indifference does not block choice and makes behaviour determined anyway. Hesitation, on the contrary, can block it, resulting in indeterminacy. These tricks are laid bare, and the multidimensional aspects of experience are brought to light, as soon as one recognises that human actions are affected by heterogeneous sentiments and moods. And if impulses to action are heterogeneous, why not also the emotions caused by them? This may already be more than enough. However, there is still another reason for perplexity. Psychic experience, in the scheme of choice, appears only at the beginning and the end of the choice process, in the recognition of tastes and the enjoyment of utility. The moment of choice itself is not acknowledged as a psychic experience. It is just a calculus of convenience that in principle can be performed by a computer. The real individual just does not appear in this scheme, since, once tastes and opportunities are identified, action is already precisely determined.2

Bounded rationality and intentional action The scheme of action To understand human action in a world of real people the scheme of rational choice must be abandoned and substituted with one in which the existence of sentiments, moods and emotions is admitted, as well as the influence exerted by society on the formation of human goals, the possibility of mistaking, the limitation of reason and knowledge, the ability to accept less than optimal outcomes, the tendency to conform to habitual behaviour, the capacity to adopt imitative strategies, and innovative talent. I will call it the ‘scheme of action’. It accounts for both intentional and unintentional choices, rational and non-rational decisions, autonomous and non-autonomous behaviour. In this section the scheme of action will be sketched. In the next three sections it will be used to build a simplified theory of intentional and rational action. Finally, for the sake of clarity and at the cost of some repetition, this theory will be summarised and some of its limitations brought to light. In the rest of the chapter I will consider various social and cultural phenomena which may contribute to make individual action non-autonomous, non-intentional and non-rational. Aspirations3 are the particular aims that move individuals to action. They could be likened to ‘preferences’, from which, however, they differ in a number of aspects. To begin with, they are not necessarily formulated consciously, and many of them could remain unconscious for a long time. Moreover, since they are affected by heterogeneous mental feelings, they may be incommensurable with each other. As a consequence their manifestations need not conform to any canon of logical consistency, nor need they be complete. Individuals may be moved by vague and indefinite aspirations which sometimes do not allow them to make a well determined choice among all possible baskets of goods. Nevertheless the fact that

Individuals, culture and behavioural institutions 55 actions are decided within specific institutional contexts may contribute to make them less undetermined than those taken by a social atom. Furthermore, aspirations are not defined here in a metaphysical way, i.e. they are not assumed to be determined by ‘nature’ and ingrained in the original structure of individual personality. Quite the contrary, they are regarded as being open to the influence of social relations, institutions and culture. People are seldom able to maximise, i.e. to perfectly realise their aspirations, and often they content themselves with partially satisfying outcomes. Normally they define tolerance thresholds which may take the form of reservation prices or premiums. Let me call them ‘reservation thresholds’. They measure the minimum levels of performance required to make the outcomes of action satisfying. They are associated with aspirations and vary with them. Reservation thresholds too may be perceived in an undefined and unwitting way. All decisions produce outcomes, although these are not always evident, as for example when they reduce to permanence in the status quo. The experience of satisfaction or dissatisfaction, and its intensity, derives from the degree of matching between outcomes and aspirations. To be precise, the degree of satisfaction derives from the gap between the outcomes and the reservation thresholds. The effect of this gap, though, is mediated by sentiments, for aspirations are influenced by sentiments and not all of them are reducible to hedonist egoism. And even those classified in the egoism group in Table 2.1 are not always easy to interpret in terms of the pursuit of individual welfare. An individual who feels envy or malice towards another, for example, may feel satisfied if he succeeds in doing better than him or, respectively, if he succeeds in damaging him in some way. In either case the degree of satisfaction depends more on the intensity of the sentiment than on the utility brought about by the action. An action moved by a sentiment of solidarity or love can provoke satisfaction even if it does not cause utility to someone. Different sentiments may produce different emotions and different kinds of satisfaction. The crucial point is that the experience of satisfaction varies, not only among the individuals, but also among the actions and the sentiments of a single individual. This does not cause theoretical problems, however, because we are not interested here in making interpersonal or intra-personal comparisons of satisfaction. What is really important is to know that an individual will continue to conform to a given behaviour so long as this brings him a certain satisfaction, whilst the experience of dissatisfaction will induce him to develop innovative moods. In fact the degree of satisfaction or dissatisfaction affects the formation of moods. Satisfaction encourages conservative moods. If these prevail the individual is induced to persist in his past behaviour and therefore not to modify his aspirations. Habitual conduct is formed in this way. If habits are socially diffused, they are behavioural institutions. People who comply with behavioural institutions normally make routine decisions that are accepted and recognised as valid by others. Sometimes, as already observed, habitual behaviour merely implies that no decision is taken. But it is difficult that an institutionalised conduct makes individuals completely passive and incapable

56 Individuals, culture and behavioural institutions of acting. In fact an institution does not fully determine behaviour; at most it restricts the decision options. Almost always there is still room for some problemsolving activity. Routine decisions normally stem from the solution of small problems that continually arise in the restricted option field left open by institutionalised behaviour. If, instead, dissatisfaction prevails, innovative moods are triggered off, and individuals tend to rethink their past conduct and change the actual situation. Innovative decisions differ from routine decisions, not only because they break habitual behaviour by introducing new and non-institutionalised decision options, but also because they often bring about a change in aspirations. Sometimes innovative processes are just triggered off by the formation of new aspirations. These may modify the reservation thresholds and thus transform into frustrating those outcomes which were satisfying when the old institutions and aspirations held. Innovative decisions are always intentional since the breaking of habitual behaviour cannot be moved by unconscious aspirations and the enacting of the means to reach a known goal cannot be made unwittingly. If innovations are efficacious, i.e. they produce more satisfying outcomes than routine decisions, some individuals will try to imitate the innovator. If imitation too raises the degree of satisfaction, a process of diffusion and consolidation is triggered which will eventually lead to the formation of new institutions. The culture linked to a given system of institutions is a set of socially shared beliefs, and consists of opinions (cognitive beliefs) and values (ethic beliefs).4 Culture and institutions play an important role in moulding human behaviour. They widely affect the formation of aspirations and reservation thresholds. But they also affect the formation of sentiments, on which both the degree of satisfaction and the aspirations themselves depend. Finally, culture and institutions affect the moods on the ground of which people decide whether to stick to their habitual behaviour or make innovative decisions. Behavioural institutions and the rationality of decisions An action is rational if it provides the means to reach a goal. It is intentional if there is a causal link between aspirations and decisions. There are cases in which reason fails, even with an intentional action. And this can occur either because of error or because of some contradiction of the personality, as will be shown in the next part of this chapter. The rationality of decisions is a problem for social analysis if it is assumed that individuals must cope with a substantial incompleteness of knowledge, about both the means and the goals, and an equally substantial limitation of computational ability. Individuals endowed with incomplete knowledge and bounded rationality must resort to simplification strategies to make decisions.5 To use a mathematical metaphor, the problem of decision-making can be portrayed as that of the solution of a system of equations where the unknowns represent the decision variables and the parameters embody the basic factors affecting decisions, i.e. beliefs, aspirations and environmental data. Incomplete knowledge means that the individual does

Individuals, culture and behavioural institutions 57 not know all the parameters and forms of equations. Bounded rationality means that the individual does not possess the computational ability or the time required to determine the solution. The latter problem is the most important. Even if all the information were available, the dimension and complexity of a decision problem could be too great to make it workable. A simplification strategy consists in ignoring some variables, or in treating them as parameters by making simplifying assumptions on their values. In this way a determined solution can be found. However, many determined subsystems can be obtained from an underdetermined one, depending on which variables are treated as parameters. The field of decision option is narrowed but it may still be very wide. The problem is now reduced to that of which variables must be ignored or treated as parameters and which values must be assigned to them. This kind of decision can only be arbitrary. Thus it might seem that simplification strategies are of little help to the individual who must make a decision anyway. It is here that culture and institutions, especially behavioural institutions, come to the fore.6 They help people to pursue their simplification strategies by reducing indeterminacy in the decision-making problem. Individuals, in coping with a decision problem, as a first move, tend to take on a conservative attitude, to stick to decision options which are socially accepted and well established, i.e. to comply with behavioural institutions. This is institutional compliance. It is not necessary to assume that the attitude of conforming to institutions is fully cognisant. Nevertheless, in this way decision options are only restricted, maybe heavily restricted, not completely determined. Institutions contribute to reduce the complexity of decision-making, not to thoroughly do away with it. In fact they cannot prescribe the conduct that must be held in all eventualities. Thus the individual is always left with some problems to solve, albeit simplified. It is necessary here to specify a rationality criterion. In their problem-solving activity, in fact, individuals have to decide on the means by which to pursue their goals. The basic assumption is that decisions aim at improving the situation, i.e. at increasing, or at least not decreasing,7 the degree of individual satisfaction. I will call this the ‘criterion of conditional improvement’. ‘Conditional’, because decisions are conditioned and limited by institutions. ‘Improvement’, and not maximisation, because individuals are normally unable to apply maximisation algorithms. People who adopt the improvement criterion will be induced not to abandon an institution so long as they do not observe a systematic worsening of the situation. Otherwise they will have to forsake the old simplification strategies and search for new ones. For this to be the case, the worsening must be considered systematic, it must not be imputable to error. If it is ascribed to compliance to an old institution, then that institution may be abandoned. As already mentioned, individuals who conform to a given institution maintain a behaviour of habitual adaptation which may be incognisant, but do so only for as long as the institution works. However, when this kind of behaviour produces bad outcomes, people will be induced to take on a more critical and mindful attitude, they will tend to abandon their old habits and begin to search for better simplification strategies.

58 Individuals, culture and behavioural institutions Institutional revision implies alertness. It is an action for which the decision cannot be left to chance and to the unconscious. Now, it is precisely this possibility, the possibility of knowingly abandoning the old institutions, that gives a fundamentally rational content to the whole decision-making practice, included that governed by institutional compliance itself. The possibility of abandoning a conduct of habitual adaptation when institutions no longer work, implies that the very adaptation response is rational behaviour and that existing institutions work as a social context of individual rationality. People who allow themselves to be guided by institutions are not necessarily stupid if institutions work at least partly in their interest. This kind of rationality is entailed by a mental reserve by virtue of which an individual is always ready to abandon an old institution if it does not function well. The precautionary nature of institutionalised behaviour A problem with the hypothesis of institutional revision is that of defining the psychological conditions that trigger the decision to change. Institutionalised behaviour is a form of precautionary behaviour. To cope with uncertainty and complexity people tend to build safety reserves. These help people to protect themselves from surprises caused by shocks and errors. The term ‘reserves’ is used here with the twofold meaning of ‘goods stored for future uses’ and ‘limiting conditions’. Any decision to build up inventories of goods to hedge against future oscillations of prices or incomes is a precautionary decision. The same applies to a decision to conform to a behavioural institution which constrains the decision options. And the difference between the two cases is not so marked as might appear at first sight. In fact the decision to build up inventories of goods is just a determination to follow a behavioural rule that constrains decision prospects, in particular the prospect of investing wealth in alternative outlets. When the rule is socially diffused it is a behavioural institution. Precautionary behaviour is normally associated with the definition of tolerance or reservation thresholds. These mark the limits for triggering behavioural revision. As already observed, they often take the form of reservation prices or premiums, i.e. ‘acceptation’ values for the outcomes of usual behaviour. In the case of inventory building, for example, individuals seek some protection from the possibility of the actual values of goods turning out to be worse than expected or desired, or, in other words, from the necessity of continually revising decisions. Reservation thresholds prevent that possibility or necessity at least in part. They define a range of tolerance for the gap between actual and expected values. Inventories act as buffers against decision errors and unexpected events, thus enabling individuals to continue operating without frustration even when aspirations are partially disappointed. The institutional nature of precautionary behaviour can be easily understood by reflecting on the fact that the fixing of reservation thresholds is accomplished through compliance to social conventions, habits and beliefs. These thresholds depend heavily on expectations and aspirations which are widely affected by

Individuals, culture and behavioural institutions 59 psycho-social processes and the action of collective actors. A liquidity premium, for instance, can be interpreted as a reservation rate of return which enables people to avoid modifying their liquid balances so long as the gap between the long-term and the short-term rates of interest is lower than the premium itself. It will differ among individuals, but not so widely, for it depends on the expectations on collective moods, on the degree of bearishness and the instability of markets, on monetary policy and the expectations about their changes and several other variables, all of which are socially determined and highly conventional. Routines also belong to this type of behaviour, since they are consolidated rules of conduct adopted by social groups to avoid the necessity of continually revising decisions. Even if individuals are not always conscious of them, norms and standards defining minimum acceptable levels of performance are frequently associated to routines. If routines are socially prescribed, they are behavioural institutions. When the rules are dictated within an organisation, they enforce hetero-directed conduct. In this case people are governed by regulations and authority. Their inclination to act in observance with rules or authority in fact protects individuals from the uncertainty linked to the decision to disobey or to make free choices. Hetero-directed activities are always associated with the fixing of standards of performance, like minimum rhythms or numbers of processed pieces, which mark the threshold below which an action is considered grievous or shameful. Learning, innovation and imitation The ability of individuals to revise their behaviour of institutional compliance plays two roles in the theory of social action developed here: it accounts for rationality of actions and explains institutional evolution. It is precisely this ability that makes institutions work as a social context of individual rationality. If institutions evolve under the drive of revision processes governed by the urge to do better, then the institutional context changes in a rational way: the institutions which serve individual goals well are consolidated, whilst those which produce systematic dissatisfaction are abandoned. New institutions are generated by innovation and this can be accounted for by the desire to improve. Within this context, institutional compliance can be interpreted as a rational means to the pursuit of a goal, even when individuals are not perfectly aware of their aspirations and reservation thresholds. This would be an implicitly intentional and rational behaviour. If such thresholds exist at least unconsciously, dissatisfaction can help bring them to light and trigger off institutional revision. Innovation is defined here in a very general sense: it concerns not just technical change in production activity, but any kind of social action; it consists in the activation of a new kind of conduct. By definition, an innovation is a case of noninstitutionalised behaviour, for it implies actions which were not socially known beforehand. Technological innovation is a special case applying to labour routines. Innovation can be stimulated by environmental or social changes which induce individuals to modify their actions or their aspirations. External change can affect

60 Individuals, culture and behavioural institutions the means, for example by making them unsuitable for the pursuit of individual ends; but it can also affect the aspirations, by making them unattainable with the available means. In both cases it can produce dissatisfaction, so that people are stimulated to revise behaviour. This, however, is not sufficient to give rise to an innovation. Innovation presupposes invention, i.e. the discovery of a new form of action. Invention is generated in the problem-solving activity involved in institutional compliance, and is strongly conditioned by behavioural institutions. Problemsolving consists in the search for solutions to the small problems arising from the incompleteness of institutions. As I have already observed, an institution does not fully determine human behaviour, but merely restricts the field of decision-making options. People act in this field through a process of learning by trial and error. Whenever a problem arises they search for a solution. They do not necessarily always find the best solution, but learn from experience. They observe the outcomes of their actions and, if these are successful, they repeat them. If, on the contrary, they observe a failure, they try new solutions. Success leads to consolidation of institutionalised behaviour and dampens the stimulus to try new solutions. Environmental and social change, however, always generate new problems, so that problem-solving activity is incessant and the production of new trials never dies out. Normally new solutions do not cause institutional change, for not all new actions are incompatible with those allowed by the decision restrictions imposed by existing institutions. When, however, they produce incompatible actions, then they are inventions.8 Only systematic dissatisfaction induces individuals to abandon the old institutions. Thus an invention emerging from problem-solving activity may remain inert for a long time when institutionalised behaviour produces satisfaction. When institutionalised behaviour causes dissatisfaction and no inventions are available, people may revise behaviour by complying with another institution, or may continue to act by putting up with dissatisfaction while waiting for better times, but no real innovation is possible. The introduction of an innovation implies an individual’s abandonment of an old institution. Yet in itself it is not an institutional innovation. The new kind of action is non-institutionalised behaviour so long as it remains confined to a few individuals. An innovation leads to the birth of a new institution when it pervades society to the point of becoming a socially accepted habit. This transformation is ensured by imitation, which is a kind of learning process based on the observation of the outcomes of actions. Successful actions are repeated, unsuccessful ones are disregarded. The difference to learning proper is that imitation is based in the observation of other people’s successes. When an innovation produces better outcomes than institutionalised behaviour, individuals are induced to imitate the innovators. This may occur because the outcomes obtained by the innovator stimulate a change in the aspirations of other individuals, thus generating dissatisfaction if the old means are inadequate to serve the new goals. But it may occur even when aspirations do not change, if the

Individuals, culture and behavioural institutions 61 innovation enables the individuals to better fulfil the given aspirations. With the passing of time, aspirations tend to change anyway following an innovation, for new forms of behaviour render the aspirations associated with the old institutions obsolete. Moreover an invention can also bring about a change in beliefs and values. Therefore an institutional innovation is often a cultural innovation too. As imitation involves more and more people, the new kind of behaviour consolidates and extends socially. When it becomes habitual in a consistent number of individuals, a new institution is born. Institutional change occurs through an evolutionary process. With the passing of time, ineffective institutions, i.e. those which do not help individuals to pursue their aspirations, are abandoned and substituted by other institutions. Aspirations and beliefs also change and overall, being institutional revision governed by the drive to do better, change implies the improvement of life conditions. This is a partial theory of institutional evolution, that is, a theory accounting for only part of the evolutionary process, i.e. that governed by rational decisions. So long as only rational behaviour is considered, the role played in evolution by individual decisions is at the centre of investigation. Institutions change because the individuals who innovate and imitate are moved by an effort to realise their aspirations. In a certain sense one can say institutions are selected by people, not by nature. And since in this kind of selection intentions and rational decisions play an important role, artificial selection is a better metaphor than natural selection. A problem linked to this is whether evolution implies the selection of the fittest or, in this case, the selection of the best institutions. How do you define ‘the fittest’ or ‘the best institutions’ and, if you want to avoid a tautology in defining the fittest, can you find out a principle regulating the evolutionary process, a principle that does not boil down to maximising behaviour and Pareto efficiency? The problem had already been worked out when individual rationality was defined in terms of a behaviour governed by the desire to improve. Boundedly rational individuals are constantly looking for better ways of doing things. If they comply with the given institutions, they try anyway to improve their capacity to satisfy their aspirations, although within the restricted range of options allowed by the institutions themselves. And when the environment and the institutions induce dissatisfaction they tend to abandon habitual behaviour and to innovate or imitate the innovators. Thus evolution is governed by a principle which could be called the ‘principle of improvement’. People seldom succeed in finding optimal solutions to their problems, but often, when they decide to change their conduct, they do it because they have found better solutions, while, when they stick to routine and habitual behaviour, they do so because this enables them at least not to worsen their situation. Thus institutional evolution is a process which does not imply Pareto efficiency, either at a given time, or in the long run. Not at a given time, because individuals do not maximise within the constraints posed by the given institutions. Not in the long run, because not even innovation and imitation, and therefore institutional change, are ruled by the principle of maximisation.

62 Individuals, culture and behavioural institutions Intentional and rational action: a synthesis and some criticisms In the preceding three sections I have set out a theory of intentional action which is still very abstract. Let me summarise it. The basic assumption is that individual decisions are conditioned by a fundamental incompleteness of the information set and a substantial limitation in computational ability. Institutions help to overcome these problems by restricting the field of decision-making options. People act as if their behaviour were governed by three decision criteria. The first is the criterion of institutional compliance, and consists in acquiescing to the constraints posed by institutions, i.e. in following the habitual practices induced by them. In this way decision-making is simplified. Such conduct can also be interpreted as a form of precautionary behaviour, since compliance with socially consolidated habits and practices provides some hedge against surprises, uncertainty and error. The second criterion is called conditional improvement. Institutions are behaviourally incomplete, in the sense that they cannot fully determine human decisions. There is always some room for discretion within the constraints posed by an institution, so that individuals are compelled anyway to carry out some problemsolving activity. In this activity they try to learn through trial and error. Since they are unable to maximise, they often content themselves with satisfying solutions. Reservation thresholds are normally associated with aspirations and represent minimum levels of acceptation for the outcomes of decisions. The third criterion is that of institutional revision. When compliance with an institution produces systematic dissatisfaction, people can not ascribe this outcome to chance or errors in the application of the second criterion. Responsibility must be attributed to the first criterion, that is, to the institutions complied with. As a consequence, there is pressure to abandon the ineffective institutions, and people tend to turn towards new forms of action. These may consist in compliance with another institution, the introduction of an innovation or the imitation of an innovator. Institutions evolve under the individual drive to satisfy their aspirations. The ineffective ones are abandoned when they produce dissatisfaction, and are substituted with new institutions through a process of innovation and diffusion by imitation. Institutional evolution is governed by the ‘principle of improvement’: the institutions enabling people to improve their situation survive. If a purposeful action is defined as one moved by the intention to realise aspirations, and a rational action as one consisting in the adjustment of means to goals, then this model can account for purposeful and rational action. In the case of routine behaviour purposefulness and rationality may be implicit. People may not be fully aware of their aspirations or the congruity of the means to the goals. However, the possibility of institutional revision implies that individuals maintain a sort of mental reserve towards customary practices – a mental reserve that could produce an active and cognisant choice when the given institutions are recognised as ineffective. The account of rational action set out in the preceding sections provides not just an abstract and simple theory of action. In reality this theory is also quite partial. In fact it concerns only intentional and rational behaviour. And there is a wide range of conducts which are not reducible to the choice of the means to

Individuals, culture and behavioural institutions 63 reach some goals, and in which people are not even implicitly moved by aspirations to realise. The formation of aspirations capable of motivating actions implies a certain freedom of choice and the existence of individuals who have some autonomy in defining the sense to give to their conduct. These conditions do not always hold. For instance, they do not hold for people who undergo a power relation. When individuals are not autonomous subjects of action their activity simply cannot be interpreted as action, and cannot be explained as intentional and rational. Thus the theory needs to be complemented with an elucidation of the conditions that compel some individuals and social groups to act in the absence of free choice. When I observed that institutions serve individual ends, or that behavioural innovations lead to the formation of new and more effective institutions, I did not pretend that every institution serves all individuals in the same way, nor that the diffusion and consolidation of a certain form of behaviour implies the compliance of all the individuals. Many institutions are forceful only locally, and involve the behaviour of only some groups of people. Consider, for instance, the productive routines operating only in a factory; or the habits contributing to the definition of a group or class style, which apply only to some social aggregates. Moreover there are institutions that govern the behaviour of individuals belonging to one class but are determined by the choices of individuals who belong to another. Productive routines, again, are a meaningful example. But there are many other cases, from the choices induced by advertising to those influenced by adverse ideologies. When analysis applies to the class differences which affect the distribution of benefits from institutions, it is necessary to qualify also the idea that the evolutionary process of institutional change is governed by the principle of improvement. Obviously this idea refers to people who enjoy the advantages provided by institutions. But, if these advantages originate from the disadvantages caused to other people, institutional change can be interpreted, from the latter’s point of view, as a process of ‘involution’. The analysis of institutional change must take account of the fate, not only of the social groups who come out as winners in the struggle for existence and dominance, but also of the losers. Nor can one overlook the possibility that the today’s losers could well be tomorrow’s winners, with the theoretical consequence that a certain change could be considered an improvement or a deterioration depending on the time period considered, and the class point of view adopted to interpret it. Finally, the theory of institutional change expounded so far seems to take for granted that evolution develops continuously, through micro-social variability, imitation and diffusion processes. But even biological research has now accepted the idea that many forms of change are discontinuous and macroscopic. In social sciences, at any rate, there is a particular reason accounting for macroscopic change, namely, the fact that people tend to act in groups or coalitions in pursuing their interests. Furthermore, there is a decisive reason for institutional change to take place in catastrophic jumps, namely, the fact that institutions are usually structured in system form, so that it is not possible to change some of them without simultaneously changing many others.

64 Individuals, culture and behavioural institutions All these limitations constrain the theory, for the time being, just because it has been formulated in an abstract, simplified and partial way. Most of them will be tackled and removed in the following chapters, and some in the present one. Meanwhile I shall say something on non-intentional actions, i.e. those which are not moved by individual aspirations. These actions cannot be interpreted as attempts to provide the means to pursue some ends, and cannot be considered either rational or irrational. Let me define them ‘a-rational’. In this case one must assume that some form of social pressure, like coercion, manipulation, addiction, or conformism, contributes to deprive people of some personal autonomy and compel them to act in an unintentional way.9 Culture, sentiments and moods can heavily affect individual behaviour, to the point that people might be unable to appraise the benefits of their actions. Sometimes it could even happen that no individual benefit is at stake.

The multiple self and false consciousness The influence of culture and institutions on the formation of aspirations and mental feelings As already observed, institutions succeed in conditioning behaviour, not only by restricting decisions, but also by moulding aspirations. After all, decisions concern the means to pursuing goals. If the available means are constrained, then so also are the goals. Indeed, there is a fairly strict relationship between the institutions of a given society and the aspirations that can be satisfied by complying with them. But it is not necessarily a one-to-one relationship, for a certain institution can be used to satisfy a range of different aspirations. Just consider how many ideals or needs are compatible with the observance of a certain religion: asceticism, love, self-respect, repression, sublimation, self-discipline, conformism and so on. Furthermore, it is even possible to satisfy a given aspiration by conforming to different institutions. Solidarity sentiments, for instance, can be expressed by joining a church, a party, a charitable association. Nevertheless it is still true that any institution, insofar as it restricts the range of possible actions, constrains aspirations too. Beliefs, moreover, are always associated with any given institution. Religious practices, for instance, are never disjoined from doctrines, dogmas and moral rules which give them sense. People define the meaning of their actions through beliefs. But they also derive instruments of understanding and valuation from them to define their aspirations. Thus the limitation of beliefs brought about by culture associated with institutions also implies the limitation of aspirations. In many cases institutions embody knowledge functions without the individuals being fully cognisant of them. And, to the extent that tacit knowledge refers to the definition of aspirations, the influence of institutions on their formation is even stronger than that exerted through behavioural constraints. Perhaps the most pervasive cultural system of a society is language. A language is not just an instrument of communication. At a deeper level it is also an instrument

Individuals, culture and behavioural institutions 65 of knowledge, a system of concepts capable of affecting the way of thinking. It makes thought possible by disciplining and channelling it, but also by limiting and selecting the universe of discourse and knowable truth. It is at the same time an institutional and cultural system. Institutional, in that it governs habitual forms of social communication and interaction. Cultural, in that it governs perception of the world and elaboration of knowledge, as well as the determination of the values through which individuals relate to the world. The pressure of an institutional and cultural setting on individuals would not be understandable if there were not a human predisposition to accept it. This predisposition is linked to the existence of two basic attitudes in the drive to social action. In relating themselves to each other, individuals are moved by impulses to self-assertion and desires for recognition – two contrasting but complementary impulses. Self-assertion, intended as an egoist utilitarian drive, is conceived as an imposition of the self on others, which, however, cannot normally be done without taking account of the others’ needs. It requires social consent and exchange. Social recognition, on the other hand, means that the merits and standing of an individual are acknowledged by the others and requires an even stronger social intercourse to be elicited. Both self-assertion and recognition take place through communication games. Every social act, even economic decision, even consumption choice, generally takes the twofold meaning of an action for the realisation of individual aspirations and a message of participation in a communication game. How can the tendency of teenagers to wear jeans uniforms or bank clerks to wear jackets and ties otherwise be explained? Jeans and jackets do not serve just a practical purpose, but also convey social messages which are determined by highly conventional rules. The rules of social communication consist of widespread habits embodying socially accepted values. They are institutionally and culturally determined forms of action. Here is another channel through which social pressure is exerted on the formation of aspirations. This has a less incisive influence than that exerted by language proper, since non-linguistic forms of communication are quite sensitive to social variability. However, it is also less generic. Each group, organisation or social class has its own social communication conventions; each particular social aggregate is able to impose specific rules on the individuals who want to be part of it. And people are continually induced to adapt their aspirations to the exigency of communicating with others and participating in social life. The individual needs for self-assertion and recognition through social communication can sometimes become so strong as to give rise to a sort of organisational involvement, a kind of proneness to participation or urge to belonging which is much more pressing than a simple desire to take part in the communication game. Organisational involvement may occur when a person is part of an organisation regulated by a strong internal system of institutions, a church, a party, a company. A member of an organisation may be motivated by personal aspirations and interests which do not necessarily coincide with those of the organisation itself. A normal employee does not work in a firm with the intention of contributing to the accumulation of its capital. Nor does a member of a party have to share the strategy of its leading group. Sometimes, however, the members of an organisation tend to

66 Individuals, culture and behavioural institutions assimilate its objectives, or to adapt their own aspirations to make them coincide with the organisation goals. This is organisational involvement. It may easily sway the lives of the leaders and, usually, of the members who obtain substantial benefits from the achievements of the organisation’s goals. But it may also sway subordinate members, especially through the pressure of a strong ideology, and provided the individuals have a predisposition to self-coercion. Organisational involvement may be so strong that, for example, a manager may decide to fight to save ‘his’ company from bankruptcy rather than accept a better opportunity offered by another company. What I have said so far refers to the influence exerted by culture and institutions on the formation of aspirations. But a similar argument could be expounded as regards the formation of sentiments and moods. Indeed, since mental feelings are often formed unwittingly, even more unwittingly than aspirations, they are highly sensitive to cultural and institutional influence. This might be a general influence, such as one exerted on an entire society or a whole nation. But it might also be quite specific. There are cultures and institutions relating to particular organisations, groups or social classes; and there are sentiments and moods as well which are typical of particular social aggregates. In fact the members’ mental feelings tend to form together, to affect each other through processes of collective formation in which the cultural and institutional setting play an essential role. The multiple self and the freedom of choice The influence of culture and institutions on the formation of aspirations and mental feelings poses a problem to social theory: to what extent are people really free in their actions? The risk of many institutionalist theories is to substitute the neoclassical automaton-atom with a subject which has a place in society like a cog in a machine. But both the Walrasian economic man and the Durkheimian social man are fictitious subjects. They carry to extremes two real dimensions of human conduct both of which are always present in actual choices and decisions. Individuals manage their lives through incessant struggle in a social environment. Society is the battlefield on which people try to secure their survival. It produces attrition and resistance to the achievement of human aspiration and, at the same time, it exalts labour productivity and multiplies human opportunities. It enables people to form aspirations which are inconceivable in the world of Robinson Crusoe. Beyond a certain stage of development social life becomes a primary good. Men cannot do anything outside society and their very existence acquires sense only as a social fact. To survive, individuals must centre their lives on themselves, but, to survive in society, they must orient their actions toward social construction. This is a basic reason why all individuals make the encounter with a divided or multiple self; all people are at the same time egoists and altruists. The experience of a multiple self might be distressing. Yet the division of the self is a fundamental condition of the freedom to choose. A compact self, completely absorbed in the endeavour to

Individuals, culture and behavioural institutions 67 maximise individual utility, can only belong to a subject who is as free as a beast in the jungle. One who is equally compact, but in the adoration of god-society, is a mere shell of a man, a body without a soul who lives on heteronomous impulses. Real freedom develops on the contradictions of human nature. One could object that, if sentiments and aspirations are moulded by culture and institutions, the prospects of choice are only fictitious. This objection is only partially valid because, if it is true that both egoistic and altruistic sentiments undergo social influence, it is also true that you can often choose between the two. Obviously the range of choices is always restricted by social possibilities. Thus, one should speak of relative freedom. Since society is made up of heterogeneous individuals and social aggregates, better still, of individuals and social aggregates with different and even antagonistic interests, values and beliefs, people undergo diverse and heterogeneous social pressures. Institutional plurality contributes in a decisive way to widen the range of choices. If diverse institutional influences affect the formation of human character, the division of the self opens up a more complex and vast range of choice than that polarised by orientations toward the ego or others. The self is multiple, not only because individuals are egoists and altruists, but also and principally because they participate in various social roles, are subjected to diverse and contrasting institutional influences and undergo the pressure of many different values and beliefs. To be more precise, every personality has various selves because the individual participates in many roles and social situations. A person may at the same time be a believer, an entrepreneur, a citizen, a parent, an artist. Therefore there will be a religious self, and a greedy, a political, a moral, and a sensuous one. And all of them may claim the sceptre of command over the soul. Every time a decision has to be made, each self could try to assert its will, often independently of the role or social situation involved in the choice. Worse still, institutional plurality also holds within any specific institutional need. Thus, for example, people brought up in the Christian religion know fairly well how to behave to stay faithful to tradition, but they also know that at any time they can choose to become a Buddhist, a Muslim or an atheist. Entrepreneurs can fix prices by following a well defined routine, but at any time they can choose to change the mark up, or to break a cartel, or to imitate aggressive competitors. Thus, the field of choice widens, not only because it is possible to combine in different ways the inducement coming from the various selves that emerge from the many social situations, but also because alternative institutions exist within any single situation. Furthermore the field of choice, besides being quite wide, is open, since innovation too is a possibility of choice. In fact an individual may often decide to follow a non-institutionalised behaviour. It is as if – to quote Bismarck – there were a republic of homunculi in each person. And a republic is based on freedom. Thus, it is on the division of the self that the freedom of choice is founded. People cannot always rely on external agents

68 Individuals, culture and behavioural institutions x

c b

P1" P1' P2" P2' y

Altruist sentiments Non-profit organisation Church

Family

Conservative inclinations

Innovative inclinations

University

Capitalist firm 1

Capitalist firm 2

Egoist sentiments

Figure 2.1

to solve their inner conflicts, and even when they simplify their lives by conforming to institutions and organisations, in reality they often make a choice, just because there are different institutions and organisations. Figure 2.1 illustrates the argument. The lower diagram10 represents a mental feelings space. It is a sort of a map of the social influences undergone by an individual’s soul, and records the psychological attitudes, or rather the sets of moods and sentiments, encouraged or moulded by some of the social organisations to which the individual belongs. Arrows represent some mappings from the mental feeling space into an aspirations space. The latter is portrayed in the upper diagram, where one can see the choice angst of an individual endowed with a double personality or, in other words, with two aspiration or preference sets, P1 an P2. Assume this person has to choose between two baskets, c and b, which combine goods x and y in different ways. c is

Individuals, culture and behavioural institutions 69 preferred to b with preferences P1, but b is preferred to c with preferences P2. Choice is undetermined. If the individual were to stick to a budget constraint, he would discover this is always tangent to both a P1 and a P2 curve. He may be uncertain in his choice at any given time. And at any time he will be free to jump from one preference curve to another. The assumption of choice freedom is unavoidable in any theory which considers the individual as a subject of action, but you pay this assumption with renunciation to behavioural determinacy. In neoclassical theory a problem arises as soon as one assumes economic agents have the possibility of choosing among different preference sets. If you refer the assumption of maximising behaviour to a given preference ordering, you can fully determine choices, thus making the individual similar to an automaton. But if you admit that a set of preference orderings exists, you inescapably open the theory to indeterminacy. And you cannot solve the problem by resorting to the idea of metapreference ordering, since either you assume that the latter is given and unique, in which case you go back to the automaton, or you pose the problem of the choice among metapreference sets, in which case you open the abyss of infinite regression. Regression must be halted somewhere and, if individuals are to be considered genuine subjects of choice, you cannot avoid treating them as free subjects. But freedom means that choices are open, are determined by personal judgements and irreducible to the principle of maximisation. A similar problem arises in some institutionalist theories of action. If you believe that individuals conform to institutions to cope with problems of uncertainty and complexity, you cannot assume that each action is governed by a definite and unique institution, otherwise the decision-making subject disappears. Resorting to the postulate of institutional incompleteness contributes to alleviate the gravity of the problem, since it allows individuals to have at least the ability to play problem-solving activities. But the fundamental problem is not solved, since aspirations motivating actions remain trapped by the institutional influence. However, you can resort to the other important postulate, that of institutional plurality. If many alternative institutions are admitted, so that people are able to make choices, you give them back their freedom. But then how will they cope with the unavoidable complexity of choosing among institutions? You cannot get out of the impasse by assuming the existence of metainstitutions, otherwise you reopen that abyss. Thus, you can only accept the idea that at last people must choose by using their own judgement. Sometimes they will make the wrong choices, often they will choose not to choose at all by sticking to tradition, or they will follow an innovator. But the possibility of open, albeit limited, choice must always be available, otherwise you cannot speak of the individual as a social actor. Salvation strategies To cope with the existential difficulties caused by the multiple self, people may resort to various strategies. Four strategies in particular deserve a brief detour into psychology. I will call them: ‘compulsive integration’, ‘reflective integration’,

70 Individuals, culture and behavioural institutions ‘compartmentalisation’ and ‘division’. It will be difficult to find individuals who conform precisely to these personality models. It is easier to find mixed strategies, so that there are numerous psychological types resulting from combinations of these four basic strategies. But here it will suffice to restrict considerations to the essential. With the strategy of integration individuals try to get some order into their interior republic by transforming it into a dictatorship, which they can do by building a hierarchy of selves. One of these is credited with supremacy. Some aspirations, values, sentiments are considered predominant and organised into a consistent core which directs choices and decisions. All the other aspects of personality are re-organised in relation to the dominating core, so as to give vent to them only insofar as they contribute to the pursuit of the core goals, or at least do not obstruct them. Impulses, desires and passions which contrast with the interior hierarchy are removed or repressed. In compulsive integration the dominating self is perceived as an objective and inescapable necessity, out of the reach of free choice. It is often the result of strong social pressure. It might have been imposed, for instance, by the socialisation process, an organisational involvement or a power relationship. People with this kind of personality take behavioural rules as rigid and unchangeable. Life is considered as the fulfilment of a vocation. Personality is aimed at the pursuit of goals perceived as transcending the immediate needs or interests of the individual. Institutional choices, when they occur, take the form of non-choices, as people feel they have no real ability to choose other than the institutions, values and culture in which they are embedded. Here is what could be called a ‘Kantian personality’: ‘You evidently don’t believe in the maxim: Act so that every one of your actions is capable of being made into a universal rule.’ ‘I never heard it before, but it’s rotten nonsense.’ ‘Well, it was Kant who said it.’[…] I take it that conscience is the guardian in the individual of the rules which the community has evolved for its own preservation. It is the policeman in all our hearts, set there to watch that we do not break its laws. It is the spy seated in the central stronghold of the ego. Man’s desire for the approval of his fellows is so strong, his dread of their censure so violent, that he himself has brought his enemy within his gates; and it keeps watch over him, vigilant always in the interests of its master to crush any half-formed desire to break away from the herd. It will force him to place the good of society before his own. It is the very strong link that attaches the individual to the whole. And man, subservient to interests he has persuaded himself are greater than his own, makes himself a slave to his taskmaster. He sits him in a seat of honour. At last, like a courtier fawning on the royal stick that is laid about his shoulders, he prides himself on the sensitiveness of his conscience. (W. S. Maugham, 1949: 60–61)

Individuals, culture and behavioural institutions 71 Reflective integration, is more easily understood by using the metaphor of a personality centred on the conscious self – say – a ‘Nietzschian personality’. In a sense this is the opposite of compulsive integration. Individuals feel free from any moral impositions, and choose institutions, values and culture, even sentiments and moods, on the basis of a reflective and autonomous activity. Goals and aspirations are mindfully determined and, whenever the possibility of a choice arises, it is made on the basis of personal judgements. The dictator of the interior republic, as well as the basic criteria to order it and to motivate actions, are chosen consciously. A person such as this seldom chooses not to choose, and often favours innovative and non-conformist decisions. Strongly integrated personalities are rare. Apart from some religious ascetic or Bohemian artist, it is not so easy to find examples of these two types of individual in modern societies. Perhaps the reason for this is that survival and success require adaptation to a changing environment. The Kantian personality, by virtue of its acceptance of the established institutional system, tends towards strong social integration, but is too rigid to succeed in adapting to change, especially radical change. The Nietszchian personality, on the other hand, is rather prone to change, but resists integration in the herd. A certain amount of interior anarchy, but not too much, seems necessary both for adaptation to change and social integration. In non-integrated personalities many relatively independent selves cohabit in the same individual, who, thus, is continually exposed to a decision conflict. People of this kind may find themselves having to face dissonant situations at different moments of their lives, but often even at the same time, and to perceive as inconsistent the various roles they are called on to impersonate, since they are motivated by heterogeneous aspirations, sentiments and moods. Each behavioural choice may take the form of a contradictory deliberation. Two main types of strategies seem to have been developed to cope with this sort of dissonance, one of which resorts to the compartmentalisation of selves, the other to the acceptance of division. Compartmentalisation attacks the multiplicity of selves by separating the different situations and the different roles which the individual plays and by obeying ‘local’ rules, i.e. rules that hold for the single selves. The greatest effort is made to avoid interior obstructions, so that the left hand does not know what the right hand is doing. The father will not follow the same rules as the husband, nor the benefactor those of the entrepreneur, so that the individual suffers the apparent incongruity of being at the same time a good father and a charitable man while being unfaithful to his wife and exploiting his employees. The case of Dr Jekyll and Mr Hyde springs to mind. But nobody better than Ferdinando Pessoa grasped the essence of this soul: ‘I do not want to remember nor to know myself. There are too many of us if we look at who we are’. Thus let me define this a ‘Pessoan personality’ as a tribute to the bard of ‘unitary multiplicity’. The individuals who accept the division of the self, on the contrary, cannot avoid acknowledging their interior contradictions. In any decision they have to make, all their selves crowd around the bargaining table. All have something to suggest, all pretend consideration. The hedonist who smokes a cigarette must continually face the eudaemonist who aspires to a longer lasting happiness. Interior

72 Individuals, culture and behavioural institutions conflict is permanent and unavoidable. Well, it is not granted that this kind of personality is less useful to adaptation. It depends on the ability of people to bear contradictions. If this ability is not well developed, the risk of anomy and maladjustment is strong. If instead it is well developed, then this kind of individual is apt to social integration and adaptation to change at the same time. An individual such as this could turn out to be a winner in a world where the struggle for existence requires both social participation and egoistic hubris, stability of character and aptitude to innovation. This is the personality of a ‘man without quality’. Here is Robert Musil’s (1978: 20) description: Hanging over his head he felt the threat of a maxim he had often read in art journals: ‘Tell me how you live and I will tell you who you are’. After consulting the journals at length he came to the conclusion that it would be better if he set about creating his own personality and he began to design his own furniture. But hardly had he created a solid and impressive line than it occurred to him it could easily be replaced with one that was more functional and svelte; and as he began to sketch a form in reinforced concrete style, with his vigour stripped of all essential, his mind wondered to the slender and budding forms of a young girl and he began to dream instead of deciding. This, in a field which he had never taken seriously to heart, was the discontinuity of ideas which he recognised so well, that pullulates without a central nucleus, with the incoherency that marks our times and induces strange calculations that jump from pillar to post with no trace of unity. In the end he could think of nothing but unrealisable spaces, revolving rooms, kaleidoscopic furnishing, devices for transporting the soul, and his ideas became increasingly devoid of content. Thus let me define this a ‘Musilian personality’. Among the merits and misdeeds of capitalism we must count the role it played in creating the conditions for the liberation of the individual and the proliferation of the selves. Of course the multiple self existed also in ancient civilisations. Already Plato had developed a doctrine of the division of the soul and of interior conflicts. But it seems that capitalism led to the accomplishment of a long historical process. The community solidarity, which maintained cohesion in many pre-capitalist societies, broke down catastrophically when the individual was posed at the centre of social order; since the principle of an individual subject, unbounded by traditional fetters, personal duty and metaphysical faith, is intrinsic to the life condition of people who are reduced to the summation of their fortunes. When the accumulation of wealth becomes the most important goal of human action, the sense of existence can only reside in the individual’s wishes. The result is an overflow of independent activities, and the passage from the organic order of armies and churches to the chaotic and protean one of markets. Institutions disintegrate and multiply, and with them cultures, ideologies, beliefs and values. It is here that the multiple self in the modern form of interior anarchy is born. If, among the four models of spiritual strategy, compulsive integration performs better in societies based on community solidarity, there is no doubt that

Individuals, culture and behavioural institutions 73 Musilian and Pessoan personalities prevail in the present world. If it is true that the human being is largely moulded by institutions, then it is understandable that a sort of isomorphism links the soul to culture. The structure of the self is a mirror to that of society to the same extent as the latter is produced by human activity. But in capitalism the mirror shatters into pieces, and a thousand reflections return to the social subject the image of his face. A thousand viewpoints can be taken to give sense to life. And since many and contrasting institutions frame the multifarious and contrasting actions of people, then multiple and contrasting is the social pressure on the individual. In this way the free subject is born, for there is no freedom of choice without institutional plurality. And in this way the social subject is born, since there is no individual responsibility without freedom of choice. Only with capitalism can society begin to be considered as the result of the actions of free and responsible individuals. Yet this is only a possibility. Ananke still dominates social relations and the life of any person is still overpowered by the struggle for existence. Indeed, in a world of scarcity in which human desires have been freed from the fetters of faith, man’s struggle for dominance over man is exalted. Thus, the bite of heightened exploitation and oppression is added to that of natural scarcity. And free choice remains a cause of dissatisfaction and frustration for most people. What is the use of a freedom to choose the ends of one’s actions, when the means are out of reach? Then is it not possible that desires and aspirations will be redesigned to adjust to the lack of means? So, many people will be unable to build their personalities in the search for really free subjectivity. Individuals who have to spend half their lives working under the command of another person are condemned to division or compartmentalisation of their selves. At least half their soul is devoted to activities in which they have no free choice. Capitalism tears individuals into many parts, but whilst it offers everybody the possibility or the illusion of reconstructing unity, only to some does it concede the means to do it. Salvation strategies: some examples Consider the situation to be faced by a man who is the president of a big country, and who happens to take a fancy to one of his assistants. This is the typical problem of a politician who wants to change the world but has the evil of the world deep down in his soul: what is to be done? The root of the problem lies in the fact that our president has a double ethic. There is an ethic of duty, which pushes people to commit themselves to change society and to build a just social order; and also an ethic of pleasure and love, one that makes people look for personal happiness here and now. Both ethics are diffused in modern societies and are both approved by most individuals. They are, however, intrinsically contradictory, for the conventional ethic of social commitment induces people to sacrifice themselves for the collective good, whilst the subversive ethic of pleasure induces them to transgress social conventions and to satisfy individual passions. Behaviour affected by these

74 Individuals, culture and behavioural institutions two ethics is governed by two contrasting kinds of sentiments – let me say, for the sake of simplicity, altruistic and egoistic sentiments. The president has a multiple self: he would like to strengthen his social position of leader of the nation and the prestige, the status and the power which are considered necessary instruments to serve the people; but, at the same time, he would like to live a life full of emotional, sexual and sentimental gratification. The president knows that the two personal motivations are in contrast. If he chooses to let himself be dominated by passion, he runs the risk of being discovered by his political enemies, criticised, judged and condemned, with the consequence that he will lose his social position and his ability to act for the good of the people. If, instead, he chooses to remain faithful to his wise wife and to fire or promote his assistant, his political prestige will be preserved but his life will be doomed to emotional impoverishment. Suppose, for the sake of reasoning, that the two choices can be expressed in terms of two goods, ‘prestige’, x, and ‘love’, y, which can be combined in fixed proportions in the ways described in Figure 2.2. The conventional choice, c = c(x, y), can be made according to the intensities represented by line Oc; the subversive choice, b = b(x, y), according to the intensities represented by line Ob. The president’s contrasting preferences are represented by two indifference curves. Choices are limited by several kinds of constraints, physical, psychological, institutional, economic etc. These constraints can be represented by lines which, by convention, can be called ‘budget constraints’. It must be maintained that these constraints do exist, because lines Oc and Ob cannot go to infinity. But nothing else needs to be known, in this exercise. An hypothesis of bounded rationality would prevent any tangency solution. Due to the heterogeneity of the factors affecting budget constraints, these could be conceived as non-linear. Finally the

x

c

n c'

b

b' 0

Figure 2.2

y

Individuals, culture and behavioural institutions 75 prices may be unknown or even non-existent – which an economist could justify by saying that there are no spot and future markets for love and prestige. These are some of the reasons why the budget lines have not been drawn in Figure 2.2.11 The choice made by the president will depend on his personal salvation strategy. If he has a Kantian personality, he will choose c, will adopt a conventional ethic and practice a form of self-compulsion whereby he puts order into his soul by giving the public good high priority in his aspirations. If his sexual and sentimental life is unhappy, he will put up with it. Personal utility cannot prevail over the public welfare towards which he feels the duty to sacrifice himself. This president is an ascetic of politics or a professional revolutionary. To him the answer to the question ‘what is to be done’ is not given by himself; it is given by God’s Providence or by History. A Nietszchian personality too might choose c, but not to comply with a conventional ethic, nor to fulfil the aim of a life conceived as pre-ordered by transcendental duty. He would choose it by a free and cognisant decision. Such a choice may be prompted by a desire for power unhampered by moral restraints, but also by the desire to use power and prestige to rationally pursue a common good. However, a Nietszchian personality might also choose b. Since the goals of this individual are knowingly determined, our president, if he were a Nietszchian, might decide that life has no sense other than the one he himself gives it and therefore that it is better to have a good time even at the cost of risking renounced power and social consideration. While a Kantian personality will steadfastly stick to c, the Nietszchian may be strongly affected by stochastic events and repeatedly change his choices. The first kind of man, if he were a civil servant in Machiavelli’s country, would be a fervent Fascist during the Fascist regime and become a fervent Christian-Democrat during the Christian-Democrat regime, but he would always remain a civil servant faithful to his duty and his wife. The Nietszchian type, most probably, would be an anarchist both before and after the war, but might change his wife and profession several times. More complex are the choices of the other two personalities. Pessoa would be unable to forsake either choice, and would take both c¢ and b¢ . Choice b¢ would be inferior to b because, for instance, in order to minimise the risk of impeachment, the president might decide to have a not particularly enthralling intercourse with his assistant, so as to be able to interrupt it at any time. c¢ , would be inferior to c just because of the greater risk of impeachment in the event of the affair being discovered. In fact the Pessoan type obtains n, an unwittingly mixed choice: a little love from his wife and a little from his mistress, all at the cost of some risk. n is less satisfying than both c and b with either preference orderings. However this is not the most awkward problem for a Pessoan president. The real dilemma is how to sustain the dissonance caused by the attempt to achieve two contradictory aspirations or to comply with two contrasting ethics. Pessoa will cope with this problem by ignoring the contradiction, which may be done by maintaining a certain degree of unawareness while making choices. The same problem would be faced by a Musilian type, who would also choose c¢ and b¢ .

76 Individuals, culture and behavioural institutions He would differ from the Pessoan type mainly in one thing: he would be fully aware of choosing n. The real difference between these two personalities, therefore, concerns the way in which conscience elaborates information. Self-deception devices and false consciousness Note that none of the choices considered so far is fully satisfying. The reason is that the individual entertains two contradictory aspirations, so that it is not possible to achieve either of them in a satisfactory way. Two different kinds of psychological conflict can arise from this situation. The first is an ‘emotional conflict’, and has to be faced by Kantian and Nietszchian individuals. If they choose c, they will be satisfied in their desire for prestige but will remain frustrated in their emotional life. The second kind of conflict is ‘moral conflict’. It is caused by the fact that the individual, whether Pessoan or Musilian, bends simultaneously to two ethics and pursues two incompatible aspirations, attaining a little of both. The two kinds of interior conflict produce cognitive dissonance and will be the more frustrating the higher the awareness of conflict. To cope with this difficulty, people may resort to self-deception devices, psychological mechanisms aimed at deforming the perception of reality, in other words, at building a false consciousness. Five such self-deception devices seem particularly rife: illusion, undervaluation, focusing, ignorance and idealisation. Figure 2.3 illustrates the first four devices. Idealisation will be dealt with in the next section. Illusion consists in overrating the x and y content of, say, c, so as to make it preferred to b whatever the preference ordering complied with. ci, for instance, giving the illusion of obtaining more prestige and love than that actually obtained with c, makes this choice preferred to b both with preferences P1 and with preferences P2. This does away with emotional conflict. Similar, but opposite in sign, is the undervaluation device. In consists in underrating the x and y content of c so as to render b preferred to it with both preference orderings. Focusing is quite different. It boils down to concentrating attention on only one aspect of c and b and forsaking the other. The president, for instance, might try to convince himself that his choices are dominated by only the x component and have no implication on his private life. In this case he will look at cf and choose c, which is certainly superior to b in terms of power and prestige content. Illusion and focusing can be combined so as to forsake one component of a choice and overvalue the other, as shown by point cfi . Another self-deception device consists in throwing a shadow of ignorance on the area containing one of the two choices. In this way the aspiration conflict cannot generate emotional conflict because one of the two choices is unknown. Now, by taking a step backwards, I must recall that frustration may arise even in the absence of multiple self, and only as a consequence of bounded rationality. If people are unable to maximise anything, their achievements normally fall short of their aspirations – and this causes frustration. Figure 2.3 can also be used to illustrate the way people may react to this kind of frustration. Assume that bundles Ob and preferences P2 do not exist, that point

Individuals, culture and behavioural institutions 77 x cfi

ci illusion

focused illusion

cf

c focusing

P 1' undervaluation

cr

b ignorance

P2

P 1" y 0

Figure 2.3

ci represents the individual aspiration, cr the status quo ante or a reservation threshold, and c the achievement of a decision. This individual enjoys at the same time some satisfaction and some dissatisfaction. c – cr measures the degree of satisfaction and ci – c the degree of dissatisfaction – a situation corresponding to what is sometimes called ‘relative frustration’ or ‘relative deprivation’.12 One speaks of ‘social relative deprivation’ when aspirations are formed on the ground of a comparison with other individuals’ achievements. To cope with frustration, in this case, a person can make use of two different self-deception devices, illusion and undervaluation. Illusion helps him to convince himself that the actual achievement was ci, whilst undervaluation helps him to convince himself that the aspiration level was c (in this case undervaluation takes the clear meaning of a ‘sour grapes’ device). In either case frustration is banished from conscience. Illusion, focusing and undervaluation are, normally, irrational practices. Although they serve to reduce the dissonance associated with emotional or moral conflicts, they prevent the individual from making cognisant choices. By deforming information, they may induce people to make choices and take decisions which hinder the improvement of their situation. Ignorance too can be irrational in this sense. It would be irrational, for instance, if it induced a person to forsake choice b, provided this would enable the individual to obtain more satisfaction than choice c in at least one preference ordering. However, rational ignorance is possible. This is the case in which the collecting of information on choice b is so costly that its x and y content, net of the information collection costs, is low enough to make c preferred anyway. The president, for instance, might be dubious both about his assistant’s proneness to become involved in a love affair and about the intensity of pleasure he might get out of it. To decide whether it is worthwhile embarking on the adventure, he should engage in a search process, made up of frequentations, winkings, courting etc. which, if discovered,

78 Individuals, culture and behavioural institutions might have strong negative consequences on the quantity of x-achievements. In this situation he might decide to ignore her smiles and kindnesses altogether. Self-deception devices can be used as reinforcements of the choices induced by salvation strategies. The Kantian type, for example, might strengthen a c choice by applying an illusion device to c or an undervaluation device to b or both. A Pessoan type might try to overcome the moral conflict caused by the choice of c¢ and b¢ in Figure 2.4 by resorting to a double device of focused illusion. He may consider c only for its x content and convince himself that the quantity of x contained in c¢ is equal to that contained in c. The same deformation of reality could be applied to b¢ , so that c¢ and b¢ will be chosen in the belief of choosing cfi and bfi. Doubly focused illusion transforms the two baskets of goods into two unrelated goods so that indifference curves that combine them in different proportions become irrelevant in justifying choices. False consciousness easily takes root in the souls of Kantian and Pessoan individuals. Illusion, undervaluation, focusing and ignorance enable people to reinforce choices by reducing emotional and moral conflicts. The real satisfaction obtained by actual choices might be modest, but frustration would be substantially abated, or at least maintained in a latent state. Moreover these devices may serve to justify attitudes of preventive commitment which, together with frustration reduction, may contribute to institutionalise and stabilise behaviour. Self-deception devices are not normally practised by Nietszchian and Musilian personalities, who rather tend to make cognisant choices. But people who renounce the possibility of resorting to false consciousness are doomed to bear emotional or moral conflict and to live in a state of permanent frustration. The behaviour of these individuals could turn out to be rather unstable. The Nietszchian who chooses c may be under the constant threat of regret moods, so that a slight change in the relative positions of c and b may be sufficient to trigger a catastrophic jump in behaviour. Nietszchian personalities are the most prone to mutation, innovation, revolution. The Musilian individual who has chosen c¢ and b¢ (and thus, actually, n), may have done so because c and b were highly uncertain. The decision to reduce commitment in choices in a way that leaves both roads open may be justified by an attitude of waiting in a situation whereby information on c and b is scarce or is dependent on the occurrence of some future events. Our president, for instance, may decide to invest a modest amount of sentiment in a superficial, clandestine and reversible love affair so long as he is in his first mandate. But then, once reelected, he may decide to change his attitude. A Musilian chap always hides a Nietszchian deep down in his soul, so that an external event which enriches his information set may be sufficient to prompt a substantial behavioural change. But the choice of n may also be interpreted in another way. Under the pressure of dissatisfaction with existing choice options, and therefore with existing institutions and ethical constraints, a Musilian or a Nietszchian individual might decide to introduce an innovation. In this case n is not just a combination of c¢ and b¢ , but a recombination of x and y which produces a novelty, a new bundle of goods. If lines Oc and Ob are constrained from above, any point on line cb in

Individuals, culture and behavioural institutions 79 x

c

c fi

n

c'

b

P3

b' b fi 0

y

Figure 2.4

Figure 2.4 is a potential innovation. This move could be accompanied by the formation of a new aspiration set – a change that could be represented, for instance, with the birth of a new preference curve, P3, passing in n. The new bundle is not preferred to c or b with the old preferences, but it is with the new one. With the passing of time the innovation could spread in the social system and produce a new ethic and a new institution. Alternatively it could be a normative innovation which brings about simultaneously, although perhaps slowly, a behavioural and cultural change. Our president, for instance, might attempt to overcome his frustration by convincing Parliament and his country to legalise bigamy. Institutional and cultural plurality has a twofold, contrasting effect on human behaviour. It widens the choice fields open to individuals, thus enhancing freedom and the possibility of individual autonomy. But it divides and multiplies the selves, thus undermining determinacy of behaviour, disseminating uncertainty, indecision and psychological instability. While the neoclassical homo economicus always exhibits a fully autonomous and determinate behaviour, the real man who faces the relativity and multiplicity of ethics and institutions has to cope with a tradeoff between determinacy and autonomy. If he chooses stability and safety, as Kantian and Pessoan individuals normally do, he will stick to tradition and suppress frustration by triggering some self-deception devices. In this strategy he may be helped by society, which, as I will show in the next part, has the power to reinforce false consciousness through ideology and other instruments of cultural manipulation. This kind of individual has low indeterminacy of behaviour but also low autonomy. If, on the contrary, the individual chooses to choose, i.e. to use his judgement in making his choices, as Nietszchian or Musilian people usually do, he preserves autonomy, but is exposed to uncertainty, indeterminacy and instability of behaviour.

80 Individuals, culture and behavioural institutions In closing this section, and before going ahead, let me clarify some of the terminology I have used so far. I have adopted the term ‘dissatisfaction’ to define a gap between aspiration levels and the rate of achievements of a decision within a given preference ordering. The term ‘psychological conflict’, is used to define a form of distress that emerges when choices involve two or more different aspirations or preference orderings. The latter can take several forms, from emotional to moral conflict. Finally, the term ‘frustration’ is used as a portmanteau concept comprising both dissatisfaction and psychological conflict. Idealisation There is a particular kind of self-deception device which deserves special treatment on account of its social diffusion and ideological effectiveness. I will call it ‘idealisation’. It is activated when an action or an aspiration is associated with negative social values and ethics. If the person who entertains an aspiration shares the values that condemn it, he is faced with the necessity of hiding from himself the real grounds for his actions and constructing a false consciousness so that his behaviour is ethically justified in some way. Idealisation is made possible by the existence of a socially accepted motivation that can be substituted for the condemned one without the actor being required to modify his behaviour. The former plays the role of a surrogate motivation. The actor may remove from his consciousness the bad motivation, which thus becomes a latent impulse, and bring the surrogate one to the surface. So long as the instrumental variables pursued with surrogate aspirations move in the same direction as those attainable with latent impulses, the actor can keep serving the latter without needing to reveal the bad motivation to himself. A momentous example of idealisation is the psychological device through which most wage workers avert from their consciousness the shame of renouncing freedom when they sign an employment contract. They cannot accept the idea that wage is a premium for the renunciation of freedom, compensation for giving up a part of their life, since social values attribute positive worth to personal autonomy and condemn bondage. But if they accept the idea that they are selling just a commodity, wage becomes the price of ‘labour power’ and human dignity is saved, for only a person endowed with freedom of contract can perform the act of selling a commodity. Furthermore, if they convince themselves that the commodity they sell springs from a stock of human ‘capital’, they can even be persuaded that there is no substantial difference between their own social standing and that of their employer, since both can be regarded as capitalists who exchange flows of goods. Another interesting example is offered by the psychological devices through which managers define their motivations. If their deep impulse is the lust for power and society condemns this kind of aspiration, they have to resort to idealisation to justify their behaviour. Suppose all the members of a company like working in a climate of democratic participation and labour slack. And suppose we are dealing with a manager endowed with a multiple self, with a part of his soul hankering after power seeking and the other after social consensus. The two selves would

Individuals, culture and behavioural institutions 81 give rise to contrasting attitudes, for managerial slack and power exertions are substitute goods. The laxist self would suggest playing down the use of power and conceding much slack, whilst the despotic self would encourage the opposite behaviour.13 But suppose there is a complex set of causal relations between power and firm size. For instance, power exertion favours efficiency and growth, while size growth implies accumulation of power, so that the two variables, growth and power, vary together. And suppose that both the company stake-holders and society at large value positively the aspiration to growth. Then an easy way for the managers to idealise their motivations is to keep power seeking concealed from their consciousness and to use growth as a surrogate aspiration. The case is illustrated in Figure 2.5. The North-East quadrant of the diagram shows two preference curves embodying two contrasting aspirations, one determined by a laxist self, PL, and one by a despotic self, PD. Since the lust for power is not considered a good impulse by the social values introjected by the manager, any action motivated by PD would be precluded. The South-East quadrant shows a monotonic transformation of power into growth. The South-West quadrant maps growth onto itself. Finally the NorthWest quadrant redefines preferences in the growth-slack space. The two new preference curves, P¢L and P¢D , are transformations of those defined in the powerslack space. Since growth is valued positively by society, in this space the manager can give vent to his real impulses, and feel justified in minimising slack and Slack

PL '

PL

PD '

PD

Growth

Power

Growth

Figure 2.5

82 Individuals, culture and behavioural institutions maximising power. But now he is convinced working under the pressure of an altruistic sentiment rather than an egoistic one, for power accumulation can be interpreted as an instrument for achieving high growth. Among the many possible causal relations between growth and power, the actor is induced by idealisation to pick out only one, and convince himself that the power he tries to accumulate is just a means to achieve a socially accepted goal, and that this is his real aspiration. In this case idealisation brings about a sort of psychological inversion of the meansgoals relationship. Note that a similar analysis can be applied to the explanation of the politician’s behaviour by substituting company growth with public well being. It is easy to understand that a politician need not consider himself an opportunist despot – one who tells people he works in their own interests while in fact tries to satisfy his own thirst for power. He just has to tell himself that story in order to feel morally justified in carrying out the most perverse actions he considers necessary to accumulate the power required to serve people. The autonomy of choices and its limitations The conception of the individual as a subject endowed with a multiple self contributes to the solution of four important problems of the theory of social action: defining the degree of individual autonomy; highlighting the conditions ensuring the purposefulness of actions; accounting for the conditions ensuring rationality of decisions; discerning irrational and a-rational actions. Let me define, in psycho-sociological rather than legal terms, the notion of ‘autonomy’, and distinguish it from ‘freedom’. The latter, intended as ‘freedom to choose’, refers to the possibilities open to people in defining their aspirations or the goals of their actions. In a theoretical setting in which the influence of institutions and culture on the formation of intentions is strong, personal freedom depends on the width of choice field allowed by culture and institutions. ‘Autonomy’, refers to the act of choice. Once an option has been selected, the extent to which it has been done voluntarily remains to be clarified. One could define the degree of an individual’s freedom, intended as the width of the option range opened to him in a given society, and the degree of autonomy, intended as the extent to which an individual’s aspirations have been chosen voluntarily. The difference between freedom and autonomy is the same as that between a possibility and an act, or between an objective situation which is external to individuals and a property of their subjectivity. A person is endowed with ample freedom of choice if the social and cultural context to which he belongs offers him the possibility of choosing among many options in many situations. He enjoys a wide autonomy if most of his choices have been made voluntarily. The processes and conditions determining reduction or loss of autonomy can be grouped into two classes: those in which the loss of autonomy depends on an absence of freedom, i.e. on restrictions on the range of choice, and those in which it takes place even in the presence of many choice options. Without any claim to completeness, it is possible to single out six different processes or conditions of

Individuals, culture and behavioural institutions 83 autonomy reduction, three belong to the first class, and three to the second. The former are: poverty, dependence on ideology and obedience. The latter are: selfcompulsion, manipulation and habituation. The first condition is perhaps the most important in all societies based on exploitation. Scientists often refer to it in terms of ‘situation- or state- or possibilitydependent preferences’, but it is a very simple and evident phenomenon. Its real name is ‘poverty’. It can take two forms: dependence on means and dependence on needs. From a legal point of view everybody is free to buy a Ferrari car. But how many have the means to do so? A permanent lack of the means to satisfy certain aspirations leads necessarily to exclude those aspirations from the field of choice. Many forms of restriction of the option set depend on the limitation of means – a purely economic constraint. This is dependence on means. Dependence on needs occurs when individuals are compelled to target their choices to the acquisition of the means to satisfy some unavoidable necessities. Many people would like to spend most of their life reading books, listening to music, writing poems, or just having a good time. But they have to eat, and are poor. Thus they have to spend most of their time working. Aspirations to spiritual enrichment are out of their reach. Their option field is rather restricted. These persons are not free. That is why their choices are not autonomous. Another similar process is dependence on ideology, a subject I will deal with in the following sections. Suffice it to say, for the moment, that an ideology is a set of beliefs linked to interests, not necessarily the interests of those who entertain them. In the case of an ideology sustaining beliefs which hamper the interests of the individuals who accept them, certain aspirations may be considered unachievable due to some of their intrinsic properties rather than to a lack of means; as in the case of a worker who believes he does not want to be an artist because he has no talent, rather than because a lack of education prevents him from revealing his talent. When an ideology negatively affects values, some aspirations may be shunned because they are ethically devalued; as in the case of a non-educated citizen who always votes but would never endeavour to stand for elections because he considers politics a dirty matter. The psychological conditions for the emergence of dependence on ideology reside in the individual’s need to cope with frustration and a predisposition to self-deception. Illusion, undervaluation, focusing, ignorance and idealisation are the seeds of false consciousness. An ideology may give false consciousness the strength of common beliefs. Dependence on ideology may reduce substantially the individual’s freedom of choice. Typical is the case of ‘sour grapes’ aspirations or, more ‘scientifically’, ‘adaptive preferences’. People often adapt their aspirations to their possibilities, as in the case of dependence on means. The difference is purely psychological. A poor man may be compelled to buy a Mini instead of a Ferrari, without necessarily convincing himself that he does not like Ferraris. Instead, people who resort to the ‘sour grapes’ device do precisely that. The phenomenon is, of course, less irrelevant and ridiculous than this. An example is offered by assembly line workers who love their work.

84 Individuals, culture and behavioural institutions Another case of limitation of individual autonomy is what is called manipulation. This is a highly widespread phenomenon in modern civilisation. Just think of advertising, fashion, manias, or the conduct influenced by conformism as well as that induced by imitation of reference groups. Manipulation is intended here as a process conditioning choices independently of the beliefs associated with them: it directly affects aspirations. In this case the width of potential choice field becomes irrelevant. A last interesting case of loss of autonomy in the presence of freedom of choice is habituation. A prominent example is dependence on consumption. Many forms of it, like drug addiction, alcoholism, tabagism, have a prevalently chemical cause. But the most interesting, for social analysis, are those in which habituation is induced by institutionalised behaviour. They are interesting because they highlight certain processes of social control, but also because they account for various forms of compliance to institutions in which action ceases to be intentional and/or rational. Just consider a TV-addict person who spends most of his free time in watching television independently of the real pleasure he feels and even independently of alternative recreation opportunities. Such alternative opportunities, if they exist, are in practice irrelevant for this person. One could say that he enjoys a certain ex ante freedom of choice, but also that his ex post autonomy is nil. Nothing prevents him, if he wants, from taking a walk in the mountains, going to the cinema, or finding pleasure even in more audacious pastimes. Yet TV-addiction prevents him from making such choices, thus reducing his autonomy. Habituation to institutionalised behaviour is a most powerful means of social stabilisation. People just cease to react to the frustration caused by institutions, and these go on moulding behaviour notwithstanding their inability to satisfy the aspirations that may originally have induced people to conform to them. TV-addiction is only one case. There are many more examples: always reading the same newspaper, going to mass every Sunday, always voting for the same party. Not all of them must be considered cases of habituation, of course. But there is no doubt that in many cases they work like that. In praise of disobedience Often the loss of autonomy is caused by coercion, i.e. the imposition of certain rules or orders. It is an interdiction when rules or orders establish what must not be done. It is a prescription when they establish what must be done. In both cases rules or orders are accompanied by the definition of punishment for transgression. Coercion calls for obedience: individuals obey orders or rules primarily because they are associated with punishment. And obedience implies a loss of autonomy which is normally caused by lack of freedom. But it is possible even in the presence of a wide field of choice if an individual’s personality is overpowered by selfcompulsion. All actions performed under the command of another person are nonautonomous. In the case of prescription, people have no faculty to choose either the goals or the means for their activities, which in fact cannot really be considered their own actions. This is another case of loss of autonomy caused by a loss of

Individuals, culture and behavioural institutions 85 freedom. Interdiction restricts the field of choices but does not nullify it completely. After all, the prohibition to eat a certain fruit in earthly paradise still left a lot of other fruits that could be freely harvested. The actions of an individual who decides to disobey must be considered autonomous, since they cannot be interpreted as imposed by another actor. Disobedience expands freedom first of all in the case of an interdiction, since it adds a forbidden option to the field of choice. The more so in the case of prescription, which would completely nullify any possibility of choice. Thus disobedience is a manifestation of individual autonomy. Of course it implies the risk of incurring punishment. But the individuals themselves evaluate this risk. The risk of punishment limits individual autonomy only for people who obey. There are different kinds of obedience, depending on a person’s mental attitudes. There can be passive, reluctant and active obedience. With passive obedience a person obeys commands without posing any questions about their meaning. He does not ask himself whether they are right or wrong, whether he would have freely chosen the prescribed actions, whether obedience implies a loss of autonomy or whether the loss of freedom caused by prescription diminishes his personality. The prescribed activities do not concern the obeying individual, who considers them another person’s business. This individual has a Pessoan personality and a compartmentalised self. Self-identification may take place outside the social relation in which he is subjected to the commands of another person, and the two homunculi do not interfere with each other. Interference, on the contrary, occurs in people who obey reluctantly. In this case the divided self does not come to terms with reality. Alienation is recognised and refused, but cannot be avoided so long as one has to obey. Obedience becomes a sentence and a punishment in itself. Independently of whether the orders are perceived as unjust or wrong, the obeying individuals refuse to consider the prescribed activities as an expression of their will, and feel belittled every time they execute an order. Thus obedience is associated with a mental reserve which induces people to transgress as soon as they can. It is in transgression, even only in the possibility of disobeying, that this kind of individual looks to widen his option range. The mental reserve predisposing to disobedience serves to reconcile the experience of alienation with aspirations to deliverance and self-respect, and is justified by a desire for recognition. Active obedience is a way to remove this conflict – a case in which coercion becomes self-compulsion. The acceptance of obedience is induced by inescapable internal pressure. Adam might have been induced not to eat the forbidden fruit by an introjection of authority caused by the strength of the command source. This psychological mechanism consists of a self-restriction of choices to the point of choosing voluntarily the prescribed action or avoiding the forbidden one. The divine prohibition was not sufficient to avoid the sin, since Adam and Eve were endowed with a free will. But introjection of the prohibition might have been. A self-imposed restriction of the field of choice is not the same thing as a free choice among a wide range of possibilities. Self-compulsion brings about an impoverishment of personality. Active obedience is typical of a Kantian personality, it is an instrument for the construction of compulsive integration of the self. Individuals

86 Individuals, culture and behavioural institutions of this kind have generally undergone strong social pressure through, for example, socialisation, education, moral subjugation and so on. They develop rigid characters and their choices take place in a sort of interior prison. To avoid recognising the external prison and, perhaps, also to escape the anxiety brought about by the possibility of evasion, people tie their own hands. The conduct of these persons is not an instrument to pursue their chosen aspirations. It may serve the aspirations of other persons. Subjugated individuals behave as they do because they feel they must. They do not deliberate on their moves, they just feel them necessary. Self-compulsion must not be mistaken for self-discipline, which is in reality the product of a Nietzschian personality, and not a form of obedience. Individuals of this sort decide their actions consciously, voluntarily and rationally. If they comply with rules, laws, customs, routines, it is because they consider them valid instruments for achieving their own aspirations. They preserve their autonomy. The difference between self-discipline and self-compulsion is the same as that between reasoned and imposed act, between an action decided by a person in the pursuit of an aspiration of his own and an activity not aimed at pursuing a personal aspiration. Purposefulness, rationality and their absence Now it is possible to focus on the relationship existing between the autonomy of choice and the purposefulness and rationality of actions. All actions determined by autonomous choices are intentional, since autonomy implies voluntary definition of aspirations or, at least, voluntary adaptation to the behavioural institutions affecting individual aspirations. Thus there are processes or situations in which the loss of autonomy implies a loss of the ability to act intentionally. Habituation is one of these. Obedience is another. A slave’s activity can certainly be interpreted as aiming at some precise goals, which however are the master’s goals. But a loss of autonomy does not necessarily imply a lack of purposefulness. It is possible that the definition of aspirations is not the result of an autonomous choice, yet the aspirations themselves are genuinely the decisor’s. Actions may be wittingly moved by individual intentions, albeit within a restricted choice field. Poverty and sour grapes situations are the typical cases in which this phenomenon occurs. So long as an action is motivated by an individual aspiration, it must be considered as purposeful, even if the decisor undergoes social conditioning and economic constraint. There are other, more complex cases, for instance a loss of autonomy caused by ideology, in which a lack of purposefulness may or may not be at work. By affecting beliefs, an ideology may induce some individuals to choose goals which contrast with their own interests, but do not necessarily prevent them from pursuing their own aspirations. A loyal worker who decides not to strike because he does not want to damage ‘his’ company, or a politicised blackleg who breaks a strike picket out of hatred for the communists, might be considered blinded by an idea, but their actions, while perhaps not autonomous, are genuinely intentional.

Individuals, culture and behavioural institutions 87 It may happen that an ideology operates more violently, for instance by imposing some values that prescribe or forbid certain actions independently of any individual goal or aspiration. This is the case of a Japanese worker who does not strike just because rebellion to authority is alien to his moral world: it is not a decision not to strike, it is just no decision at all. In this example, ideology does not affect the goals of an action, but acts directly on the means, by prescribing or forbidding them. The same is true with manipulation. When this affects the formation of aspirations, it restricts individual choices, yet it does not necessarily make actions unintentional. However, it can also affect the formation of beliefs, or even act on will in such a way as to induce or obstruct certain actions independently of their goals. Some forms of advertising, for instance, aim precisely at this effect: to induce a certain consumption in an unintentional way. The use of violence, as a means of bringing up children, is another instance of this class of dehumanising processes: it aims at determining human behaviour in the same way as with Pavlov’s dog. Self-compulsion, in this respect, is similar to manipulation. It can be interpreted as a sort of socially conditioned self-manipulation. The child who has been brought up to perceive disobedience as a painful experience, may grow up by introjecting paternal violence and convincing himself that disobedience to authority is a mortal sin. He may become a good soldier, but the actions he carries out voluntarily in executing an absurd order could hardly be considered as motivated by a free personal aspiration. A good soldier is like a good slave or a good dog: his behaviour responds only to the goals of those who command him. Thus there is a class of behaviour including all the non-autonomous and unintentional ones: those determined by obedience and habituation, as well as a part of those induced by ideology, manipulation and self-compulsion. Since these are not actions determined by decisions concerning the means to reach the actor’s goal, the very notion of ‘rationality’ is impossible to apply. This kind of behaviour is neither rational nor irrational: it is a-rational. I have shown that there are actions which are intentional and rational, and others which are unintentional and a-rational. But these two classes of behaviour do not cover the whole field of human conduct. There is a third class comprising the actions which are intentional but irrational, in which decisions about the means turn out to contrast with the attainment of the ends. Apart from cases of irrationality caused by error, the most interesting complications are those linked to institutional plurality and the multiplicity of self, phenomena that may cause a sort of multiintentionality of actions. Often an individual takes a decision stimulated by different internal pressures and, in trying to satisfy more than one sentiment at a time, obeys several selves simultaneously, like a religious man who does charity work both out of love for his neighbour and to earn a place in Paradise. And often a given action actually succeeds in satisfying more than one aspiration. But not always. Indeed, the most interesting cases of irrationality are those caused by multiintentionality. When an action is moved by contradictory goals there is no hope: if

88 Individuals, culture and behavioural institutions it is rational with respect to one of them, it is necessarily irrational with respect to the others. Contradictory goals generate two forms of irrationality, which could be called ‘synchronic’ and ‘diachronic’. Synchronic irrationality occurs when two or more contradictory goals simultaneously motivate a decision. The Christian banker who must decide whether to grant a loan to a very risky borrower faces an insoluble problem. If he grants the loan at a 40 per cent interest rate – assuming this is the profit maximising risk premium – he takes a rational decision in his role of entrepreneur. This, however, would be an irrational decision from the viewpoint of the Christian self. In fact 40 per cent is an usurious interest rate for the Catholic church, so that the banker’s decision alienates him from Paradise. This kind of problem often spoils the life of Pessoan and Musilan individuals, whose multiple selves are unable to integrate in a coherent personality. Not so with Kantian and Nietzschian persons. If the Christian banker organised his soul by imposing a hierarchical order to his many selves, for instance by putting God’s grace at the top of his aspirations and allowing himself to accede to more earthly choices only when they do not contrast with the supreme end, then he will not demand a 40 per cent interest rate. If he contents himself with 15 per cent, provided this is the threshold to divine benevolence, then his action will be rational. Profits will be low, of course, but profit is considered a legitimate goal, in this case, only if it does not contrast with God’s grace. Integrated personalities, however, might meet with another kind of irrationality. Let me assume that that loan is to be repaid in ten years, and that the borrower goes bankrupt after five years. Our Kantian banker will perhaps kick himself when he realises that his losses would have been lower had he applied a 40 per cent interest rate. His entrepreneurial self might rebel to the gullibility of his Christian self, and might seize power in that interior pseudo-republic, even if only temporarily. Then the decision to apply a 15 per cent instead of a 40 per cent rate, which was considered rational ex ante, would be considered irrational ex post. The same problem must be faced by the hard drinker who forces himself not to drink more than one glass of wine a day. Let me assume he succeeds for a day, but that in the evening he feels very disheartened. He will ask himself whether his moderate self was perhaps too hasty in taking such a childish commitment. Maybe he will consider the decision irrational and, to correct it, will decide to drink some more glasses of wine. We have here two decisions that the very person who took them might consider both irrational at different moments of his life. Diachronic irrationality is often revealed by regret. It is explained by the multiplicity of selves. Summing up, synchronic irrationality occurs at the time of decision-making, i.e. the same time the decision is made, whereas diachronic occurs after the decisionmaking process has been completed and implemented. Another kind of irrationality may be induced by ideology in those cases in which certain values are associated with the means to pursue aspirations. The most adequate means to reach a given goal may be morally condemned, or, alternatively, some means may be considered good or better than others even if in reality they are not. For instance deference towards the boss is often considered the best road to a successful career. In certain cases this belief may work well for some

Individuals, culture and behavioural institutions 89 Table 2.2 A classification of human actions

ì Rational ï ïï ìIntentional í ìCaused by error ï ïor ideology ï ï ïï ï ï ïîIrrational í ï ïCaused by Kinds of action í ï ï ïî multi-intentionality ï ï ï Unintentional and a-rational î

ì Diachronic ï irrationality ïï í ïSynchronic ï ïî irrationality

individuals, but, if it is shared by many workers, most of them may turn out to be worse off, for instance, because career opportunities are open to few individuals in a hierarchical organisation. In this case ideology causes an error, inducing irrational behaviour, i.e. an action which brings about dissatisfaction instead of satisfaction. Table 2.2 schematises what I said so far about the variety of actions. In conclusion, one could say that non-rational actions span a wide field of human behaviour, maybe wider than that spanned by rational actions. In fact, besides including all the actions in which reason, in spite of the actors’ good will, fails due to logical error or ideological deception, they also embrace a-rational behaviour, in which the will itself fails. But what should induce us to believe that the set of non-rational actions might be very great indeed is that, due to multiintentionality, many rational actions can also be considered as synchronically or diachronically irrational.

The ideological conditions of social reproduction Ideology An ideology is a system of socially diffused beliefs developed in relation to material interests. Let me expound separately on these three aspects. The beliefs are organised into a system. Thus, although part of a culture, an ideology does not coincide with it. A culture is the set of all the beliefs belonging to a society. It contains ideologies but also other things, for example, opinions justified by knowledge interests rather than material interests, or opinions and values that are not linked to any interests at all. And since many diverse and even contrasting beliefs may co-exist in a society, a culture is never truly self-consistent. An ideology tends instead to be just that. The various beliefs belonging to an ideology strive to organise themselves so as to sustain each other and to build an all-encompassing view of the world which is capable of self-justification. They are beliefs that develop in relation to material interests. A scientific doctrine, insofar as it is determined by a knowledge interest, is not necessarily ideological. It is difficult, for instance, to find an ideological component in

90 Individuals, culture and behavioural institutions Einstein’s theory of relativity. However, it cannot be excluded that a science is imbued with ideology. It is always possible that a material interest lurks behind a knowledge interest or widely affects it. To bring out the ideological component of a theory or a world view, one should be able to single out the material interests it sustains. They are socially diffused beliefs. A personal system of thought is not an ideology, at least until it is assimilated by a social group. An ideology is a system of thought shared by a social aggregate. Therefore I should not deal with it in this chapter, which is dedicated to individual behaviour. But it is impossible not to do so, since ideologies not only affect individual behaviour, they also influence the definition of individual interests, especially when these become class interests. Thus the problem of social classes as producers of ideologies, and hence the problem of defining class interests, cannot be avoided. The definition of the interests of a class raises a serious philosophical dilemma. A class in fact defines its own interests with the help of ideologies. Is ‘shirking’ explained by opportunism or by a desire for deliverance? If you reply by asking to reveal the hypotheses lurking behind the notions of ‘opportunism’ or ‘deliverance’, you may trigger off a process of regression which sooner or later ends up with an axiom or an ontological proposition. Then you are back to ideology: how metaphysical is the doctrine of homo economicus? And whose interests does it serve? The problem is even more complicated than that. A class ‘in itself’, i.e. as an objective social aggregate, is made up of different individuals. When you speak of class interests you must refer to the class ‘for itself’, i.e. as a collective subject. Interests are defined in relation to goals, and these belong to social actors, either individual or collective (organisations, movements etc.), which do not necessarily coincide with the class in itself, if at all, because they are only single parts of it. Setting aside the problem of the multiplicity of individual goals, the question immediately arises: which individuals’ or organisations’ goals and interests can be identified with those of the class? To assume that all people belonging to a class have the same goals is ingenuous metaphysics. To assume that some organisations or groups of people are in a privileged position to define class interests is a metaphysical bet that especially political organisations normally make to constitute themselves as ‘instruments’ of collective action. This kind of bet is also made by the theoreticians of political programmes and philosophies. One can even attempt to justify it ‘scientifically’, but metaphysics remains. Only the historical success of the bet can prove its validity and, often, just as a self-realising expectation. Validation of assumptions on collective interests can only take place ex post. In other words, to escape the ‘hermeneutic circle’ in this matter one must resort to axioms which, ex ante, are inevitably arbitrary. Let me now define the class interests of workers and capitalists without fear of metaphysics. Let me begin with the ill-famed distinction between immediate and fundamental interests. The immediate interests of a class are the set of interests of the individuals or groups belonging to the class in a given historical and social situation. They may be multifarious and even contrasting, and may change with social evolution. Fundamental interests are common to all people belonging to the

Individuals, culture and behavioural institutions 91 class and are determined by the position of the class in the fundamental structure of society. They are permanent, so long as that social structure persists. Since capitalism is defined as a system in which the workers are oppressed and exploited to make them produce the surplus value required to sustain accumulation, the working class has an immediate interest in the reduction of oppression and exploitation and a fundamental interest in their abolition. More specifically, the workers may have the most diverse immediate interests in a given situation. The unemployed might be interested in an expansive fiscal policy, metal workers in a reduction in working time, chemical workers in an improvement in workplace safety. The fundamental interests of the whole class, though, are common and coincide with the abolition of the employment relationship. The immediate interests of big businesses might diverge from those of small industry, the retail trade or the banks. Yet all the capitalists have in common a fundamental interest: capital accumulation. Fundamental class interests are defined axiomatically. Returning to ideology, let me now pose some problems. There may be a mismatch between the ideology espoused by a social class and its material interests. This may occur for two reasons. First, a contrast might arise between immediate and fundamental class interests, in which case, whatever the ideology sustaining class action, it cannot back both of them. Second, a class might be unable to build its own ideology autonomously. These and other related problems have been elucidated with great clarity by Jon Elster (1983), whose four fundamental propositions on ideology I am now going to recall and slightly modify. First proposition. The ideologies accepted by a certain social group do not necessarily favour the interests of the individuals belonging to it. This is Marx’s thesis in his interpretation of religion as an opium for the people. Engels in turn interpreted the Calvinist doctrine of predestination as a rationalisation of the fact that commercial success depends on some over-individual economic force; and Kolakowski (1978) observed that this is an ideology of impotence rather than an instrument favouring the class interests of the capitalists. Second proposition. The ideologies commonly accepted in a given society do not necessarily favour the interests of the dominating classes. An opium for the people which justifies the condemnation of usury does not stimulate formation of the financial conditions of capital accumulation. So much so that many Mediterranean countries have had to wait for completion of a culture secularisation process before modern industrial take-off could begin. Third proposition. The ideologies linked to certain interests do not necessarily favour those interests. Recalling an example in the preceding section, some clerks, to get higher incomes, might tend to concentrate efforts on their career and to assimilate an ideology based on collaboration with the owner and the boss. Yet not all clerks can be promoted to command positions. If the diffusion of this ideology discourages the distributive conflict and, as a consequence, causes a worsening of the average salary, it brings about an outcome which is the opposite of that aimed at by the majority of clerks. Fourth proposition. The ideologies favouring certain interests are not necessarily determined by the aim of furthering those interests. Japanese workers who perceive

92 Individuals, culture and behavioural institutions ‘their’ company as a community moved by a collective aspiration to growth will tend to behave in such a way as to foster that growth. As a consequence, their wages could increase rapidly. Yet this is not the reason why they accepted that ideology. The most interesting of these propositions are the first and the fourth. In fact it is not so important to understand which interests originate an ideology as to know which are the ones favoured by it. Thus it is possible to classify ideologies into two kinds: conducive and adverse. An ideology is conducive (adverse) for a class if it favours (hinders) the achievement of its interests. And since it is possible to define both immediate and fundamental interests for each class, an ideology can be immediately or fundamentally conducive (or adverse). Thus, for instance, a company ideology attributing positive values to class collaboration, effort, career and conflict resolution, is immediately and fundamentally conducive to the capitalists and fundamentally adverse to the workers. Rather than understanding the interests from which an ideology originates, it is important to discern the mechanisms sustaining its formation and its diffusion. The formation mechanisms concern the socio-psychological processes that induce the individuals belonging to a group to elaborate or accept certain beliefs shared by the group. The most interesting of these are: social reasoning, rationalisation, conformism, collective sentiments and moods. The diffusion mechanisms pertain to the social communication apparatuses through which a certain ideology spreads among the masses. The mechanism of ideology formation Praxis is the mother of ideology. Opinions, in fact, are elaborated to understand the world in which people act, while values are built to give a sense to action. Ideologies have an essentially pragmatic import, and even their scientific elements attain significance as instruments of praxis. An ideology knowingly elaborated to accomplish an action must contain some scientific beliefs. Action, is the provision of the means to pursue a goal, to fulfil an interest. It requires an understanding of the means, of the relation between means and ends and of the social context which may affect the choice of the means. Since the very elaboration of such beliefs is a means to reach some goals, the construction of an ideology knowingly aimed at the pursuit of certain interests cannot set rationality aside. Marxism, for instance, as an ideology of working class action, in order to be a conducive ideology must be ‘scientific’ socialism, i.e. an instrument of social and political knowledge. But a conducive ideology requires rational choices not only in its production. Also its acceptance implies some degree of rationality. The worker, the shop steward or the revolutionary who espouses an ideology must do it through the use of judgement. To accept an ideology means to decide that the beliefs it conveys are valid, that they serve to sustain and guide action. Thus reason is a requirement of the elaboration and the acceptance of an ideology, if it is to be a conducive ideology. In reality the production and the acceptance of a reason-based ideology cannot be separated if this is to be shared by a group of people who elaborate the ideology to pursue their common interests, for it must be ‘decided’ rationally by all individuals belonging to the group. Obviously there may be some division of knowledge

Individuals, culture and behavioural institutions 93 in the process of ideology construction. Certain actors may play a more active role in the elaboration of abstract and general ideas, whilst others contribute more practical and specific pieces of knowledge. But all individuals normally take part, in varying degrees, in a social exchange of suggestions, criticisms, assessments, evaluations – in a social reasoning process – which eventually leads to the formation-acceptance of a common world view. This process is observed at its best in a democratic party, where discussions continually take place in all its layers, theoretical choices are made in all cells of the political body and flows of ideas continuously travel up and down the hierarchical ladder. Reason, however, does not suffice. Also moods and sentiments matter, and the culture and institutions that contribute to mould them. Interests can be interpreted as arguments of a utility function only in a utilitarian approach. But to what extent can you reduce a sentiment of solidarity or one of hate and envy to a utilitarian choice or the collective moods that sometimes unexpectedly drive a mass of strikers to demand ‘everything and at once’? It is especially in the definition of the action motives that the formation of an ideology is affected by the social context. The interests moving the action of a class are not the interests of all the individuals belonging to the ‘class in itself’, but only those of the people who adopt the ideology governing collective action. In awakening the awareness of such interests, collective moods and sentiments play an important role. Collective moods and sentiments are mental feelings common to all the individuals belonging to a social aggregate. A collective mood is at work, for example, when a mass of speculators rushes to buy shares under the drive of a bull mania. Yet a social aggregate is made up of individuals. The problem then is: what induces many individuals to develop common moods? There is no room here to delve into social psychology, a discipline which has produced several models of collective moods formation, from epidemic processes to mass psychology. Nor is it necessary to do so. Suffice it to assume that an individual attitude exists which, in certain circumstances, drives many individuals to adjust their beliefs, sentiments and moods to those of other individuals with whom they share some cultural affinity and a similar social position. This attitude is conformism, which is a kind of precautionary behaviour. As such, it serves to cope with uncertainty and the limitations of reason. The awareness that a belief is accepted by a great mass of people can help a single individual to overcome the difficulty of judging its validity. The fact that it is shared by the collectivity becomes a factor of positive judgement. Thus conformism is not necessarily a negative attitude. To the extent that it is a rational precautionary form of behaviour, it can turn out to be an effectual instrument of class action. Conformism, though, may lead to the adoption of conducive ideologies only when it develops within a process of social reasoning. When, by contrast, it sustains a rationalisation process, it usually leads to the acceptance of adverse ideologies. Social rationalisation acts on beliefs on the basis of self-deception devices and serves, indeed, to give a collective content to them. It is a communication process whereby many persons help each other, or are helped by some informationally privileged actor, to resort to common self-deception devices.

94 Individuals, culture and behavioural institutions The fox might decide not to eat the grapes it cannot reach by convincing itself it does not like them, regardless of whether they are sour or not. Alternatively, if it does not like sour grapes, it might convince itself they are just sour. This is a subtle but substantial difference. The former mechanism affects the formation of aspirations and constrains individual autonomy by limiting the definition of individual goals. The latter influences the beliefs and may induce irrational actions, since it engenders opinions and values which might hinder the pursuit of personal interests. Two different self-deception devices are called upon in these cases, undervaluation and ignorance. One serves to devalue goals that are judged unattainable, the other to neglect possible choices that could cause frustration. In realising the impossibility of reaching a certain objective, people might be assailed by frustration. To avoid this, they elaborate or accept ideologies which deform or limit their perception of the world. An example is the case of the employees of a firm who, faced with the failure of negotiations to obtain a high wage increase – a failure caused, say, by the inability or bad faith of their union officials – let themselves be convinced of the validity of a (false) management declaration that the high wage increase was not possible because of bad market conditions. Undervaluation would help the workers to convince themselves that the wage increase they obtained is the most they could achieve. Ignorance would induce them to believe that market conditions are really bad. Quite different is illusion, by which people elaborate or accept wrongly positive beliefs about the state of the world or the objects of aspiration. While undervaluation causes an underrating of a certain state of the world, illusion produces an overvaluation. The practical consequence may be the same: the definition of ‘wrong’ interests due to misinformation. Referring again to the example of those employees who gave up a heavier wage increase, if an illusion mechanism is at work, they might convince themselves that the small improvement they obtained is in reality big, whereas undervaluation would have convinced them that it was the most they could expect. Finally, focusing induces people to deform their choice fields by concentrating their interests on some opportunities and foregoing others. Normally ‘realistic’ opportunities are privileged whilst ‘unrealistic’ ones are forsaken. By this device, for example, the employees who obtained a low wage increase might convince themselves that wage is less important than job preservation and that the low income improvement they obtained was a good price to pay for employment stability. The company management might encourage the spreading of this conviction by declaring that, in that particular conjuncture, a high wage increase would cause a fall in employment. An ideology sustaining this view is that embodied in the theory according to which capital-labour substitution is brought about by changes in relative factor prices. Ideology works by giving the strength of common sense and conventional wisdom to the wrong opinions formed on the ground of self-deception devices. The diffusion of these opinions takes place through social rationalisation and is often sustained by conformism. This, now, works as a behaviour aimed at avoiding the frustration of potential failure. Insofar as it induces people to limit their choices in a way that hinders the pursuit of their interests, it turns out to be an irrational

Individuals, culture and behavioural institutions 95 kind of behaviour. Social rationalisation affects the definition of interests: it produces deformation of reality to the point of inducing people to pursue the wrong interests. There would be no need, here, to know what the ‘right’ interests are. What counts is that the actual interests are defined on the basis of wrong perception and false beliefs. Diffusion apparatuses and ideological hegemony Something remains to be said on the apparatuses of ideology diffusion, which are reducible to two basic structures: network and centralised communication systems. A network communication system is a structure in which no branch point, or, in this case, individual, has a privileged position in the transmission of messages, anyone being able to play the role of both transmitter and receiver. Communication takes place through personal contacts, which can be either verbal or mediated by technologies. Examples range from a pub conversation to e-mail forums. Network communication plays a fundamental role in the formation of common sense and collective moods since, as already observed, confirmation of shared opinions, values and mental feelings and observation that most of our neighbours feel and think in the same way, is a decisive factor in corroborating the validity of opinions. In centralised communication systems the transmitter and the receiver of messages are structurally distinct, there being a systemic asymmetry in the direction of information flows: the transmitter elaborates and emits the beliefs; the receiver receives but does not transmit them. Even if there are some feedback mechanisms, they are generally weak and inefficient: just think of the reduction in sales of a newspaper which is out of favour with its readers. Moreover the transmitter communicates simultaneously with many receivers, while the latter may be connected to only one transmitter at a time. Mass media is the most important case of centralised communication. Another case is that of cultural communication apparatuses, i.e. structures dealing with the diffusion of literature, art, music and the like. Centralised communication is used by all hierarchical organisations, schools, churches, parties, trade unions and firms, since an organisation continually transmits internal messages. The diffusion processes working through centralised communication need complex organisational apparatuses and can be set up only by investing huge amounts of capital. It follows that capitalists and political actors have privileged access to their control and therefore a greater capacity to diffuse ideologies conducive to their interests. Since the working classes are hindered in acquiring control over the means of production, they are also excluded from the branch points of centralised communication apparatuses, and this reduces their capacity to diffuse their own ideologies. This is not an absolute asymmetry, of course. It is a historical fact that democratic action continually contributes to the formation of political, union and cultural organisations which aim to serve the interests of the oppressed classes and strive to build centralised communication apparatuses to diffuse their ideologies. But without doubt their strength is incomparable to the overwhelming control of capital and the state.14

96 Individuals, culture and behavioural institutions Now it is time to open up the problem of ideological hegemony. And let me start by formulating a ‘general law of capitalist ideological hegemony’. The oppressed classes are normally induced to accept ideologies which are conducive to the interests of the dominating classes. The dominant classes control the branch points of centralised communication apparatuses, besides the economic and political apparatuses, and tend to use them to diffuse ideologies which are conducive to their own interests. Thus, if the lower classes have no access, or greater difficulty in access, to the control of centralised communication apparatuses, the probability of their being induced to accept adverse ideologies is higher than the probability of elaborating and diffusing conducive ideologies. A probability is not a certainty. Yet the history of capitalism shows that only in some exceptional circumstances the dominated classes have been able to rid themselves of the ideological hegemony of the dominating classes. A capitalist system is dominated by a capitalist class. Thus the normal regime of cultural production in this system is one of capitalist ideological hegemony. The struggle for existence and dominance selects the winners, among other things, on the ground of the moods and sentiments governing their actions, egocentric and enterprising attitudes being conducive to success. At the same time it tends to disseminate contradictory mental feelings among the losers. Altruistic and conservative attitudes, which are responsible for bad performance in individualistic competition, tend to prevail among the lower classes, if at all, because these classes gather the losers. The drive toward social recognition and personal deliverance fosters the fascination exerted by the winners’ sentiments over the loser’s mind. It is true that in certain special social and economic conditions solidarity and revolutionary feelings could spread among the oppressed classes, thus sustaining the collective liberation movements that break out from time to time in capitalist society. But in normal times prevalence goes to a combination of conservative moods and egocentric sentiments. These attitudes, in the face of individual miscarriage, feed an accumulation of frustration and a consequent predisposition to self-deception and false consciousness which render the oppressed classes rather prone to accept adverse ideologies. The ‘winners-are-the-best’ and ‘money-doesn’t-make-happiness’ ideologies could well go hand in hand in plaguing the minds of the working classes and contributing to let them passively accept the status quo. Thus the political and organisational conditions for the ideological hegemony of the dominating classes – i.e. control of centralised communication apparatuses – is reinforced by a strong psycho-social condition – the spreading among the masses of the frustration, self-deception and false consciousness that predispose them to the acceptance of adverse ideologies. Ideological hegemony as a condition of social reproduction At the end of Chapter 1 I dealt with the conditions of social reproduction and observed that there are four types: financial, technological, cultural and political. The first two have been dealt with in Chapter 1, whilst the fourth will be in Chapter 3. The cultural conditions will be handled here.

Individuals, culture and behavioural institutions 97 They can be summarised in the concept of ‘ideological hegemony’. A hegemonic ideology is shared by both the dominant and the dominated classes. If it is conducive to the dominating classes’ interests, it will be called the ‘dominating class ideology’. It is adverse for the dominated classes if it induces them to accept their social condition as normal. In Chapter 1 I showed that the financial and technological conditions of reproduction presuppose some form of social asymmetry in the classes’ cultural endowments. For example, credit rationing can be class-biased also because the workers are not endowed with the managerial abilities which make a debtor trustworthy for the banks and financial markets, while the capitalist control of production is more efficient than the workers’ control because knowledge and information endowments are distributed among classes in such a way as to make it more difficult for the workers to control the managers than vice versa. Moreover both technical progress and cultural apparatuses tend to privilege the specificity of the managers’ ability and competence rather than that of the workers. Finally, the latter tend to develop motivations (better working conditions, higher wages, employment stability, instead of higher productivity and flexibility) which make them less suitable than the managers to direct production in view of accumulation. I also observed that all these psycho-social conditions tend to reproduce themselves independently of the existence of any ideological hegemony, just as a consequence of the social distribution of income, wealth and productive control: the proletarians are unable to make the investments in ‘human capital’ required to break the conditions of their dependency, while they are incapable of affecting technical progress so as to raise their managerial and control ability. Ideological dependency also performs an important function. The dominating class ideologies are swallowed by the oppressed classes to lull frustration. They nurture rationalisation processes which induce people to accept their social condition as normal or even just. Then the ideologies themselves favour proletarian proneness to make choices which hinder the change of that condition both at individual and collective level. For example they may induce a manual worker to convince himself that his subordination to an employer is the consequence of a poor individual endowment of natural abilities rather than the effect of a low investment in ‘human capital’, and, by the same token, that the dominant position of the capitalists is justified by a natural predisposition to command and organisation activity rather than the result of a high investment in ‘human capital’. Such convictions may reinforce the worker’s decision not to invest in ‘human capital’, thus reproducing the existing distribution in individual endowments. Therefore, even if the purely economic conditions of this distribution did not operate – supposing, for instance, that some administrative device existed which equalised the total costs of education of all individuals – the distribution would tend to be reproduced anyway. The influence of ideology is even stronger when it affects the choice of goals and the formation of aspirations rather than decisions about the provision of the means of action. Also in this case the ideological reinforcement of self-deception devices is decisive. Workers can easily be convinced that it is not worthwhile to aspire to more than a safe job, a tranquil life, a happy family and a long life as a

98 Individuals, culture and behavioural institutions pensioner. Why should such a person invest in the effort to acquire managerial abilities? The adverse effects of dominating class ideologies become dramatic in the sphere of collective choices. A question that all revolutionists have raised at some time is the following: why do the proletarians not normally rebel collectively against their condition, if they know that their numbers would give them the strength to change the world? It is easy to answer with another question: why should you make any effort at all to change the world if the one in which you live is the best of all possible worlds? No doubt this attitude is at least partially motivated by a precautionary choice, given the uncertainty involved in revolutionary change. But equally doubtless is the role played by the dominating class ideologies in augmenting the subjective perception of this risk and focusing choices on risk aversion, in nurturing the illusion of living in the best of all possible worlds, in deforming historical knowledge and disseminating ignorance so as to render irrelevant and undesirable the possible alternatives. Moreover dominating class ideologies also work by associating moral condemnation with values and motivations (e.g. extensive social equality and real freedom) contrasting with conservation of the existing order, or by deforming their definition so as to make them sustain resistance to change. For instance, equality is said to be good only if it is meant as referring to the legal position of citizens; and freedom only if it is intended as freedom of contract, of speech, of organisation and of deciding if your president shall be Al Gore or George Bush. But a dominating class ideology does not serve only to disarm the dominated people. It also arms the dominators. For them, the existing world really is the best of possible alternatives. Therefore ideology does not just offer them justification for their social position, their aspirations, choices and decisions. It also offers them arguments for actions aimed at preserving the existing order. A capitalist could easily be convinced by a liberal ideology that he must resist a strike and fight a union, not only to guard his profits, but also to preserve free competition. And, if he attaches a strong moral value to this freedom, he could even be convinced to fund organised groups of blacklegs, vigilantes and fascists to defend democracy. This observation gives rise to another problem. There are several dominating classes, which may have different interests. In a complex social system, certainly in a modern capitalist system, the dominating classes in civil society are different from those who dominate the state. In Chapter 3 I will define the ‘political class’ as the group of actors who have political power. For the moment let me just anticipate an important argument by stating that the political actors, as members of the political class, have particular interests, motivations and aspirations that render them relatively autonomous from civil society. They are able to develop particular ideologies or pieces of ideology to justify, determine and sustain their action. The problem is: if these differences exist among the dominating classes, what do we mean when we speak of a ‘dominating class ideology’ and what about ‘ideological hegemony’? One of the most important characteristics of ideology is that it is, or tends to become, a system of beliefs. A dominating class ideology must be able to organise

Individuals, culture and behavioural institutions 99 social beliefs into an organic, self-consistent and all-encompassing system. It must do so by justifying the position and sustaining the action of all social groups. And it must do it so as to favour the interests of the dominating classes, whoever they are. This implies that a dominating class ideology does not have just the two functions of arming the upper classes and disarming the common people. It has a third one which is no less important: reconciling the motivations and integrating the actions of the different components of the dominating classes so as to make them serve each other. In particular, a dominating class ideology acts by building links among the social and political actors. These may be widely autonomous in their choices and decisions. Yet, if they share a common ideology, they can act in each others interests while pursuing their own specific goals. Just imagine a parliament which passes an Anti-Combination law, not to conform to a mandate of the bourgeoisie, but to comply with a moral pledge toward the ideology of free competition and political democracy. This latter function of ideology is very important because it sustains the role played by the state in providing the political conditions of social reproduction, as will be shown in Chapter 3. In conclusion and more generally, it can be said that a dominating class ideology is adverse to the dominated classes and conducive to the dominating ones because it induces social rationalisation processes and false consciousness in the dominated classes and rational choices in the dominant.

Conclusions Homo relativus A realist ontology of the social being must be an ontology of indeterminacy and relativity. In concluding this chapter it is possible to argue with greater precision on the sense in which people are the summation of their fortunes, intended as the circumstances and vicissitudes that mould their lives. There are three senses: people are conditioned culturally, institutionally and historically. Individuals acquire their values and opinions from the culture in which they are embedded. Culture selects or produces the information required to make decisions, choices and to evaluate the outcome. Culture conveys the sense of the actions of the other people with whom each individual interacts. Not only does it determine the elements of knowledge required to define individual aspirations, but it also sustains the formation of collective and individual sentiments and moods. The cultural context provides the conditioning through which beliefs are organised into adverse ideologies, and its influence is especially effective in all forms of behaviour in which people are compelled to suffer a loss of freedom and autonomy. All the existing institutions limit and orient the choice of aspirations, whilst the specific institutions with which the individuals comply limit decisions concerning the means. Conditioning is stronger when people behave in an unintentional and non-autonomous way, since this kind of behaviour ceases to be motivated by freely chosen aspirations and is reduced to hetero-directed and socially induced activity.

100 Individuals, culture and behavioural institutions Finally, historical conditioning occurs because the cultural and institutional settings which hold sway in a given historical situation are always the result of a path-dependent evolutionary process. People often act together with the precise intention of compelling history to serve their interests and goals. Yet it is also true that each social aggregate always interacts, by co-operating or conflicting, with other aggregates, and therefore it is unlikely that the outcomes of a certain collective action correspond exactly to the aspirations which moved it. Moreover, since the goals of all the individuals who participate in a collective action do not always coincide, the outcomes of the action may turn out not to have the sense and the value each individual had expected. These relativistic conclusions on people’s ability to be the subjects of their actions and their history might be thought not to be valid at least for that part of human behaviour which can be interpreted in terms of intentional and rational action. However, rational action covers only a part of individual behaviour. Even when institutions effectively function as a social context of rationality, the rationality of human action remains bounded. Thus the evolutionary process governing institutional change can hardly be given the meaning of a march toward perfection. Boundedly rational individuals rarely maximise anything; they only tend to improve or at least not to worsen their living conditions. The criterion of conditional improvement implies that, within the constraints posed by institutions, individuals take decisions which are only partially satisfying. The beliefs acquired through culture are not just objective knowledge, but rather tend to express partial viewpoints on the state of the world. After all, the ‘state of the world’, from the point of view of an individual, is the set of actions of the other individuals with whom there is interaction. Therefore its sense is always impregnated with subjectivity and plurality of interpretations. Since every interpretation is affected by sentiments and moods which are multifarious, contrasting and mutable also within each individual, it is impossible to maximise anything, even independently of the degree of individual rationality and knowledge: information remains ambiguous, since it is formed within hermeneutic circles which impair its objectivity and stability. One might think that, insofar as the evolution of culture and institutions is governed by innovation and imitation and the process of behavioural revision is governed by the drive to do better, a succession of institutionally conditioned, yet rational, decisions, should bring about optimal outcomes at least in the long run. This is a vain hope. Since any given institution limits the decision options, it may well prevent some possibilities of action which are better than those it allows. Moreover, although its change may lead to the institutionalisation of a behavioural improvement, it may also modify aspirations, sentiments and moods and therefore restrict the field of decision options in an arbitrary and non optimal way. Institutional change, even when it is governed by the principle of improvement, remains sub-optimal, path-dependent and non-teleological. People are doomed to make their own history by running short of meaning. It is unlikely that the sense they eventually might succeed in finding out will come up to their expectations.

Individuals, culture and behavioural institutions 101 This is not to say that people have no freedom of choice. They do have some autonomy, especially because in most decision-making situations they have to face institutional and cultural plurality and a multiplicity of selves. But this kind of freedom brings about a new form of frustration. Besides being short of satisfaction due to bounded rationality, people also have to bear the distress caused by the necessity to comply with contrasting ethics, beliefs and sentiments. This might cause indeterminacy of behaviour, uncertainty, indecision and instability of choices. Indeterminacy is the price of freedom. Since people sometimes resort to selfdeception devices and false consciousness to cope with frustration, and since society often helps them in this endeavour by supplying easy solutions with ideologies, mythologies, religions and other forms of cultural subjugation, a usual way out of indeterminacy is to allow oneself to undergo a certain loss of autonomy. Whilst homo economicus, by virtue of his Olympic rationality and his simplistic utilitarian preferences, always exhibits a strong determinacy of behaviour, homo relativus normally has to cope with a trade-off between determinacy and freedom. One might ask whether homo relativus is really a different animal from homo economicus. The answer to this question would be negative if the latter species were correctly defined. If rationality is just the attitude to provide the means to achieve individual goals and if this achievement is evaluated by the individual in terms of personal satisfaction, then there is no doubt that many human actions in capitalist society are economic, i.e. egoistically rational. Rationality may be bounded and aspirations socially and culturally conditioned, in which case homo economicus gets a more human face and becomes a nicer chap, yet the value of a person of this sort undoubtedly tends to be the summation of his fortunes. A more interesting question is: are we dealing with a natural and, so to say, universal characteristic of the human being, or are we instead observing the product of a particular historical conjunction and a special cultural situation? Karl Polanyi grasped the essence of this problem very well. For most of the history of humankind and, still today, in the indigenous people’s reservations that resist capitalist colonisation of the soul, social relations are dominated by the urge for community reproduction, while human actions are inspired by a need to belong, by sentiments of solidarity, loyalty, faith and even servilism, but not of opportunism. By reducing social relations to economic transactions, the modern Prometheus freed humans from all traditional bondage and, with the gift of contractual freedom, he made them individuals, i.e. relatively, and at least potentially, autonomous subjects. He also made them homines economici. For the first time in history egoism becomes a positive value, a value. Yet this metamorphosis is neither complete nor stable. Those indigenous people’s reservations do not resist only in the Trobiand islands, provided they still resist there, but here and now, in every pore of the capitalist system and in the soul of every individual, if for no other reason that, together with the gift of legal freedom, the industrial revolution also conveyed to people the desire to make that freedom real. It is true that the daily struggle for survival, in the forms it takes in modern markets, continually feeds the development of the economic brand of human sentiments and behaviours: by awarding success to the greedy, by

102 Individuals, culture and behavioural institutions condemning the naive and guileless people to oppression and exploitation, by diffusing the ideology of profit as a prize for ability, by nurturing the growth of needs and the need for growth. But it is equally true that in this way, paradoxically, whilst it reduces the experience of absolute deprivation in an ever increasing number of persons, it generates a new spiritual experience: relative deprivation. Capital accumulation succeeds in levering on individual satisfaction only insofar as it produces dissatisfaction. And the stronger the accumulation the more extensive and unbearable becomes frustration. But the growth of relative deprivation is a factor of spiritual catastrophe. It is in this contradiction of capitalism, rather than in those sustaining its ‘laws of movement’, that we should perhaps place our hopes. Servant-master dialectic and beyond There is a particular form of loss of autonomy which is typical of the capitalist mode of production: wage work. The workers engaged by a firm with an employment contract undertake an obligation to perform labour activity under the command of another person. The institutional role of the other party, an owner, an officer, a manager etc., is of no import; what really counts is that, from a technical point of view, the employer assigns tasks to workers in the labour process and controls their activity, whilst, from an economic point of view, it works as a functionary of capital and determines the goals of labour activity to serve capital accumulation. Formally, the loss of autonomy of the workers during labour time is total: the obligation to obedience properly means renunciation of freedom, of action, of determining the ends of their own activity. Labour activity is not a worker’s action because it is not moved by a worker’s aspiration. The ensuing behaviour, as a consequence, is unintentional and a-rational. Unintentional, because it is not moved by the individual’s goals; a-rational, because it does not consist in providing the means to achieve an end. Labour activity, of course, is a means – a means of fulfilling a productive goal – and therefore it is intentional and rational, but not for the worker. The rationality and purposefulness of labour activity are in reality the properties of the plan which gives it form and rule. They belong to the capitalist, not to the worker. To paraphrase Hegel: what the worker does is properly the capitalist’s doing, which is pure and essential acting in the labour process, while the worker’s doing is an inessential acting. A worker’s activity is hetero-directed activity. This means that it is ruled by coercion, through rules, commands and punishments. In a pure employment contract, in which no form of sharecropping arrangement is admitted (piecework, profit-sharing, performance pay etc.), not even incentives are used as instruments of work regulation. Wage is not an incentive to work hard, it is a premium for obedience: it is commensurate to labour time, not to individual productivity, and is paid regardless of the way the workers perform their jobs. In reality, non-payment of wages, in the form of fines and firing, is an instrument used to rule work activity. But this is a punishment not an incentive. Yet a worker is a person, a being who aspires to recognition and self-realisation, an object who aspires to be a subject. The problem, then, is: how does this denied

Individuals, culture and behavioural institutions 103 subjectivity react to the negation of subjectivity, to labour coercion? In a preceding section I showed that obedience can take three forms, depending on the way it is elaborated by conscience. And indeed there are three forms of acceptance of the worker condition. Active obedience is developed by the workers who get trapped into an organisational involvement with their firms. Workers of this kind suppress the distress of subjectivity annihilation via a process of identification with the enemy. If the boss forces them to perform operations the overall sense of which is lost on them, they reconstruct that sense by an act of faith: the boss’s authority is an embodiment of the general rationality of their activities. If the capitalist deprives them of the right to suit their actions to their aspirations, they take over the capitalist’s aspirations as their own. This kind of worker, not only obeys willingly, but tries to anticipate commands and give the best of himself. Obedience is transformed into selfcompulsion. The firm’s hierarchy enters into the deep structure of the self, thus contributing to create a compulsive integration of personality, to put fideistic and authoritarian order into the interior republic. Identification with the company becomes a psychological mechanism of individuality transcendence which presupposes the activation of some sort of self-deception device of the idealisation or illusory kind. These persons live for a superior cause. It might be the ‘construction of socialism’, as in Stakhanovism; or it might be community growth of the paternalist firm, as in productive Shintoism. The workers who find fulfilment in their jobs through self-compulsion are workers in their lives not just in their factories. Passive obedience, by contrast, is typical of a compartmentalised personality. In this case, individuals behave as objects in the labour process and as subjects outside it. Their lives are not in the factories, but elsewhere: Dr Jekyll in the firm, Mr Hyde at home. Possibly, they beat their spouses and children, perhaps they write novels and poems during their free time. But when they are at work, at any rate, they comply with their duties. They execute orders with normal diligence, without enthusiasm, but also without resentment. Maybe they do not understand the purpose of their jobs? Never mind: it is not their concern. Do they have to suffer harassment by the boss? It does not matter: they are paid for this too. Their labour ideology is marked by indifference. This perhaps is the most representative model of the modern worker, at least in times of social peace. Not all individuals succeed in self-coercing or compartmentalising their souls. Some people insist on looking for reasons to believe in themselves even when their self is annihilated by work. Workers of this type are unable to reconcile themselves with the reality of alienation and coercion, and develop a divided self. When they work, their minds are always somewhere else, like the Gramscian worker, whose body is on the assembly line while his soul is in another world. They too might write poems, but try to do it during working time. The only way they can accept their reified self is by countering it with an active antagonist. But since the latter does not have much say in labour activity, work becomes an unbearable pain. Then an inner conflict arises in which the self aspiring to autonomy, fulfilment and recognition opposes the one who must conform to coercion. Workers

104 Individuals, culture and behavioural institutions of this kind obey with a mental reserve, and practice reluctant obedience. Whenever possible, they resort to defence devices which take the form of varying degrees of disobedience, from shirking to sabotage. What dismal science calls ‘opportunism’ is in reality a repeated attempt at salvation. For this kind of worker the poet wrote: ‘freedom is the pleasure of not performing a duty’. In fact, if obedience annihilates personal autonomy, disobedience aims to re-establish it. Musilian workers may not need either a political or a company ideology, and tend to develop bad sentiments, rage, rancour, envy, maybe even class hate, but do not know what to do with them. Musilian workers work and dream, and sing with the voice of Pedro Pietri (1973: 21–22): Everybody hates their jobs everybody hates their take-home pay everybody hates New York state taxes everybody is praying for better days everybody is waiting for messiah ………………………… everybody must work until they have saved enough money for a good down payment on a semi-decent credit card funeral A jump in quality takes place when reluctant obedience is transformed into political self-discipline. Then a new figure of worker is borne who is able to change class hate into class action. The immediate relief brought about by a small act of disobedience is recognised for what it actually is: a simple expedient without ends and without prospects. The drive to resist the loss of autonomy is elaborated and transformed into an aspiration to freedom. The necessity to carry on an activity which serves the economic aspirations of some other party is transformed into a means to achieve a self-determined political aspiration. Workers of this kind realise that work and coercion can be overcome, but also that they cannot be overcome in an immediate way or with an individual action within the confines of the assembly line. Thus they tend to develop a historical and social vision of their being – a vision by which the tragedy of their life condition acquires a sense in the perspective of a social action aimed at making history. These workers realise the impotency of shirking, it being an expression of impotency and an ineffective means of liberation. Therefore they may even decide to practise passive obedience in the labour process. But, if they do so, it is to serve a superior goal. Nietzschian workers practise preventive commitment. They resist alcohol, drugs and company ideology because they know that the causes of their condition and the way out must be sought elsewhere. If the three abovementioned forms of obedience are three kinds of conscience elaboration for the acceptance of the worker condition, political selfdiscipline differs from them in that it lays the foundations for the rejection of that condition. The consciousness of servitude – says Hegel (1977: 117) – ‘has been fearful, not of this or that particular thing or just at odd moments, but its whole being has

Individuals, culture and behavioural institutions 105 been seized with dread; for it has experienced the fear of death, the absolute Lord’. The servant-master relationship is born out of this fear. The shame of coercion as it unfolds daily can only be understood by looking at an initial moment in which the ‘fear of death’ generated the relationship.15 The loss of autonomy occurring in the labour process presupposes a more fundamental loss of autonomy – that which compels the worker to enter an employment contract. The condition originating the reduction in freedom of choice to the alternative between wage work and unemployment has been called ‘poverty’. This is almost a euphemism, but less ironical than ‘dependence on the state’ used by scientists. And it is obviously also a synecdoche, as it means much more than ‘poverty’. It means that a social condition exists whereby the great mass of people are precluded access to control of the means of production, that is, of the means of existence. Workers who practise self-discipline transform class hate into class consciousness and understand that rebellion to the assembly line leads nowhere if it does not pass outside the factory gates. These workers understand that the reduction of labour to commodity is only an ideology – an ideology which rationalises obedience in the labour process by making it formally compatible with a freedom of contract in the market. They know that the basic conditions of subjection and liberation exist in civil society and in the political arena, and cannot be confined in the labour process. They fight to conquer their freedom, they also fight in the factory, but do not allow themselves to be confined with it. Countering Hegel, they aspire to free themselves, not through work, but of work. As Marx said, in criticising a young Hegelian: labour is already legally free in all civilised countries; the problem is not to free labour, but to abolish it. On sources The human essence is no abstraction inherent in each single individual. In its reality it is the ensemble of the social relations. Feuerbach, who does not enter upon a criticism of this real essence, is consequently compelled: (1) To abstract from the historical process […] and to presuppose an abstract – isolated – human individual. (2) The human essence, therefore, can with him be comprehended only as ‘genus’, as an internal, dumb generality which merely naturally unites the many individuals. (Theses on Feuerback in 1978: 145) The premises from which we begin are not arbitrary ones, not dogmas, but real premises from which abstraction can only be made in the imagination. They are the real individuals, their activity and the material conditions under which they live, both those which they find already existing and those produced by their activity. (The German Ideology in 1978: 149) Marx dug deeply into the problem of human agency, especially in his early works. He elaborated a profound criticism of the metaphysics of the individual agent,

106 Individuals, culture and behavioural institutions including that of homo economicus; an alternative social ontology in which he focuses on the influence that institutions, culture and the social and material condition of production exert on the formation of ‘human nature’; a conception of the individual agent as a historically and socially conditioned subject, who, however, is seen neither as a social atom, nor as a cog in the machine; finally a theory of culture as an ideological projection of class interests, and science as an instrument of praxis. Marx is one of the four pillars on which the theoretical construction developed here rests. The other three are Thorstein Veblen, John M. Keynes and Herbert Simon. It may seem a daring combination. Yet I am convinced that the doctrines of these four thinkers are the tesserae of a mosaic, that they complement each other, and that they can be used to build a complete theory of social action. Thorstein Veblen is one of the most radical critical economists, certainly the most radical critic of neoclassical economics and its homo economicus fake. Among his most important works, as far as the themes developed in this chapter are concerned, I must refer to some essays collected in (1919), especially ‘Why is Economics not an Evolutionary Science?’ In Veblen there is an attempt to explain human behaviour as institutionally conditioned and institutional evolution as a process governed by natural selection. Yet there is no evolutionist ideology of the kind which was so fashionable at the time of Herbert Spencer and which is being reawakened and refined by much of contemporary neoclassical evolutionary economics. In particular, there is no pretension that evolution always marches toward the maximisation of social welfare or toward optimal adaptation. To Veblen evolution is not teleologically determined. It is a cumulative process of change which, because of the prevalence of predatory instincts and ceremonial attitudes in the ‘industrial civilisation’, may well lead to the establishment of disserviceable and even imbecile institutions. Only his moderate optimism, just as in Marx, induced Veblen to believe that in the very long run man would prevail over the thing. Among the limits of the founder of institutional thought, two in particular deserve to be cited here. There is a tendency to reduce human action to the product of base impulses or instincts, so that the fictitious hedonist psychology of neoclassical economics is often substituted with an equally fictitious psychology of instinct. There is, moreover, a dichotomy between behaviour induced by thought habits and that induced by matter-of-fact knowledge, with the pitfall risk, in which some of his followers fell, of seeing in behavioural institutions only a support to non-rational action and in technology only the product of rational action. Apart from the presence, in Veblen’s thought, of a sort of determinism which is both technological and instinctual, the major lacuna in his thought concerns the justification of institutionalised behaviour: why do individuals tend to comply with institutions if they are not completely without reason? It is this lacuna that accounts for Veblen’s incapacity to understand the fact that the evolutionary selection of institutions is predominantly artificial rather than natural. This lacuna is overcome by Herbert A. Simon’s (1955, 1957, 1959, 1982) theory of bounded rationality. An institutionalist use of this theory must be based on the

Individuals, culture and behavioural institutions 107 argument that people tend to conform to institutions because, in the impossibility to elaborate all the relevant information, they try to simplify problems by following rules of thumb, familiar procedures and behavioural routines. These are nothing less than habitual forms of behaviour and, when they are also socially accepted, are behavioural institutions. Simon’s thesis on procedural rationality seems less interesting and general and more ambiguous than that on bounded rationality, since the sense in which it is rational to follow a procedure is not immediately clear. In fact rational behaviour must be defined in terms of attempts to provide the most suitable means to achieve a given end. On the contrary, to follow a procedure implies sticking to the rules, in a rather Weberian sense, independently of the goals they serve. However, if tolerance thresholds or aspiration levels are attached to procedures, the behaviour governed by them becomes implicitly rational, at least insofar as conformity to the thresholds gives some satisfaction. But then procedural rationality means nothing more than satisfying behaviour. I developed this institutionalist application of Simon’s theses in Screpanti (1995). More recently, Simon (1990) tried to develop a theory of altruism as an attitude determined by socially prescribed rules of conduct. Docile individuals tend to conform to these rules to avoid the shame and guilty feelings caused by transgression. Socially prescribed behaviour consists in institutional practice and gives to action the meaning of implicit altruism to the extent that it embodies social wisdom and information which aim at rendering individual decisions collectively beneficial. A theory which is opposite and complementary to Simon’s has been developed by Keynes (1921, 1936): the information we need to take decisions, far from being too abundant with respect to our computational ability, is often in reality too scarce and sometimes completely absent. Uncertainty, in Keynes, means no knowledge, no information. Even probabilistic information is lacking, and this is a serious problem when crucial decisions have to be taken. In these cases institutions do not help: they induce habitual behaviour, but when you have to decide on an important investment, a speculation, risk-taking linked to the abandonment of money, in a word, when decisions are involved that imply a prominent change in behaviour, routines cannot assist. People must form some outlook on the future state of the world, and on this basis they must decide whether it is worthwhile changing their habitual conduct. Keynes maintained that, in this kind of situation, individuals act under the pressure of collective moods, the state of confidence, animal spirits, long-term expectations, optimism, pessimism etc. These moods are formed through some psycho-social interaction processes of a fleeting nature, like the ‘beauty context’, in which social conventions and highly volatile mental feelings play a decisive role. Only when moods do not encourage change, do people stick to some kind of precautionary behaviour, for instance those induced by liquidity preference. This type of conduct too is affected by social conventions, but its routine and conservative nature likens it more to institutionalised behaviour. Among all Keynes’s followers, the one who developed his theory of decisionmaking under uncertainty more creatively, is perhaps George L. S. Shackle (1969, 1972, 1979). What Shackle taught us is that mood-determined economic actions are the opposite of routine behaviour, since they involve decisions which are

108 Individuals, culture and behavioural institutions intrinsically innovative, in which social actors try to ‘impose upon the material chaos a psychic order of (their) own invention, not seeking to solve a problem but to conceive a work of art’ (1966: 755). In more recent times the ideas of these great thinkers have been reformulated and recombined in various ways to lay the philosophical foundations for a contemporary institutionalist revival. A new approach is arising from this renaissance that could be labelled ‘radical institutionalism’ – radical, because it is subverting the very roots of mainstream economics. There is no room here to do justice to all the interesting works which are contributing to this revolution. Suffice it to refer to the most stimulating ones. The first is the work of Richard R. Nelson and Sidney G. Winter (1982), a book following in Schumpeter’s footsteps, and focusing on the evolution of technological routines and the growth of firms, rather than the evolution of institutions in general. More general explanations of institutionalised behaviour have been attempted by B. L. Loasby (1976, 1999), P. E. Earl (1983), R. A. Heiner (1983), G. M. Hodgson (1988, 1993) and S. Hargreaves Heap (1989). I found the latter particularly enlightening in the way it tackles and subverts the traditional approach to rational behaviour. A peculiar tendency of modern institutional thought, especially the brand developed by sociologists and organisation theorists, is to propose a constructivist view of social reality. Some prominent contributions in this field are collected in W. W. Powell and P. J. DiMaggio (1991). Enlightening is that of W. R. Scott (1994). Post-Marxian thinkers are today also developing constructivist approaches to social theory. See, for instance, D. F. Ruccio (1986, 1993), B. Hindess (1987), J. L. Amariglio and T. Callari (1993), J. L. Amariglio and D. F. Ruccio (1994), T. Carver (1998). A theory of ‘superior needs’ was developed by D. C. McClelland (1987), who focused on what he calls the ‘need for power’, the ‘need for affiliation’ and the ‘need for achievement’. The originator of this line of research was A. H. Maslow (1970), who discovered that human actions are moved, not only by the basic, lower needs privileged by utilitarian economists and functionalist sociologists (survival, safety and belongingness), but also and especially by two higher needs, namely the need for recognition and the need for self-actualisation. F. Fukuyama (1992) drew attention to two different kinds of need for recognition, one centred on ‘virtues’ and one on ‘standing’, and then distinguished two forms of the latter, ‘megalothymia’ – the aspiration to superior status – and ‘isothymia’ – the aspiration to equality. J. O’Neill (1998) showed the positive role that aspiration to deliverance and merit recognition plays in the working of a post-capitalist society. U. Pagano (1985) considered the possibility that some kinds of labour activity may cause gratification beside disutility, and highlighted the implications for social organisation. Particularly enlightening are some essays by J. Elster (1979, 1983), who, in spite of an obsession that continually drives him to measure the real forms of behaviour on the benchmark of the unreal neoclassical approach to rationality, has contributed, among other things, to call attention on the role played by sentiments and impulses not reducible to the utilitarian angst.

Individuals, culture and behavioural institutions 109 Elster’s essay on Emotions and Economic Theory (1998) contains a promising attempt to overcome the traditional economist’s approach to sentiment-induced behaviour by taking into account some recent findings of cognitive psychology (de Sousa, 1987; Johnson-Laird and Oatly, 1992; Damasio, 1994; LeDoux, 1995, 1996) on the role played by emotions in determining rational behaviour. The fact that current literature still does not clearly distinguish between sentiments and emotions is a source of confusion which can probably be attributed to the immaturity of this field of research. A branch of cognitive psychology I have found particularly useful is that dealing with cognitive dissonance, which is at the base of the theory of self-deception devices. The classical work is L. Festinger (1957). A few recent contributions from which I have taken inspiration are A. Tversky and E. Shafir (1992), W. Doise (1993), M. K. Johnson and S. J. Sherman (1990), P. Legrenzi, V. Girotto and P. N. Johnson-Laird (1993), E. Shafir, I. Simonson and A. Tversky (1993). Elster (1986) also contributed to call attention on the problem of the multiple self. I. Steedman and U. Krause (1986) put forward stimulating and innovative arguments on this theme. One must not forget that Adam Smith had already implicitly elaborated the view that human personality is endowed with contrasting sentiments – sympathy, altruism, egoism, vanity, avarice and a love of power and social esteem. However a first rough but explicit elaboration of a theory of the multiple self can be found in Maffeo Pantaleoni (see N. Bellanca and N. Giocoli, 1998). I have previously pointed to the multiple self as a major reason for behavioural indeterminacy. Some students have tried to cope with multiple preferences by resorting to concepts like ‘second-order desires’ (H. G. Frankfurt, 1971) or ‘metaranking of preference orderings’ (A. Sen, 1974) or ‘preferences among preferences’ (R. Jeffrey, 1974) or ‘metapreferences’ (A. O. Hirschman, 1982), all of which convey the idea that people have the ability to choose rationally among preference orderings. This line of attack is not so convincing, if at all, because it leads to the question: why people who enjoy multiple preferences do not also experience a personality division on metapreferences? It is even less convincing if one considers that it helps to get rid of the real problem rather than to cope with it – the problem of indecision and indeterminacy of behaviour. But the problem is there, and is today being acknowledged by an increasing number of theoreticians, as is proved by the growth of literature on the ‘weakness of will’, ‘regret’, ‘self-deception’ and ‘irrationality’ – all cases in which the existence of metapreferences is irrelevant. See, for instance, J. Elster, 1979 and A. Rorty, 1980. Finally, to return to Marxism, the contribution by Antonio Gramsci (1975) on the theory of ideology cannot be ignored. With his notion of ideology and culture as practical instruments of collective action and as means of formation of class consciousness and the collective will, Gramsci furnishes the necessary correction to the positivist conceptions which were so fashionable in the Second and Third International Marxism. Ideology is not just a deformation of ‘objective knowledge’, nor is it simple ‘superstructure’.

110 Individuals, culture and behavioural institutions The pejorative meaning of the word has become predominant and this has modified and distorted the theoretical analysis of the notion of ideology. The process of this mistake can be easily reconstructed: (1) ideology is identified as separate from the structure and it is maintained that ideologies do not alter the structures but vice versa; (2) it is stated that a certain political solution is ‘ideological’, that is, it is not sufficient to alter a structure, it is useless, stupid etc., even if it is believed it can alter it; (3) then it is stated that any ideology is ‘sheer’ surface appearance, useless, stupid etc. Instead it is necessary to distinguish the historically organic ideologies, that is to say, those which are necessary to a certain structure, from the arbitrary, rationalistic, ‘deliberate’ ones. When they are historically necessary they have a validity which is a ‘psychological’ validity, they ‘organise’ human masses, they form the ground on which men act, acquire consciousness of their position, struggle etc. (Gramsci 1975: 868–869) Elster (1983), although entrapped in the notion of ideology as deformation of ‘true knowledge’, developed an acute analysis of the role of ideology in affecting human behaviour.

Notes 1 This is one of the reasons why it is difficult to distinguish between emotions and sentiments. The joys of love, for example, are often experienced simultaneously with love itself. Yet they are only the consequences of actions determined by the love sentiment. The difference between emotions and sentiments immediately comes to the fore when one realises that the same sentiments can bring about quite different emotions. For instance, an action initiated under the urge of passionate love could end up by producing angst instead of joy. 2 Provided there is an equilibrium which is unique and stable. Otherwise other kinds of criticisms could be raised, and the very notion of ‘rationality’ would become blurred. 3 Note that the term is not used in Simon’s sense. 4 If one wished, beliefs themselves could be considered as institutions. It would be sufficient to refer to a common definition of institution as a ‘way of thought’ or ‘mental habit’. Then one could speak of ‘cultural institutions’ as socially diffused mental habits, and even of ‘ideological institutions’ as a subset of them. Although I think this extension of the meaning is acceptable, I prefer to avoid it. The reason is that, while transaction, behavioural and normative institutions (defined respectively in chapters 1, 2 and 3) directly affect human behaviour, culture, beliefs and ideologies affect it only indirectly, i.e. because they influence the perception and image of the world on the ground of which motivations are formed. 5 Note that, to avoid confusion, I use the term ‘decision’ when speaking of the selection of means, the term ‘choice’ when dealing with the selection of goals. 6 Of course transaction institutions too have behavioural implications, since any socially sanctioned manifestation of will binds the individual to follow some determined forms of behaviour. Behavioural institutions in the strict sense differ from transaction institutions because they do not necessarily imply manifestations of will that produce effects on other parties, nor the acceptance of social obligations. 7 Thus, since the maintenance of the status quo may be one of the available options, one should speak of ‘non-worsening’ rather than ‘improving’. If preservation of the status quo is to be considered satisfying, the reservation thresholds have to coincide with the given situation. 8 A great deal of inventions are produced by ‘pure’ scientific research. The process of invention generation, however, is always institutionally conditioned, even in scientific activity. Science

Individuals, culture and behavioural institutions 111

9 10 11

12 13

14

15

develops in a social and cultural context which is strongly institutionalised. A theoretical system accepted by the scientific community is none other than a system of mental habits regulating research activity. A theoretical system reduces the complexity of the world of ideas and restricts the field of theoretical decisions accessible to scientists thus channelling their research. Most of normal science consists of problem-solving activities undertaken in the context of consolidated research routines, and produces small solutions to small problems which do not change the scientific institutions. However it is always possible that research produces epistemological breaks, i.e. scientific inventions in the full sense. However, the loss of autonomy, as I will show later on, does not always lead to unintentional action. This representation was suggested to me by Luca Fiorito. The reader who does not feel at ease with these assumptions and who prefers to stick to a more traditional line of reasoning could modify my treatment by making the following changes: 1) re-label goods x and y as ‘potatoes’ and ‘lemons’; 2) assume their relative price is constant; 3) draw a budget constraint which is tangent to both preference curves in c and b; 4) interpret lines Oc and Ob as paths of consumption expansion induced by income growth. Note that, when achievements are lower than or equal to cr, the individual suffers absolute frustration, i.e. he experiences no satisfaction at all. If one prefers to assume that power accumulation is not a real impulse, maintaining that the deep aspiration is valorisation of the manager’s human capital, the result of this analysis does not change in essence, for there is a contrast anyway between managerial slack and the desire to build a reputation of managerial efficiency. I must add that, with the passing of time, democratic organisations tend to assimilate the dominating class ideologies, thus transforming themselves into diffusion apparatuses of adverse ideologies, and compelling the oppressed classes to continually re-found their own organisations. At the basis of this tendency there are two processes which are accounted for in social theory by means of the median voter theorem and the iron law of oligarchy. Since I will deal with them at greater length in later chapters, I will only briefly mention them here. According to the median voter theorem, the left wing parties who aim at conquering power democratically are normally compelled to attract the moderate voters, even going against the fundamental interests of the classes they should defend, and therefore to modify their political programmes and ideologies so as to make them compatible with maintenance of the existing social, economic and institutional structure. The so-called iron law of oligarchy, on the other hand, shows that there is a built-in tendency of organisations, even democratic organisations, to transform their leading groups into power élites interested in maintaining their domination. According to Hobbes (1931, chap. 20), but even to Locke (1960, § 23), it is ‘the Vanquished’s’ will to live that generates, in a more or less implicit covenant, the right of ‘the Victor’ over the slave. Hobbes and Locke strengthen this thesis on the ‘voluntariness’ of the enslavement pact (let us set aside the subtlety of calling it ‘drudgery’ or ‘servitude’ instead of ‘slavery’), by observing that the slave preserves anyway a complete exit freedom: in fact, who can prevent him, if he dislikes his slave condition, from rescinding the pact by committing suicide or disobeying the master (thus justifying him in giving death)? This theory looks like the argument by which many scholars ‘prove’ the real freedom of the wage worker with the observation that he preserves anyway the faculty to leave his job.

112 The state and normative institutions

3

The state and normative institutions

The incompleteness of institutions All institutions are incomplete, in the sense that they cannot fully determine the social actors’ behaviour. This is the principle of institutional incompleteness. I have already used it in the preceding chapter, but I will now focus more closely on it. The main reason for incompleteness lies in the unpredictability of change. One of the primary causes of change is stochastic variability in the natural environment. This is inescapable: however strong the human effort to dominate nature, a significant residue of surprise will always remain. But even more important is social change. Individuals are heterogeneous, since they are endowed with different knowledge and different aspirations, sentiments, interests and values. Therefore the outcome of social interaction remains intrinsically unpredictable. Moreover egoism contributes to complicate things, since the propensity to co-operate cannot be taken for granted, with a consequent higher probability of the emergence of conflicting behaviour. In a society where social relations are fundamentally antagonistic there is strong uncertainty. In fact co-operation implies also the sharing of knowledge, the co-ordination of actions and the definition of collective goals. Non-co-operative behaviour prevents all this and contributes to the unpredictability of the social environment. The incompleteness of transaction institutions implies that transactions are unable perfectly to determine the claims and obligations of social actors and therefore their actions and their interactions. However precise the manifestations of will expressed in a transaction, they never succeed in defining behaviour in every situation. The same is true, as observed in Chapter 2, of behavioural institutions, which bind social actors to habitual forms of conduct. The unpredictability of change weakens these constraints and makes them unable fully to determine human behaviour. There are several forms of incompleteness, but five deserve particular attention: civil, political, operational, disciplinary and behavioural. Civil incompleteness results from the fact that the prerogatives constituted in a transaction are founded on the obligations undertaken by the parties who directly engage in it, so they are not valid universally, and cannot count on the compliance

The state and normative institutions 113 of all social actors. This kind of incompleteness is particularly relevant in defining property rights. In Chapter 1 I assumed that the foundation of these rights lies in transaction institutions. But a foundation is not a complete building. Insofar as it tends to constitute rights, a social action initiated by the parties to a transaction needs to be completed by normative action from a superior social sphere. A right is a universal claim, it must hold for all members of a given society. Private parties are unable to constitute rights under an agreement if the transaction institution they use is not complemented by law. Political incompleteness, does not pertain only to third parties but also affects the relations of the contracting parties in a transaction. In a society in which conflict and competition prevail, the obligations regulating a relation are normally the result of bargaining power. This is evident in the relations set up in employment contracts. But it is also apparent in many other relations, if one considers that the principle of equivalence, assumed for the sake of abstraction in Chapter 1, does not in fact operate in any real society. If one removes this assumption, it is easy to understand that even a simple sale contract establishes a transaction equilibrium which is none other than a balance of forces. But bargaining powers can change in time and become incompatible with the obligations undertaken in a transaction. This may cause the re-negotiation or breach of the agreement. However detailed the clauses of a contract, it is impossible to foresee all the possible actions of the parties. And even assuming perfect loyalty from all the social actors, the transaction clauses are unlikely to be so precise as to avoid contrasting interpretations. A third source of incompleteness has received much attention in contemporary neo-institutional literature. A transaction constitutes obligations pertaining to the economic goods which are the object of the transaction. But a change in environmental conditions may induce or necessitate a change in the characteristics of the object. And it is difficult to define obligations envisaging every state of the nature or society. Again in this case there is no need to assume that the contracting parties are not loyal to account for institutional incompleteness. Even if they are convinced they are respecting their commitments, the parties may formulate different interpretations on the best way to cope with the effects of unexpected change. It is unlikely that the terms of a transaction are so well defined that they solve all the problems in an equally acceptable way for all the parties. This is operational incompleteness. In addition, if one removes the assumption of perfect loyalty, another problem arises, that of enforcement. In the case of a change in the interests and aspirations of social agents, a party may find it opportune to break the agreement signed in a transaction and decide not to honour its obligations. Few transactions are completely self-enforcing. Since agreements are manifestations of the parties’ will, they are enforceable only so long as that will persists, unless an external arbiter is appealed to. But this does not always help to solve the problem, if the arbiter’s authority depends on the consensus of the parties. In this sense transaction institutions are burdened with disciplinary incompleteness. Finally there is a kind of incompleteness mainly involving behavioural institutions. Since these constrain a person to follow a habitual conduct, they do not help

114 The state and normative institutions much in coping with environmental changes. Even the problem-solving activity which normally accompanies habitual behaviour may be insufficient to overcome all obstacles, especially when these are caused by relevant changes in the situation. Moreover there are certain kinds of behaviour that social actors may be unable to adopt even when it is in their interests to do so, for example when they do not possess all the relevant information or the necessary means to make the right choices. Not all people, for instance, choose to use safety belts when driving a car, or take preventive health measures, or invest in their own ‘human capital’, even though it would be worth their while to do so. In these cases behavioural institutions are incomplete in the sense that either they are not adopted as normal forms of conduct by those who should be interested in doing so or they are not adopted in the right way. This is behavioural incompleteness.

Where the state comes from Social conventions Normative institutions are born out of the need to reduce the incompleteness of transaction and behavioural institutions. They are the socially recognised rules which regulate relations in a social aggregate. Some kinds of normative institutions originate from inside civil society and are considered valid by social actors without needing the sanction of a superior legal body. These are social conventions. They consist in civil society’s attempts at self-regulation and hold so long as the social actors comply with them. Two kinds of conventions are especially important: customs and regulations. Customs hold in informal communities like small towns and districts and arise in a spontaneous and non-deliberate way. They are unwritten but generally recognised rules. Their rationale lies in the interest that an actor develops in the consideration shown by other social actors of his manifestation of will. Every actor desires the prerogatives acquired through transactions and the outcomes of actions achieved through behavioural institutions to be recognised and respected – not only by the specific parties encountered in any transaction, nor only by the actors involved in direct interactions, but by everybody. In reality many people tend to conform to this desire for reasons of reciprocity. Most individuals try to behave towards others in the same way as they expect others to behave towards them. Perhaps the reason for this attitude lies in the well known folktheorem by which, in a context of repeated interaction, an actor will react to the actions of another in the same way as the other has reacted to his actions. Conformity to a convention in reality implies loyalty on the part of an actor, not just towards any other single person with whom he directly interacts, but toward all the members of the community in which the convention holds. This attitude of conformity to social conventions is justified by the exigency of every actor to maintain a reputation of sociability which stands as a sign of his disposition to fairness and correctness in any specific interaction. In other words, through his reputation, an individual manifests his willingness to start any interaction with a

The state and normative institutions 115 co-operative move, expecting in this way to obtain initial co-operative actions from others. Since the strongest and most widespread interactions are those regulated by transaction institutions, it is mainly in the activation of these that social actors produce social conventions. A transaction is a social act: it does not involve the interested parties alone, but is manifested to many other third parties. A transaction institution conveys to the social aggregate a message regarding the obligations and claims constituted through the transaction and demands everybody’s recognition of them. And, since it is a socially accepted form of transaction, it often induces that recognition. Institutionalised transactions are usually associated with social conventions. Social actors try to overcome the civil incompleteness of transactions without resorting to legal bodies which are external to the community. However, the folktheorem also implies retaliation, that is, non co-operative reaction to non co-operative moves. Transgression of a custom is interpreted by people as an attack on the community: not just an offence to the specific person who directly bears its negative consequences, but a threat to all the other actors who might interact with the offender in the future. Therefore the latter may experience a non co-operative reaction from all the members of the community, and not just from the individual who suffered the damage. Ostracism, boycotting, exclusion and contempt are some of the most efficacious instruments used by a community to enforce compliance with customs. In this way social actors try to overcome the disciplinary incompleteness of transaction institutions. It would seem that civil society is fully capable of self-determination: it is selfconstituted through transaction institutions; it induces conformity to behavioural institutions; it spontaneously produces the rules which complete those other institutions. Unfortunately this is not the case. The real problem is that most social conventions are valid only within rather small communities. And there is a good reason for this limitation. A single actor conforms to a custom expecting that the same attitude will prevail among the other persons with whom he thinks he has a high probability of interacting. On the other hand, a group of persons may enact non-legal punitive action only towards individuals with whom a repeated interaction has been set up. Although an actor does not necessarily have to interact directly with all the others, he must interact with a series of other individuals to whom he attributes a high probability of interaction with everybody. Only in this context can one expect a reputation to gain credit and a punishment to acquire efficacy. Social conventions regulate relations also in collective movements and organisations. Movements are informal sets of social actors who act together in the pursuit of common ends. They differ from communities mainly because their members are motivated by common goals and therefore have a strong reason to co-operate. Yet they too require social conventions and coercion instruments, especially for coping with free riding. Since they are informal aggregates, their normative institutions are unwritten and often not even explicitly formulated as well-established customs. Nevertheless the pressure of the group on the individual is strong. The enforcement mechanisms are broadly the same as those operating

116 The state and normative institutions in other social aggregates that are not dominated by an external authority – boycotting, ostracism, mockery, exclusion – although they are often used with greater efficacy and cruelty in a movement. Collective movements differ from communities in another aspect: since they are constituted for the pursuit of a given goal, they are short-lived, and tend to disband or to transform into organisations after achievement of their aim or recognition of its impossibility. Communities have a long life, longer than that of the individual actors belonging to them. Therefore their customs are sustained by the strength of tradition. In movements, by contrast, social conventions die with the dissolution of the social aggregates or are transformed with them, and typically must be reinvented each time a new movement is created. Another kind of collective actor which is capable of at least partial selfconstitution and self-regulation is an organisation, which differs from communities and movements in several respects. Like movements and unlike communities, organisations are social actors moved by collective goals, although in many cases, as in capitalist firms, not all the members have the same goals. Unlike movements and like communities, organisations often last longer than their members. Therefore their conventions are well-established and long-lasting. In addition, a division of labour operates in organisations that justifies the existence of social bodies endowed with deliberative and judicial functions – functions which are exerted either by mandate of all the members (in associations) or by virtue of the power created by the members’ obligations (in capitalist firms). Therefore social conventions normally arise through explicit and formal deliberation rather than in a spontaneous way, and take the form of internal regulations. These, unlike customs, are written rules. They may include formal instruments of coercion that are quite different from those contemplated by customs, as for instance fines, firing, transfers, warnings. Finally, as in communities and movements, the conventions operating in an organisation have local scope, since they cannot be imposed on external actors. Summing up, one could say that the self-determination ability of civil society is not confined to the relations constituted through transactions, but, in the attempt to overcome the incompleteness of transaction and behavioural institutions, it manifests itself in various processes of rule production. Social conventions try to reduce the disintegrating effects of civil and disciplinary incompleteness of those institutions by imposing to all the members of a social group a correct behaviour. But social conventions are also able to develop mechanisms to overcome operational and behavioural incompleteness. In organisations, for instance, regulations define procedures for the solution of new technical problems. In communities there are customs enabling the members to prevent or cope with disasters or unexpected adversities. However, customs and regulations are valid only within small groups, so that the problem of incompleteness remains open. Particularly in the regulation of relations among groups or members of different groups it is necessary to resort to normative bodies which are external to the groups themselves. Conventions, moreover, may not even locally resolve the problems caused by political incompleteness. In fact, when the bargaining strength among members of a group changes, there may be a tendency to modify the institutions on which the structure

The state and normative institutions 117 and the working of the group itself depend, so that it is difficult to cope with disintegrating drives caused by change. The state, the laws and government action The inability of civil society to completely define its normative institutions makes it necessary to resort to superior rules – these are laws – and therefore to an organisation which is external to the civil society itself and which is able to produce and enforce laws – this is the state. A transaction, giving rise to obligations and claims, potentially produces rights. Property rights are especially important. These are absolute titles and should hold universally. A transaction also produces prerogatives and obligations of contracting parties which often have effects on third parties. They may produce losses and advantages to actors who did not participate in the transaction. The civil incompleteness of a transaction consists of the possible production of externalities on actors who are not directly involved in the transaction itself. Furthermore transaction institutions imply some freedom of social actors and, first and foremost, freedom of contract. The very idea of a transaction as a manifestation of the parties’ will presupposes that the parties are endowed with personal rights. The problem is that, in a social system in which egoism dominates individual behaviour and competition social relations, the actors tend to develop a tendency to expand the sphere of their claims even to the detriment of others. The state tries to solve these problems through laws. But only insofar as it is not part of civil society can the state act toward it as a superior legal body. Only by virtue of this characteristic can the institutions resulting from its manifestations of will be considered stronger than transaction and behavioural institutions. This principle is important to understand the reason why the state is a fundamental component of the capitalist system. Since freedom is a basic institutional condition of the capitalist mode of production, and since civil society is prevented by its competitive nature from establishing and guaranteeing completely the freedom of all its social actors, the state, which endeavours to do this, is a necessary institutional condition for the existence of capitalism. No model of ‘pure capitalism’ represents the social and institutional structure of capitalism correctly if it is only based on two subjects – the capitalists and the workers. However pure and abstract the model of capitalism one adopts, it must be based on at least three autonomous subjects: the workers, the capitalists and the state. Laws are considered the source of rights. They constitute and give form to rights, and define the way in which these can be applied. The state carries out a precise activity of institutional delimitation. It defines the boundaries of civil society over which it performs its action; and it defines the limits of the action it can exert on social actors. In reality, insofar as the state claims to be the source of rights, it produces a sort of legal hypostatisation of rights. If it is true that the ultimate foundation of rights lies in manifestations of the social actors’ will, what the state actually does is give legal sanction to the institutions created in civil society. On the contrary, the law

118 The state and normative institutions presumes to be the primary source of rights. This is at the same time both an ideological and a practical fact made necessary by the inability of civil society to constitute itself completely. Legal hypostatisation is necessary in a practical sense because the state’s capacity to make up for the normative deficiency of civil society depends on its ability to appear as a superior legal body. The institutions it produces must be stronger than those produced by civil society, so that the latter must be considered valid only if they are compatible with the former. Laws, therefore, must be considered as primary institutions, as the real sources of the rights and duties resulting from the interactions of the social actors, and not just as legal sanctions of them. Nevertheless this phenomenon implies an ideological deformation because in reality the state does not constitute civil society, but only tries to complete its self-constitution. The state’s legislative bodies are parliament and the judicial authorities. Judges, too, produce norms, through judgements, sentences and orders, either by interpreting the laws issued by parliament and constitutional laws or by making up for their deficiencies. This function is particularly important in common law systems, in which the courts are continually called upon to sanction or legalise transaction institutions. But it also works in civil law systems due to the incapacity of general laws to adapt promptly and precisely to the evolution of civil society. In fact even laws, despite their function of completing the institutions generated in civil society, leave open an unavoidable residue of incompleteness. All institutions, including normative ones, continually try to keep up with social and environmental change but, since this is incessant, never succeed in fully regulating it. Therefore, notwithstanding the state action aimed at institutional completion, the principle of institutional incompleteness is ever present. A state, however, does not reduce to its legislative and judicial bodies. Equally important are its coercion and government apparatuses. And the rationale of these articulations also lies in the necessity to make up for the incompleteness of transaction institutions and social conventions. The judicial and repressive apparatuses try to cope with disciplinary incompleteness. In this case an enforcement service is offered by the state – a service which the social actors are for the most part unable to perform by themselves. In dealing with political incompleteness, the state assumes the role of an arbiter, which it does by using both the judicial and the government apparatuses, and particularly the economic ministries. The courts perform an arbitration as well as an enforcement and a legislative function. Whenever the changes in bargaining strength of the parties involved in a transaction give rise to contrasting interpretations of claims and obligations, the courts settle the differences of opinion by emanating judgements which are considered valid interpretations of agreements and rules. The validity of judicial interpretations does not depend on their intrinsic equitableness, but only on the fact that they are manifestations of the will of state bodies and therefore are preponderant with respect to the will of the social actors. An arbiter’s judgement may be perceived by a party as arbitrary and unjust, yet, once pronounced by a judge, it is considered valid and binding.

The state and normative institutions 119 As to behavioural incompleteness, there is a vast series of events in which the state claims to know the social actors’ interests better than they do themselves. This is often a well founded pretension, given the basic asymmetries in the distribution of information and cultural endowments that infect all capitalist societies. In addition the state has the capacity to offer to social actors, or impose on them, choices they would not be able to make if left to their own devices, especially because of the inequality in the distribution of wealth and income. Just think of the role played by the state in supplying merit goods, health services, public insurance, education etc. In these cases the state tries to compensate for the incompleteness of behavioural institutions by legally enforcing social habits, by reforming others, or by offering the means required to make them accessible. Finally, operational incompleteness is the most difficult to overcome. Recall that it consists in the impossibility of defining the obligations in a transaction regarding unpredictable events. The state can do a lot of things to cope with this kind of problem, both ex ante and ex post. At micro-economic and micro-social level, for example, it can issue bankruptcy laws, it can pay unemployment benefits, assist the disabled etc. But it can do even more at macro-social and macro-economic level, since it can govern change in such a way as to reduce its negative effects on the mass of social actors. More generally one could say that the state tries to make up for institutional incompleteness in two ways: by normative and government actions. With the former it contributes to limit and regulate the social actors’ behaviour by means of laws. Note that normative action does not consist in determining the actions of social actors, which are in fact decided within the civil society in the pursuit of aims and motivations that cannot be decided by the state. Normative action, in reality, boils down to the production of rules which normally take the form of prohibitions and limitations to social action. For example the transformation of a sale contract into a legal transaction renders the effects of a sale valid for everybody. A sale transfers the ownership of goods. But a person’s property right over the goods is sanctioned by the law and is equivalent to prohibiting all other actors from interfering with the former actor’s use of the goods. Government action acts on the effects of institutional incompleteness rather than on the incompleteness itself. It does not try to pose limits on social actions, but endeavours to cope with their negative consequences. In a society based on competition and antagonism, actors often perform actions which cause losses to others. These actions cannot in general be effectively dealt with from within civil society. A government action is required, which is accomplished by the state through its executive and administrative bodies, rather than legislative and judicial ones. Before further analysing the structure and working of the capitalist state, it is necessary to touch upon a notion which will be dealt with in more detail in Chapter 6, that of an institutional system. The institutions holding in a given society are normally not all consistent with each other. This is an unavoidable consequence of individual heterogeneity and institutional plurality. Yet a society cannot work and survive for long if the institutions regulating it are chaotic and contrasting. In particular, any given mode of production is characterised by specific structures of

120 The state and normative institutions social relations and therefore by particular institutions which tend to form an organic whole, that is, to be organised into a system. An institutional system is the set of institutions moulding a mode of production or a specific form of a mode of production. Some of them are organised in a consistent way so as to sustain each other in regulating social relations that are typical of the mode of production itself. In an institutional system it is possible to single out the core institutions and distinguish them from those considered to be peripheral. The former govern most of the transactions in a given society. They cannot be modified singularly or independently of each other. They cannot be modified without changing the entire system. For example, it is not possible to regulate the productive use of labour under an employment contract on a wide scale without endowing workers with freedom of contract and without creating a legal and political system which attributes some fundamental rights to all citizens. Peripheral institutions are not necessary for a society to work, they are often at odds with core institutions and can be changed singularly without affecting the structure of the system. They are sometimes useful in achieving a better adaptation to particular environmental and cultural circumstances, but they may also hinder the working of a society. They are tolerated for as long as they do not dramatically impair it, sometimes even longer. For example, a religion which severely condemns usury may encumber capitalist development since it impedes efficient use of liquid assets and credit. Yet, as we learned from history, capitalism can survive and prosper even in Christian and Muslim countries, where usury is considered a mortal sin. The state plays an important role in the stabilisation of a society and is the principal guardian of coherence of the institutional system that regulates it. Since institutions serve to constitute social relations and to keep civil society together, the state, with its normative and government action, works at systematising institutions and guiding institutional change. This it does by continually producing and modifying the rules so as to make them compatible with each other and adapt them to environmental and cultural changes. The political class and the state Functionalist theories of the state, those accounting for its existence on the basis of its functions, are not very enlightening, unless they indicate the social forces that enable it to perform those functions. Functionalism is a defect of some Marxist theories, but also of many neoclassical and new institutional approaches, for instance those which try to explain the state as an instrument for the maximisation of social welfare in a context in which markets miss this target. To justify the latter argument, it is not sufficient, let alone reasonable, to assume that all social actors maximise their individual utility, since social welfare simply does not derive from the summation or a particular combination of individual preferences, especially if individuals are assumed to be opportunists. Nor is it sufficient to assume a postulate of maximisation of total value in any transaction. Apart from the problems this postulate poses in the presence of wealth effects – that is when wealth, income,

The state and normative institutions 121 cultural endowments and information are distributed asymmetrically – it still remains to be understood why the parties in a transaction should aim at total welfare rather than individual advantages. In either case, however, there is a lack of analysis on the social forces underlying the state formation. In sociology and political science there is a great tradition of thought, started with Machiavelli and continued with Mosca, Pareto, Michels, Wright Mills, Lasswell and many others, which has worked out the problem of the state formation in the simplest and most realistic way, that is, by postulating the existence of a class of particular actors who are devoted to political activity. This class makes a profession out of the conquest and management of political power. As a homage to Gaetano Mosca, I insist on calling it a ‘political class’. It is the head of the state. The state’s body, i.e. the administrative apparatus, is the instrument through which it exerts its action. Note that this theory does not begin with the state and its functions to account for the nature and action of the ruling class. Quite the contrary, it begins with the action of the political class to account for the existence of the state. The political actors are the individuals and groups of individuals who devote themselves to the conquest of political power. The political class is the set of all the political actors who have conquered power and therefore have title to issue and impose laws and government measures, that is, to pose limits and constraints on the actions of social actors. It is composed, first, by the top management of the executive and legislative powers; second, by judges at all levels; third, by senior bureaucrats, civil servants and public managers who have discretionary power to apply the legislative and government measures; fourth, by high ranking officers of the army and police force; and finally by the so called ‘vestal class’,1 that is, the experts and intellectuals who are part of the state apparatus and carry out the activity of ideology production. The manifest motivations of political actors, i.e. those they declare to themselves and the world, are always of an ideological nature and boil down to the pursuit of some collective good. Ideological, because the definition of the collective good cannot remain unaffected by the real interests of the actors who formulate it, nor by the vision of the world through which the problems and the interests of the community are interpreted. Political actors, unlike social actors, are officially motivated by objectives which are manifestly not their own. Therefore altruism, rather than egoism, predominates among them, at least ideologically. After all, self-deception devices, especially of the idealisation type, are normally used by people who make a profession of power handling. Yet there are latent impulses, which are properly reducible to power. Of course rent seeking and the desire for status recognition are also important. In Chapter 4 I shall investigate further the nature and function of power, as well as its relation to other motivations of human activity. Here it is sufficient to postulate that power, status, prestige and income are isomorphic variables and that they are increasing functions of the position occupied by an individual on a hierarchical ladder. This enables me to reduce all the egoistic motivations of political actors to one called ‘power’.

122 The state and normative institutions The existence of latent motivations beside manifest ones seems to pose a twofold problem in political analysis. There is the problem of reconciling the egoistic nature of latent motivations with the altruistic nature of manifest ones, that is, to explain how it is possible that the action of political actors is simultaneously moved by two contrasting motivations. There is also that of accounting for the existence of impulses which are hidden by definition and which the actors often do not manifest even to themselves. In reality, if interviewed on this problem, the political actors would hardly deny their interest in power. But they would hasten to add that, to them, power is a means, not an end, a means used in the pursuit of an altruistic end. This is no problem. The existence of two apparently contrasting motivations does not render political action contradictory, since the two impulses sustain each other and concur to justify the following postulate of power-seeking behaviour: political actors aim at the conquest and conservation of power. It is irrelevant whether the manifest and surrogate motivations prevail over the latent or are just the product of an idealisation device, for in most cases of political action the goals pursued by either motivations are isomorphic. And those who prefer to stick to Machiavelli can easily justify their tastes with Occam’s razor: if the social actors who live in a capitalist system are assumed to be prevalently egoistic, why should political actors be thought to be any different? The postulate of power-seeking behaviour is only a provisional assumption, the weakest assumption required to found the analysis of the political actors’ behaviour. Equally weak is the Occam’s razor justification. In Chapter 4, where I deal directly with power, a stronger and more general assumption will be made and the postulate will be transformed into an axiom. Basically there are three instruments for the conservation of political power: ideology, force and service. Ideology has already been dealt with in the preceding chapter and more will be said about its utilisation by the state in the next section. Force, deserves little space. Suffice it to observe that its effectiveness is stronger than it appears, since the state has a legal monopoly on it and can therefore use it sparingly. Social actors see the state as an omnipotent subject and tend to observe its laws and measures also because they are accompanied by force. The monopoly of force, however, cannot be considered the fundamental, nor even the most important, factor of the state’s existence in a capitalist mode of production. In this system the social actors are legally free subjects, and most of their transactions take place within civil society as effects of decentralised decisions. Since the state does not create civil society, but presupposes it, its monopoly of force is not sufficient to account for its relation to it. Moreover it must be observed that, however strong a state may be and however pervasive its ideology, no political class can succeed in perpetuating its power by resorting to these two factors alone, if one maintains that social actors are, at least relatively, autonomous subjects, besides being legally free. The state’s ability to offer services to civil society is a necessary condition for its existence. This is true for three reasons. First, because there is competition among the different political actors. Different groups tend to develop different ideologies, so that the ruling group’s ideology might in the long run be ineffective in holding rival groups at

The state and normative institutions 123 bay. Second, because the grip of state ideology on people’s minds is weakened if it is not accompanied by some deeper reason for consensus. Last and most important, because the rules of political struggle in a capitalist system imply preservation of rights of freedom on the part of citizens, so that the consent of at least a part of civil society is a condition of power conservation. Consensus is obtained through the production of normative institutions and government actions aimed at coping with the incompleteness of the institutions generated within civil society. This is the basic service offered by the state to social actors. Consensus and resources are what it demands in exchange. The present theory of the relationship between the state and civil society is not inspired by the principal-agent model. The political class has basically no other principal than itself. Even if one would believe that political action is moved by the desire to pursue collective welfare, it should not be forgotten that this goal is defined anyhow by the political class itself, by its ideology and its values. As I will show in the next section, political actors may have some reasons for considering themselves endowed with superior capacity in the definition of collective welfare. So that they always consider themselves responsible to their own conscience in the first place. The interaction between the state and civil society may thus be likened to an exchange rather than an agency relation. The state must supply the services social actors require because the conservation of power depends, in the long run, on their consensus. Procedural democracy and class ignorance Democracy is today the dominating political form in capitalism, one could even say: the typical form. I have already observed that, among the fundamental institutional conditions of capitalism, there is the guarantee of personal freedom. Why is this condition so important? Basically because there can be no capitalism if individuals are not endowed with freedom of contract; which consists in the faculty of autonomously deciding to enter a transaction and to define its form and content. This faculty cannot be limited to just a few individuals, for all social relations, in a capitalist system, are based on transactions, contracts and agreements. All people must be recognised as free and equal in their capacity to create social relations by means of transactions. Of course individual rights are not limited to freedom of contract. Yet this is the hard core of the complex juridical and philosophical apparatuses on which the whole theory of fundamental rights hinges. Freedom of contract necessarily calls forth many other kinds of freedom in order to be fully enjoyed, and a concentric series of rights springs and expands from it, from economic rights – e.g. freedom of enterprise and freedom to associate – to the whole set of positive freedoms – of thought, of speech, of the press etc. If individuals are free and equal in the sphere of economic transactions they must also be free in that of political expression. Of great importance are the ‘negative rights’, which consist in the limitations of the state’s faculty to interfere in private decisions, for example, freedom of movement, freedom of domicile, habeas corpus etc. Obviously these are not

124 The state and normative institutions absolute freedoms, since the state organs can indeed search and arrest people. Yet there are strong procedural limitations on these kinds of interference: they must be carried out in conformity with law and by subjects who have legitimate power. An entrepreneur cannot interfere with the correspondence of his employee, nor a Union officer with that of an entrepreneur. The procedural nature of negative rights must be stressed. Understanding this nature gives us a clue as to the meaning and functioning of democracy in capitalism. Democracy, which in principle is the government of all citizens, in a social setting in which people cannot really govern tends to boil down to selection procedures for the ruling class. These procedures ensure that political actors acquire power legitimately, that is, as legal representatives of the citizens’ will – a condition required by the principle that all individuals are equally endowed with fundamental rights which exclude their subjection to any political actor. The real problem is not so much that of ascertaining whether politicians endeavour to be elected to represent the voters’ will or whether instead they strive to represent that will in order to be elected – as some perfidious economists have insinuated; for, even in the second case, the representatives would undergo an agency relationship if the voters were able effectively to control them. The real problem is: do the electors have this ability? Real capacity to control does not depend on formal procedures alone. It also depends on the possession of cultural and information endowments. The ability to control requires some substantial conditions. Without these, democracy reduces to a vacuous form. And the relationship between the electors and the elected is not a principal-agent relationship precisely because the substantial conditions of democracy are lacking in a capitalist society. There is now a vast amount of literature on the subject of ‘rational ignorance’. The tragedy of democracy in a capitalist system is not just that the collection and elaboration of information, particularly the information required to exert an effective control on the political class, are too costly for the typical voter. The real problem is that, in a society characterised by marked asymmetries in the distribution of income, wealth, power and alienated labour, there is also a tendency of ignorance, misinformation and political impotency to be unequally distributed. There is, first of all, a basic asymmetry, which could be defined ‘professional’, between the political and the social actors. The former have made politics their profession and devote 100 per cent of their time to it. Therefore they are better equipped than the latter, who dedicate most of their time to other business. This kind of asymmetry is particularly strong in advanced capitalist systems due to the complexity of their political structures. Thus the ability of social actors to control the political class is rather limited. Nor can the problem be resolved by assuming that people create organisations, like political parties, to which they delegate the control function, since the problem then arises of who controls the controllers. After all, the parties are normally controlled by professional politicians not by the rank and file. But there are also strong inequalities within civil society. There is an education inequality, which is linked to that involving income and wealth. And there is a

The state and normative institutions 125 more subtle form of inequality, which could be defined ‘cultural’, and which entails the capacity, the will and the interest in collecting information – a form of inequality linked to values and opinions affected by tradition and ideology. At the basis of all this, however, there is a kind of inequality which is typical of capitalism, since it depends on its particular form of division of labour: the inequality in the distribution of time devoted to alienated labour. It affects all people who are doomed to spend most of their life earning their bread by working under the command of another person. Subordinate work is the negation of real freedom. The time spent in working is time in which people give up all aspirations to free activity, time during which they do not exist as free subjects. But it is also time subtracted from the acquisition of cultural instruments for the exercise of freedom. How can a worker fully gather and elaborate the information required to produce political action if he has to devote ‘his muscles and his brain’ to producing commodities? Thus the phenomenon of rational ignorance cannot be understood without getting a grasp of its class nature, and perhaps a better definition would be ‘class ignorance’. The classes who are at an economic disadvantage are also culturally and politically disadvantaged. And it is among those who are less equipped and battle-trained that moderate moods tend to form. Moderatism is a precautionary behaviour. The lack of instruments for political knowledge raises the perception of uncertainty and the fear of change. The moderate citizen may be dissatisfied with his state, but does not know and cannot figure out what he can expect from any attempt to change it. He is less informed and perhaps more risk averse than the other citizens, and tends to overvalue the negative aspects of change. He normally lowers his political aspirations and tolerance thresholds and, in the end, contents himself with the status quo. Thus he tends to support the ruling class, whichever it is, provided it guarantees stability. In addition, the classes who are badly equipped culturally are also easier to influence. And this exposes them to ideological control – a control which is exerted by two different types of political force. The first comes from political and cultural organisations of the capitalist class. The social actors who enjoy income, wealth and power concentration are also better equipped with cultural instruments and are able to control most of the ideological apparatuses. The second comes from the political class in power, who enjoys an advantage in producing ideology simply because it has power. The moderate citizens, in their desire to avoid change, tend spontaneously to attribute legitimacy to the state and the ruling class. After all, force itself performs an important function in producing the ideology of power. Saint Paul’s principle (that he who bears the sword is entitled to it, as is proved by the fact that he bears it), perfectly expresses the concept that the possession of force is a factor of its legitimisation. And he who is compelled to obey the laws of a state is often induced by coercion itself to believe in their validity. By recognising their legitimacy, he justifies his own impotency and resigns himself. Most moderate citizens tend to accept the opinions of the ruling class independent of their content and simply because they are the ideas of those who have power. Moreover the ruling class controls a substantial part of the instruments of consensus production,

126 The state and normative institutions schools, parties, mass media, cultural institutions, and has in addition privileged access to those it does not control directly. Finally, let me recall that a part of the political class, which I have called the ‘vestal class’, is composed of professors, intellectuals, preachers, Nobel laureates, Pulitzer Prize winners and other kinds of institutionalised geniuses, who are paid to produce consensus. Most of them are workers of the mind in the position of state apparatus officers. They are what Anthony Downs calls ‘experts’. They are specialised in selecting, analysing and evaluating information. People normally trust them – a trust justified by the authoritativeness of their judgement. But the experts earn their bread by producing opinions, and they wish to increase their prestige, status and income. Since these goods are produced by the power system to which the experts belong, these are particularly prone to supporting the establishment. The oppressed classes are disarmed before this social group since, besides being poorly equipped with critical defence instruments, they do not even have the benefit of doubt. In fact the legitimacy ascribed to the vestal class’s thought derives not only from its position in the social division of knowledge, but also from the fact that its authoritativeness emanates from prestigious pulpits. All this contributes to reinforce the effects of the asymmetry in the distribution of ignorance. Poor and disarming cultural models, fuelled by soap operas and light songs, are cultivated among the ignorant people so that, in a dreadful vicious circle, the moderate mass is held captive in a condition of cultural deprivation. Note that, although irrationality cannot be excluded, it is not necessary to assume that the social actors belonging to the oppressed classes are irrational. It is sufficient to observe that the inequality in the distribution of power and income brings with it an inequality in the distribution of cultural instruments. Then precautionary behaviour and the high cost of cultural defence naturally predispose the dominated classes to suffer the ideological hegemony of the dominating ones; and this reinforces the initial inequality. Rational ignorance is a class phenomenon produced with the reproduction of class relations. To say it again with Downs’ words, the division of labour and knowledge paves the way for a division of political information, so that power inequalities are endemic in modern democratic societies.

The state action The two-party system and the gathering of consensus The forms of democratic procedures are multifarious, as are the ways in which political actors organise themselves into parties and coalitions for the gathering of consensus. History shows, however, that the two-party system is the prevailing model in advanced capitalist countries. To understand why, it is necessary to study the forms of distribution of political moods. These are defined as the mental attitude of people towards the conservation of the existing political system and government. Consider a social system at a given time. A possible distribution of individual moods is portrayed in Figure 3.1. Moods, M, are represented on the horizontal axis. Their frequencies, V, i.e. the number of individuals sharing any given value

The state and normative institutions 127 V

M0 M1

MC

M2

M

Figure 3.1

of moods, are represented on the vertical axis. M0 is a threshold value, call it the ‘neutral moods’. Conservative moods, i.e. those prone to accept the existing political system, prevail on its right. Revolutionary moods, i.e. those manifesting a desire to change the system, are on its left. Conservative moods can be divided into three subsets: centre-left, at the left of M1; centre or moderate, between M1 and M2; right, at the right of M2. Mc represents the centre, i.e. the median, of political attitudes. Moods do not consist of preferences for any particular political programme. Political programmes are too complex for us to be able to measure individual attitudes towards them by a single variable. After all, the hypothesis of class ignorance implies that most social actors are unable to form a precise and firm opinion on the political measures proposed by the parties. Political moods are not opinions. They are rather mental feelings towards the existing institutional system. All values of M greater than M0 express acceptance of the status quo. Obviously there are varying degrees of enthusiasm and firmness of convictions. Higher values of M represent greater enthusiasm. Let me assume for the moment that the V distribution is normal and Mc is its average, so that this value of moods corresponds to those of the modal and median people. Modal is the individual whose mood is shared by the greatest number of people; median the one who divides all the people into two equal parts, with 49.999… per cent on his left and 49.999… per cent on his right. Suppose that the political actors are organised into two parties or two party coalitions and that, to conquer power, they have to gather support from the majority of the people. The gathering of consensus takes place through an electoral competition. The political class transforms collective moods into a political will. This is not a programme, but the political orientation sustaining it, that is, the orientation sustaining the policy proposed by the winning party and implemented by the government and parliament. Therefore it is a single-value variable. The voting process transforms the distribution of mood values into a single value of political will. This must thus correspond to only one of the many values of M. It is easy to foresee which one will prevail in this example. The party who

128 The state and normative institutions wants to win the election must gain more than 50 per cent of the votes. Therefore it must gain the support of the median voter. As a consequence there will be a tendency for both parties to converge towards the centre. The so-called ‘median voter theorem’ is a paradox of democracy: notwithstanding the division of political actors in two opposing parties, and in spite of the fact that citizens express a vast range of moods, the moderate attitude tends to prevail anyway. In this example it is impossible to forecast which party would win the elections, yet one can easily forecast which political will would prevail. It is that corresponding to Mc or one near it. This is the fundamental reason why the two-party system predominates in capitalist societies: it crowds out the extremists and allows the general political will to depend on moderate moods. One of the most dangerous effects of the two-party system – dangerous from the point of view of democracy – and most advantageous from the point of view of power stability, is the high rate of abstentionism. This mainly hits the extreme fringes of electors, who do not feel well represented by parties who court the moderate mass. In theory it should also hit the moderate mass, who could feel well represented by both parties and therefore have no reason to prefer one or the other. In reality this does not occur because the moderate electors are among the less informed, more ignorant and easier to influence, so that a small difference between the parties, even a specious one, may be sufficient to induce them to vote. Abstentionism may be a weapon available to the extreme fringes to defend themselves against moderate dictatorship – a weapon which is efficacious, however, only when moderate moods are not very strong. An extremist voter might decide to vote for a party he does not fully like only if the divergence of his moods from those pursued by the party is not very marked. This simple fact could weaken the attraction of the median voter. Assume there is a fixed value for the highest divergence a voter is prepared to tolerate between his own moods and those pursued by ‘his’ party. And suppose the moods distribution is asymmetric, that is, there is a predominant mass of, say, left wing voters. There will be a left wing modal voter. In this case the left wing party could lose the elections if it tried to attract the median voter, for the number of votes it would lose from the left is higher than that it would gain from the centre. In order to avoid this, it must try to attract the modal voter. If the extremist citizens systematically practice abstentionism, when the distribution of moods is asymmetric, the median voter theorem no longer applies. Instead, another one holds which could be defined the ‘modal voter theorem’: when political moods are asymmetrically distributed and the extreme fringes of voters may systematically abstain, the victory of a party will be decided by the frequencies near the mode.2 The effects of this theorem, however, are not very different from those of the median voter, if distribution is unimodal and asymmetry is not very strong, because in this case it is probable that the modal voters are near the moderate mass. The two theorems yield different results when the mood distribution is bimodal, that is, when public opinion is strongly divided. This can be seen in Figure 3.2, where a wide divergence of collective moods is shown. In this case there are two modal voters, a left-wing one (on M2) and a right-wing one (on M5). The median

The state and normative institutions 129 V

δ

M1

δ

M2

δ

M3

δ

MC

δ

M4

δ

M5

M6

M

Figure 3.2

voter’s moods, Mc, have a low frequency. Let d be the highest divergence a voter is prepared to tolerate between his own moods and those of the voter courted by ‘his’ party. If both parties look for the consensus of the median voter, they converge toward Mc and receive few votes. The left-wing party obtains the votes represented by the area over Mc – M3, whilst the right wing party those represented by the area over Mc – M4. In this situation there is room for the birth of two new parties, an extreme right-wing one, which would converge on M5 and gather the consensus represented by the area over M4 – M6, and a left-wing one, which would converge on M2 and gather the consensus represented by the area over M1 – M3.To prevent this, the moderate parties will tend to move toward the extreme position. The median voter will no longer exert any force of attraction. The parties will be attracted by the modal masses and abstentionism will probably become higher among the moderate voters, who will no longer be courted by the parties. One of the most anti-democratic effects of the two-party system emerges when the distribution of moods is unimodal and the median voter rules, and consists in the tendency of the political class not to honour its electoral commitments and to ignore its electors’ expectations. If a party, say, a left-wing party, has won the election by appealing to the moderate voters, it will have no reason to pursue a left-wing policy. In fact most of its left-wing voters voted for it even though they did not feel well represented and would in all probability continue to do so. The extremists who abstained can be attracted only by abandoning the moderate voters, who, however, are those who determined victory. A similar reasoning applies to a right wing party. But the winning party is not constrained even by its moderate voters, who have a low level of political participation3 and are easily conquered by propaganda, the leader’s personality and other ‘communication’ techniques. This implies that the winning party feels free to do what it likes. However, if the primary goal of political action is the maintenance of power, the choices of the ruling party will be somewhat constrained by the political and social situation. The distribution of collective moods depends on the satisfaction and frustration generated by the social and economic system and the institutional setting sustaining it. The political class will therefore be constrained by its ability

130 The state and normative institutions to regulate the degree and social distribution of satisfaction and frustration and, thence, by the state of the power and conflict relations among social classes in civil society. Now it is necessary to distinguish between a short run and a long run horizon, which, for the political class, are, respectively, the duration of Parliament and the length of a long wave phase (25–30 years). The adoption of a myopic horizon might convince a ruling party to honour its ideological vocation rather than comply with the moderate voters’ expectations. Alternatively a myopic horizon might induce the party to favour the classes and the groups who are considered stronger in a certain period. These may be the Unions and the workers’ movement, if, for example, the left-wing party was brought to power by the hegemony generated by a wave of social struggles. Such hegemony might push part of the moderate mass toward the left, which might be sufficient to induce the political class to make choices favouring the workers. But an hegemonic drive might be exerted by the capitalist class, as it occurs in a period of social reaction. Often the strongest social actors try to affect political choices through lobbies, a practice which is propitiated by the two-party system. Disengagement of the ruling party from the commitments with its non moderate voters predisposes it to pay heed to the lobbies. Its dependence on the moderate voter’s consensus may compel it to pay their support, the reason being that lobbies often supply two effective instruments of re-election: money for the election campaign and control of mass media. Such policies, however, are dangerous for the survival of the political class if they are not sustained by long run vision. By favouring particular groups or classes they may contribute to sharpen class conflict. In particular, if they produce deleterious effects on the evolution of the economy and institutions, they may cause strong divisions in collective moods. And the situations in which collective moods become bimodal are the most dangerous. They disarm the moderate voter and destabilise the processes of collective will formation. Worse still, a bimodal distribution may evolve toward a situation in which trespassing of the revolutionary threshold is feasible and appealing for a great mass of social actors. A situation like this can cause catastrophic changes in the institutional system, changes which could undermine the power of the ruling class, not just for an electoral session, but forever. Therefore a far-sighted political class may make ample use, in the short run, of the freedom of action ensured by the dictatorship of the moderate mass, but will always work towards a long run objective, stabilisation of social relations, to prevent class conflict from evolving in an uncontrollable way. And, to stabilise social relations, it is necessary to feed the increase of satisfaction and the diminution of frustration, which can easily be done by sustaining economic growth. If the short run freedom of manoeuvre is used to pursue long run stabilisation, then the political class plays the role of a servo-mechanism in the capitalist mode of production. And it does it precisely by virtue of its relative autonomy from civil society. The ability to perform this service ensures legitimisation of conditions: the moods of social actors will remain distributed in a way that will support the political class.

The state and normative institutions 131 Government action: the state as a conflict and accumulation regulator As already observed, the distribution of moods depends on the degree of satisfaction and frustration generated by social and economic conditions and therefore by the institutional system sustaining them. In Chapter 2 I showed that an institution can cause satisfaction or dissatisfaction or both, depending on the results it allows social actors to attain in relation to their aspiration and tolerance thresholds. I also showed that, when an institution generates dissatisfaction, processes of institutional revision may be triggered off which drive social actors to abandon an old institution and adopt a new form of behaviour. But institutions are mainly organised into systems in which the core ones cannot be modified independently of each other, that is, without changing the entire system. Normally an institutional system causes both satisfaction and frustration: first, because people are unable to maximise their achievements; second, because institutional plurality and the multiplicity of selves cause emotional or moral conflict; third, because many institutions, even if unsatisfactory, cannot be abandoned without disrupting the whole system. The simultaneous presence of both satisfaction and frustration can apply to any single social actor, who may be induced to tolerate certain unsatisfactory institutions if the system keeps other institutions alive which generate satisfaction. But it can also apply to society at large, which may tend to split into social groups who receive different doses of satisfaction and frustration. Collective moods, then, may tend to dichotomise when there are some social strata in which satisfaction overwhelms frustration and others in which the opposite occurs. Strongly conservative moods tend to prevail among the former and strongly revolutionary ones among the latter. The state, under the drive of the political class toward power conservation, operates to prevent processes of mood divergence and therefore tries to raise collective satisfaction and lower frustration. I already showed that the state acts on civil society through the production of rules and government action in the attempt to cope with the incompleteness of conventions and institutions. Now it is necessary to go beyond a study of the form of state intervention and to investigate its nature. This can be done by focusing on the functions of a capitalist state, i.e. those undertaken by the state as part of the capitalist mode of production. The state regulates the dynamics of civil society by producing software and hardware. It performs a cultural action, by intervening in the sphere of ideology production; and an economic and social action, by intervening in the three circuits of capital reproduction, the monetary, productive and commodity capital circuits. In the sphere of ideology production, state action endeavours to make up for the deficiencies of civil society. The state does not just produce self-legitimisation ideologies, but also offers support to the processes of civil society stabilisation. Because of the antagonistic nature of social relations, civil society tends to generate unfeasible aspirations and conflicting visions of the world. The aspirations of the lower classes are raised by their observation of the dominating class’s achievements, by the firms’ manipulative and persuasive action on the consumer and by the

132 The state and normative institutions expectations of improvement brought about by economic growth. The situation of permanent conflict in which the exploiter and the exploited interact may cause the formation of class ideologies, values and opinions capable of justifying and sustaining social actors in class struggle. The state intervenes in this sphere by using a wide range of instruments for the production and diffusion of ideologies – schools, universities, research foundations, cultural agencies, political parties, churches etc. – and does it by diffusing ideologies that hinder any tendency to diverge collective moods. Dominating class ideologies try to produce moderating effects on the formation of aspirations. The lower the aspiration and tolerance thresholds, the smaller the degree of frustration and the greater the degree of satisfaction generated by existing institutions. The more modest the welfare level aimed at by the working class, the lower the claims and grievances. In this way the increase in frustration and the decrease in satisfaction is hindered, given the level of economic achievements. In addition the state also works on the diffusion of interclass ideologies, that is, cultural and philosophical constructs which, by exalting national, ethnic, religious, linguistic and historical identities, counter class antagonism and sustain a sense of belonging to a community. In this way the divergence of collective moods is hindered, for a given degree of frustration and satisfaction. Consider now the state action in capital circuits. The monetary circuit was perhaps the first in which state intervention took place. There is a strong reason for this priority: availability of money and credit is a fundamental condition for capital accumulation and social reproduction; and, as a consequence, a general and centralised control of credit is indispensable to overcome the limitations that self-finance and the collecting of the individuals’ savings poses on accumulation. The use of money as an instrument for feeding accumulation and stabilising the business cycle has not always been practised effectively, and state intervention in the monetary circuit has often contributed to destabilise accumulation rather than the reverse. But with the passing of time policy techniques have been refined and the role of the central bank as lender of last resort and guardian of the banking system has gradually emerged as a primary tool of every policy aimed at regulating accumulation. There are two kinds of state intervention in the circuit of productive capital: productive and distributive. As a production regulator the state tries to furnish civil society with goods that private capital is unable to produce or produce in socially acceptable quantities and at socially acceptable prices. In all capitalist systems there is a vast range of goods and services, including national defence, police, education, health services, transport and communication networks, energy supply, the production of which is either controlled or regulated by the state. What characterises these goods is that their production benefits the community but does not offer adequate stimuli to efficient private investment. There are several reasons for this gap between private and public interests: non-excludability and non-rivalry in consumption, strong production and consumption externalities, high investment costs, natural monopolies, lack of adequate information. If social actors were left completely free to organise production, most of these goods would not be produced,

The state and normative institutions 133 or at least not to the desired extent, and collective frustration would increase dramatically. State intervention in production aims to supply not only low-price consumer goods to consumers but also low price investment goods to firms, so that it also increases, one might say, capital satisfaction, i.e. the firms’ profits, and thus sustains capital accumulation. In fact, in most public goods and services, it is difficult to separate the benefits that go to the exploited classes from those that go to the dominating classes. Just think of the advantages brought about by economic infrastructures, transport and communication networks and education and health service systems. Particularly important is the role played by the state in regulating income distribution and, more generally, employment relations. The state intervenes systematically in the industrial relations game and the productive use of labour. A very important example, which dates from the very beginning of capitalism, was the legislation on labour disputes, working time, working conditions and unemployment. More recently states have also learned to intrude effectively, by means of incomes policy, in the conflict over wage determination. Furthermore, through taxes and subsidies, the state is able significantly to affect the secondary distribution of income and wealth. Starting in the 1930s, and especially after the Second World War on the basis of Keynesian theory, a new kind of state intervention developed: intrusion in the circuit of commodity capital. It is in this circuit that realisation crises take place. In certain periods of economic depression the instability caused by effective demand problems was responsible for disastrous phenomena of accumulation setbacks and employment contractions that gave rise to waves of social unrest capable of threatening the very existence of capitalism. Keynesian macroeconomic policies basically aim at coping with this kind of difficulty and, when applied systematically, they succeeded. The logic of political regulation of effective demand is based on the principle that the state can and must do what the private capitalists are unable to do, i.e. supervise and take care of the accumulation process in the interests of collective welfare, or of capital in general. In this kind of policy the state properly acts as a general capitalist. Collective welfare mainly refers to the living conditions of the working class in terms of social security, health services, wages, consumption and employment. The two objectives, collective welfare and capital accumulations, are compatible with each other because the achievement of one is a condition for the achievement of the other. This conclusion can be generalised: the regulation of conflict and accumulation is not confined to state intervention in one circuit or another, but pertains to its basic role as an equilibrating mechanism. The stabilisation of class conflict is an important condition for capital accumulation because economic growth can be impaired by social disorder. When the working class is too strong, income distribution changes in such a way that the investment stimulus is weakened; an effect that may be exacerbated by depression caused by acute class conflict and lack of social discipline. When the working class is too weak, however, income distribution changes in such a way that consumption and aggregate demand shrink. Moreover

134 The state and normative institutions a softening of the spur provided by workers militancy can slow down innovative investments and technical progress. A stable accumulation process requires, not abolition of conflict, but its regulation, so that it develops in an orderly way within the given rules of the game. The state performs the function of conflict regulator in civil society. It is therefore intimately capitalist, social stability being an essential condition of capital accumulation. But it is also true that capital accumulation is an important condition of social stability. In fact, the higher the rate of growth, the easier it is to satisfy the social classes’ conflicting claims on national income. Therefore the state, acting as conflict regulator, is also induced to sustain the accumulation process. In this kind of intervention, since it is interested in capital accumulation as a whole, the state tends to assume a position of supremacy towards individual capitalists. Again by virtue of its autonomy, the state acts as a guide and a supervisor of capital, it behaves as a general capitalist. Normative action: the state as the political structure of capital In a capitalist mode of production the state is capitalist, first of all, in a functional sense, in that it performs the function of conflict and accumulation regulator. But it is capitalist in a deeper sense too: it does not limit itself to ensuring that conflict and accumulation evolve in a orderly way within the rules of the game; it plays a fundamental role in the very definition of the rules. The state, as already observed, tries to make up for the incompleteness of institutions arising from civil society, by sanctioning and generalising them, and giving them form and legal foundations. Yet, the self-constitution of civil society, however incomplete, is always at the base of the state’s normative action. The raw material of state action consists of spontaneous institutions and conventions, the habits and customs, produced by social actors. Therefore the fundamental social relations through which civil society tries to constitute itself cannot be impaired by the state; quite the contrary, they are reinforced and perfected by its action. But the fundamental relations of the capitalist mode of production are those ensuring the command of capital over labour. Thus the state is capitalist also in a structural sense: in a capitalist society it operates for consolidation of the institutions guaranteeing capitalist exploitation. And since the exploitation relation is based on a power relation, it is necessary to investigate power in more detail to be able to understand this aspect of the state’s nature. The basic form of capitalist power was dealt with in Chapter 1. It consists in the prerogative of a social actor to determine the activity of another in the pursuit of his own goals. This is what is called ‘command’. The employment contract was shown to generate the capitalist’s power in the form of command over labour. It is an economic form of command, in the sense that the activities of the commanded subject are directed to the achievement of an economic end. There is also a political kind of command, in which the dominating actor is motivated by political goals. This is peculiar to the state and, to be precise, it represents the way in which the political class exerts power in the state apparatus, in the army, the police, bureaucracy etc.4

The state and normative institutions 135 Here, however, it is necessary to focus on another form of power: the faculty of a political actor to impose normative limits on the social actors’ choices and decisions. I will call this kind of power ‘domination’. The dominating actor does not determine the goals and the actions of the dominated actors. Domination merely implies that the social actors’ actions must remain within the boundaries posed by the dominating actor, who normally exerts domination by fixing prohibitions. The domination exerted by the state over civil society is based on the issue and implementation of laws. Political domination is quite a particular form of power, in modern democratic systems. The state, in fact, is a political actor at the service of civil society. It cannot determine the social actors’ aspirations. Therefore, in principle, and apart from what has been said on direct intervention in the productive circuit, the state does not exert command over civil society. This, in a capitalist system, is incapable of complete self-constitution principally because its social relations, being based on the exploitation of labour and the power of capital, are fundamentally antagonistic. In a transaction regulating a conflict relation the obligations entered into by the parties are not taken for granted and accepted once and for all. The institutional character of the transaction marks its form but cannot determine its content in a precise and immutable way. In particular, the terms of an employment contract reflect the bargaining strength of the parties and therefore are continually open to transgression, breach and renegotiation, so that the power relation set up under this contract is inherently precarious. Therefore the most dangerous form of institutional incompleteness in capitalism is political. Civil society is unable to hold together on its own, not so much because of a generic bellum omnium contra omnes, but because of the difficulty of stabilising the social relations based on the capitalist command over labour. Civil and disciplinary incompleteness in a capitalist system also acquire a sense from the inadequacy of capitalist command. The workers’ obedience in the labour process, for instance, generates the employer’s claim over the produce of labour but, alone, is unable to transform this claim into a property right which is enforceable to third parties. To avoid the permanent threat of breakdown, a capitalist society needs a political structure which is more solid than the one it produces itself through transaction institutions and social conventions. It needs a political body capable of sanctioning its institutions, thereby legitimising the self-constituted power relations, enforcing observance of obligations on the parties who have undertaken them in a transaction, transforming into property rights the claims generated by transaction institutions, that is, making them valid universally. Only the state is able to perform these functions, because it presents itself to civil society as a public authority to private agent, as the bearer of law, as a legitimate monopolist of force. This is why it is an essential component of the capitalist mode of production: it is the supporting structure of capitalist power. The state is the political structure of capital. As such, it works as an institutional system aimed at the conservation of power relations. Capitalism cannot exist without it. Recent neo-institutional literature has produced the concept of ‘inequalitypreserving institutions’, a concept which aptly describes most state institutions. Yet inequality-reducing institutions also exist. Just think of progressive taxation,

136 The state and normative institutions wealth taxation, transfers and subsidies for the poorer social groups. Although it might seem paradoxical, both the institutions aimed at the preservation and those aimed at the reduction of inequalities are, as a system, capitalist institutions. All of them serve, directly or indirectly, to perpetuate exploitation, since all operate to sustain the power structure of capitalist relations. In this fundamental sense the state is capitalist in a capitalist system. In his important work on law and the state, Evgeny Pashukanis (1978) refers to the existence of a dialectic relationship between the state and the individuals. Perhaps ‘dialectic’ is not the right word, but as a metaphor it serves the purpose: it conveys the message of an interdependence between the state and civil society. In fact, it is not only civil society, as the place of capitalist power, that needs the state. It is also the dominant political class, and therefore the state, that needs a civil society which is structured by power relations. It needs it to perpetuate itself as an autonomous political actor. First of all, state power is founded on social relations which make that power necessary. Without class antagonism there is no necessity for a capitalist state. A civil society built on the capitalists’ command over the working class is the social foundation of the capitalist state. Second, since the state has the monopoly of force, the very possibility of its existence is ensured by a social situation in which economic power is unequally distributed. A fully democratic society is capable of complete self-constitution. The emergence of an autonomous and separate body endowed with political domination would be impossible in such a society. The existence of a capitalist state is made possible by a pre-ordered limitation of the decision-making ability and autonomy of choice of the great mass of social actors. Third, the conservation of the political class power implies that democratic representation procedures can be used to deprive most of the social actors of any real political mandate and control faculty. I already observed that this result can be obtained by virtue of that peculiar social and cultural phenomenon called ‘class ignorance’, and that inequality in the distribution of cultural instruments is mainly determined by the same division of labour on which capitalist power is based. In a certain sense, one could say that the capitalist nature of social relations produces the means, the cultural instruments, of political domination and its perpetuation. Finally capital also produces the economic resources required by the state for undertaking its action and by the political class for remunerating its activity. The state, insofar as it does not control the productive circuit, does not produce value, and therefore must be financed by the surplus value produced in civil society. If the state succeeds in extracting more resources from labour activity than it returns to it, there is a form of state exploitation. But exploitation is perpetrated in civil society. Thus the conditions for the extraction of the resources the state needs for its existence are founded on the capitalist nature of social relations. The political conditions of social reproduction The idea that the state constitutes the political structure of capital clarifies an argument I put forward in chapter 1 about the conditions of social reproduction.

The state and normative institutions 137 In that chapter I observed that the distribution of property rights does not generally play an essential role in creating those conditions, which were in fact accounted for by a theorem on class-biased credit rationing. The question concerning the civil society’s capacity to reproduce itself was left partially open. The point is that, due to institutional incompleteness, civil society in a capitalist system is unable to create the full conditions for its reproduction. I can now observe that the state, by furnishing a service of institutional completion, performs an essential role in the creation of those conditions. Without the state, capitalism is unable to reproduce itself. The social structure of capitalism is a power structure, but transaction institutions and social conventions are insufficient to sustain this structure. It is the state that provides the residual power structure required to stabilise all the social relations on which exploitation and command over labour are based. However, there is a more specific sense in which the state is an essential condition of social reproduction: it controls the production of base money. Credit is the lever of accumulation, since it enables the creation of capital from nothing. Those who control it, control accumulation. Those who are excluded from it, are excluded from wealth accumulation. But base money is the lever of credit. Those who control its production regulate the conditions of social reproduction. Now, monetary control is a specific function of the capitalist state. The whole history of the state role in financial markets is one of struggles for the acquisition and extension of the instruments of monetary control, regulation of the banking sector and enforcement of class-biased credit rationing. The central bank’s activity as a lender of last resort does not reduce to the control of business cycles. It is a financial function, as it serves to guarantee the availability of finance to the social actors who use it to accumulate capital. The central bank’s activity as controller and regulator of the banking sector also has a class discipline function. It serves to prevent access to credit by social actors who are unable to use it efficiently in capital accumulation. After all, money is an institution or, rather, a system of normative institutions. And the state is a money producer in its position as creator of normative institutions or, more specifically, as a guardian of class-biased credit rationing. The monetary authority, as a state branch, fixes the rate of discount, through which it affects the banks’ loan rates in a decisive way. As shown in chapter 1, the higher the interest rates, given the risk spread, the more class-biased is credit rationing. When the monetary authorities raise the discount rate, they play a precise role in class conflict. To begin with, they lessen the workers’ welfare by making consumer credit more costly. Furthermore they inflate the rentiers’ incomes by boosting the returns on financial assets. Finally, and most important, they exacerbate the workers’ dependence on capitalists. In fact the slow down in effective demand brought about by higher interest rates reduces employment and depresses the workers’ self-confidence and ability to resist capitalist pressure, whilst the tightening of credit rationing restricts the workers’ chances of finding alternative earning opportunities. In its role of conflict regulator the state alternates monetary measures. In some periods it may decide to soften economic pressure on the working classes in order to blunt the distributive conflict. At other times it may decide to intensify pressure in order to discipline the workers. Base money also furnishes a sort of risk insurance to the banks and, through

138 The state and normative institutions them, to the entire economy. Any cut in the discount rate helps the banks to stock up base money and to hedge against loan defaults and the danger of bankruptcy. At the same time it helps them to supply cheap and abundant credit, thus socialising investment risks. Class-biased credit rationing also means that the capitalists’ risks are hedged better than the workers’. Yet this kind of insurance cannot be extended too far, even to capitalists, if the disciplinary role of financial markets has to be preserved. Discount rate manoeuvres must balance the disciplinary and insurance functions of the banking and financial systems. Therefore there is a structural dimension of monetary policy, besides a regulatory one, which is perhaps even more important than the latter. Interest rates cannot all be nil5 in a capitalist economy, since the discipline that markets impose on workers and capitalists is an essential condition of accumulation. The state’s grip on the base money supply results in a form of political control on the conditions of social reproduction. The meaning of state autonomy A widely debated theme in political theory is the extent to which the state is autonomous from civil society. Attitudes oscillate between two opposite positions. On one extreme there are the state-centred theories, which claim that the state is able, by virtue of its monopoly of force, completely to dominate civil society and to constitute it. At the other end of the scale there are the socially-centred theories, which maintain that civil society dominates the state. These can be divided into two groups: some Marxist theories, according to which the state is controlled by the bourgeoisie, the social class which dominates civil society; and some contractarian theories, according to which all the citizens of a society use the institutions of democracy to impose their preferences and their will on the state. State-centred theories reduce civil society to an amorphous body totally incapable of selfconstitution and autonomy. The socially-centred conversely tend to reduce the state to an agent of social groups. In either case the problem of state autonomy cannot even be properly posed. More interesting is the group of theories eschewing both these extremes, and to which the one developed in this chapter belongs. The basic idea is that the state is constitutionally autonomous and structurally dependent. Constitutionally autonomous, because it constitutes itself as a domination apparatus of a specific class of actors, the political class, whose specific goals are defined autonomously with respect to those of the social classes. Structurally dependent, because the political class’ survival requires some ability in relating the state to civil society as the supporting structure of its political cohesion. In the same way that neither of the two fundamental classes of civil society, the workers and the capitalists, is able to realise its own interests other than by contributing to the realisation of those of the other class, so the political class relates to the social classes as a subject which is at the same time both autonomous and accommodating. The capitalists aim at profits and accumulation, but are unable to achieve either of these goals unless they furnish the workers with the means to produce their income. The workers aim at wages and welfare, but are unable to achieve either unless they produce profits for the capitalists. The political

The state and normative institutions 139 class, finally, aims at political power, which it cannot conquer and maintain unless it furnishes civil society with the political services it needs to reproduce itself. Structural dependency compels the state to operate as an instrument for the achievement of goals which are not those of the political class, and is the condition for the state’s ability to work as the political structure of capital. Constitutional autonomy enables the state to privilege the interests of one or another group in different times, depending on how social influence evolves in civil society. Therefore there is also a situational dimension of the autonomydependency relationship, besides a structural one. A state is situationally instrumental to some specific social groups in a given historical period if it helps them to achieve their goals. It is situationally functional to some specific social groups if it contributes to satisfy their interests even when the groups themselves are unable to manifest them in the form of goals to pursue. Finally a state is situationally adverse to some social groups if it hinders the pursuit of their interests or goals. For example, a state that decides to increase unemployment benefits, when the Unions have exerted pressure to adopt that measure, operates instrumentally in favour of the worker movement. A decision to expand public expenditure to sustain aggregate demand, even if no social group has exerted any pressure to do so, operates functionally with respect to the immediate interests of both the capitalists and the workers. And a state that emanates an anti-trust law acts adversely to the interests of monopoly capitalists, but functionally with respect to overall capital accumulation. In the short term, however, both constitutional autonomy and structural dependency appear to be fables rather than facts. The political class needs social support at all times and naturally tends to undergo the influence of social forces and therefore to intervene in social conflict by favouring some particular interests. But it does not always privilege only a particular group or class. It may favour the stronger groups in any specific situation. And situations evolve in time with the evolution of class conflict. Nothing in principle prevents the state from favouring the workers’ movement at times when its organisations and struggles enable it to exert strong political pressure. While it is true that the state’s structural dependency serves the fundamental interests of capital in the preservation of social setting, it is not true that it is always and under any circumstances at the service of the specific interests of the bourgeoisie. It may even in certain situations operate against the immediate interests of some bourgeois groups or even against those of the entire dominating class.6 It is however also true that no privilege sustaining the interests of particular groups can be pushed to the point of exacerbating class conflict and endangering capital accumulation. Political classes who go too far in a particular direction risk going under. In actual fact those who make the mistake of committing themselves too heavily with the immediate interests of a particular class, may sooner or later be swept away by the social consequences of their partiality. Whether by virtue of the struggle for power and the selection of the ruling class, or by virtue of the ability of the political class in power to develop a strategic vision, the state, in the long run, tends to operate in view of social stabilisation.

140 The state and normative institutions On sources Machiavelli – according to G. H. Sabine (1951) – created the modern theory of the state: the state as the most powerful institution of modern society, an organised force invested with the task of regulating and controlling all the other social institutions and the right to direct them according to its own interests. Some alleged flaws in Machiavelli’s thought, its unsystematic construction and the weakness of its philosophical foundations, in reality bear on its principal merit: the refusal to found politics on a strong metaphysics of the social being. Not only is there no reference to natural law rationalism in Machiavelli’s thought; there is indeed no general philosophy of man, no anthropology abstracting from the historical and institutional determinations of human action. The basic condition of social antagonism from which the modern state emerges, and the related vices of greediness, disloyalty and deceitfulness characterising the individuals’ motivations and actions, are referred by Machiavelli to a state of institutional corruption and decay which is always conceived as historically determined. This argument can best be appreciated by reading The Prince – in which Machiavelli theorises despotism as the most effective form of government in an epoch of war of everybody against everybody – together with the Discourses – where he shows some predilection for the republican system but also the awareness that this presupposes a type of human being endowed with moral strength, civic sense and love of freedom. Machiavelli treats the human character as strongly influenced by institutions, but he also treats the latter as the product of human action. We have here the premises of a weak ontology of the social being based on a sort of institutional relativism which succeeds in avoiding both the risk of holistic metaphysics and that of utilitarian and individualist reductionism. Machiavelli tends to reduce the problem of power legitimisation to one of consensus, and this brings him to formulate a sort of theorem of inter-class consensus, whereby the state action is approved both by the privileged classes and by the lower classes, by the former to defend their privileges, by the latter to defend themselves from over-abuse. Machiavelli’s thought represents the cultural background and the distant origin of the Italian school of power sociology, whose leading figure is Gaetano Mosca. The tendency to reduce the problem of legitimacy to one of consensus is stretched to the extreme consequences by Mosca. Both his theory of the ‘political formula’, which reduces democracy to a mere instrument of formal and ideological legitimisation, and his thesis of ‘sound administration’, which sees good government as an instrument for rising consensus, express a deep distrust of the very idea of democracy as popular self-government. Yet they contain a cogent criticism of the modern forms of procedural democracy. As such, they could turn out to be analytically useful, provided they are supplemented by an analysis of class struggle in modern capitalist systems. The principal criticism one can make of Mosca is that he lacks precisely this kind of analysis. Other important and valid criticisms that have been raised against him, and which bring to light the confusion that sometimes entangles his arguments, his tendency to look at political domination

The state and normative institutions 141 as a natural fact, his reduction of social change to the change of political élites and, more generally, the strong smell of conservatism that rises from all his work, must not induce us to reject the substantial grain of truth contained in his doctrine: that democracy is just the specific formula of political domination in modern states. More precisely, three important arguments of G. Mosca (1884, 1923) deserve to be highlighted: first, that the principal motivation of the political class is power; second, that the condition for fundamental asymmetry in the distribution of political power in modern societies lies in the organisational, technical and professional abilities required for victory in the political arena and success in state administration; third, that reproduction of political dominance is ensured by a mechanism of social reinforcement which makes the distribution of political abilities depend on that of power itself. Since the élitist theory of the state can easily provide grounds for a conservative ideology, it is understandable that it has been rediscovered, if not reinvented, by various orthodox economists. Apart from the complex sociological contributions made by V. Pareto (1916), some important sources of the modern economic theory of democracy are H. Hotelling (1929), A. Smithies (1941), H. R. Bowen (1943) and especially J. Schumpeter (1950), in which the following definition of ‘the democratic method’ appears: ‘that institutional arrangement for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people’s vote’ (p. 269). A book that has now become a classic is that of A. Downs (1957), in which one can find both the median voter theorem and the assumption that state functionaries maximise their personal interest. I dealt with the median and modal voter theorems in E. Screpanti (1999a). Other important contributions to this theory of the state are: W. A. Niskanen (1971, 1975), W. M. Landes and R. A. Posner (1975), T. E. Borcherding (1983), J. E. Stiglitz (1989). Finally J. Cullis and R. Jones (1987) have much to say on the notion of rational ignorance. Marxist approaches to the state have the opposite drawbacks and merits of the élitist theories. The analysis of the economic, social, institutional and historical conditions, on which the latter is rather lacking, is instead at the centre of the Marxist discourse, which however tend to neglect the possibility of ‘the state functionaries developing their personal interests and pursuing them, often at the expenses of both capital and labour’ (C. Pitelis 1991: 121). It is true that Marxist students, especially those following an instrumentalist approach, generally tend to reduce the state to a more or less complex epiphenomenon of the social and economic conditions of production and to explain its rationale in functional terms. Marx and Engels themselves often seem to believe that the conditions for social reproduction are fully developed in civil society and that the state is not fundamentally necessary for ensuring them (Clarke, 1983). This view led Marxist research astray for a long time. Nevertheless I must explicitly state that the theory developed here is akin to the Marxist approach, and that the reference to assumptions and concepts of the élitist theory has mainly served to account for the constitutional autonomy of the state and to explain its action in behavioural rather than functional terms.

142 The state and normative institutions And let me recall a few insights of the founding fathers’ thought. In The German Ideology Marx and Engels argued that the state is an apparatus capable of compacting the entire civil society of an epoch, since all institutions pass through its mediation and receive a political form from it. Also illuminating is the idea, developed by Engels in Anti-Dhüring, according to which the state is the ideal collective capitalist: it represents the organisation that society generates to defend the capitalist mode of production from the disruptive effects produced by the workers’ and the single capitalists’ actions. In addition, in The Origin of the Family, Private Property and the State, Engels made it clear that the state arises to cope with class conflict, to control it and maintain it within the boundaries of political and legal order. Finally, Marx observed, in Contribution to the Critique of Hegel’s ‘Philosophy of Right’, that the goal of the state becomes the private aim of state functionaries in their pursuit of status. Half a century later Engels (1978: 627) tried to be more precise: Society had created its own organs to look after its common interests, originally through simple division of labour. But these organs, at whose head was the state power, had in the course of time, in the pursuance of their own special interests, transformed themselves from the servants of society into the masters of society. This can be seen, for example, not only in the hereditary monarchy, but equally so in the democratic republic. Nowhere do ‘politicians’ form a more separate and powerful section of the nation than precisely in North America. There, each of the two major parties which alternately succeed each other in power is itself in turn controlled by people who make a business of politics, who speculate on seats in the legislative assemblies of the Union as well as of the separate states, or who make a living by carrying on agitation for their party and on its victory are rewarded with positions (emphasis added). I bet you Schumpeter and Downs were well acquainted with this piece of Engels. Very important too are some ideas put forward by Gramsci in his Prison Notebooks. Although scourging Mosca rather fiercely, Gramsci used his postulate on the tendency of state functionaries to transform themselves into a caste to criticise the liberal theories of the state. One can find in Gramsci the conception of the capitalist state as a subject of hegemony and as a centralised educational will. Moreover there are arguments about the role of state action in equilibrating class conflict and, to avoid social uprisings, in improving the life conditions of people – a goal which is achieved by protecting them from economic crises and controlling the level of employment. But the most important of all is the thesis on the state as an autonomous force and, at the same time, as the embodiment of the unity of dominating classes and their co-ordinator in view of economic growth. As such, the state, by means of its planning activity, becomes an instrument of rationalisation and acceleration: it stirs up, stimulates and punishes. Evgeny B. Pashukanis (1978) gave a decisive contribution to the criticism of law and the state in a capitalist system that should be included among the classics of Marxist thought. He clarified that, in a system of market exchanges, the

The state and normative institutions 143 contracting parties must be defined as legal personalities endowed with freedom of contract and property rights on goods; that property rights, as well as the civil and criminal law that regulates and sanctions them, are the basis of the entire juridical structure of the capitalist state; that state power, as guarantor of the exchange process, takes the form of an impersonal and neutral authority founded on law; and, finally, that state mediation transforms class power into legitimate authority. Among the most recent contributions to Marxist thought, of particular interest are some belonging to the so-called ‘structural-functional’ approach, especially N. Poulantzas (1969, 1975) and I. Gough (1979). The most interesting argument here is that class struggle is not limited to civil society, but always involves the state, which systematically and continually intervenes in class conflict by playing the role of supporting accumulation and social cohesion. Reproduction of the capitalist mode of production is not guaranteed by the economic structure alone, but presupposes the existence of a political actor as a factor of order and organisation. This actor is the state, which can thence be interpreted as the organisational structure sustaining social reproduction. A basic proposition of the structuralist theory is that the state works for capital by virtue of its function in the social reproduction process, and independently of the ideological preferences of the political class. The question then arises: how the political class’s autonomous decisions are constrained to serve the functions they are assigned in the structure? A convincing answer to this question was given by A. B. Bridges (1973), F. Block (1977) and C. E. Lindblom (1982), who suggested that state legitimacy is dependent on the good working of the economic and social system. And that state action must sustain capital accumulation and social stability if the political class is to maintain its power. These ideas have been formalised in C. Offe’s (1975, 1984) principles of ‘dependency’ and ‘legitimisation’. Finally, a decisive criticism of instrumentalist conceptions of the state emerges from the ‘logic of capital school’, particularly from the works of E. Altvater (1973), J. Hirch (1977) and J. Holloway and S. Picciotto (1978). Here the state is seen as an autonomous actor and, as such, an essential condition of the reproduction and accumulation processes, since no individual capitalist, nor any group of them, is able to produce the general normative system capable of guaranteeing the conditions for accumulation. This view has been pushed to its extreme logical consequences by T. Skocpol (1979, 1985), F. Block (1980), G. Poggi (1990) and other post-Marxist scholars, who recognise that state officials are moved by their own interests and goals and, as a consequence, that the state is an autonomous subject with respect to civil society. However, it is recognised that, although this autonomy may be rather strong in the short run and in certain ‘exceptional periods’, in the long run and strategically ‘states usually do function to preserve the existing economic and class structures’ (Skocpol 1979: 30) and ‘the requirements of continuing capital accumulation’ (Poggi 1990: 97). According to C. W. Barrow (1993: 150), ‘the Marxian paradigm of state research should be understood as an analytic constellation that is not reducible to any

144 The state and normative institutions particular set of conclusions; nor can it be identified with any single methodological approach’. Yet there is ground to believe that most of the results of this research are less contradictory than might seem at first sight. Some recent attempts at a synthesis are noteworthy, for example those of B. Jessop (1982, 1990), D. Held and J. Krieger (1984) and R. R. Alford and R. Friedland (1985), from which I have drawn much inspiration. In particular I share Jessop’s principle that the political and economic aspects of society must not be reduced to one another nor must they be treated as fully independent of each other.

Notes 1 A vestal is a virgin charged with the service of watching the sacred fire keep burning upon goddess Vesta’s altar. By extension she is an officer who guards and defends the fatherland ideals and principles. 2 It is not precisely the modal individual who matters, since the distribution can be asymmetric near the mode too. Yet the masses near the mode will certainly count more than those near the median. 3 This proposition is corroborated by empirical findings. For instance, the percentage of leftwing people who often take part in political discussions was, in 1998, 16.3% (France), 22.5% (Germany), 14.5% (Great Britain) and 28.3% (Italy). These percentages drop to 6.6% (France), 13.7% (Germany), 8.6% (Great Britain) and 12.9% (Italy) among centre-oriented people. Rightwing people’s scores are comparable (although a little lower) with those of the left-wing in all four countries. Moreover, in 1995, only 6.8% (France), 12.8% (Germany), 22.7% (Great Britain) and 25.5% (Italy) of centre-oriented people showed some attachment to a political party. This percentage rises to 26% (France), 27.8% (Germany), 40.9% (Great Britain) and 37.9% (Italy) for left-wing people and 34.2% (France), 29.8% (Germany), 49.9% (Great Britain) and 43.8% (Italy) for right-wing people. The figures exhibited in this note have been elaborated by Cannoni (1999) on Eurobarometer data. 4 The state also has economic command insofar as it directly exerts control in the productive circuit. In this case the capitalist state generates state capitalism. It is worthwhile noting that the existence of a state command over labour transforms the nature of the relationship between the state and civil society, since in this case the former, as an entrepreneur, becomes part of the latter; or, rather, it tries to subsume it in itself. 5 Yet it may happen that in some special junctures the prime rate in real terms is zeroed. This may occur in times of strong economic distress, when accumulation stagnates and workers are angry. Then the accumulation insurance function and the social disciplinary function of monetary control can be decoupled – an easy thing to do if the banks’ perception of social and economic disorder is strong and, as a consequence, the risk spread of interest rates is wide. In a situation like this, a drastic cut in the discount rate might induce the banking system to zero the real prime rate for Mr Boss while raising the usury rate. 6 It is even possible for the state, in an attempt to sustain capital accumulation in general, to practice a form of euthanasia of the bourgeoisie. This has occurred for instance in many underdeveloped countries where, because of the cultural and political backwardness of the national bourgeoisie and the inadequacy of its entrepreneurial ability, industrial take-off required expropriation of the means of production and the launching of an accumulation process sustained by state capital.

Forms of co-operation and power 145

4

Forms of co-operation and power

The struggle for survival and the struggle for power The social actors’ reason is bounded; their aspirations are mostly determined endogenously and are strongly conditioned by externalities of any kind; their actions are moved by heterogeneous mental feelings in which altruistic and egoistic sentiments can be simultaneously present; their egoism often induces them to transgress social rules and even the obligations voluntarily undertaken; their knowledge and information are limited and asymmetrically distributed; their social environment is continually changing and radically uncertain; finally, their relations are constituted through transactions in which the principle of equivalence does not hold. Perfect and atomistic competition cannot exist in this world, even as a limit case – I mean, the neoclassical kind of competition that eliminates all inefficiencies and power hubris. Yet the social environment of the capitalist system is a highly competitive one: social actors are compelled to continually compete with each other in order to pursue their ends. The word ‘compete’, though, must be understood with the meaning of ‘trying to obtain together in a context of rivalry and antagonism’, whereby the term ‘competitor’ becomes a synonym of ‘rival’. Capitalist co-operation works through competition, and this is the specific form of the struggle for survival and domination in the human species in a particular historical era. Thus you have to look at evolution rather than equilibrium if you want to understand. In political economy the evolutionary theory has been used in different ways. In some approaches it works as a metaphor aimed at accounting for institutional change. This I have mostly done in the preceding chapters. Institutions are the units of analysis and their environment consists in social actions, for which they serve as instruments. Their change occurs through artificial selection. This is the theory of institutional evolution. In the theory of social evolution, the units of analysis consist of collective actors, firms, associations, states, nations, classes etc. It is still a metaphoric use of evolutionary theory, for collective actors are not ontogenetic units of the human species. And also in this case artificial selection prevails, as most collective actors are created and destroyed by deliberate actions.

146 Forms of co-operation and power Finally, evolutionary theory can be applied to the relations among individuals. This is the theory of human evolution proper. It is no longer a metaphor, but a correct application of Darwin’s view to the human species. I maintain that individual actors approach the individuals of other animal species and other individuals of the human species with a prevalently competitive attitude, i.e. as the objects, the instruments and the conditions for their own existence. I have no pretension that things work like that by force of ‘nature’, that they have always been and always will be like that. But no doubt that they work like that at least in the capitalist mode of production. Natural more than artificial selection prevails in human evolution since, while institutions and collective actors are created by people, the genetic structure of individuals is still a far cry from being knowingly selected by human intelligence. Moreover, evolutionary theory, as applied to institutions and collective actors, is Lamarkian, as the transmission of acquired characters occurs through culture. The theory of human evolution is basically Darwinian. Since institutions and organisations are created and used as instruments for the pursuit of human ends, the first two kinds of evolutionary theory must be integrated in the third. In this vision, application of the principle of the struggle for existence to institutions and organisations ceases to be a metaphor, for the production of instruments for individual self-assertion is part of the practices through which that struggle takes place. It might seem that the differences between the evolution of the human species and that of the other animal species are only of a quantitative nature. Other animals, for instance, like humans, are able to create more or less complex organisations to increase the individuals’ capacity for survival and adaptation. They are able to give rise to structures for the governance of social aggregates, structures which are based on power and dominance relations, just as occurs among humans. And they can also succeed in producing tools for work and rules for social aggregation. Obviously the software and hardware produced by humans are more complex than those observed in animal societies. But these in fact are differences in degree not in substance. Yet a substantial difference does seem to exist. In animal species power and subjugation are developed as means, while with humans they enter the sphere of goals and aspirations. Both in animal and in human societies power is an instrument for social cohesion, a relation sustaining the governance structures of social organisms. Moreover, subjugation is an instrument for individual self-assertion, a tool used by individuals to increase their capacities for survival and pleasure, both in animal and human species. Nevertheless it is arduous to believe that the leader of a wolf herd strives for power as an end in itself, whilst it is not so difficult to believe of humans. I think it should not be hard to accept this view in the light of common experience. What is strange is that it does not appear in the list of fundamental axioms of political economy. At any rate, I posit this idea as an axiom. And, since Bertrand Russel was the first to bring its full philosophical implications to light, I will call it the ‘Russell axiom’: an insatiable desire for power is one of the most important drives of human

Forms of co-operation and power 147 beings. This is an extremely meaningful axiom in the explanation of human behaviour, but is still a little too generic. To make it more useful, it must be qualified, which I shall try to do by translating it into what could be called the ‘Perroux lemma’: capital accumulation is driven by the desire to acquire power.

Co-operation The nature of co-operation Co-operation is the harmonisation of various individuals’ actions in a group aimed at the achievement of some intrinsic benefits for at least one individual. It is not necessary that the benefits accrue to all the co-operators, nor even to only one of them. It is sufficient they accrue to somebody, even one who does not participate in the co-operation. The possibility that some co-operators might not gain any specific advantage is not a sufficient reason to exclude co-operation. Many individuals might be compelled to participate or might have no real freedom of choice. There are many ways in which co-operation may produce benefits, but they can all be reduced to three basic modalities: exploitation, specialisation and mass force. These elementary modalities are rarely observed in isolation, since they normally appear combined with each other in a more or less complex way. However, it may be useful to investigate them separately in, so to say, their pure shapes. Exploitation, in its elementary form, can be seen at work in the abstract case of a capitalist who succeeds in extracting a surplus value from workers who produce no more than they would produce by working in isolation. This is co-operation between an actor who can take decisions and some other actors who execute them. In elementary exploitation, co-operation among workers does not produce any benefits. It is the co-operation between the workers and the capitalist that produces a benefit in the form of a surplus value accruing to the latter. The second kind of co-operative benefit is specialisation, which occurs when several actors constitute reciprocity relations enabling them to transfer at least a part of the results of their actions to each other. Specialisation improves the returns from individual activities since it enables the routinisation of actions, the reduction of errors and wasted time, and the possibility of learning from experience and imitation. The simplest form of specialisation occurs in transaction, a particular case of which is the exchange of commodities. In exchange, two or more actors transfer to each other the results of specialised actions and obtain, as a co-operation benefit, the surplus return that the specialised activities yield over and above the non-specialised ones. Again in this case it is not necessary for the transaction to be mutually beneficial. Since exchanges do not necessarily take place in the absence of power, certain actors may appropriate some co-operative benefits in excess of what they give in exchange. A third kind of benefit is mass force (Massenkraft), a concept elaborated by Marx in his analysis of co-operation. It arises when the returns of the concerted actions of various non-specialised actors are greater than the summation of the

148 Forms of co-operation and power return the actors would obtain by acting separately. This kind of benefit may be brought about by economies in the use of space, time and capital indivisibilities. A classical example of space economies is that of an oil pipeline whose pipe diameter is doubled. If labour and raw materials are roughly proportional to the circumference of the pipe section, direct costs will double, whilst the oil flow, and therefore the return, which is a function of the square of the diameter, will more than double. As an example of time economies consider the crew of a rowing boat who, by rowing synchronously, reaches a higher speed than by rowing in an uncoordinated way. An example of scale economies linked to indivisibilities is that of a group of artisans who, to economise on fixed costs, decide to work in a collective laboratory. Another classical example is that of a group of workers who have to lift some blocks of marble. The dimension of blocks that can be lifted by several workers together is greater than that of the blocks which can be lifted by the same individuals working separately. Note that in these examples no specialisation is assumed. The production of co-operation benefits poses some co-ordination problems. For instance specialisation within an organisation inevitably poses problems of assignment. When different workers have different tasks requiring different competencies, it becomes necessary to assign the tasks to them. A similar problem arises with mass force: when several workers have to do a job requiring a combination of efforts it becomes necessary to synchronise actions. Thus, for instance, a rowing team can try to synchronise their movements by singing a rhythmic song together. In current economic literature the benefits of co-operation are often accounted for by resorting to concepts like ‘returns to scale’, ‘team production’ and ‘division of labour’. Some of them, like the first two, are quite useful in formal analysis. Others, like the third, are useful from a philosophical point of view. Behind those concepts, however, lurks one or another of the above-mentioned specific benefits of co-operation or a combination of them. Increasing returns to scale, for instance, are normally defined as emerging from mass force or specialisation. Team production recalls the complementarities linked to the co-ordination of specialised jobs. The division of labour, finally, occurs when the tasks of co-operating actors are assigned in a co-ordinated way so as to obtain the benefits of specialisation and/or exploitation. Elementary forms of co-operation Transaction institutions, through which collective actors are constituted and cooperation prompted, rarely succeed in defining the individuals’ claims and obligations with full accuracy. Nor are they always fully effective in enforcing people to honour their obligations. In a world in which people are moved by egoistic sentiments and information is asymmetric, the possibility of opportunistic behaviour inevitably arises. And opportunism is a strong corrosive acid for co-operation. It undermines cohesion and exacerbates distributive conflict. This is however not the worst that may happen. The most serious damage occurs when opportunism leads to reductions in co-operative benefits.

Forms of co-operation and power 149 Individual actions are not necessarily moved by egoistic sentiments alone. Altruism may be mixed up with egoism. This is true in three senses. First, egoism or altruism may prevail in different individuals. Second, as a consequence of the multiple self, any single individual, and even any single action, may be simultaneously moved by contrasting sentiments. Finally, an action may be moved by an intention which can be interpreted as both egoistic and altruistic in different time spans, and the individual who performs it may be incapable of disentangling his intentions. Altruism is an important cement for co-operation, and would eliminate the very causes of the problem, given that opportunism is occasioned by a combination of egoism and information asymmetries. It may take different forms and be manifested with different intensities: from benevolence to sympathy, from loyalty towards partners to dedication to a superior cause, from moral integrity to selfrespect. To investigate co-operation, at any rate, it is not necessary to specify the different forms and degrees of altruistic behaviour. It is sufficient to define a critical motivational threshold separating the actions which corrode co-operation from those which sustain it. To do this, a sort of minimal altruism test could be envisaged, on the ground of which one could say that a person behaves altruistically, at a given time and toward a given social group, if he avoids procuring benefits for himself by causing losses to other individuals in the group. This is minimal altruism in three senses. First, just like Hume’s ‘confin’d generosity of men’, it does not require people to sacrifice themselves for the good of others, that is, to suffer personal disadvantages in order to procure advantages for others. Second, it refers to a particular group, allowing the possibility that, through the group, people act egoistically in a wider context. Third, it applies to a given time, without excluding the possibility of a behaviour which is altruistic in the short run and egoistic in the long. The persons who pass this test are not opportunists, at least locally, and a cooperative activity in which these kinds of people predominate has a high probability of not dying for internal causes. The problem is that in capitalist co-operation, where exploitation is rule, individuals are not encouraged to comply with the minimal altruist test. At any rate, independently of the kind of sentiments that move people, cooperation requires some forms of production governance structures, i.e. social structures regulating collective action in production. Governance structures are necessary to give rise to co-operative benefits, assign tasks to people and co-ordinate activities. This is a technological use of governance structures, and must be present even if all the co-operators are altruistic. Then, if there are problems of distribution, control and enforcement, in other words, all the problems of power exercise brought about by social antagonism, there is also a political use of governance structures. Depending on the way decision-making faculties are distributed, governance structures can be divided into hierarchical and network structures. The former type is based on subordination relations, i.e. on the separation between decisionmaking actors and subservient actors. In the latter type, by contrast, all individuals preserve their decision-making autonomy. Both these kinds of structures may give rise to different forms of co-operation in which technological and political uses are mixed up in varying degrees, depending on the nature and combination of

150 Forms of co-operation and power egoist or altruist sentiments prevailing among co-operators. And, since in fact sentiments are manifold and subordination relations combine in many ways with network relations, the forms of co-operation can be many. Yet it is possible to define some elementary forms of co-operation – forms which are rather abstract and simple and in which only one type of governance structure and only one of the two kinds of sentiment appear. These are shown in Table 4.1. Governance structures are frameworks of social relations through which orders, dispositions and information are conveyed and by means of which the management and control of co-operative actions are exerted. The social relations sustaining governance structures are defined as ‘governance relations’. Since there are four elementary forms of co-operation, there will be four elementary types of governance relations. These are shown in Table 4.2. It goes without saying that in reality one rarely encounters cases of co-operation forms and governance relations corresponding to the pure models shown in Tables 4.1 and 4.2. But, for the moment, it is better to stay in the abstract, even at the cost of dealing with rather utopian forms of co-operation and social relations.

Table 4.1 Elementary forms of co-operation Prevailing sentiments Egoistic

Altruistic

Capitalist firms

Co-operative firms

State bureaucracies

Associations

Goods markets

Social movements

Companies markets

Forums

Hierarchy Types of governance structure Network

Table 4.2 Types of governance relations Prevailing sentiments Egoistic

Altruistic

Hierarchy

CommandSubordination

DelegationAdministration

Network

Exchange

Cohesion

Types of governance structure

Forms of co-operation and power 151 Some utopian forms of co-operation and governance relations The most utopian form of co-operation is association which is constituted with a partnership contract. This is a multilateral contract through which many individuals contribute their labour capacity or simply their dedication and initiative. The partners’ contributions are in principle equal. The association’s goals are decided unanimously at its constitution and coincide with those of its members. Once the association has been set up, decisions are taken by majority vote, and the members who dissent are free to conform to the ‘general will’ or leave. A worker co-operative or a political association belong to this kind of organisation. Whatever the benefits of co-operation, they must accrue to all partners in equal shares if their contributions are equal. If the organisation is complex, the partners may give a mandate to a small social body to perform a decision-making and co-ordinating function. Yet decisions are taken in the pursuit of the goals defined by the partners. Therefore exploitation and accumulation of internal power are excluded anyway. The fundamental kind of governance relation in this form of co-operation is delegation-administration. It consists of the combination of two different social relationships. The first, delegation, is the mandate through which the partners assign to the directors the task of managing the association in their own interests. Directors are agents of the participants. They do not acquire power over the principals because the latter maintain the right to take autonomous decisions. Directors take administrative measures which are passed to the partners through directions, instructions and regulations. The partners, knowing that the decisions pursue their own interests, comply with them voluntarily. They conform to an administrative instruction because they consider it an indirect emanation of their own will. This is the second relation, administration. An association or a co-operative works effectively in the interests of its members only if egoistic sentiments are kept at bay and both the principals and agents have passed the minimal altruism test. Otherwise some co-operation benefits may be lost if there are information asymmetries. If the directors do not restrain their egoism, the information and knowledge asymmetries they almost inevitably acquire may be exploited to gain personal benefits. This will tend to transform the administrative governance relationship into a power relationship and the association will lose its nature of an organisation aimed at the pursuit of all partners’ interests. More generally, if certain partners practise some form of free riding, by trying to appropriate benefits over and above the proportion of their contributions or to furnish lower contributions than their proportion of benefits, a problem of control arises which makes it necessary to set up governance structures with political uses. But control activities are costly, and the co-operation benefits are reduced. Yet it seems reasonable to believe that the gravity of this kind of problem can be limited. In an association interactions among partners tend to be long-lasting, so that everybody knows that any non-co-operative action has a high probability of stimulating non-co-operative reactions. It is in the personal interest of longsighted partners to adopt forms of behaviour that pass the minimal altruism test.

152 Forms of co-operation and power Moreover all the partners know that the production of co-operation benefits depends on the fairness of all of them. Therefore every one has a personal incentive to personally control the opportunistic behaviour of the other partners. This condition does not hold in a capitalist firm, in which the main co-operative benefit consists of exploitation, and in which the workers have good reason to shirk, solidarise with free riders and even to enact collective opportunism. Finally, to the extent that a co-operative firm is actually controlled by its members, it is probable that technologies and knowledge are used which contribute to reduce information asymmetries, so the control of opportunism becomes easier. Another utopian form of co-operation is a social movement. This is a group of individuals who act together in the pursuit of a common goal, without giving rise to a formal organisation. It exploits mass force as the main co-operation benefit: the effects of a common action are greater than the summation of the effects of individual actions. All the participants undertake the same action simultaneously, for instance the interruption of work during a spontaneous strike, so that there is no specialisation. However, co-ordination is necessary, if for no other reason than individual actions must be synchronised. Since there is no functional organisation, with explicit rules, division of tasks and formalised decision-making procedures, co-ordination and decision require a committed and voluntary governance relation, a relation, that is, which involves all members in the same way and springs directly from their will. This is cohesion: all participants strive to conform spontaneously to collective action, to be part of it and to lose themselves in the mass. Social movements do not have directors whose functions are defined by an agency relationship. They may have leaders, i.e. individuals with whom the participants identify and to whom they attribute, by following them, the role of guide and spokesman of collective action. Leaders are selected by the participants’ action. They must be able to bring to light the collective will, which they do by giving it voice. When they succeed in this endeavour, they obtain the participants’ consensus. If, instead, they do not succeed, they lose their followers. In certain moments of inaction the participants’ will may be larval and hazy. Yet action requires it to be manifested. The leader plays a maieutical function: he interprets the participants’ implicit will, enabling it to become an effective collective will. But it is the participants’ will anyway. In the cohesive governance relationship it is the movement that creates the leader, not the other way round. Collective decisions are taken spontaneously and without a vote, there being no formal rules to sanction them. Usually they are taken unanimously by the simple act of participating in the action. There can be no dissenters or, if there are, they cannot manifest themselves as such, since a minority of dissenters who do not conform to collective action are immediately perceived as anti-social and counterproductive. In fact they would undermine the production of the main cooperative benefits of social movements, i.e. mass force. Dissenters either keep silent and conform or leave. In these social aggregates, even more so than in associations, egoistic behaviour must be kept at bay. The crucial problem, here, is not opportunism, since it is

Forms of co-operation and power 153 difficult for strong information asymmetries to arise when all participants do the same things. Sly people are easily recognised and expelled or induced to conform. The most serious problems are created by the leaders’ behaviour. If these are moved by the thirst for power or other egoistic sentiments and have access to some asymmetric information and knowledge, they may be easily transformed into charismatic heads, i.e. into actors endowed with influence. And the leaders can indeed acquire specific competencies which enable them to personally exploit their position. But since the movement survives only by virtue of cohesion and the free will of participants, it tends to disband when cohesion transforms into a power relation. Such a transformation implies that, beside the benefits brought about by mass force, a new particular benefit arises which the participants judge extraneous to the original goals of the movement. Alternatively, instead of disbanding, the movement can transform into a formalised hierarchical structure, for instance a Party or a Union. A genuine social movement, i.e. one pursuing the goals of its participants, is a rather unstable collective agent, at least if it moves in a social context in which egoistic sentiments prevail. In a social movement, intense flows of communication pass among the participants. People continually talk to each other and transmit information, ideas, beliefs, proposals, projects and dreams. It is through these flows that that larval form of general will is created which the leaders then transform into a collective decision. In a movement, therefore, the form of co-operation called ‘forum’ also operates. But a forum is not necessarily a movement. It is defined as a group of individuals who communicate with each other without a bond of subordination. Social relations in a forum have a network structure. Participants are equal in that they all have the same right to communicate. And they participate, not only to appropriate the contents of messages transmitted by others, but also to put forward their own convictions. Thus an altruistic impulse or a desire for merits recognition or both are normally present in a communicative co-operation. People participate driven by the impulse to enrich the ‘collective mind’, and transmit their convictions to others because they believe they are collectively useful and valid. Anybody can enrich themselves spiritually by receiving others’ messages, but communication would cease if everybody tended only to receive and not to transmit. People participate in forums also to show off their merits, and this implies the donation of personal contributions. In forums, altruism and egoism tend to coalesce into sentiments and attitudes that favour spontaneous co-operation. Therefore the governance relationship prevailing in this co-operation form is again cohesion, which however, unlike in a social movement, aims at uniformity of thought, rather than unity of action and will. Without an intrinsic drive to uniformity of thought there is no motivation to communicate in a forum. The main co-operative benefit of a forum is specialisation in knowledge production. Obviously, also in a forum egoistic motivations and sentiments can produce destructive effects. Malevolent and selfish individuals may try to transmit wrong and dangerous messages. Just think of computer viruses. Even worse, individuals in privileged positions in networks, or endowed with capital or authority, may try to deform the network structure of a forum and transform it into a centralised

154 Forms of co-operation and power communication structure, and then use this to pursue their own particular goals at the expense of other participants. Think, for example, of the chairman of a scientific or political conference who uses his authority to select the order of speeches according to his own preferences. However, such a kind of problem which normally occurs in centralised communication structures, is less likely to occur in a forum or, at least, can be coped with more easily by the participants’ action. This is mainly because, although messages contain information which is asymmetric ex ante, i.e. before communication, they become public knowledge once communicated. In fact a forum can be interpreted as a communication structure aimed at the reduction of information asymmetries or the enrichment of individual information sets. Moreover, people who participate in a forum normally have, besides a specific interest in the kind of information transmitted, also some common competence in the subject matter of communication, so that it is difficult to deceive them with false or misleading information. One might find it strange that associations, movements and forums are considered utopian forms of co-operation. Whilst I showed that problems can be caused by opportunistic behaviour, I also argued that they can be worked out or controlled by the participants, and that it is probable that a minimal altruism test will be passed by most individuals so long as co-operation is moved by the will to pursue a common end. The point is, however, that not all problems arise from internal causes, i.e. the participants’ behaviour. Many of them are caused by external forces. In a capitalist system co-operative groups operate in a social context of struggle for dominance, a context in which egoistic behaviour is necessary for survival and self-assertion. As a consequence the minimal altruism test might be difficult to pass for most people. It is possible that the social context exerts such strong pressure on co-operative groups as to push them to modify their internal structures and dynamics to the point of altering their nature. This phenomenon will be investigated in detail below, after the problem of power has been elucidated. At any rate, the term ‘utopian’, as I use it here, does not mean ‘unrealistic’ or ‘impossible to achieve’. It rather means ‘alternative’, and intends to convey the idea of a form of co-operation working on the grounds of institutions and governance relations which are at variance with those prevailing in a capitalist society. A utopian form of co-operation might be difficult to survive in such a society. But the world changes. Markets and capitalist firms A market could also be conceived as a utopian form of co-operation. Even if one does not adopt all the neoclassical hypotheses warranting the beneficial effects of atomistic competition, it would be sufficient to assume that all exchanges take place among actors who have passed a minimal altruist test. Then commodities would be sold at their production cost, since nobody would appropriate a value over and above what is offered in exchange. If the cost of production included a fair remuneration of the labour used in production, prices would coincide with embodied labour and the principle of equivalence would regulate all transactions.

Forms of co-operation and power 155 But these hypotheses are not appropriate for describing a capitalist economy. In market exchanges, actors aim at the benefits of specialisation and enter into transactions in which everybody preserves some decision-making autonomy. Thus the possibility of formal subordination relations is excluded. Individuals are moved by egoistic sentiments, so that this form of co-operation can be used to perpetrate exploitation. Then some kind of power must be at work which is not reducible to formal subordination, but makes the extraction of a surplus value from exchange possible. Normally people do not participate in market exchanges in the pursuit of collective goals. And no law can compel people to pass a test of minimal altruism, so that nothing can prevent the possibility of unequal exchanges. In a capitalist firm the worker who does not obey can be fired – a sort of forced exit resulting from a free exit decision of the employer. In markets there is no possibility of forced exit. Exit remains a free option exerted only when there is some incentive. Yet this freedom may be imperfect, for instance because an actor who dislikes a transaction might have no better alternative options. Normally changes in the distribution of information, asset specificity, wealth and market power take place under conditions of asymmetry. With information asymmetry, better informed parties can enact opportunistic exploitation. The other party’s only form of defence is exit, but, as is well known, this option could cause the market to disappear – a situation which makes everybody worse off. Therefore the exploited party might accept the situation even if he knows he is buying a lemon. A poor car driver might prefer to run the risk of buying a bad car rather than be left without any car at all. Asset specificity too can cause market exploitation. If a seller wishes to sell a commodity which is demanded only by a specific party, while the latter offers in exchange goods, e.g. money, which are desired by many other actors, then the seller is exposed to the buyer’s opportunism. He has no alternative but to deal with that buyer, and thus he has no real exit freedom. Similar to this is the deformation in the terms of trade that can arise from an asymmetry in market power, which occurs when a contracting party has no competitor who sells (buys) the same commodities at better conditions, whilst the other party has competitors who would buy (sell) that commodity at those conditions. In this case the weak party has no alternative but to accept the exchange at the terms established by the strong party. The two parties do not have the same degree of exit freedom. These forms of asymmetry, however, are not sufficient conditions for market exploitation. A necessary condition is that there is some asymmetry also in the distribution of wealth, so that some actors are in a situation of poverty. If there is no poverty, all the actors can take advantage of that particular exit form which consists of the refusal or postponement of exchange. The greater the individual wealth, the wider is the exit freedom and the lower is the possibility by other parties to use market asymmetries to extract surplus value from exchange. The greater the poverty, the narrower is exit freedom. ‘Wealth effects’, from which neo-institutional literature often abstracts for the sake of simplicity, are widespread in capitalist systems. But they should properly be called ‘poverty effects’.

156 Forms of co-operation and power Summing up, the governance relationship regulating co-operation in markets is exchange. Through it, people try to produce the benefits of specialisation. Yet in reality several forms of asymmetry exist which transform markets into battlefields where the actors endowed with power and privileges can also obtain the benefits of exploitation from exchange. Now look at the capitalist firm. Its production governance structure is based on hierarchy. This has technological and political uses. The technological uses are basically two: co-ordination and knowledge-production. Co-ordination aims at producing the benefits of specialisation and mass force. It is in hierarchical structures that roles and tasks are attributed, assignment and synchronisation problems worked out, scale and scope economies prompted. Since a firm always endeavours to develop its specific capabilities, improve the use of techniques, innovate and adapt to a changing environment, its internal hierarchy is also used in knowledge-production processes. The firm’s know-how is mostly embodied in machines and is, so to say, crystallised in labour routines. Yet knowledge is not something that is acquired once and for all. Rather, it is continually modified, since the environment is competitive and permanently changing and the innovation process is incessant. The use of techniques requires a ceaseless problemframing and problem-solving activity which spans the entire hierarchical structures of firms. Knowledge-producing hierarchy is the articulation of the firm’s collective mind. Besides these two uses of hierarchy, which relate to any kind of firm in general, including democratic and co-operative firms, there are another two, which are typical of the capitalist firm. These are the uses in exploitation and control activity. All kinds of co-operative benefits can be produced in a capitalist firm, but exploitation is the most important. Hierarchy shapes all social relations in a capitalist firm, and the fundamental form of hierarchy is that which binds the people who obey to the people who command. This is a pure power hierarchy. Its existence is a necessary condition for capitalist exploitation in the production process, as I showed in Chapter 1. The firm’s goals proper do not coincide with the goals of all the actors who participate in its activity and cannot therefore be considered as collective goals. Since exploitation is one of the firm’s goals, there is no reason to expect the workers to co-operate eagerly in its achievement. And, given the technological dynamism of the capitalist firm and the necessity to articulate collective knowledge throughout the entire hierarchy, it often occurs that information and knowledge asymmetries enable the employees to defend themselves from the excesses of exploitation and, so to say, limit the damages from work. This is what makes control necessary. Monitoring, metering and disciplining the workers aim at limiting the damages from shirking. Thus the fundamental governance relation in a capitalist firm is command. But there are others. Since a firm entertains exchange relations with external actors, customers, suppliers, contractors, creditors and so on, its social dynamics is also regulated by the exchange governance relation. Exchange often regulates internal relations too, especially in big companies organised in M-form, where transactions

Forms of co-operation and power 157 among factories, departments and divisions can be implemented in internal markets or quasi-markets. The forms, functions and working of production governance structures in market and capitalist firms have only been sketched here. They will be investigated more deeply in Chapters 5 and 6.

Power Power and its forms Power is a social relationship between two actors, a dominating and a dominated one, by virtue of which the former determines at least in part the behaviour of the latter and defines his goals. The capacity to define goals is fundamental because, if the goals of actions are defined by the individuals who perform them there is no power relationship. For instance the workers who execute a director’s instructions in a co-operative firm are not subject to a power relation if the director pursues the goals defined by the workers themselves. Nor is it necessary for the goals to coincide with the interests of the dominating actor, who could take decisions as an agent of a third actor and impose their execution on the dominated actors. A worker who obeys the orders of a boss or an officer or a director is subjected to a power relation even if the ends of his activity are not those of the persons who directly exert power. The social actor, in this definition, is defined in a general sense: he can be an individual or a collective agent, a firm, a committee, a government. A firm can have power over another firm, the government of a nation over that of another. They must be actors, however, i.e. social subjects who are able to take decisions and define goals, not just social aggregates. A social class ‘in itself’ is not a social actor and cannot establish power relations with other classes. A class, however, can give rise to organisations and movements that pursue its interests. Organisations and movements are social actors and, since they enter power relations with other actors, one can metaphorically say that they give rise to class power relations. I said that a dominating actor has the capacity to determine the dominated actors’ behaviour. Since the definition is rather general, the term ‘determine’ is intended in a wide sense, to the point of embracing the meanings of ‘conditioning’ and ‘limiting’. Therefore different forms of power can be defined, depending on the meaning attributed to that term. In particular, research on power has singled out three different forms, which are denominated in various ways by different authors and which I will call ‘command’, ‘influence’ and ‘domination’. Command is the capacity to determine the activity of the dominated actor by means of orders. The dominating actor has the prerogative of taking decisions and ordering the subordinate to execute them. The latter has the obligation to obey. The relation constituted by the employment contract is a typical example. Formally the worker is obliged to comply with the employer’s will.

158 Forms of co-operation and power Thus the contract deprives him of decision-making capacity. Nevertheless a worker is left with some autonomy of action, for it is impossible to determine precisely the care and effort with which orders are executed. Moreover the freedom of contract enables the dominated actor to resign or to claim a revision of the contract terms. The right to strike, complain and protest derive from this prerogative. A strike in fact, as I will better argue below, can be intended as an attempt to impose a contract revision by transforming the exit faculty into a voice faculty. The employment relation normally is a contested terrain: many terms of the power relation can be contested, discussed and revised in the very process of power exercise. Yet it must be made clear that this partial autonomy does not create a dominated actor’s power. The dominated actor can resist the execution of orders, practise passive or reluctant obedience, but he cannot give orders to the employer. His autonomy is reduced, at most, to the capacity of limiting the dominating actor’s power, never to that of acquiring power. I will call this capacity ‘counter-power’. Subordinates, however, may have power. A firm’s hierarchy does not consist just in the relation between an employee who obeys and an employer who directly gives commands. There is normally a long ladder of directors, officers, bosses – cadres, for brevity – who have been engaged with an employment contract and therefore are subordinate actors, but who are used by managers to transmit orders, to exert control, to meter performances, to find solutions to production and organisation problems. Cadres undoubtedly exert command over workers in the lower layers of the hierarchy. Their power derives from the normative structure of the organisation. Its base is authority, a property assigned to their role by internal regulations and the managers’ decisions. The orders cadres give to their subordinates consist in the transmission and translation of the orders they receive from their superiors. Theirs is ‘delegated power’, power delegated by the superiors. Asymmetric information can here give rise to a form of partial autonomy. Cadres know the abilities of their direct subordinates better then their superiors and can use this knowledge to opportunistically acquire power over the former and limit the power of the latter. More importantly, an efficient use of hierarchies for working out production and organisation problems requires that cadres be enabled to take some autonomous decisions. A cadre is not just a machine for the transmission of orders, he is also a device for the solution of problems. As such, he must be able to find solutions, develop ideas, give his own orders and have them executed. He must be put in a position to exert a genuine power of command and decisionmaking, even if within the boundaries posed by the directions he himself has received. Thus not all power exerted by cadres is delegated power. Part of it is ‘residual power’, that is, the capacity to autonomously determine at least part of their subordinates’ actions to make them pursue goals defined by the cadres themselves. For the time being there is no need to distinguish between the residual power a cadre conquers opportunistically and that granted him by the authority structure of an organisation. Suffice it having clarified that a cadre’s power is normally the summation of two components: delegated power, by which he transmits orders from the upper to the lower layers of the hierarchy, and residual power, by which he is able to implement his own autonomous decisions.

Forms of co-operation and power 159 A different form of power is influence, which consists of the capacity of a dominating actor to condition the behaviour of a dominated actor without using explicit orders. Behaviour must be influenced in the pursuit of goals defined by the former. Yet, unlike command, influence is exerted on an actor who is legally autonomous, i.e. who has no obligation to obedience. The decisions taken by the dominated actor are taken by himself, but they are taken to pursue at least partially the goals defined by the dominating actor. Influence relations are typical of, although not exclusive to, markets, and occur whenever a transaction transgresses the principle of equivalence. Whenever an unequal exchange takes place, influence is lurking behind the transaction. Contracts for services, of agency, of sale fit this kind of power very well. We undergo the influence of the plumber when, by using his asymmetric knowledge, he makes us pay too much for his services. We undergo influence when we buy goods at monopolistic prices. Normally, since the dominated actor in an influence relation maintains a certain degree of autonomy, he is also able to take decisions to partially pursue his own goals, yet he must do so in such a way as to reconcile the pursuit of his own goals with that of the dominating actor’s goals. Thus this form of power, even more so than with command, is an ideal field for the contested exchange game. And since, for reasons I will set out in the next section, the distribution of the sources of power is strongly asymmetric in a capitalist society, François Perroux is perfectly right in defining the market as a network of influence relations. How can a dominating actor in an influence relation impose the pursuit of his own goals if he is not able to give orders? He can do it in different ways. Ex ante, he can use privileged possession of the power sources, for instance, to induce the dominated actor to accept unfavourable contract conditions. Ex post, he can do it by manifesting his wishes during the contract implementation. If the dominated actor knows that the other party has the capacity to break the contract and cause him losses, and if these are higher than the losses that would derive from complying with the dominating actor’s requirements, then these requirements will be accepted. But often it is not even necessary for the requirements to be explicitly manifested. C. J. Friederich formulated the well-known rule of anticipated reactions, according to which dominated people are induced to conform to the dominating actors’ unexpressed wishes in anticipation of their possible reactions. Influence is a rather elusive form of power. Given that there is no formal bond of subordination between the parties, and that their relation provides benefits to both of them, how is it possible to reveal the existence of influence? A simple criterion would be exploitation: when a contracting party obtains a higher value in a transaction than he gives in exchange there is exploitation and we may surmise this is a fruit of power. For a better understanding of the social meaning of influence, it might be useful to compare it with command. If one observes the ex ante conditions of the transactions giving rise to the two forms of power, no difference will be noted: in either case the transacting parties are endowed with freedom of contract. Yet if one looks at the ex post conditions, i.e. at the relations arising during implement-

160 Forms of co-operation and power ation, interesting differences come to light. In the command relation the dominated actor is not autonomous, but acquires some de facto partial autonomy by exerting counter-power in the attempt to limit the dominating actor’s power. In the influence relation the opposite situation arises: the dominated actor is legally autonomous but undergoes real conditioning which may make him lose some of his decisionmaking autonomy. The substantial similarity between command and influence accounts for the possibility of extracting a surplus value by using either form of power. A dominating actor, instead of engaging workers with an employment contract, could mobilise them by using a contract for services or a sharecropping contract. If he can impose contract conditions enabling him to extract a surplus value, then there is exploitation and influence. It is, however, a kind of exploitation perpetrated through control of the exchange process instead of the production process. Often this kind of transaction is preferred by dominating actors to cope with monitoring and control difficulties, and is used to complement the employment relation. Piecework, for instance, is a case in which the employment relation is complemented with clauses from a contract for services or a sharecropping contract. When the production process gives rise to high monitoring cost, it might be profitable for the employer to offload this cost on to the employee. He might pay an income higher than a normal wage, in this way, but if the difference in pay is lower than the cost of monitoring, it is worth disbursing it. The worker might find it profitable to accept a form of contract envisaging a sort of over-exploitation. He might produce a higher surplus value (net of monitoring costs) than he would produce under a simple employment contract. But, since he will exert self-control, he will have no monitoring costs at all, and will consequently obtain a net income that is higher than the normal wage. The third fundamental form of power is domination, which consists of the capacity to constrain the actions of dominated actors by issuing norms to which the latter are compelled to conform. A rule does not determine behaviour. It is not a decision. Yet it poses limits on decisions by imposing prohibitions or prescriptions. Social actors are autonomous decisors, but, since any actor’s decision determines actions which may affect or alter the decision-making context of other actors, general rules are needed. If all actors in a social system contribute to the production of rules, so that the norms ends are shared by everybody and defined by the collective will, then the individual’s compliance to the normative system does not imply coercion and power. But normally rules are not produced with the participation of all the actors. In an indirect democracy, laws are an expression of collective will only if the actors delegated to produce them are effectively controlled by those who delegate them. And, as I showed in Chapter 3, in a capitalist system there are usually substantial information and knowledge asymmetries that considerably alter the delegation relationship between the political class and the social actors. Therefore there is no guarantee, in such a system, that norms are functional to the pursuit of the goals of all people. In a system in which not all the actors contribute to the production of general rules, those who do contribute acquire power: not only are they able to restrict the decision-making field of social actors,

Forms of co-operation and power 161 but they can do it in the pursuit of goals that might not coincide with those of the community. Social actors, on their part, can try to condition the political class, i.e. to exert some political influence, in the attempt to obtain rules enhancing their power in social transactions. In this way an influence deriving from political domination can be added to the power some social actors already directly exert in the social arena. Power can be used to accumulate power. Those who have influence and command in the spheres of markets and production activities can use it to increase their power bases and transform them into political influence. Thus political domination may work at reinforcing the bases of economic power, in a vicious circle which often contributes to stabilise power relations and social structure. The same phenomenon may occur in a private organisation. In an association or a co-operative, for instance, some members may develop a certain influence over some others and acquire the power to produce regulations or to condition their production so as to make them serve goals that do not coincide with those of all the members. In this way, the power exerted in the form of influence is complemented by that exerted in the form of domination. Both forms of power might sustain each other and accumulate to the point of altering the structure and goals of the association. A similar process normally occurs in a capitalist firm, but, obviously, without the need to alter anything. From their position of authority the managers obtain the prerogative of giving orders to the workers, as well as the faculty to produce internal regulations. Therefore they exert domination, beside command. And it goes without saying that domination can be used to make command more effective. Bases, sources and instruments of power According to Hobbes (1931: 43), power is a ‘present means to obtain some future apparent goods’. This definition has been criticised by the theoreticians of ‘power as a social relation’. It seems to reduce power to a thing, an object which can be possessed, wealth being an example. Criticism of this kind, however, is somehow off target, since, as I will argue below, the power relation can indeed be seen as a good, namely a positional good. But, apart form this, there is undoubtedly a sense in which wealth, for instance, is power: it can be used to employ workers. To say it in Hobbes’ own words – ‘Riches, joyned with liberality, is Power; because it procureth friends, and servants’ (1931). True, wealth as such – say, in the form of a money balance – is only a sign and does not at that stage represent command over labour. It is a form of potential power. So long as it is idle, it is only paper. Yet it can be invested and transformed into actual power. Thus, there is less contradiction than might seem at first sight between the two notions of power. In reality Hobbes’ definition relates, not to power proper, but to its bases, that is, to the socio-economic conditions from which power can originate. In the literature on power, the two definitions are formulated in terms of a ‘dispositional’ and a ‘relational’ conception. Propositions like ‘the president has the power to dissolve Parliament’ or ‘a lobby has the power to condition Parliament’

162 Forms of co-operation and power refer to dispositional power. Many scholars believe the dispositional conception is more important than the relational one and, anyway, that propositions referring to the latter can be translated into propositions referring to the former. This is an ambiguous idea, though, and often produces confusion rather than clarity. It tends to obscure the simple fact that power is substantially the power of man over man. It can, however, be accepted if reformulated in the following way: the dispositional conception refers to the bases of power rather than to effective power. There is thus a precise sense in which the relational conception can be reduced to it: there can be no effective power if there are no bases for it. Then one can say, not that ‘the president has the power to dissolve Parliament’, but that ‘he has power over Parliament because his office gives him the authority to dissolve it’, and not that ‘a lobby has the power to condition Parliament’, but that ‘it has influence on some members of Parliament because it has enough wealth to bribe them’. There are three kinds of power bases: wealth, force and authority. Wealth is a socially recognised stock of value, i.e. a set of goods, material, immaterial or symbolic, to which social actors attribute value. Physical goods, like land, means of production and consumer goods, belong to the material form. Intellectual endowments, knowledge, information, skills, to the immaterial. Money to the symbolic. Force is the capacity to exert physical coercion over things or persons. It is exalted, in the modern world, by the development of increasingly potent weapons, but especially by the development of organisations in control of social coercion, armies, police forces etc. Authority is the property of a social role constituted by a normative system, and consists in the faculty of determining the activity of people located in subordinate roles. It must not be confused with power: this pertains to social actors, whilst authority pertains to social roles. The social actor’s power can be constituted with transaction institutions, social roles with normative institutions. The former is a social relation, the latter the property of a role, i.e. a thing, however abstract. Authority is a base of power.1 The individual who holds it gets power by virtue of it, but it exists independently of the particular person assigned to the authority role. Authority is a symbolic object, it can be possessed, acquired or lost by an individual, it cannot be created by individuals. It is created by the rules governing social relations. To understand the differences between power and its bases, on the one hand, and wealth and authority as bases of power, on the other, consider a small enterprise whose owner uses capital to hire a few workers. The owner’s power in the production process directly takes the form of command. Its base is the wealth that allowed the capitalist to pay the workers. Now compare this case with a big joint stock company. The firm has legal personality. Its capital is supplied by shareholders and bondholders; but its transactions are stipulated by the company itself, not by its shareholders. It engages workers and constitutes itself as an employer. The company, not its shareholders, is the party to the employment contract. Who, then, has command over the workers? Clearly, the company’s functionaries. But these are not the owners of capital. What is the base of their power? To answer this question consider that the employment contract, by instituting the workers’ obligation to obedience, gives the employer the prerogative of emanating the rules

Forms of co-operation and power 163 required to regulate production activity. Thus it is the company’s normative system that creates the roles of command, i.e. the authority structure of the firm. The functionaries acquire command over labour by virtue of the role they are assigned. Therefore authority, not wealth, is the base of their power. Authority relates to roles, not to the individuals who hold them, and individuals obtain power because they are endowed with authority. The workers obey the functionaries, not because the latter are owners, but because, having undertaken the obligation to obey the company, they credit it with the right to emanate the rules governing their productive activity. Although the institutional base of their obligation to obedience is the employment contract, their obedience to the managers is founded on and legitimised by an internal normative system. The firm, as a legal person, uses the shareholders’ and bondholders’ wealth to constitute itself as an employer and the internal regulations to generate the authority of its functionaries. The firm transforms external wealth into internal authority, translates one power base into another. Wealth, force and authority are bases of power in a rather generic sense, and only under some particular conditions can they be used to generate effectual power, i.e. can become sources of power. Since power is a social relation, the existence of an actor able to acquire it is not sufficient for its emergence, the existence of an actor who has to endure it is also necessary. In a system in which all individuals possess the same amount of wealth, this cannot be used to generate power. The same is true of force: if all individuals have the same force, nobody can use his own to subjugate others. Likewise there can obviously be no power relations between two persons who have the same authority. The bases of power can become its sources only if they are unequally distributed. If a person does not like a particular social relation he can abandon it, i.e. can exert an exit faculty. And, if an actor does not like his subjection to another, only a limitation of his exit faculty may induce him to accept such a relation. The bases of power can be used to constitute a power relation only if the asymmetry of their distribution gives rise to asymmetry in the exit faculties. Wealth as capital sustains the employment contract because the workers are inadequately endowed with it. If they do not possess any means of subsistence, they cannot avoid working. If they do not control any means of production, they cannot avoid working under the command of an employer. By contrast, since the latter controls the means of subsistence and production, he can fire an insubordinate worker and substitute him with another or reduce production. In addition, wealth as capital determines the size of the firm and can affect its market share. The greater the size, the greater the potential market power. A monopolistic firm has of course an advantage over atomistic customers, for the latter have no exit faculty if they need the firm’s products. On the contrary the firm has some exit faculty, since, if a customer does not want to pay its price, it can offer its products to another customer. Privileged information also is a wealth form that can be used to give rise to power relations. The directors of a co-operative firm get information on suppliers, on the firm’s structure, on investment opportunities etc. A worker who observes a firm’s bad performance but does not know the reasons for it, has little exit faculty

164 Forms of co-operation and power in his relation with the directors: he does not know their faults, nor whether there are better managers, let alone how to reorganise production. The directors have this information. Thus it is easier for a director to fire a worker than vice versa, and it is possible that the delegation-administration relationship transforms into a power relationship. Finally an individual’s resources can be more or less specific with respect to alternative uses. Consider a supplier and his customer. If the supplier has resources which can be used only in connection with the customer’s plant, whereas the plant can use alternative suppliers’ resources, the exit faculty is asymmetrically distributed and that customer is able to exert some influence over the supplier’s decisions. A particular form of asymmetry in exit faculty is generated by authority. A subordinate worker who dislikes the authority of a superior can exit only by leaving the company. Conversely, an officer who does not like the behaviour of a subordinate worker need not leave: he can fire, move or demote him. Moreover, since authority is instituted by normative systems, a superior always has greater ability than a subordinate to produce, interpret and apply rules. The bases of power are their sources only as ex ante conditions of its existence, i.e. as social and economic conditions for asymmetrically distributed exit faculties in a transaction. However, their capacity to sustain a power relation is even better understood by looking at social action from an ex post perspective, that is, by observing the circumstances guaranteeing implementation and enforcement. In this perspective the bases of power work as the instruments for its exercise. What ensures that a dominated actor will stay loyal to his obligations, will execute the orders of the dominating actor or accept his conditioning or respect his rules? There are two kinds of instruments of power: sanction and legitimisation, each of which may be internal or external to the power relation. An internal instrument can be used without resorting to actors or rules that are not directly involved in the relation. Perhaps the most effective form of sanction is what could be called ‘forced exit’, for instance the firing of an employee or the termination of a contract. This would not be a sanction if there were no kind of asymmetry. But, since a power relation is based on asymmetry, exit is always reduced to a punishment when exerted by the dominating actor. Most transactions generating power relations in a capitalist system are self-enforcing precisely because they are based on a fundamental social asymmetry. A worker obeys principally to avoid being fired. A supplier dominated by a customer strives to satisfy his demands because he does not want to lose outlets for his business. A monopolist’s customer pays exorbitant prices because he does not want to give up the commodity sold by the latter. Forced exit, however, is an extreme form of sanction. It has a drawback: it kills the relation. Yet its very availability is an instrument of power. In most cases the simple threat of being fired is enough to induce the dominated actors to conform to the will of the dominator. Another extreme type of sanction is punishment, a kind of instrument which is sometimes instituted by the contract itself, for instance in the form of provisions for fines, penalties, advance payments, deposits etc. A dominated actor who accepts

Forms of co-operation and power 165 a transaction providing for punishment, is then compelled to accept this when it is inflicted. Refusal to undergo it can cause termination of the contract. This means that the efficacy and enforceability of ‘minor’ sanctions are sustained by threat of the major one, i.e. forced exit. In other words it is always the bases of power that sustain its exercise. In the majority of cases, though, the efficacy of a disciplinary apparatus is sustained by a normative system which is external to any specific power relation. This occurs, as shown in Chapter 3, because transaction institutions are completed by normative ones. Actually all modern legal systems tend to codify transaction institutions. As a consequence, the self-enforcing efficacy of a transaction is reinforced by a general normative system. This gives us further enlightenment as to the class nature of modern legal systems: since they strengthen the self-enforcing efficacy of transactions which constitute power relations, they play a supporting role for bases of power and render the consequences of their unequal distribution more effective. Thus the problem of legitimisation as an instrument of power is posed. The enactment of a sanction when a transgression occurs is an exceptional case. Normally there is little need to resort to it. Usually the dominated actors comply voluntarily. Why do they comply? Well, often they do it because they consider the relation to be legitimate. All transaction institutions are potentially self-legitimising in a capitalist system, since they are used by actors who are endowed with freedom of contract. From a legal point of view a worker enters into an employment contract of his own free will. He undertakes obligations as a free subject. Thus it is the manifestation of will that makes the relation legitimate. But also in this case the external instruments of power play a fundamental role. The capacity to make contracts is sanctioned and legitimised by the general legal system. General legal systems produce a sort of meta-legitimisation of the power relations arising in civil society. In addition, legal systems also institute the authority roles of the state’s officers and the rules constraining the citizens’ actions. Therefore the law contributes not only indirectly, but also directly, to generate asymmetry in the distribution of one of the most important bases of power, authority. And since public authority sustains the authority generated within civil society, the state builds the foundations for the entire structure of social power. Accumulation and the desire for power The bases of power assume the meaning of sources when considered as ex ante conditions for power and that of instruments when considered as ex post conditions for its exercise. But they are also the means for power accumulation. Economists are accustomed to look at accumulation as a process directly involving wealth and few of them have investigated the accumulation of power. ‘Accumulation’ is usually intended as a process whereby a stock of goods increases. But power itself can be considered as a good. This may seem strange, having spent so many words explaining that it is a social relation. No wonder. It is only a

166 Forms of co-operation and power matter of looking at the same object from another angle. The world is full of social relations that are goods, i.e. goods which can be enjoyed just as social relations. Teaching, learning, loving, playing, attending an art performance, are all social relations that produce satisfaction. They are desired by social actors and therefore considered as goods. In general, all social relations in which there is transmission of information and knowledge or manifestation of emotions and sentiments are relational goods. A particular class is represented by positional goods. These are commodities whose fruition on the part of one person causes another to consume a negative quantity. Winning a literary competition, being a member of an exclusive club, winning an honour, showing off a status symbol, these are all positional goods. Prestige and power belong to this class of commodities. Thus, if power is a good, one can desire its accumulation. After all, its consumption does not imply destruction of the object: the enjoyment of power over an individual takes place only so long as the social relation persists. Therefore if it is considered a consumer good it must be likened to a durable commodity, and in this form it can be accumulated. This is the reason for the importance of the Russell axiom and the Perroux lemma. The Russell axiom says that the desire for power is one of the most important of the insatiable human desires.2 It could be considered as a self-evident axiom if it were not so generic. The problem is that it does not specify the social and cultural conditions that make it significant; nor does it specify the particular kind of actors to whom it applies. To make it more meaningful one has to qualify it. First of all, power is not necessarily desired by all actors. Yet there are at least some whose actions are inspired by the quest for power. Second, that axiom, whilst not implying that the desire for power is a natural characteristic of human beings, holds true in many historically observed societies and certainly in a capitalist system, in which it can be reformulated in a particular form: capital accumulation is driven by the desire to acquire power. This is the Perroux lemma. The axiom is useful in accounting for the dominating actors’ behaviour in any hierarchical society. The lemma serves to explain the entrepreneurs’ behaviour. Here is a problem in which neoclassical economics is rather lacking. While a utilitarian philosophy seems well suited to account for consumers’ behaviour as driven by the desire to maximise utility, it is unable to justify a motivation for profit maximisation. What utilitarian pleasure is achieved by the accumulation of wealth? Capital does not consist of consumer goods, thus it can give no direct satisfaction to its owner. One could say that it serves to buy consumer goods and that its utility is indirect. But then the profit motive should be expressed as a function of the capitalist’s consumption preferences, so that the investor’s behaviour should not be explained as different from that of the saver. In this case, however, the maximisation of utility in consumption cannot imply the maximisation of profit at any time. Accumulation implies earmarking a part of income which is subtracted from immediate consumption. And, unless one assumes entrepreneurs always prefer to postpone consumption, it is difficult to understand why they should maximise profits indefinitely. Whereas beyond a certain level, wealth accumulation cannot

Forms of co-operation and power 167 be motivated by maximisation of expected utility, even in a non-satiety hypothesis. Life does not last forever, and everybody, sooner or later, reaches an age when the life expectancy makes further postponement of consumption stupid. Moreover it seems plausible to believe that that age will be lower the higher the accumulated wealth. Thus, unless one assumes that entrepreneurs retire precisely when they reach that age, and therefore retire earlier the richer they are, the hypothesis that entrepreneurial actions are permanently driven by a profit motive aimed at future consumption is untenable. Nor can one accept the idea that wealth is desired for itself. This is the miser’s motivation. It explains some forms of behaviour, certainly not that of the entrepreneurs. The miser in fact enjoys not spending his money. The capitalist, instead, enjoys spending it (to buy investment goods). Finally one could resort to that ‘spontaneous urge to action rather than inaction’ which Keynes referred to in The General Theory, or a special predilection for risk bearing with which certain actors might be endowed. Now, it is true that there are many people who love risk, otherwise lotteries, casinos and stock exchanges would not exist. But apart from the fact that this hypothesis is normally rejected by neoclassical ideologists, and pour cause,3 it would not explain the wide variety of refined risk hedging practices which are normally associated with entrepreneurial activity. A joint stock company is not a casino. The question remains open: what motivates entrepreneurial activity? The Perroux lemma gives us a precise and fully satisfactory answer. It is not a simple ‘urge to action’, but an ‘urge to social action’, a passion for power accumulation. Power is a desirable good, one that is not destroyed in consumption, one that is enjoyed when it is ‘taken away from other people’. A person enjoys it the more, the less it is enjoyed by others. Furthermore the more you have of it the more you desire. For power, rather than for potatoes, it seems sensible to assume non-satiety. So this lemma accounts very well for a kind of behaviour consisting in the reduction of expenses for the purchase of normal consumer goods and the increase of expenditure for the purchase of goods which procure no direct utility. This lemma alone is able to justify the prophet’s maxim: ‘accumulate, accumulate!’ If the social meaning of wealth accumulation lies in power accumulation, one can risk an audacious generalisation of the very notion of ‘accumulation’. Since wealth, in a capitalist economy, is a base and a source of power, and since there are other bases and sources of power, like authority and force, then it is possible to conceive the entire process of social development, including the growth of state apparatuses and the size of hierarchical structures, as an immense accumulation process. This view calls forth a generalisation of the notion of ‘exploitation’ too. If it is true that, in a capitalist society, power is used to extract surplus value, it is also true that surplus value is used to extend the bases of power. One grasps the essence of the accumulation process by observing that it unwinds through a threefold metamorphosis. First, there is the transformation of effective power into value, and this is the process of extraction of surplus value from production. Then there is the transformation of value into a base of power, and this is the process of

168 Forms of co-operation and power saving. Finally there is transformation of the bases of power into effective power, and this is the process of investment. Power investment occurs when a wage fund is used to engage workers; or when authority is used to order a worker to perform some activities; or when market power is used to induce a firm with specific resources to accept an unprofitable sub-contracting relation; or when legislative power is used to emanate laws or measures that limit the citizens’ freedom of action. Note that market power, which is a form of wealth, and legislative power, which is a form of authority, are considered here as bases of power. And note also that there can be exploitation in all three cases: when you use command over labour to raise labour productivity; when you use influence on a sub-contractor to lower his remuneration below the value of his services; when you use political domination to extract a fiscal surplus from the tax-payers’ pockets. ‘Investment’ is intended here as a generalisation of the notion adopted in economics. This concept refers to bases of power in general, and therefore it applies, not only to the decisions aimed at the expansion of capital stock, but also to those aimed at extending the sphere of political domination or of social and economic influence. Investment in the means of production, for instance, can be intended as an operation which enables a firm to increase employment and extend command over labour. In addition, since it aims to increase the firm’s size, investment tends to raise its market share, i.e. a condition for its market power. Finally, by swelling the firm’s size, investment also contributes to lengthen and widen its hierarchical ladder, thus increasing its managers’ authority. Likewise, one could observe that investment in R&D aims at raising the asymmetry in knowledge endowments on which firms build their market power. In the same way, an investment in information collection, for example in market research, industrial espionage, staff chart-rating etc., aims at building the economic influence depending on particular information asymmetries. Also note that in this view lobbying is intended, not only as a rent-seeking activity, but especially as a rent-investment activity: social actors spend their wealth to extend their influence on political actors. A final example is provided by manipulation, which in many investigations is considered as a form of influence. Here it is intended as a kind of investment in certain bases of power. With manipulation, money and energy are spent to induce some social actors to modify their actions in such a way as to favour some dominating actors. Advertising, for instance, aims at altering the consumers’ aspirations to induce them to buy a greater quantity of certain products, or to accept higher prices or to strengthen loyalty to a certain brand, all changes that contribute to increase the manipulator’s power. Ideological propaganda may aim at modifying the values and beliefs of social actors to induce them to accept the legitimacy of some authority and domination relationships. And recall that legitimisation is an instrument for the exercise of power: an increase in it implies strengthening of power efficacy. More generally, one could say that manipulation aims to modify the aspirations and cultural endowments of dominated actors in order to reduce their exit faculties. The choice fields are deformed and some options

Forms of co-operation and power 169 which might otherwise enable the dominated actors to get rid of the influence relation are precluded. In this way a source of influence is constituted or consolidated. Accumulation and the systemic effect A generalisation of the concept of ‘capital accumulation’ into one of ‘power accumulation’ contributes to bring to light an important aspect of the accumulation process, the systemic effect. Power differs from normal goods, not only because it is a positional good, but also because it produces strong externalities capable of transforming the competitive process into an organic process of social growth, a process in which the struggle of anyone against everyone is transformed into a more or less voluntary form of social co-operation. Competition, by feeding egoistic sentiments and the practice of opportunism and violence, should turn social life into a chaos of subjugation and collective self-destruction. Yet this does not normally occur. Capitalist societies have widely demonstrated that they are able to produce stability and growth in co-operation. It is precisely power accumulation that produces this miracle. Power enables the winner of the struggle for existence to subdue the loser instead of destroying him, to transform him into a means for the pursuit of the winner’s ends, to compel him to co-operate in this pursuit. But the principal instrument in the exercise of power is legitimisation. No power relation can last long if the dominated actors are not convinced of the legitimacy of their obligation to obedience. Any attempt by a social actor to expand his own power contributes to enhance the conditions of power legitimisation and generates externalities for the exercise of power by other social actors. In particular there are at least three kinds of power externalities. There is, first of all, a psychological externality. Power, perhaps more than any normal good, is able to produce Veblen effects, to stimulate sentiments of envy and induce imitation impulses. Why more than a normal good? Because it is a positional good, so that the people who endure it are in a position, not just not to consume it, but to consume a negative quantity. And, although the position of a dominated actor may nurture the development of meritocratic ideologies which legitimise the power relation, it also fosters frustration and requital attitudes and a search for compensation. A dominated actor in a certain relation may wish to imitate the dominator in another. A soldier who suffers the impositions of his senior officers may try to take it out on his younger comrades. A worker who undergoes the arrogance of his boss in the factory may try to get even with his wife at home. Psychological externalities foster the extension of power in all spheres of social action, its penetration in all pores of the social body. It is psychological externality that generates that kind of power proliferation which induced Michel Foucault to speak of the micro-physics of power. The process of power extension through imitation effects is an important social stabiliser, since it turns any dominated actor into a virtual dominator, any slave into a potential instrument of enslavement.

170 Forms of co-operation and power There is, secondly, an ideological externality. When the winner subdues the loser, and compels him to co-operate and to contribute to collective welfare, social competition and the power relations it generates tend to become legitimised in the dominated actor’s mind. The ideology of the jungle law as an instrument of efficiency catches on, as well as that of transactions as instruments of co-operation, of power as the just prize for ability, of obedience as a moral obligation of the weak. Any firm which increases the number of its employees, does not just extend its managers’ power, but also nurtures the conviction that the capitalist firm is a socially useful and therefore legitimate form of co-operation, thus contributing to extend power legitimisation throughout the whole society. Since the power accumulation of a social actor widens inequality in the distribution of power bases, it augments the conditions of dominance in general, and, since it brings about power legitimacy, it enhances the instruments for its exercise in the entire society. Any individual who accumulates power fights not only for himself, but also for the whole power élite. Finally there is a normative externality. When a judge issues a sentence punishing a worker’s disobedience or a sub-contractor’s opportunism, he generates a rule which can be extended by jurisprudence to the whole system. Any judge’s sentence, any parliamentary law, any government measure, even if prompted by specific problems, must be formulated as general rules. Therefore they generate normative legitimacy for every kind of power relation provided for in the rules. A social actor who stimulates the production of a norm might do it to consolidate his own personal power or the power of the social group to which he belongs. But he also contributes to extend the instruments of power of other social actors or groups. The ‘systemic effect’ was brought to light by some radical critics of Dahl, who noted that it is perversely reductive to limit the analysis of power to decisionmaking situations in which the actors’ aspirations are taken as granted and autonomously formed. Most power in modern capitalist societies is exerted, not so much by determining decisions within an open choice field, but by restricting the choice fields, by reducing the alternatives open to the dominated actors. Through the processes of culture and rule production, beliefs, values and normative institutions enable social actors to define and delimit their choice fields. Through them, the legitimacy bases of social relations are built; the conditions are created for inducing the dominated actors to voluntarily accept power relations. It is important to understand the organic character of the process. Just like all externalities, the systemic effect functions through the action of single actors, but not necessarily by virtue of their intentions to produce it. A capitalist who undertakes lobbying to obtain privatisation of a state monopoly he wants to buy, acts in the pursuit of his individual goals. Yet, if his action is successful, it creates a precedent that may pave the way to other privatisation measures. Thus that capitalist’s action is a class action. Capital accumulation cannot be fully explained at microeconomic level and cannot be appreciated in full depth if it is not referred to a broader process of power accumulation. This is an holistic process. In spite of the egoism and individualism of single persons, every dominating actor works at consolidating the power structure

Forms of co-operation and power 171 of society. It is the process in its entirety that gives stability to power accumulation in the single firm or single church. Not just the entrepreneur who introduces an innovation, but also the minister who issues a decree, the priest who gives a penance, or the husband who disciplines a wife, in other words, everyone who accumulates personal power works for capital accumulation. Even the extraction of surplus value cannot be fully understood if it is not seen in a systemic perspective. If the social base for the exploitation of a manual worker is power, then even the intermediate officers and cadres who control and meter the worker’s performance contribute to the production of surplus value. Their action fosters an increase in labour productivity. If the bases of workers’ obedience rest on the existence of a police force, a cultural industry, a mass media apparatus, a church etc., then even a policeman, a poet, a journalist and a priest contribute to the extraction of surplus value. Capital, as a social relation, is a complex, holistic relation. So is exploitation. Power infects everything Power is the most effective governance relationship in co-operation set-ups where egoistic sentiments prevail and exploitation is practised. It should not be necessary in co-operation set-ups in which the participants pursue common goals. But, as argued above, power relations can also arise in these set-ups if egoistic sentiments are strong and information asymmetries are substantial. In reality this would not be such a serious problem if asymmetries affected only some individuals who perform similar jobs. Since the knowledge and information endowments of actors who have similar skills cannot be very different, asymmetries should not be so strong as to prevent collective control of opportunistic behaviour. Power relations rarely develop among peer actors. The problem, however, can become severe if asymmetries involve people in different layers of a hierarchy, especially if they belong to relations between officers and the rank and file. If the members of an association or a co-operative do not have the knowledge required to effectively evaluate the officers’ activity, then only the passing of a minimal altruism test can ensure subordination of these officers to the pursuit of collective goals. Otherwise it is highly probable that administrative and cohesive governance relations tend to transform into power relations. A co-operative group that needs to interact with external actors inevitably undergoes the influence and pressure of external behavioural factors which tend to deform internal governance relations. There is, first of all, a sort of co-operative pressure. A self-managed company, for instance, co-operates not only in internal production, but also in external relations, with suppliers, customers, government bodies, consultancy firms etc. External co-operation aims at taking advantage of the benefits of specialisation. The outputs and the inputs of the self-managed company will tend to adapt to the external environment, the customers’ requirements, the conditioning by suppliers and the constraints of rules. The nature of products, the techniques in use and the

172 Forms of co-operation and power organisational forms will have many of the characteristics of exogenous data. But, according to the theorem of class-biased control, in a capitalist environment production techniques and organisation tend to generate information and knowledge asymmetries that favour the firms’ managers and functionaries. Thus, any effort to mould internal organisation and knowledge so as to propitiate democratic control will be hindered by the necessity to adapt production to external demands, influences and pressures. Moreover, markets are highly competitive environments. If a self-managed firm has to compete with capitalist firms, it must conform to a greater or lesser extent to their performances. And since the latter play their fight with strokes of productive reorganisation, innovations, productivity increases, dismissals etc., the former, to avoid failure, will undergo the same pressures. Even more pressing is competition in financial markets. According to the theorem of class-biased credit rationing, a self-managed firm is hindered in the struggle for fund raising. Therefore selffinance is more essential for a self-managed firm than for capitalist firms. But self-finance implies the production of a surplus value. The task of ensuring the firm’s prosperity is assigned to the managers, who need power to impose wage cuts, collective dismissals and work intensification, when these measures are necessary for survival. Nor would it be difficult for them to idealise their motivations and interpret their mandate in terms of an accumulation goal, if they operate in a competitive environment in which growth is a condition for survival. Firm growth may imply an increase in the managers’ power. Thus, while external co-operative pressures favour the formation of internal information and knowledge asymmetries, competitive pressures tend to encourage accumulation of the managers’ power. Both kinds of pressures encourage transformation of the delegation-administration governance relationship into a power relationship. However, these thrusts might still not be sufficient to cause goal-displacement. After all, the firing of an opportunist worker or the request for an exceptional production effort might be unavoidable if the firm is to survive. If the managers passed a minimal altruism test, they would need to use a whip, e.g. when free riding makes it necessary, just in the workers’ interests. And if the latter understand the situation, they may willingly adapt to it, in which case the administrative nature of the governance relationship would be preserved. The transformation of an administrative relationship into a power relationship calls for another condition, namely, that accumulation of power is an element of the managers’ objective function. And there are psychological pressures that work at the creation of this condition. The managers, in a capitalist economy, are involved in power transactions in all their relations with external actors, and need to acquire the ability to handle these transactions. To avoid a monopolist supplier or a monopsonist customer extracting surplus value from their firm they must learn to achieve monopolistic and monopsonistic power, as well as to practise lobbying, and to play strategically with other firms; in other words, to augment their influence. To this must be added the strong imitation effect of continual contacts with powerful capitalists and politicians. Once the managers have assimilated a taste for power,

Forms of co-operation and power 173 in a situation of strong information symmetries, there will be no more difference between a capitalist firm and a co-operative. Co-operation, power and conflict Power sustains co-operation, but also generates conflict. There is an intrinsic reason at the basis of this fact: since power is a positional good, if there are some actors who enjoy positive quantities of it, there must be others who ‘consume’ negative quantities. And not everyone is prepared to passively accept subjugation. Of course, imitation effects may induce some people to compensate oppression in one field with personal power in another. But compensation for an evil is not the same thing as its elimination. Since the desire for deliverance is widespread in modern culture, revenge sentiments may be stronger than envy. At any rate, either kind of sentiment nurtures the class conflict that always conditions the relationship between dominating and dominated actors. Power, however, does not just generate conflict from co-operation, it also generates co-operation from conflict. By compelling the oppressed to co-operate, for instance in production, it unites them, enables them to communicate and moulds them in the same culture. It might induce them to co-operate in struggle. Since a power and exploitation relationship is based on some form of asymmetry in the exit faculties, the common action of oppressed people will tend to act on these faculties. For instance a strike aims, among other things, to cause a financial loss to the firm, i.e. to undermine, in its capital, the bases of its power. But an employer’s wealth is a condition of his privileged exit faculty. The more impoverished a firm becomes the weaker is its ability to practise lockouts to discipline the workers. Furthermore a workers’ coalition aims to reduce competition among the workers themselves, i.e. the capacity of a firm to substitute an undisciplined worker with a more docile one, a striker with a blackleg. Again, the effect is to reduce the counterpart’s exit faculty. Indeed numerous struggle practices can be used by worker organisations to reach this goal. Just think of the closed shop, by which a firm’s unionised employees prevent the employer from engaging non-unionised workers. Or reflect on the economic meaning of picketing: it is a form of struggle aimed at reducing the exit faculty of a firm during a strike, i.e. its ability to substitute a striker with a blackleg. But think especially of collective negotiation. By preventing individual agreements, it prevents firms from fixing wages by playing on competition between workers, that is, by exiting from an individual employment relation and exploiting a more favourable alternative. As to exit action proper, it is rarely exerted individually by the workers. The power relation itself implies that exiting seldom pays off for a dominated actor. In reality the workers unite to undertake collective action precisely because a satisfactory way of individual exit is precluded. Hirschman’s ‘hydraulic model’ of social action shows that, when the dissatisfaction caused by a relation cannot be manifested through exiting, it tends to be voiced. This simple model makes us understand an important truth: that protest is intrinsic to power relations, that

174 Forms of co-operation and power wherever there is power there is revolt. Since power is not possible if its bases are not distributed unequally, dominating actors enjoy an exit premium, so that the exit option always leaves a residue of discontent among the dominated actors. The residue can only be expressed by voice. After all, protest action is usually undertaken by the oppressed people, not by the oppressors, to whom the exit doors are always wide open. But protest action is more efficacious when undertaken collectively. Thus the dominated actors are induced to unite their forces out of the necessity to express their grievances effectively. Since any form of protest hits, directly or indirectly, the counterpart’s bases of power, there is no voice in a power relation which is not an action against power itself. The oppressed always take action, knowingly or unwittingly, to modify the power relation, to reduce the counterpart’s power or to increase their own freedom space. The widening of freedom spaces is attained at least partly in an immediate way through the protest action itself. The refusal to work is a rejection of oppression and is realised directly in a strike. The strikers no longer obey the bosses, but act in the pursuit of their own goals. As strikers, the workers are free subjects. During a strike the command relation is overthrown. This, by the way, is why a collective action of protest is often perceived as gratifying in itself. Collective action, though, is first and foremost a means to reach a goal and aims at obtaining indirect effects. It strives to modify contract conditions, reduce the bosses’ intrusiveness, slow down production rhythms, reduce labour time, abate the counterpart’s power and the ensuing exploitation. The means used to reach these goals often consist in an attempt to realise them immediately, even though only partially, and boil down to direct attacks on power and exploitation. Shirking and collective disobedience, all forms of strike action, sabotage etc., weaken the disciplinary and control capacity of factory hierarchies, cut labour time, lessen labour effort and slow down productivity. In this way the employer’s choice field is diminished. If he accepts the workers requests, in reality he agrees to reduce power and exploitation in the near future. If he does not accept them, he has to suffer an immediate reduction in power and exploitation. The higher the employer’s immediate damage from the struggle, compared with that caused by accepting the workers’ claims, the higher is the probability of his capitulating. The workers’ bargaining strength depends on this ratio, which is a measure of their counter-power. The concept of counter-power deserves comment, to avoid confusion with Galbraith’s notion of ‘countervailing power’ – a notion conveying the idea of the dominated actors’ capacity to build up power to oppose that of the dominators. According to the definition of power I have adopted, a power relation cannot be bilateral. If one actor has command over another, he cannot in turn be commanded by him. To understand why, look at power as a positional good. Since one of the two parties consumes a positive quantity of power, the other party must consume a negative quantity. The confusion over this issue is sometimes fuelled by two strange phenomena: cross-powers and rivalry of powers. The cross-powers occur when two actors are linked to each other by more than one power relation, either directly or indirectly.

Forms of co-operation and power 175 For instance the president of a republic might have power over parliament on some issues, whilst parliament has power over the president on others. These must be different issues, so that what really happens is that two different power relations are crossed. Rivalry of powers, occurs when two actors try to command a third actor, in which case the two powers might tend to contrast each other. An example is offered by the case of blacklegs in a strike. They undergo twofold pressure, from the factory owner, who uses wealth as a base of power, and from the strikers, who use force. A blackleg cannot obey both actors, but some may decide to obey the owner and others the strikers. In this case one might have the impression that a power and a countervailing power are at work. In reality they are only two different power relations. In fact they determine the behaviour of different blacklegs. Counter-power is not power, but the ability to limit a power. It is exerted, for example, by workers who disobey their bosses or practise shirking. With these actions the workers do not acquire the capacity to command the counterpart, but obtain the effect of reducing the latter’s capacity to command them. Again, the idea is easy to grasp if power is regarded as a positional good. By building up counter-power, a subordinate does not ‘consume’ a positive quantity of this good, but reduces the quantity of negative consumption by cutting down the positive consumption of the counterpart: workers who disobey reduce their own subjection by the same ‘quantity’ to which they reduce the boss’s power. From the movement to organisation The workers’ co-operation in a struggle usually takes the form of a social movement – a form I showed to be quite utopian in a capitalist society. Now I can also show that the utopian nature of liberation movements generates an instability of collective action which bears the main responsibility for the brevity of their life or – to be precise – for their tendency to give way to goal-displacement and to transform into some less utopian form of co-operation. The workers’ collective action exploits a particular advantage of co-operation: mass force. People unite because they understand that the set of their individual resistance actions (exit, shirking, disobedience, sabotage etc.) is far less effective than the actions they can enact collectively. The governance relation prevailing in a collective movement is cohesion – a unity of intention and will that everybody can promote without needing to resort to great organisational efforts. However, if a collective movement purports to continue a certain action for a long time, it is compelled to enter into strategic interactions with other actors, in which case it cannot avoid organisational activities. Strategic decisions aimed at enhancing collective action endeavour to exploit other co-operation benefits, besides mass force. In particular they try to take advantage of specialisation. In this way a struggle organisation is created. It may be a Party, a Union or another form of association. It will however be an organism with an internal division of labour, an organism which assigns roles and specialised functions to its members. If there is some guarantee that the leading groups adhere to the mandate they receive from the members and act in the pursuit of their goals, the governance of the organisation would be ensured by a delegation-administration relation and not

176 Forms of co-operation and power by power. The members would implement the official leaders’ directions and decisions because they know they have been taken in their interests, and would conform by self-discipline rather than by obedience. But is that guarantee easy to obtain? And under what conditions? A first condition consists in the loyalty and honesty of the officials. They must pass a minimal altruism test, which usually occurs at the beginning of the processes of transformation from a movement into an organisation, when the organisation leaders are chosen by the movement itself. They became leaders by virtue of their capacity to identify and express the collective will and their ability to translate it into collective decisions, but also by virtue of an implicit moral quality: their loyalty to the movement. They may maintain their moral attitude after the transformation, but this is not guaranteed, especially when the organisation consolidates and attains political strength. There are various reasons for this pessimistic consideration. Some are intrinsic to the organisational abilities of the oppressed classes. Others are extrinsic and depend on the relations the organisations and their officials entertain with external actors. As general conditions of the utopian character of democratic cooperation, the former reasons have been dealt with in preceding sections. Here I will re-elaborate them with reference to the struggle organisations of the oppressed classes. The first reason relates to the unequal distribution of freedom which is typical of any society where there is a division of labour between people who command and people who obey, actors who decide and actors who work. Exploited and oppressed individuals have to spend most of their lives working. They cannot dedicate themselves to organisational activities, not, at least, to the extent of being able to do it efficiently. In fact union officials who come from the rank and file stop working when they operate full time in the union organisation: they cease being workers and become organisers. The second reason is linked to the above. To manage an organisation, political and administrative competencies are required which are attained through long years of study and experience. But also motivations deriving from a political vocation are needed. These qualities rarely develop among members of the working class, since workers have to spend most of their time on the assembly line. Historical experience shows that the official leaders of an organisation usually come from the vestal class or the petty and middle bourgeoisie. They are people who have been able to study to make a profession out of a political vocation. This is a cross permanently born by workers’ organisations. Alas, it has even been theorised as a revolutionary thesis, e.g. by Lenin. The third intrinsic reason concerns the control capacity of dominated actors. The division of labour in a capitalist society causes intellectual, knowledge and information endowments to be unequally distributed among people. Oppressed actors are also fouled by class ignorance. They have little motivation and ability to control their representatives, who thus have the opportunity to transform a delegation-administration relation into a power relation. The extrinsic reasons relate to the fact that the organisations of oppressed people operate anyway in a capitalist context, in a highly competitive environment, in a

Forms of co-operation and power 177 system of social relations in which power is pervasive and power externalities are strong and multifarious. Indeed power infects everything: imitative effects induce the official leaders of democratic organisations to assimilate the culture, habits and love of power of the officers of those organisations in which power is the dominant governance relation; ideological externalities induce officials to believe that power is the just prize for their ability, especially when this is manifested as the capacity to ensure the organisation’s survival and growth; normative externalities induce the officers to comply with, and therefore to legitimise, the existing normative institutions. All these effects tend to favour goal-displacement in the workers’ organisations, and therefore to deform the relationship between the officers and the rank and file. There is, in particular, a special phenomenon of power externality which infects the struggle organisations of oppressed people and their goal-displacement processes. With organisational involvement the officials tend to identify themselves with an organisation to the point of considering its survival as the most important goal. Since they perform their activity as a profession, they cannot allow their organisation to die. To survive, an organisation has to establish stable relations with other collective actors, including the organisations of the ‘class enemy’. But the stabilisation of these relations implies congruency of goals. Either a struggle organisation aims at the destruction of the enemy, in which case it must cease to exist once the goal has been achieved or it has realised that it is impossible to achieve; or it must survive at any cost, in which case it cannot aim at the destruction of the enemy. This, for instance, was the tragedy of ‘revolutionary syndicalism’. Therefore such an organisation will tend to pursue the immediate interests of the oppressed people only to the extent they are compatible with the survival of the existing system of social relations. The survival of a revolutionary organisation, whether the revolution succeeds or fails, is a form of power externality, because its officers put it at the top of their goals, which can occur when the officers’ goals are put before the members’. Thus there are many reasons at the basis of what Robert Michels, on the grounds of his experience in the German workers’ movement, called the ‘iron law or oligarchy’. One might hope that this law is not so iron as it was thought to be at the time of the Second and Third International. After all, the long run tendency of working time to shrink and of real wages to rise, work at a reduction in knowledge asymmetries and therefore at an increase in the workers’ control capacity. Yet history shows an unequivocal fact: that the construction of struggle organisations is an activity condemned to recurrent re-foundation. Appendix: some ideas for an operational measurement of power The problem with power measurement is not that we lack a yardstick, but that we have too many. There are several reasons for this. To begin with, various students propose different definitions of power and, as a consequence, different measurement criteria. Furthermore there are different forms of power, each constituting a particular kind of social relation for which only particular kinds of measurement are suitable. In addition, power is manifested by its capacity to determine the

178 Forms of co-operation and power actions of an actor, so that, depending on which actions are considered and which indicators are used to highlight the influence exerted on them, different power indices can be used. Finally, there is no universal numeraire for power, as money is for economic values.4 There are indices, formulated in terms of utility or individual preference ordering, which I do not find particularly interesting, not only because utility is scarcely observable, nor because interpersonal utility comparisons are impossible, but especially because these indices tend to treat power as a form of exchange in which subjection is seen as a voluntary choice aimed at utility maximisation. Other indices are formulated in terms of probability, and these are scarcely operational because it is difficult to measure ex ante the probability of undefined events such as the possible actions that can be influenced in a power relation. More interesting are the partial ordering indices of power relations, which measure ordinally the positions of actors in a power hierarchy. Their main drawback is that they make us know whether an actor has power over another, but not how much, so that it becomes impossible to compare the power of many actors even if they are located in different layers of the hierarchical ladder. This drawback becomes particularly serious when you want to compare power positions in different fields or, even worse, when you want to measure power relations in different scopes. A partial ordering measure, for instance, does not allow us to say that the chief executive of General Motors has more power than the owner of a mechanical workshop. When power consists in command, it is possible to use labour time as an index. This is a cardinal measure. Consider the simplified case of the firm whose hierarchical structure is represented in Figure 4.1, and assume all the employees work the same amount of hours a year. The ‘man-year’ can be used as a numeraire for measuring power. There are 14 actors, lh (h = 1,…14), i.e. 13 workers and one employer. Let me define the following symbols: Lj Lkj

mij m kj

= man-years spent in the firm by employee lj = man-years spent in the firm by employee lk under the command of actor lj = man-years spent in the firm by employee lm under the command of actor lk, who is under the command of actor lj = coefficient of actor lj’s power delegated by actor li = coefficient of actor lk’s power delegated by actor lj

r kj

= coefficient of actor lj’s residual power exerted on actor lk

rkjm

= coefficient of residual power exerted on actor lm by actor lk (who is under the command of actor lj) = summation over all j = summation over all actors lk who are under the command of actor lj = summation over all actors lm who are under the command of actor lk who is under the command of actor lj = amount of direct power of actor li

Lmkj

Sj S kj S mkj

DPi

Forms of co-operation and power 179 j

EPi EPjk RPjk RPkjm

TPi FTP

= = = =

amount of power delegated by actor li to actor lj amount of power delegated by actor lj to actor lk amount of residual power of actor lj over actor lk amount of residual power of actor lk (who is under the command of actor lj) on actor lm = amount of total power of actor li = amount of total power in the field

The amount of direct power of li is the summation of all man-years of the actors under his command:

DPi = S j Lj The amount of direct power of lj is equal to the summation of all man-years of the actors under his command. It can be broken down into the summation of the power delegated by li to lj and the residual power of lj. Delegated power is EPi j = mij S kj Lk j

while residual power is RPjk = r kj S kj Lk j

Since mij + r kj = 1 , it holds that DPj - EPi j + RPjk = mij S kj Lk j + r kj S kj LK j = S kj Lk j

The amount of total power of li is equal to the summation of his direct and indirect power. The latter is equal to the summation of the power delegated by li to his direct subordinates and that delegated by the latter to their direct subordinates: TPi = DPi + EPi j + S j Lj + S j mij S kj Lk j + S j S kj m kj S mkj Lmkj

Finally the total power in the field is equal to the summation of the total power of actor li and the residual powers of all his direct and indirect subordinates: FTP = TPi + S j RPjk + S kj RPkjm = S j Lj + S j mij S kj Lk j + S j S kj m kj S mkj Lmkj + S j r kj S kj Lk j + S j S kj rkjm S mkj Lmkj









= S j Lj + S j mij + S j r kj S kj Lk j + S j S kj m kj + S kj rkjm S mkj Lmkj = S j Lj + S j S kj Lk j + S j S kj S mkj Lmkj

A numerical calculation based on the data in Figure 4.1 and referred to the power exerted in one year (all the Lh have unit values) shows that FTP is equal to 13, i.e.

180 Forms of co-operation and power Man-years in the firms L1

L2

L6

L7

L3

L8

Hierarchical layers i=1

L4

L9

L5

L 10

L 11 L 12 L 13 L 14

j = 2, 3, 4, 5

k = 6, 7, 8, 9, 10

m = 11, 12, 13, 14

Figure 4.1

the number of the firm’s employees. Note, by the way, that, if L1 is an employer who does not work but only commands, FTP coincides with the overall ‘living labour’ used in this firm. These results have been obtained with a simple example, but could be extended and refined. For instance, they could be applied to a hierarchy guided by a collective actor, like a committee or a board of directors. One could measure the committee’s total power and then, by dividing it by the number of members, calculate the fractional power of each of the members. If different members, or coalitions of them, have differential voting power, the actual power of each of them could be obtained by multiplying their fractional power by a coefficient of voting power. Another extension could elaborate the hypothesis that the residual power of intermediate cadres can also be delegated, in which case the total residual power of a cadre could be measured in the same way as that of a top manager. Still another extension could take into consideration the view that power weighs more if it is exerted on skilled work. I am not so convinced that power exerted on eight hours of an engineer’s life is different from that exerted on eight hours of a labourer’s life. Certainly it may have a higher worth and be used to produce a higher value, but it is power in itself, the faculty of command over an individual, that has to be measured here. After all, however you look at it, it is still power over eight hours of an individual’s life. Nevertheless, if one likes to comply with that view, he could elaborate Marx’s analysis of complex labour and multiply a worker’s man-years by a measure of his labour complexity. This kind of measurement has a limitation, as it can only be applied to hierarchical organisations in which power is exerted in the form of command, capitalist firms, bureaucratic apparatuses, military bodies etc. It cannot be applied to cases of domination or influence. However it also has a merit. It makes it possible to measure power differences and compare the power of actors who are not in a subordination relationship with each other, even when they operate in different organisations. With this measure, for instance, you can say that the General Motors’ CEO has more power, and how much, than the owner of a mechanical workshop.

Forms of co-operation and power 181 On sources Through the co-operation of numerous wage-labourers, the command of capital develops into a requirement for carrying on the labour process itself, into a real condition of production. That a capitalist should command in the field of production is now as indispensable as that a general should command in the field of battle. This passage comes from the chapter on ‘co-operation’ of Marx’s Capital (vol. I:448). In that chapter Marx lists a series of co-operation benefits, from scale economies, to specialisation, from team production to mass force. He elaborates arguments of various classical economists on the advantages of co-operation and division of labour, but he takes his distance from his sources when he focuses on exploitation as one of the benefits of co-operation. This theoretical innovation enables him to understand that power is the fundamental governance relation in capitalist production. Power in fact, according to Marx, is an essential instrument for the perpetration of exploitation, whilst a power hierarchy is the base for the co-ordination and control functions of production organisations. Moreover Marx grasps the external effects that production mass force has on the workers’ social and political co-operation. He looks at workers’ coalitions as instruments for limiting labour competition, reducing the capitalists’ faculty of exit in the employment relation and contrasting their social power. His analysis of worker movements, however, is somewhat lacking when he fails to recognise the relevance of the anarchist critique of political organisations as characterised by an intrinsic tendency to produce internal power relations (Bakunin, 1963). We had to wait for Robert Michels’ conversion from socialism to sociology to get an accurate analysis of this phenomenon. His book Zur Soziologie der Parteiwesens in der modernen Demokratie, which contains an indepth investigation into the inexorable oligarchic drive of organisations, is particularly interesting also because it is based on Michels’ experience in the German Social Democratic Party. There is now a huge mass of literature on the benefits of specialisation in a firm, which, moreover, is growing exponentially in these times of institutionalist revival. Therefore I will limit myself to citing, in this field, only two insightful and comprehensive works by P. Milgrom and J. Roberts (1990, 1992), which focus on the problematic aspects of co-ordination as well as the way organisation contributes to cope with them. The role of organisational hierarchies as problem-framing and problem-solving devices has been investigated by M. Egidi and L. Marengo (1995), M. Egidi and M. Messori (1995) and M. Egidi and M. Ricottilli (1997). The minimal altruist test I proposed in this chapter draws from Hume’s notion of a ‘confin’d generosity of men’ (1888: 495), but especially from its re-elaboration proposed by G.A. Cohen (1995: 141), who defines it as ‘the artificial virtue that restrains our natural tendency to appropriate what a generally welfare-promoting property institution assigns to others’, and suggests it is a minimal subjective condition for communism to work. Bertrand Russell wrote a splendid book on Power (1938), the best idea of which seems to me to be the one I have synthesised here in the ‘Russell axiom’. The

182 Forms of co-operation and power ‘Perroux lemma’, which specifies this axiom by referring it to a capitalist economy, can be found, together with other stimulating observations, in Pouvoir et économie (1974). The same book and the (1950) article, also develop a conception of the market as a network of power relations. A. Reati (1968) elaborates this conception by applying it to the analysis of market forms. My use of the ‘hydraulic model’ to account for power-induced conflict is based on the ideas put forward by Hirschman (1970, 1993). Given the nastiness of the notion of ‘power’, it is understandable that it did not enjoy great success in mainstream economics, either neoclassical or Austrian (Young, 1995). It received greater consideration, however, in radical political economy, where the role of power analysis is indeed crucial. Among the vast radical literature on this subject I will cite only a few contributions, and, to start with: S. Marglin (1984), who stresses the historical aspects of power; R. Heilbroner (1985), who reinterprets the Marxian theory of accumulation on the basis of the assumption that ‘the drive for power and domination becomes sublimated into the desire to accumulate capital’ (p. 141); and S. Bowles and H. Gintis (1992). The latter propose a definition of power as ‘the capacity of some agents to influence the behaviour of others to their advantage through the threat of imposing sanctions’ (p. 325), which is compatible with the one adopted here. My definition, though, displaces the ability to treat sanctions on the backstage: this ability might be a structural property of a social system, rather than a faculty of individual actors. Further I believe it is not necessary for a dominating actor to obtain a personal economic advantage from the use of power: it is sufficient for him to be able to determine the goals of the dominated actors’ actions. On the problem of power definition, economists should be more attentive to the sociologists’ and political scientists’ works. Clarifying contributions were furnished by political science in the 1950s and the 1960s. The definition adopted here draws particularly, but not without qualifications, from the works of R. A. Dahl (1957), D. Easton (1953) and C. J. Friedrich (1963). Among the contributions of political scientists, some which I have found particularly enlightening are those of H. D. Lasswell (1930, 1936) and, more important, that of H. D. Lasswell and A. Kaplan (1950). In these books one can find a development of Mosca’s, Pareto’s and Michels’ theory of élites; a development which innovates with a psychoanalytic attempt at explaining the personal motivation to power accumulation and with the adoption of Marx’s view that any change in the political structure of a society and its ideological justifications tends to reflect the prevailing social division of labour in the production processes. Lasswell treats power as a means to accumulate values, but, at the same time, he considers power itself as a value, so that his analysis is not confined to the political arena but is open to the investigation of the whole social and economic process. Enormous influence has been exerted on political science and the sociology of power by the theory developed by Max Weber (1922), a theory which has subsequently been twisted, especially by Talcott Parsons and the functionalist school in sociology, to serve a vision of power as a legitimate instrument for the pursuit of collective goals, and to confine the problem of decision-making conflict

Forms of co-operation and power 183 to political issues. Radical thought convincingly contrasted both these approaches. A. Giddens (1968) is exemplary for his critique on the former, P. Bacharach and M. S. Baratz (1970) and S. Lukes (1974) for their critique on the latter. We also owe to Lukes an elucidation of the ‘systemic effect’ in the exercise and accumulation of power. P. Morriss (1987) and B. Barry (1989), although privileging the political field, offer comprehensive and insightful treatments of power. A. A. Berle (1967) and J. K. Galbraith (1983) offer deep investigations from an institutionalist point of view. B. De Jouvenel (1962) digs into the political philosophy of power. In the analysis of power as a positional good a path-breaking contribution has been given by F. Hirsch (1977). A. Sen (1983) examines the role of positional goods in the emergence of the profit motive. U. Pagano (1999) offers, besides an acute analysis of power as a positional good, some applications of this notion to various social problems, like the power structure in capitalist firms and worker co-operatives and the definition of social classes. R. Bartlett (1989) focuses asymmetric access to knowledge as a source of power. The vision of the labour process as a contested exchange was elaborated by S. Bowles and H. Gintis (1988, 1993). My personal view is that any transaction is a contested exchange if it involves power relations, and that this is so for three main reasons: first, because power implies an asymmetric distribution of the exit faculties, so that the practice of voice becomes unavoidable as a means of bargaining and grievance manifestation; second, because their nature of positional goods makes power relations essentially antagonistic; third, because in a capitalist economy power is typically associated with exploitation, so that this kind of governance relationship necessarily generates a distributive conflict too. I must stress that the basic reason why the labour process is a contested terrain lies, not in the incompleteness of contracts, but in the fact that the employment relationship is a power and exploitation relationship. As to the problem of measurement, finally, I have been unable to find satisfactory cardinal measures in the literature on power. Ordinal approaches, instead, abound. Let me recall the partial ordering measure proposed by H. A. Simon (1953), who however applies it to the calculation of power changes rather than to power itself. A. Weale (1976) develops Simon’s ideas with acumen, only to abandon them after showing their limitations. Two highly refined analytical proposals about power measurement have been recently put forward by J. B. Miller (1990), who uses lattice theory, and S. Vannucci (1999), who resorts to game-theoretic instruments.

Notes 1 Note that the notion of ‘authority’ is used here with a different meaning from that developed by Weber, who uses it as a synonym of ‘power’, precisely as ‘legitimate power’. 2 Russell includes glory – one could say, status or prestige – among the insatiable human desires. This is a kind of desire for social recognition. Status is a positional good and is strictly cognate to power, but in an asymmetric way. While it is true that any power position is always a position of high social status, at least in a society permeated by the struggle for dominance, it is not true that any position of high status or prestige involves the exercise of power. A great artist has great prestige, but, often, little power. In fact an artist mostly aims at merits recognition rather

184 Forms of co-operation and power than status and power. Since in this chapter I am mainly interested in the accumulation of capital and power, I will avoid any consideration about status as an action motive. Insofar as status is associated with power it can be considered as an isomorphism of it. Insofar as it is not, it is of no interest in this analysis. 3 If all entrepreneurs are risk lovers, their actions would be a prize in themselves. What then is the social utility of profits? 4 In reality there is a particular case, the voting power in committees, in which a natural numeraire is available: the number of votes. But an index of voting power does not measure power proper. It only measures the way a committee’s power is distributed among its members. To transform it into a measure of power, it is necessary to multiply it by the power of the committee itself, which brings us back to the original problem.

Production governance structures 185

5

Production governance structures

A taxonomy of governance structures The notion of a ‘governance structure’, introduced in Chapter 4, is defined as a set of institutions and social relations aimed at regulating collective action. Now it is necessary to highlight the types, uses, functions and operating mechanisms of governance structures. There are two types of governance structures, which differ in the kinds of social relations they establish and the form of co-operation to which they give rise. Hierarchical structures are based on relations of command-subordination or delegation-administration and are present in capitalist firms and bureaucracies, in co-operative firms and associations. In these organisations there is neat, if not complete, separation between decision-making and implementation functions or between command and subordination roles. Network structures, are based on exchange or cohesion relations and are present in markets, social movements and forums. Any social actor preserves at least formally a complete decision-making autonomy, although various influence mechanisms can limit the substantial autonomy of some subjects to the advantage of some others. Obviously mixed forms of governance structures are possible, like trusts, cliques, industrial districts, M-form companies, keiretsu and so on and so forth. But I prefer to remain at a high level of abstraction. Since capitalism is a socio-economic system in which labour is used in production to extract surplus value and surplus value is used to feed capital accumulation, the two fundamental processes that have to be regulated are production and accumulation. Therefore there will be production governance structures and accumulation governance structures, in either of which both hierarchical and network apparatuses could be at work. Governance structures have both a technological and a political use, in that they help prompt the technical advantages of co-operation and, at the same time, they serve to co-ordinate the actors’ behaviour so as to bend them to serve goals that might be contrasting with those of the actors themselves. They play three social functions: disciplinary, whereby actors are induced or compelled to conform their behaviour to the co-operative exigencies; selective, in which actors are allocated jobs and social positions to obtain the best of their

186 Production governance structures capacities; finance, through which actors are provided with the means required to carry out their activity. These three functions are played by means of various social devices or operating mechanisms, the most important of which are: control, that is, the exercise of power to compel an actor to behave according to the expectations of another; incentive, the provision of rewards that stimulate a voluntary conformation of individual behaviour; ideological deformation, through which influence is exerted on the people’s values and beliefs so as to prompt a spontaneous conformation of behaviour. Accumulation structures are explored in Chapter 6. In the present one I deal with production structures. The second part is dedicated to an analysis of the control, incentive and ideological schemes used in capitalist firms to align individual behaviour to the pursuit of the company goals. The third part is instead dedicated to markets. In particular, since co-operation and exploitation take place through exchange in these kinds of structures, I focus on the conditions of price formation and the social meaning of the price system. Finally, in the fourth part, I deal with the way in which knowledge production can be regulated in market processes and in organisational apparatuses.

The capitalist firm Conflicting co-operation According to Edith Penrose a firm is a set of resources which can be used productively. In this definition it is not the accountancy value of the means of production that matters, but rather the capacity to produce services. The set of labour capacities available to the firm constitutes its peculiarity, its specific competence, that which makes it a unique thing, and justifies its existence as a productive entity. Pushing the definition one step further, one can look at the firm as an organisation of social actors who take part in productive activity. The social actors embody the specific competencies from which labour services spring. They pull the means of production out of inertia, transform them into instruments of labour and make them productive. This social definition shifts the essence of the firm from the objective to the subjective dimension, and induces one to interpret it, not as the wealth it owns, but as a set of social actors it does not own. Different actors, besides being endowed with different capacities are also moved by different motives. Thus a constitutional heteronomy is at the base of the firm. Since the ends of participating subjects do not coincide with each other, they cannot all coincide with those of the firm. It follows that the emergence of internal conflict is an unavoidable characteristic of productive organisations. Now the firm can be redefined as a group of potentially antagonistic actors involved in the pursuit of a co-operative goal. The problem arises: is the firm a social actor? An actor is a subject who takes intentional decisions and performs purposeful actions. Are there goals of the firm as such, goals defined independently of those of the actors who participate in its activity? One could answer that the firm’s goals coincide with those of at least one

Production governance structures 187 participating actor, and are decided by him. But this answer poses more problems than it actually works out, two of which, in particular, are quite important: that of the ‘dominating goal’ and that of the ‘alignment of behaviour’. The ‘dominating goal’ problem addresses who or what determines the firm’s goal proper; the ‘alignment of behaviour’ problem addresses how it is possible to obtain from heterogeneous actors the behaviour required to attain the firm’s ends. The former problem is easy to work out in the case of a personal enterprise, for here the ownership is individual and therefore the firm is seen as a means to reach the owner’s ends. However, the firm is regarded as a thing rather than an actor, a conclusion which contrasts with the observation that firms do have legal personality. Things are more complicated in the case of many owners. Which of them has the prerogative of defining the firm’s goal? One answer could be: the majority. But the firm remains a thing and cannot be considered as a social actor. Even more complicated is the case of a firm which, due to absentee ownership, is controlled by managers – a going concern endowed with legal personality and capable of entering various kinds of transaction with all its participants (including creditors and shareholders). This kind of firm is formally and substantially a social actor proper. But who defines its goals? Certainly not the owners, who do not have control. Nor the managers, at least legally, who merely hold the position of functionaries or agents, and therefore act formally to pursue ends which are not their own. Here is the paradox of a social actor whose goals nobody can define. The problem of the dominating goal is easier to work out if an effort is made at historical and institutional specification. After all we are interested not in a generic firm, but in a capitalist firm. The capitalist firm can be defined as an organisation which furthers the extraction of surplus value from the production process and the use of surplus value for the valorisation of capital. The capitalist firm, in this wider vision, is regarded as an organisation that controls and co-ordinates heterogeneous social actors and uses their productive services to increase the value of its assets. Thus the dominating goal of the capitalist firm is capital accumulation. It may coincide with that of some of the participating actors, for instance the managers’, but is not necessarily determined by them. It is more likely that the ends of these actors are formed as a function of those of the firm itself. In this case, by using a notion introduced in chapter 1, one could refer to the capitalist firm as a ‘corporate actor’, an actor whose goals are not determined subjectively, yet who acts as a subject. This kind of actor undergoes internal and external pressures in the form of social and economic processes capable of affecting the choices and decisions of its participating actors. Most of the individuals participating in production activity in a capitalist firm are dominated actors. The problem of alignment of behaviour arises with them. Its solution, as a practical problem, requires the setting up of different kinds of production governance structures. Individual behaviour can be influenced through various devices. It can be directly determined by means of commands. This defines the control-based governance structures. But behavioural alignment can also be induced by playing

188 Production governance structures on individual motivations, that is, by furnishing incentives for a spontaneous alignment. This defines the incentive-based governance structures. Finally, individual behaviour can be affected by altering the systems of beliefs so as to induce people to willingly pursue ends that coincide with the firm’s goals. In this case too a spontaneous alignment is aimed at, but here ideologies rather than incentives are used to mould individual action. This defines the ideological or culture-based governance structures. Governance structures must be at work in any kind of productive organisation, independently of its institutional and social setting. There is a technological rationale that makes them necessary in both a capitalist and a democratic firm. But in a capitalist firm, in which the corporate actor’s goals do not coincide with those of the employees and exploitation is the most important co-operation benefit, command and control structures are called on to direct the individuals’ actions. These structures have an intrinsically political nature, as they serve to regulate the power relations among the participants to productive activity. The technological uses of production hierarchies: co-ordination Co-ordination and knowledge-production are the two principal technological uses of hierarchical governance structures. The latter will be dealt with in the fourth part of this chapter. In the present section I focus on the former. Co-ordination is necessary whenever the technological benefits of co-operation are aimed at. As clarified in Chapter 4, these benefits boil down to two: mass force, which results from any economy in the use of space, time and indivisible factors, and specialisation, which emerges from the combined use of workers endowed with heterogeneous capacities. Both benefits pose co-ordination problems. Since they consist in returns exceeding the summation of the single workers’ contributions, cooperation cannot result from simple individual initiatives. It presupposes the sharing of private information and the creation of new information. It calls for the formulation of a supra-individual plan of production through which individual actions are co-ordinated. One kind of co-ordination problem is synchronisation, which arises prevalently with mass force. This is the case of the oarsmen rowing a boat or of the workers who try to move a block of marble. Their activities must be harmonised, so that all workers move in the same direction, at the same time and with the same amount of effort. Also consider the case of non-specialised artisans who use the same plant: their shifts must be organised so as to avoid over- or under-utilisation of capital. A second kind of co-ordination problem is assignment, which mainly arises in connection with specialisation, when heterogeneous individual capacities have to be distributed among different jobs or tasks. In this way each capacity will be utilised to bring forth that surplus product the specialised activities can generate over and above the productive capacities of the single individuals. Obviously assignment problems also emerge with mass force. For example, the workers who move the marble block must all have roughly the same strength.

Production governance structures 189 And synchronisation problems may emerge from the co-ordination of specialised activities. The actions of different workers, for instance, must be synchronised so as to avoid any loss of time in shifting intermediate goods from one process to another. Co-ordination problems also arise with technological complementarities. These may involve relations among workers and relations among workers and the means of production. In the presence of specialisation, the productivity of a worker depends on that of the others to whom he supplies outputs and from whom he receives inputs. Production is the result of team work and labour productivity can be observed and measured with accuracy only for the team. And, since it is impossible to observe the specific productive contribution of any single worker in the team, it is impossible to co-ordinate actions through decentralised decisions. One method for working out co-ordination problems would be a collective decision. All individuals reveal their own private information. Together they all decide the goals of their common action and elaborate a production plan. Then they implement the plan by complying with collective will through self-discipline. Apart from free riding problems, which for the moment I shall set aside, the fact remains that a collective decision becomes unfeasible as soon as the team reaches certain dimensions, as occurs in the overwhelming majority of modern enterprises. If each individual had to transmit and receive messages from all the others, an enormous flow of information would be generated; and the decision-making process would become long and complex. It is easier to cope with this problem by assigning to a particular individual or a restricted group of people the tasks of collecting and elaborating information, taking decisions, formulating a production plan, giving instructions for its implementation, in other words, playing the role of a central co-ordinator. In this way a new job arises, a new specialisation; and, with it, the productive organisation takes a hierarchic structure. But a hierarchical organisation gives rise to various special costs, not least, those for remuneration of the bosses. These costs grow with the complexity of the production process, the dimension of teams, the quantity of information to be gathered and elaborated. Since the benefits of co-operation are also positively correlated with these variables, a hierarchical organisation expands so long as the co-operation benefits it elicits are higher than the costs it generates. This is also why firms normally tend to internalise, through vertical and horizontal integration, the production processes which produce stronger synergies. Often co-operation benefits take the form of scale economies. Mass force tends to grow with the firm’s size because the constraints of technical indivisibilities weaken, while the opportunities of economising on the use of space and time widen, when the scale of production expands. Growth in size enables the firm to push specialisation ahead. Besides scale economies there are the economies of scope, which result from extending the production of inputs used in the production of several outputs. These economies are obtained by multiplying the number of joint products produced by means of those inputs. Thus, economies of scope also originate from the benefits of mass force and specialisation.

190 Production governance structures The role of the division of labour in arousing increasing returns to scale had already been recognised at the time of Cesare Beccaria and Adam Smith. The nature of the virtuous circle that can be triggered off by the division of labour and which sustains the expansion of the firms’ size had also been recognised. Then one might legitimately ask why, after more than 200 years of industrial development, total monopoly is not held by a single big company. A possible answer is that any attempt to expand the range of co-operation causes ever increasing and more complex co-ordination problems. Since a hierarchical organisation is set up to cope with these problems, growth in the firm’s size brings about a superfetation of hierarchical apparatuses and a rise in organisation costs. Suppose, for example, that a certain production can be efficiently obtained by a team made up of three workers and a boss, and that there is a firm composed precisely in this way. Now suppose the firm wishes to triplicate its scale of activity by setting up another two teams. Most probably its personnel will more than triplicate, because now the activity of the three teams will also have to be coordinated. This could be done by creating a new authority role with the task of directing the three bosses. In other words the bosses of the three teams constitute a new team in need of co-ordination. A hierarchical structure tends to grow as an apparatus of nested teams: teams that co-ordinate teams. Therefore growth of the firm normally brings about not only an increase in the number of workers, but also an increase in the number of co-ordinators which will most probably be more than proportional to the former. In the foregoing example the number of productive workers grows by 300 per cent (from 3 to 9), while the number of bosses grows by 400 per cent (from 1 to 4). If a hierarchy has a pyramidal structure, with the number of labour tasks being its ‘base’ and the number of command layers its ‘height’, then expansion normally takes place through an increase in ‘volume’. This example is rather simple and does not allow for any generalisation. However, considering that the complexity of co-ordination problems increases with production size and that the remuneration of command tasks increases with the height of hierarchical ladders, it would not be difficult to accept the idea that organisation unit costs are an increasing function of the firm’s size. It follows that the increasing returns brought about by co-operation benefits are often contrasted by the increasing costs brought about by hierarchical expansion. The political uses of production hierarchies: exploitation and control Its uses in co-ordination and knowledge-production give the firm’s hierarchy a rationale of technical efficiency. Hierarchy emerges from the nature of the production process and serves to exploit the technical benefits of co-operation. This kind of organisation must be present in any firm that adopts techniques which are complex and advanced enough to render individual independence impossible or inefficient. And it must be present regardless of the prevailing institutional and social setting. Hierarchical organisation is present, therefore, even in self-managed

Production governance structures 191 companies. What changes from one institutional setting to another is the kind of governance relation. In a self-managed firm a delegation-administration relation should prevail in which the actors located at the upper layers of the hierarchy operate as agents of those located at the lower layers. The workers-principals fix the goals of the firm, the managers-agents freely decide on the means to achieve them; the workers control the managers’ activity and at the same time carry out the instructions received from the cadres, which they do by practising self-discipline rather than obedience. The co-ordination and knowledge-producing functions are not sustained by a power relationship. In a capitalist firm, on the contrary, the company’s goals do not coincide with the workers’. The capitalist firm aims at a very special co-operative benefit, one which is not intrinsic in technology: exploitation of the workers. This benefit is attainable even when the other advantages of co-operation are absent. In Chapter 1, where the problem was tackled at the highest level of abstraction, I showed that the firm can make profits even if there are no scale economies, mass force, specialisation etc. This, broadly speaking, is the idea Marx tried to convey with the notion of ‘production of absolute surplus value’. For this to be possible the employer must have a prerogative of command in the labour process, a prerogative typically established by means of the employment contract. On this contract rests the power structure of the capitalist firm. Now it can be added that the same power structure is used to exploit the technological benefits of co-operation. If capitalist direction is thus twofold in content, owing to the twofold nature of the capitalist process of production which has to be directed – on the one hand a social labour process for the creation of a product, and on the other hand capital’s process of valorisation – in form it is purely despotic. (Marx, 1990 I: 450) In a capitalist firm the co-ordination and knowledge-production functions are sustained by a command-subordination relation rather than a delegation-administration relation. The firm can use its hierarchical structure to appropriate, not only the ‘absolute surplus value’, but also the ‘relative surplus value’, the surplus brought about by technological advances – an argument which was investigated in depth even before Marx, for instance, by Pierre-Joseph Proudhon. To the extent wages are not geared to labour productivity, especially in the presence of strong complementarities, the firm can appropriate all the technological benefits of co-operation. And can do it, not only de facto, but also de jure, because, due to the workers’ obligation to obedience, the labour services obtained through command result to be produced and owned by the firm. But the profit motive immediately generates a conflict of interests, whose solution requires the firm’s hierarchy to be used to exert control.

192 Production governance structures The control exercised by the capitalist is not only a special function arising from the nature of the social labour process, and peculiar to that process, but it is at the same time a function of the exploitation of a social labour process, and is consequently conditioned by the unavoidable antagonism between the exploiter and the raw material of his exploitation. Similarly, as the means of production extend, the necessity increases for some effective control over the proper application of them. (Marx, 1990 I: 449) One cannot expect workers to be devoted to a firm which exploits their work and establishes a power relation with them. This is not a question of altruistic or egoistic sentiments. Even the world’s most honest worker may be moved by his sense of justice to refuse the power structure of a capitalist firm, and therefore to practice shirking, reluctant obedience, disobedience, sabotage and, more generally, to conceal personal information the firm might use to exploit him. It must be added that the very aspiration to freedom and justice might induce the workers to solidarise with ‘opportunists’ rather than to act against them. Why should a worker blame or dissuade a loafer if he knows that his behaviour helps to reduce profits and not wages? And if a loyal worker realises that the low productivity of a shirker negatively affects that of his colleagues, causing loyal workers to increase their effort, he will likely react by imitating the free rider rather than trying to make up for his misdemeanours. Class antagonism permeates the whole organisation of the capitalist firm and tends to feed the mass ‘opportunism’ of workers, a behaviour that constantly tries to build counter-power spaces in the factory, to undermine the bosses’ power and the entire hierarchy of command. The workers’ counter-power, as an instrument for abating the capitalists’ power, also serves to reduce exploitation, besides oppression. Any act of disobedience, any negligence in the execution of orders, any concealment of labour capacities, any slackening of discipline, leads to a diminution in the rate of exploitation. According to some modern economists, the information asymmetry on labour capacities would not be sufficient to affect profits if it were not accompanied by another, more serious, problem: contract incompleteness. As I argued in Chapter 1, this opinion is based on an erroneous conception of the employment contract, a conception which likens it to a contract for services. If there were contract completeness in a transaction by which the workers sell the firm some specific labour services, a completeness extended to the point of providing rigorous enforcement devices, no worker would find it worthwhile to hide information. Since specific services are sold, and since these are remunerated at a fair value and enforced rigorously, any infringement of contract obligations would produce economic effects that would fall upon the transgressor himself. But the employment contract is not a contract for services. It defines the limits of authority and not the characteristics of labour services. These are generally not known ex ante in a contract of employment. Now, even if an employment contract were complete in its domain, i.e. if the limits of authority were fully defined, the

Production governance structures 193 information asymmetries on labour capacities would succeed in affecting profitability. The prerogative of a boss to order a worker to carry out certain operations may be as clear and well defined as you like, but if the boss has no exact idea of the worker’s capacities, he will be unable to obtain the performance he wants. The efficacy of ‘opportunism’, of course, would be stronger if the employment contract were incomplete, i.e. if the prerogatives of the employer were badly defined. This kind of incompleteness, as I showed in Chapter 1, has a twofold origin. It is determined by the vagueness and imprecision of social norms and customs, a vagueness necessarily linked to the universality of normative institutions. Any specific contract can only partially make up for this flaw, first, because most employment contracts are collective rather than individual and therefore cannot adapt to the idiosyncrasies of individual actors, second, because the specific commands that can be given in a specific labour process will depend on the particular services to be produced, but these services are unknown at the time of contract stipulation. Furthermore, the definition of authority boundaries also depends on the workers’ technical capacities, i.e. on their competencies and abilities, which are precisely the kind of endowment that show strong information asymmetries. No contract will be able to specify with any precision the maximum weight of marble blocks an employer can order a specific worker to lift. If the imprecision of authority boundaries gives vent to ‘opportunistic’ behaviour, the conflict of interests gives it motives and makes it inevitable. The workers constantly tend to underestimate their individual capacities, the capitalists to overestimate them. The times and tempos of work, the length of down time, the evaluation of physical effort and mental dedication are constantly made objects of bargaining and grievances. The result of the production process is always uncertain. Therefore it is necessary to exert organisational control and establish a hierarchical structure to undertake the functions of measuring the capacities and the effort of individual workers, monitoring their performances and disciplining their behaviour. Control, in turn, gives rise to some particular organisational problems. Its function is to compel the workers to reveal their hidden information and discourage hidden action. It must be exerted by bosses who are close to their subordinates and have the required competencies, who know the labour process and the single individuals who operate it. The bosses must be able to interpret individual behaviour so as to extract relevant information from it. And they must be able to refer this information to their superiors. These capacities are limited, and the number of workers that can be efficiently controlled by a boss is bounded. If there are many workers and many teams at work, there must be many bosses. To accomplish their tasks, the bosses need delegated authority, an authority constituted through the firm’s normative system and legitimised by an appointment conferred by the superiors. Delegated authority enables the bosses to give their orders as consequences of the orders they in turn have received. But a boss must also have some discretional power if he is to be able to cope with situations his superior does not know, in other words he must have some residual power. Since this implies a certain autonomous decision-making capacity, the power of bosses

194 Production governance structures itself is exposed to the risks of information asymmetries. In fact the bosses too might try to conceal information, both information concerning their own capacities and that concerning their subordinates’. Situations of collusion with free riders might ensue, as well as attempts by the bosses to use their residual power to obtain, both from their subordinates and from their superiors, some personal advantages that conflict with the pursuit of the firm’s goals. The exercise of control gives rise to two kinds of difficulties. The bosses’ capacities are bounded, and the bosses may be tempted to behave opportunistically. The well-known problem of who controls the controllers arises. Thus the organisation of a firm tends to take a pyramidal form also in performing its exploitation and control functions. And in this case too, as in that of co-ordination, the height of the pyramid will tend to be greater the wider is its base. But cadres cost money, so that again an increase in a firm’s size leads to increases in organisation costs. The firm is always faced with the necessity of having to compare the advantages it obtains from the use of a control hierarchy with the costs it has to bear to pay the controllers. Incentive-based governance structures One way of economising on control is to transfer it to the workers, which can be done by offering them incentives correlated with labour productivity. If they are individual incentives, they serve to induce self-control in the single worker, to encourage him to quench his impulse to rebel and his aspirations to freedom, so that he voluntarily accepts to align his behaviour with the pursuit of the employer’s interests. If they are collective, they aim at inducing group control, at coaxing workers to control each other, at convincing each of them that the others’ aspiration to freedom conflicts with his own interests. Group control is based on the divide et impera method: each single worker is bribed to believe that his colleagues’ behaviour is compatible with his own interests only when it is conducive to those of the employer. Incentives create in the worker a real interest in over-exploitation, and are accepted if the increase in personal income they bring exceeds the increase in effort they induce. Firms, may find them advantageous if the reduction in control costs they allow is greater than the wage increases they call for, given the expected growth in productivity; or, if the growth in productivity they give rise to is greater than the increase in wages, given the cost of control sustained by the firm. Since self-control does not cost the workers anything, while group control costs very little, it should not be difficult to design incentive schemes which are acceptable to both the workers and the firm. The difference between the cost of control to the firm and the cost of self- or group control to the workers, given the expected growth in productivity, generates a surplus which can be divided between the two classes of actors. Both of them must profit from the scheme if it is to work at all. Information asymmetries on labour capacities take two different forms. Ex ante there is a problem of hidden information: the worker is interested in widening the area of his claimed capacities beyond all reality, so as to increase his probability

Production governance structures 195 of being hired and his expected income. Ex post, he tends to reduce that area, e.g. by declaring that he does not have the abilities and competencies to carry out certain orders or to do it with the skill required by the firm. This is a problem of hidden action. Thus control has to be exerted both ex ante and ex post, in the form of pre-selection before recruitment and in the form of monitoring during the labour process. In control-based governance structures, for example, pre-selection takes place through interviews, exams and tests, while ex post control, as shown in the preceding section, is implemented through the power structure of the firm’s hierarchy. In either case wage remunerates obedience and does not depend on labour productivity. With incentive-based governance structures things are quite different. Control is transformed into self-control or group-control by means of pay schemes which differ greatly from the typical wage system. Wage of course is not abolished, for workers have to be compensated for obedience so long as they are mobilised with an employment contract, and a substantial part of their pay must consist in a fixed wage independent of productivity. But this remuneration can be supplemented with various forms of incentive pay that do depend on labour productivity. Incentive pay schemes which are used in conjunction with wage are reducible to three principles embodied in three kinds of contractual relations, sale (of services), sharecropping and agency. Since the power structure inherent in the employment relationship remains in force, there is no possibility of the worker having substantial decision-making autonomy. Therefore, in studying these principles as bases for an incentive integration of wages, it is better to refer to ‘incentive schemes’, rather than ‘contracts’ proper, and, to be precise, to a ‘sale incentive scheme’, a ‘sharecropping incentive scheme’ and an ‘agency incentive scheme’. The sale scheme is based on the pretence that the worker sells a commodity to the firm. Why is it a pretence? Because there is no commodity or service to be sold. The employment contract avoids any ex ante definition of labour services, worse still, it attributes their production and property to the employer. Sale incentive schemes in reality consist in systems of wage differentials defined in accordance with some of the workers’ characteristics which are passed off as ‘work’ characteristics. For instance workers who give stronger guarantees of loyalty to the firm’s goals are paid more. Since these characteristics eventually result in higher productivity, their remuneration gives the impression of rewarding a flow of labour services. In reality this impression has an ideological meaning and only by virtue of this meaning does it also have an economic value. The sharecropping scheme is based on the principle of attributing to the worker a part of the surplus value he obtains by exerting self-control. By arousing in the worker a personal interest that can only be pursued as a complement of the firm interests, his behaviour is aligned to the pursuit of the firm’s goals. The advantage for the firm is twofold, since the sharecropping scheme reduces control costs while, at the same time, raises labour productivity. The agency scheme differs from the sharecropping scheme in one essential, although not easily discernible, characteristic: behavioural alignment is induced, not by arousing in the worker a personal interest which is complementary to that

196 Production governance structures of the firm’s, but rather by making him take on the firm’s goals as his own. In this case too the ideological content of the scheme is important. Consider for example a boss engaged by the firm’s owner and paid a high wage. The scheme works well if a fiduciary relationship develops between the boss and the owner by which the former voluntarily pursues the latter’s interests. In practice it is difficult to distinguish clearly between the three kinds of schemes in an incentive pay programme, because they are often used jointly. For instance, a sharecropping scheme is clearly present in piecework pay, where the worker is remunerated with a share of his personal product. But the sale scheme may also be present in this case, if the kind of product expected from the worker’s activity is technically well defined. Wage differentials as instruments of pre-selection Given the technique, every firm knows, at least generically, the jobs and tasks required in its labour process. If they do not know ex ante the flow of services they will need during the labour process, they have an idea of the stock of labour capacities they require. But, since jobs are normally used as complements to each other, it is impossible to define separately the individual productivity expected from the appropriate competencies. When team production prevails, as is almost always the case in modern industry, only team productivity counts and all specific competencies are equally useful. Therefore the firm might be tempted to offer the same pay for all competencies.1 The problem is that different investments in ‘human capital’ are necessary to acquire different kinds of competence. If wages were uniform, the workers would tend to even out their investments at the lower level. Wage differentials offered ex ante have, first and foremost, the effect of rewarding the investments in ‘human capital’ demanded by firms, and are normally geared to schooling and experience. They are not commensurate with the kind of school. If we abstract from any difference in individual abilities and from the authority roles inherent in some jobs, two workers with a first degree will be paid in the same way, whether a biologist or a statistician, while a worker with a master’s degree will be paid more than either of them. In this way wage differentials give the workers an incentive to present the number of years spent at school as an information about their competencies. These pre-selection incentive programmes mimic the sale scheme. Since wage differentials are commensurate with schooling, they give the impression of remunerating the flows of labour services produced by labour capacities. Now consider the other aspect of labour capacity, ability, i.e. the intensity of effort and diligence provided in the application of a certain competence. In part this characteristic is psychophysical in nature and depends on the natural endowments of individuals, in part it has a moral and cultural nature, and depends on the workers’ proneness to align their behaviour to the firm’s goals and the bosses’ commands. Both aspects of ability can be observed by the firm, but only ex post, that is, only when the effort has been made or when it has produced its effects. Ex ante they are unknown to the firm.

Production governance structures 197 Also with abilities wage differentials may be used to induce the workers to reveal their information. But, in order to function as pre-selection instruments, they must be accompanied by some self-enforcing mechanisms, i.e. by some kinds of ex post incentive schemes or by organisational control. The workers who reveal greater abilities are paid higher wages, thus they are induced to show off the best of themselves. But the privileges they obtain are maintained so long as the application of abilities comes up to expectations. If loss of privileges is considered worse than receiving no privilege at all, as occurs for instance when privileges signal a reputation, then the workers will have some interest in correctly revealing their information. This case has been observed in Japanese labour markets, where most of the workers are engaged on open-term contracts, almost like tenures, and where being fired enormously reduces the probability of finding another job. When the required abilities have mainly a moral or cultural nature, an effective self-selection instrument is again represented by schooling. I argued that schooling signals the intensity of investments in ‘human capital’ and that schooling-based wage differentials induce workers to invest in the acquisition of capacities. But, while it is easy to understand why people wish to be paid in relation to their investments in themselves, it is not clear why firms should pay them in this way, given that abilities are effective when used as complements. Why should an MA be paid more than a BA? The answer to this question seems to be that investments in capacities requiring many years of school presuppose a great ability and therefore a higher expected productivity. An elegant model of schooling as an indicator of ability was proposed by Spence. It is based on the principle that better endowed workers find it easier to overcome the pains of school. If wage differentials are linked to schooling and the costs of studying are a decreasing function of individual ability, school can indeed act as a pre-selection device. In fact wage differentials can be fixed in such a way that better endowed people would find it more convenient to face the cost of study. The pay advantages that talented individuals would expect to receive would be sufficient to compensate them for the cost of studying, which for them is low; whilst less endowed people, for whom the cost of studying is high, would prefer to accept a low wage and avoid a long and distressing investment in schooling. This model abstracts from the effects that school may have on improving labour capacities and focuses on its effects of signalling ‘natural’ endowments. One might legitimately ask what these endowments actually are. Milgrom and Roberts gave a plausible answer to this question: the required endowments concern the ability to concentrate on boring activities, to patiently endure the bosses’ harassment, and to show due respect for authority. In other words it is the moral and cultural aspect of ability that is selected in this way. A young man who finds it easy to adapt to a school hierarchy, would find it just as easy to adapt to that of a capitalist firm. And this is precisely what employers are looking for, for proneness to obedience keeps at bay moral hazard and any temptation to indiscipline. Schooling at least in part signals such a precious quality, and the wage differentials based on it aim at inducing disclosure of this information.

198 Production governance structures Another pay scheme conceived to produce the same pre-selection effects is the seniority system. Although the wage premiums paid to senior workers seem to reward fidelity ex post, in reality they serve to induce the ex ante disclosure of some information. Undoubtedly firms prefer loyal workers. Here loyalty has a twofold meaning: first, the workers prefer not to quit, i.e. they will not abandon their firm if another firm offers them the chance to put to better use the ‘human capital’ accumulated in the former; second, they try to avoid being fired, and therefore tend to conform voluntarily to the requirements, the orders and even the interests of the employer. This kind of worker will eagerly accept a long-term or an open-term contract based on a seniority pay system, i.e. a system whereby low junior pay is compensated by a fidelity premium which increases as retirement age approaches. Workers who enjoy a greater love of freedom, knowing that they run the risk of being fired before they reach seniority, that is, before any fidelity premium matures, rather tend to prefer short-term contracts which guarantee high junior wages. A widespread, though not always so effective, pre-selection technique is that based on exams and tests, in which candidates for a job are asked to show the best of themselves. Pay is then commensurate with the capacities revealed in the exams. It is in the worker’s interests to reveal his information correctly, if he knows that he will be able to keep his job and wage only if his work performance comes up to the firm’s expectations. Workers would not be induced to cheat at exams if they knew the benefits they could obtain in this way would be rather precarious. Sometimes the risks of dismissal, demotion or pay-cuts act as enforcement mechanisms capable of inducing the workers to correctly pre-select themselves. But their effectiveness depends on the ability of the firm to precisely meter and control individual performances ex post – a condition that does not occur very often in modern industry, where team production is the rule. This is why candidates normally try to cheat at the exams. Another pre-selection device associated with pay differentials is based on management by objectives, a sort of bargaining in which performance standards and evaluation criteria are jointly negotiated by the employer and the employee. Higher wages are associated with more ambitious objectives. The worker is not exposed to a vocational or aptitude test but is requested to decide on the pay scheme he demands in relation to the labour activities he feels able to offer, thus revealing his capacities. If the performance is easy to evaluate ex post, the worker is interested in pre-selecting himself correctly, i.e. choosing the most suitable work objectives for his capacities. A similar method is based on menus of contracts, in which a worker is offered the possibility of choosing from among various complex contracts in which fixed wages are combined in different degrees with performance-based pay. Each contract is characterised by a particular wage level and a particular incentive intensity. The worker chooses the contract he prefers. If he has a low risk aversion, for instance, he will choose a remuneration consisting of a low wage and a high incentive. In this way he is induced to reveal correctly his capacities.

Production governance structures 199 Individual incentives as instruments to induce self-control As already observed, if workers try to widen the area of the capacities they claim to possess at the time of contract stipulation, once the labour process is underway they tend to restrict it. Since this area defines the technical limits of authority, its restriction results in a reduction of the power scope, i.e. the set of labour activities that can be determined by the employer with an action of command. In this way the workers set up a counter-power through which they obtain, not a reduction of formal labour time, but certainly reduction of the effectiveness of power, the capacity of bosses to determine their activity. Since the rate of exploitation, given the wage rate, is an increasing function of labour productivity, and this is determined by the capitalist control on labour activity, the restriction of the power scope also results in a reduction in the rate of exploitation. It might be thought that the benefits accruing to the workers from this result is nil, if the wage is fixed. But productivity is an increasing function of labour effort, i.e. of the care, diligence, alacrity and strength with which the orders are executed. And effort causes toil and sacrifice. Thus the counter-power action aimed at abating labour productivity is not justified by the mere desire to reduce profits, although this motivation might be present in a conflicting social relationship; it is mainly justified by the desire to reduce sweat. There are two conditions that make it difficult for an employer to obtain maximum effort. One has to do with the collective nature of labour activity and emerges in situations in which team production prevails, a problem I will deal with in the next section. The other has to do with the individual character of labour activity, and depends on the worker’s information privilege over his personal capacities. In principle it is always possible for an employer to obtain all the effort he requires, provided he is prepared to push the control, monitoring and metering of the labour activity as far as necessary. The problem is that the more you push, the higher are the costs of control. Hence the necessity to pass part of these costs onto the workers’ shoulders. The fundamental condition that makes self-control feasible is that labour performance is easier to observe than effort. In this case it is possible to link at least part of labour pay to the labour outcome. Let me define an incentive wage, wi, as the summation of a fixed wage, wf , and an incentive linked to performance, wp: wi = wf + wp(Q, F) where Q is performance, expressed as some measure of labour productivity, and F a measure of performance variability. wp is obviously an increasing function of Q. Less obvious, but equally certain, if the worker is risk-averse, is the fact that wp depends on F too. In fact Q will vary in relation to stochastic events which are both objective, like variations in the efficiency of labour instruments, and subjective, like variations in the evaluation criteria or in the bosses’ moods. Therefore the incentive must contain a risk premium which will be higher the greater the performance variability.

200 Production governance structures For wi to be a proper wage, some conditions must hold true. The first is a subordination condition, and is wf > 0. Only when the worker is remunerated with an obedience premium is it possible to preserve the command structure enabling the employer to maintain control of the labour process. A command prerogative is necessary also when incentive pay is applied, if a surplus is to be extracted from labour activity. In fact it enables the firm to determine the nature of the labour process and products, as well as to fix the level of incentives. An incentive feasibility condition must also hold. If productivity is an increasing function of effort, e, that is Q = Q(e), Q ¢ > 0 , the incentive feasibility condition is that Q is easier or less costly to observe than e. By rewarding Q, incentives induce the worker to practise self-control and provide the required amount of effort. Incentive pay does not eliminate exploitation, even when it gives the impression of remunerating the workers for their products. Thus it is possible to define a marginal exploitation condition, which is Q ¢(e) > wi¢ = w p¢ (Q ¢(e)) . The marginal product of effort is greater than the marginal incentive. When incentives are defined with a linear formula, for instance w p = a Q + b F , it is easy to measure marginal exploitation. In fact, if wi = w f + w p = w f + a Q + b F , and assuming F does not depend on Q, it is wi¢ = a Q ¢(e) . Therefore the marginal exploitation condition is Q ¢(e) > a Q ¢(e) or a < 1

The rate of marginal exploitation implicit in incentive pay is

Q ¢(e) - wi¢ = Q ¢(e) - a Q ¢(e) = 1 - a If the net value of an increase in production obtained with piece-work is $100 and the worker is paid a $40 incentive, it is a = 0.4 and the rate of marginal exploitation is 60 per cent. Linear incentive formulas are quite common and are easy to interpret as applications of the sharecropping scheme. There are various forms of pay that follow this scheme, sale commissions, royalties etc. Piecework pay is their prototype. And one might ask why, in spite of its simplicity and apparent effectiveness, piecework is not so widespread in modern industry. There are many reasons for this. Some of them have to do with the workers’ rejection of piecework systems, others with the firms’ rejection. A single worker has two good reasons for refusing piecework. One is the risk inherent in production and therefore in remuneration. As already mentioned, labour performance depends on events which are not completely controllable by the worker: a tool or a machine may break down, raw material may be defective, the worker’s health may deteriorate; worse still, performance may be evaluated by bosses who are inclined to use highly subjective criteria and harassing methods. When these conditions are present in abundance, the risk premium acceptable for a worker may be higher than the risk premium the firm is prepared to pay, in which case piecework is rejected.

Production governance structures 201 Let me show this in a simple case. Suppose F is an increasing function of Q, meaning that the higher the labour productivity the riskier the job, and assume, for the sake of simplicity, that F = g Q . Then the above formulae change in the following way. The marginal incentive is wi¢ = (a + bg )Q ¢(e) , while the marginal exploitation condition is

Q ¢(e) > (a + bg )Q ¢(e) or a + bg < 1 The rate of marginal exploitation is Q ¢(e) - wi¢ = 1 - (a + bg )

Marginal exploitation will be lower the stronger is the incentive, the higher the worker’s risk aversion and the riskier his labour activity. Very risky jobs and high risk aversion would induce workers to demand marginal incentives that kill the incentive feasibility condition. Firms would pay lower incentives than such exorbitant premiums. Therefore the workers would refuse piecework. Another reason concerns the high degree of effort required by piecework systems, obviously higher than that necessary to get a fixed wage. And workers are not always prepared to work harder in order to earn more. Moreover there is always a possibility of the performance obtained with piecework providing the firm with some information on the workers’ abilities that may subsequently be used to raise work standards. In this case there is a risk of producing the wellknown ratchet effect whereby the earnings obtained today with piecework might bring about an increase in the amount of effort expected tomorrow by the firm in normal work conditions. Whatever the pledge given by the firm regarding the utilisation of information collected through the piecework system, the risk of recontracting is always open. And this is a risk that weighs heavily on the worker’s shoulders, given that it is his privileged information that is revealed, and given that the employer can fire an employee but an employee cannot fire his employer. The reasons for rejecting piecework are even stronger from a collective point of view. In fact a single worker may find it worthwhile anyway to accept incentive pay if he is convinced that the greater amount of money he immediately gets on risk and incentive account makes up for the greater effort he is required to provide. But his colleagues might try to obstruct this practice if they think that the exceptional performances of one or a few individuals might be used by the firm to fix the standards for all or most of the employees. This is the basic reason for the apparent contradiction often observed between individual and collective attitudes towards piecework, the single workers often being more prone than their Union delegates to accept incentive pays. It must be added that unionised workers are normally more conscious of the fact that piecework brings about a sort of overexploitation of the working class, that it swells profits more than wages. But firms too might have good reasons for avoiding piecework. To start with, they avoid it when they produce commodities with a strong trade-off between quality and quantity, if for no other reason, because quality is an aspect of perform-

202 Production governance structures ance which is difficult to measure, so that pay cannot be efficiently linked to it. If incentives are mainly commensurate with quantity, the workers would tend to privilege it to the detriment of commodity quality. Moreover firms shun piecework in all processes requiring complex activities that call for the provision of different and heterogeneous kinds of effort or entail the production of composite commodities characterised by different kinds of performance. If the worker is paid only in relation to a particular kind of performance, he would tend to privilege a specific kind of labour effort. In this case the principle of equal compensation, whereby all kinds of effort must be equally remunerated at the margin, would suggest cancelling out all marginal incentives, thus paying a fixed wage. A particular case is that of the correct maintenance of machinery. When the means of production utilised by a pieceworker have a high value and require accurate and diligent use and upkeep, the principle of equal compensation would suggest motivating machine maintenance to the same extent as production. But this is not always possible. As a consequence, the only way to induce the worker not to neglect the correct use of machinery is to discourage him from over-utilising them, in which case a fixed wage turns out to be the most efficient form of remuneration. Collective incentive schemes However, the most important reason for the poor diffusion of individual incentive schemes is team production, in which it is not possible to define wage premiums on the basis of individual effort because it is impossible to observe individual performance. Note the difference between this condition of hidden action and the one dealt with in the previous section, where I assumed that individual performance can be observed. In normal conditions, any member of a team would try to form solidarity with free riders, to react with reduced effort to the reduced effort of his colleagues. The reason is that the benefits produced by greater effort accrue to the firm and not to those who provide it, nor to their colleagues, so that no team member has any interest, nor valid justification, to discipline a shirker. The productivity of the effort of each worker is positively correlated to the productivity of the other team members. If worker A works less, worker B should work more to keep up the team’s production. Why would he do so? And if B has no valid reasons for compelling A to work more, the spontaneous reaction would be to reduce his own effort too. In this way a sort of unintentional group shirking emerges which can be interpreted as spontaneous counter-power action. Team production, however, offers firms the possibility of reversing the complementarities of work activities to their own advantage. It offers them the opportunity of stimulating reciprocal control among the workers. This can be done by means of a collective incentive scheme, a kind of premium rewarding the performance of the whole team. Group incentives are commensurate with team productivity and are normally distributed among the members in some way they consider fair. There are various kinds of group incentives. One consists in bonuses paid to the members of a team (a work group, a division, even an entire factory) depending

Production governance structures 203 on the quantity produced by the team. Another kind is that in which the bonus is commensurate with savings in labour costs. Once a target is fixed for unit labour costs, the workers are paid a percentage, often quite high (50–70 per cent), of any difference between target and actual labour costs. Since the target labour cost is calculated on the basis of a fixed wage and a standard or normal productivity, the only way to cut real labour costs is to raise labour productivity, i.e. to work harder. Still another kind of group incentive is profit sharing or profit-based premiums. Workers receive annual bonuses calculated in proportion to the profitability of the whole team. Alternatively they could be apportioned part of the company shares, so that they acquire title to the distribution of dividends. Finally, a kind of group incentive which is widespread in some strongly unionised systems is that based on collective agreements whereby automatic wage increases are linked to growth in average labour productivity. The intention of these agreements, at least in the Union’s view, is to stabilise the wage share, but its main effect, from the firm’s point view, is to encourage workers to improve productivity. How and why do group incentives succeed in transforming worker solidarity into an instrument of reciprocal control? The crucial point is that individual effort, which is not observable from outside the team is sometimes easily observable from within. Since the productivity of any team member is an increasing function of the productivity of the other members, and since productivity depends on effort, each member directly suffers the hidden actions of his colleagues and, if the team is not too big, he is able to identify the workers who are responsible of any production setbacks. If four workers push a trolley and one of them does not push hard enough, the other three have to push harder and immediately recognise the slacker. An outside observer cannot do that. Now, a worker who notes the poor effort of a colleague will not be tempted to form solidarity with him if he knows that in this way he will negatively affect team production, team incentives and therefore his own personal income. It is true that the benefits a worker obtains from his shirking is normally higher than the losses he sustains on incentives, given that these are divided among all the members of the team. But it is also true that the cost of controlling a colleague is often lower than the income loss brought about by not controlling him. Therefore, while each individual worker might find it worthwhile to shirk, no one would be interested in allowing the others to do so. That is why group incentives transform worker solidarity into group control. Fixed wages sustain the workers’ indifference to the firm’s profitability and motivate forms of behaviour aimed at cutting exploitation, work toil and the bosses’ power. Group incentives give rise to an individual interest in the team’s productive achievements and induce the workers to control each other. In this way they succeed in aligning the workers’ behaviour with the employer’s interests. But they also obtain some other important results. Repeated and prolonged interaction takes place within a team. Therefore each worker tends to assume an attitude of co-operation with his colleagues, expecting in this way to obtain cooperation from the others. This means that opportunistic behaviour is discouraged even independently of control. Each worker will tend to practise individual selfcontrol in the hope of inducing the others to do likewise. A consequence is that the

204 Production governance structures cost of reciprocal control shrinks even further; for it is easier, when there are only a few shirkers, to identify the loafers who hang about the factory. It must be added that the life of a company is always longer than that of its teams, which can be repeatedly disbanded and re-organised, especially when the labour process changes. Therefore a worker knows that at any time in the future he might have to interact with any other worker outside his present team. As a consequence, when the time to disband a team approaches, the temptation to resort to opportunistic behaviour is kept at bay by the desire to accumulate a reputation for co-operation which can be used in the future. Moreover co-operation induces the sharing of information and the ensuing production of collective knowledge. The information a worker would hide from the boss is spontaneously made available to colleagues. As I will show in the fourth part of this chapter, knowledge-production is a process of collective elaboration of information and calls for members of a group to reveal their own information, to integrate it with that of the others, to modify it on the ground of their own and others’ experience, to correct mistakes, to imitate efficient practices. All these forms of behaviour are naturally stimulated by the co-operative attitudes that may arise in a team. Group incentives also carry out this function: they transform a team of employees into a group of persons capable of giving the best of themselves precisely because they are part of a group. It transforms the natural tendency of individuals to conceal their information (from the boss) into a tendency to reveal it correctly and readily (to colleagues). Group incentives are also effective in reducing control costs. They serve, not only to pass part of these costs on the workers’ shoulders, but also to allot them efficiently. The firm retains for itself the control of team performance, on which it has an advantage over the single workers. In fact, because of the collective nature of production, its value can be better measured by an outside observer than by an internal one. The workers, instead, are saddled with the costs of control of individual actions, which, as already said, are easier to observe from inside than from outside the team. This allocation of control costs is efficient because it assigns to each subject that part of control he is better able to exert. Finally, group incentives also have a significant effect on risk bearing remuneration. If pay is to include a risk premium, this would be greater the higher the risk and the workers’ risk aversion (or the lower their risk tolerance). Group incentives contribute to reduce this premium. First of all they lower individual risk, and the more so the greater the size of the team. In fact, since incentives are paid to any individual in relation to the production of the entire team, the negative effects of an individual error are shared by all the team members. In other words the group provides some insurance for the individuals. Moreover it raises the collective risk tolerance, for a team may tolerate more risk than all its members. The risk aversion of a team is lower than that of its single members. Since incentives are paid for team production, the risk premium is also defined as a function of the risk borne by the team. Therefore it will be lower than would be paid to the single workers in relation to their individual risk tolerance.

Production governance structures 205 Efficiency wages Group incentives too prompt the question: why, in spite of the evident benefits they offer, are they not so widespread as theory would forecast? The answer is that the conditions for their efficiency are often lacking in large firms. It is true that workers are used in conditions of strong complementarity in modern factories. But it is also true that this complementarity usually involves the entire productive process of a firm, and modern industrial companies are rather big. And when an entire big firm constitutes a team, there is little sense in applying incentive schemes to single individuals or small groups. It must be added that in big modern companies there is also strong complementarity between technical and ‘human capital’. Machines are very advanced from a technological point of view and continually exposed to the disrupting effects of innovation. In these circumstances, the tasks, competencies and labour activities required of the workers are somewhat complex. Workers, among other things, must have the ability to adapt to technical change and take care of machinery efficiency. A skilled worker who works in these conditions is not ordered to produce always the same piece or to carry out always the same operations. He is normally required to undertake several activities simultaneously or in rapid succession, including machine upkeep. In these cases the principle of equal compensation is applied, so that each worker is paid a fixed wage. It might be thought that, if the whole firm enjoys team production, incentive schemes should be applied to all employees to the same extent. But in this case the negative effects of a worker’s shirking on total production is small, so that the cost another worker has to bear to reproach the shirker tends to exceed the benefits. Any incentive to group control fades away and the solidarity in shirking once again prevails. Firms can no longer count on the workers’ reciprocal control. Not even profit sharing is convenient. Modern industrial companies make massive use of fixed capital and their capital-output ratio is high. Thus the incentive that could be provided with employee stock ownership plans is problematic. If the quantity of shares distributed to the workers is too small, they provide insufficient incentive. If the part of capital so distributed is sufficiently large, organised workerowners could exert substantial control on management, especially if external shareholding is widely dispersed. In this case the conditions for the managers’ control on workers would be transformed into those for the workers’ control on management, which is why worker share-holding is not very widespread in modern large firms. Indeed, when they are used, they serve mainly to reinforce organisational involvement and company ideology rather than to furnish a proper economic incentive. Most of the employees in large firms are skilled workers who hold productive capacities of some value. This may be partly firm-specific, partly trade-specific, and partly generic, but is mainly the result of investments in ‘human capital’ made in a firm. The management tries to avoid losing this capital and, in periods of production slowdown tends to dismiss fewer workers than justified by demand conditions. Conversely, as a consequence of the labour hoarding practised during

206 Production governance structures recessions, new recruitment during boom periods grow less rapidly than production, quite independently of technical progress. This is one of the major reasons why average labour productivity tends to vary pro-cyclically in modern economies. In these conditions labour becomes a relatively fixed cost. Finally, it must be noted that, both because of the huge size of modern companies and because of the high skills required of the employees, workers are organised into big Unions which are capable of bargaining and substantially affecting wage rates and work conditions. Thus the determination of wages is not under the complete control of management, and this is another reason why there is poor diffusion of individual and group incentive schemes. So, hierarchical control cannot be avoided in large firms. Now, if it is not possible to substitute control costs with incentive schemes, is it not at least possible to minimise them by letting the workers themselves exert some kind of self-control? One rather widespread device for coping with this problem is what is called the ‘efficiency wage system’, a kind of pay arrangement which looks like a macroeconomic incentive scheme and which is capable of inducing some individual self-control. Let me stress this point: in spite of their collective nature, efficiency wages stimulate individual self-control rather than reciprocal control in a team. I will show this later. There are several versions of the efficiency wage theory. The one I am going to present here is a personal elaboration of mine which has received some empirical confirmation. It differs from mainstream theory especially in three ways: (1) it treats efficiency wages as a sort of collective incentive scheme whereby wages vary in accordance with aggregate labour productivity; (2) it accounts for wages being fixed, through a bargaining process, by the firms and the Unions; (3) it demonstrates the irrelevance of unemployment levels in disciplining the workers, and explains at the same time the hysteretic nature of unemployment rates. The increase in efficiency wages paid to the ‘representative’ worker is dwi dw f dwp = + wi wf wp

with dwp wp

 dL dp  = f  × ,H  L p  

where dwf / wf is the annual rate of increase of the ‘outside ‘ wage, i.e. the wage a worker would receive outside his firm once fired, dwp / wp a sort of a ‘fidelity premium’, i.e. an employment rent paid to all the workers in a firm, dL / L the rate of change of employment, dp / p the rate of inflation and H the degree of the workers’ autonomous militancy. Note that the equation is defined in terms of rates of change in wages and not in terms of their levels, an assumption I make for the sake of realism. In fact wage negotiations normally deal with wage increases. The basic reason for such practice is that average labour productivity grows permanently in modern industrial systems, so that the workers feel collectively entitled to a systematic participation in the improvement of their performances. This does not

Production governance structures 207 mean that every claim or every wage increase concession is immediately linked to short-run productivity changes. It merely means that the distribution conflict has wage increases as its main stake rather than wage levels. Wage fixing is the result of a collective bargaining process, a form of class conflict in which indiscipline and militancy are the workers’ weapons, whilst hierarchical control and dismissals are the management’s. In reality the firm uses the good old method of the carrot and the stick, hierarchical control being the stick. In order to reduce its cost, control is exerted in an incomplete way, i.e. not on all the workers but only on some chosen stochastically. The worker who likes to shirk knows he may get away with it, but also that there is some probability of being caught out and fired. This probability, as well as the cost of control, is an increasing function of the extension of control. The carrot consists in a premium paid to all employees over and above the outside wage. In this way the worker is disciplined by the fear of losing his job and the fidelity premium. Fear is stronger the more effective is hierarchical control, i.e. the probability of being fired if caught shirking. It is stronger the higher the premium, i.e. the income loss from being fired. Finally, it is stronger the higher the probability of being dismissed independently of individual worker effort. In fact the longer the expected permanence in a firm and the higher the fidelity premium, the higher is the present value of expected future income. The workers’ fear of losing his job independently of personal effort is a decreasing function of the rate of change of employment, dL / L. When this is positive, with new recruitments greater than dismissals, fear is low; when it is negative, fear is high. The level of unemployment does not play any relevant role in determining efficiency wages. This is because workers have to cope with a kind of information asymmetry which is complementary to that coped with by the firms. The latter do not completely know the labour effort of their employees, but know better than them the demand for their goods and their demand for labour. The workers, on the contrary, know their labour capacities very well but not the demand for labour and goods. Since they are endowed with a ‘human capital’ which is at least trade-specific, in the event of dismissal they would search for a new job in the same trade in which they have worked up until then. To know how unemployment would affect their income, they would have to know the trade-specific rate of unemployment, but this information is not easily attainable. Moreover, because of the irregularity of production and employment cycles, the level of skilled labour unemployment tends to be perceived as temporary, and this does not encourage emigration. Finally, the systems of direct and indirect social assistance in advanced capitalist countries, in addition to the relatively high income of the worker family, allow unemployed people to survive for a long time. This accounts for the so-called ‘rational emigration’ phenomenon brought to light by recent research. Workers emigrate only when they are faced with a situation of permanent unemployment in the presence of below subsistence incomes. Thus emigration from underdeveloped to developed countries is strong. Conversely, internal emigration in advanced countries

208 Production governance structures is moved by pull factors (better incomes and labour and life conditions) rather than push factors (dismissal and temporary unemployment). The consequence of all this is that the rate of general unemployment in an advanced capitalist country, which is public knowledge, does not affect the workers’ fear of losing their jobs and is hardly relevant for their decisions on the convenience of shirking and wage claims. Knowing that there is high unemployment in the South of Italy does not affect a worker from the North, who does not expect strong labour competition in his district anyway. Nor is the fear of a skilled engineer much influenced by the knowledge that there is high generic unemployment in Italy. This worker would be interested in knowing the rate of unemployment in the mechanical industry, but the rate of general unemployment does not provide this information. The only kind of relevant information workers can obtain to estimate the income loss brought about by dismissals, is that conveyed by their firm’s engagement and dismissal policies. The rate of change of company employment is the variable that more directly affects the workers’ income loss risk and their fear. Thus dwp / wp is an increasing function of dL / L. When new recruitments exceed dismissals, employees may correctly assume that the risk of losing their jobs is lowering and that the probability of finding a new job in the event of being fired is on the increase. Therefore, to maintain discipline, firms are compelled to augment the fidelity premium. The function linking dwp / wp to dL / L also depends on the effort required of the workers. Since effort is sacrifice, the more you ask of it the higher the value you must give to the job, so as to render job loss more costly for the worker. Finally the magnitude of the fidelity premium will depend on the psychological effectiveness of dismissals, i.e. on the effect they have on the workers’ fear of losing their jobs. The stronger the fear, for any given rate of employment change, the lower efficiency wages can be. Taking into account all these parameters, the cost of control, the intensity of required effort and the psychological efficacy of dismissals, companies will calculate the minimum increase in efficiency wages. The efficiency wage increase is not unique, since all increases which are higher than minimum contribute anyway to induce workers to slacken their resistance to the firm’s requests. The minimum increase constitutes the firm’s targets in wage negotiations. The firm’s wage increase proposals will be linked to this target. The workers’ targets, which are determined by taking the firm’s proposals into account, will obviously be higher than these. They will also take into account the rate of inflation, if workers exert real wage resistance. The price elasticity of wages is a measure of the efficacy of real wage resistance. The final result of wage conflict will depend on the workers’ militancy as expressed by the intensity and extension of industrial actions. Militancy is a partially endogenous variable. In fact industrial actions serve, at least in part and among other things, to check information communicated by firms about market conditions. I have already observed that companies, in order to save the human capital accumulated in their labour process, tend to dismiss fewer workers during recession phases than market conditions would justify, asking for wage moderation in exchange. But, when they announce the necessity to resort to

Production governance structures 209 wage restraints to tackle fall off in production, they arouse the resentment of the workers, who are led to suspect moral hazard. The workers’ immediate reaction is to strike. If there is really a drop in demand, however, the workers’ resistance might induce the management to react by starting dismissals. Then the workers realise that the information received was correct, so that they stop their strike and accept the wage restraints. Thus the degree of endogenous militancy is low in recession phases, when employment growth is negative. If firms try to be clever, by announcing the necessity for wage cuts or resisting the workers claims’ for higher wages in boom phases, they cannot react to strikes by starting dismissals. The arm wrestling between management and the Unions could go on for a long time and would probably end up with the workers’ victory. The degree of endogenous militancy tends to grow with employment. The rate of employment change affects efficiency wages, not only because it comes into the firms’ calculations as an instrument of discipline, but also because it influences the efficacy of the workers’ resistance and initiative. The degree of autonomous militancy, H, consists in the workers’ proneness to embark on industrial actions independently of any changes in demand conditions. Endogenous militancy can be defined as a function of the rate of change of employment and estimated, for instance, by regressing on the latter variable the number of working hours lost in strike activity. Thus the degree of autonomous militancy can be calculated as the difference between the actual hours lost and those determined endogenously. It can be negative or positive. Autonomous militancy expresses the workers’ autonomy. It sustains social behaviour which is not affected by the economic cycle, and renders wages a partially independent variable. The efficacy of autonomous militancy, i.e. the parameter linking dwp / wp to H, measures the influence of exogenous industrial actions on wages. To sum up, efficiency wage increases are basically a function of three independent variables, the rate of inflation, the rate of employment growth and the degree of autonomous militancy. They also depend on a set of parameters, the most important of which are: the required effort, the intensity of control, the psychological efficacy of dismissals, the price elasticity of wages and the efficacy of worker actions. Since efficiency wages do not give the impression of being linked to performance, one might ask whether they can legitimately be likened to an incentive pay scheme. In reality that link exists, since the tendency of companies to treat labour as a quasi fixed factor urges them to moderate dismissals in recession phases, with the consequence that labour, like plants, will be under-utilised in these phases and over-utilised in prosperity phases. Thus aggregate labour productivity will vary pro-cyclically, exactly as efficiency wages do, and these will eventually turn out to be correlated to macroeconomic labour performance. In other words efficiency wages can be interpreted as functioning according to a macroeconomic incentive scheme – an understandable result in the light of the observation that each firm in its entirety works as a team. A factor contributing to weaken the cogency of this interpretation is worker militancy, which renders the wage a partially autonomous variable and makes us interpret it as a magnitude determined not only technologic-

210 Production governance structures ally but also socially. In reality, as I will show shortly, wages are also determined politically. Two final considerations remain to be made. The first is that, since the level of unemployment does not work as an instrument of labour discipline, there will be no ‘equilibrium’ rate of unemployment. The equilibrium reached in the distribution conflict is in reality a situation of balance between militancy and fear, between the workers’ autonomy and the capitalists’ animal spirits, and not between the demand and supply of whatever. When the workers are angry, indiscipline increases both inside and outside the factory. As a consequence, wages rise rapidly. But then the investors’ confidence tends to become depressed, investments slow down and the economy falls into recession. The ensuing decrease in employment paves the way for wage moderation. In an ‘equilibrium’ situation the rate of employment growth and the degree of autonomous militancy must balance in such a way as to keep inflation constant, so that wage negotiation actually deals with real wages. In this state the ‘natural’ rate of unemployment may be whatever, and may even vary continually, both stochastically and deterministically: it will depend on past history. The second consideration concerns the role that the state may play in the process of wage and employment determination. The political actor who governs the public budget and the supply of money is in a crucial position in the industrial relations game. He can determine aggregate demand and push unemployment to whatever level he wishes. He cannot, however, fully control inflation, which mainly depends on the degree of militancy and the firms’ price policies. A state endowed with those powers might decide not to exert them, leaving the social classes free to fight their struggle in their own way. Sometimes this is not the most responsible policy, for the conflicting classes might reach a ‘natural’ equilibrium with dramatic rates of unemployment and inflation. A state that decides to exert its macroeconomic powers might do so by adopting an incomes policy in which the unions are called upon to moderate militancy. In exchange they will be offered macroeconomic management aimed at stabilising unemployment to a targeted rate. Company ideologies and organisational involvement Control-based and incentive-based governance structures have various flaws: to start with, they are costly. But often they are also difficult to apply, especially in labour processes which use refined technologies and which are continually exposed to technical innovation, so that workers cannot settle down into routine practices. When the evaluation of effort (in the case of control) or performance (in the case of incentives) is dependent on the subjective judgement of bosses, there is a danger of exacerbating problems, for complaints over the evaluation of effort and performance may add to the normal causes of conflict. One kind of governance structure which does not present these flaws is that based on organisational involvement, i.e. on assimilation, by the workers, of the firm’s goals and the rules governing its internal life. Whilst control- and incentivebased governance structures appeal to the workers’ own interests and try to align them with the firm’s goals by means of punishments and rewards, in organisational

Production governance structures 211 involvement the very perception of the workers’ interests is altered so as to induce the employees to accept the firm’s goals as their own. When this result is obtained, the necessity for control and incentive pay fades out, for the workers have no more reasons to conceal information or actions. The workers know why they must apply themselves, since they share the firm’s objectives, and know how to do it, since they know the rules, traditions and conventions regulating company life. Organisational involvement is normally not a problem for managers, who, being in control of the organisation, the labour process and the investment decisions, may also largely control the condition for the accumulation of their income, power and various other kinds of privileges. This is true not only for the managers of firms which are not controlled by their owners, but substantially also for those of firms controlled by external owners, given that the profit motive tends to coincide, in a long run horizon, with that of accumulation, in which the managers are particularly interested. Indeed it is highly probable that the managers’ interests spontaneously tend to coincide with the firm’s goals. Not so with the workers, who have to endure the power and exploitation relations underlying capital accumulation. Therefore, in order to induce organisational involvement in them, it is necessary to trigger off a process of formation of beliefs which has all the characteristics of an ideology production mechanism. A company ideology is an instrument through which the workers’ organisational involvement is constructed. It consists in the inculcation of a false consciousness whereby the relation between the firm’s means and ends is inverted. In the real world labour activity and production organisation are the means for pursuing profit and accumulation ends. In ideology, conversely, profit and accumulation are presented as the means for attaining the aspirations of employees. When the workers are convinced they are working for a social good and their own self-realisation, and that the company objective coincides with their own interests, organisational involvement has been obtained. It goes without saying that, in a capitalist firm, the company’s ideology is a conducive ideology for the social actors who command and exploit and an adverse ideology for those who obey and are exploited. It is in fact the instrument used by managers to overcome problems of hidden action and information at little or no cost, apart from that incurred for remunerating ‘human relations’ and ‘communication’ experts. Most workers are psychologically exposed to the acceptance of company ideologies, since factory life is a source of continued and varied frustration. The work in itself is frustrating. The productive discipline, through which the worker is used to pursue an end which is not his own, is also frustrating. The alienation brought about by a contract whereby the workers ‘freely’ sell their freedom is frustrating. And similarly frustrating is the feeling of exploitation the worker suffers every time they observe increases in productivity which do not result in wage increases. Frustration may manifest itself as an effect of the multiple self that develops almost naturally in a social context wherein the individual simultaneously participates in different and contrasting roles in the firm, in the family, in the union, in the church and so on. But it may also be caused by the incapacity to obtain goods and rewards which come up to the workers’ aspirations. Many people

212 Production governance structures try to cope with frustration by resorting to various self-deception devices, like illusion, undervaluation, focusing, idealisation and ignorance, which are often sustained by processes of social rationalisation. It is precisely through these processes that adverse ideologies are formed and diffused. A company ideology must be inculcated in the workers’ minds. Various socialisation mechanisms2 accomplish this task. Some of these mechanisms are external to the firm, and take place in the family, at school, in society as a whole. Others are internal, and occur especially in the apprenticeship and training period. External socialisation is a very important precondition for internal socialisation. The acceptance of a dominant ideology is a necessary condition for settling down well in company life. This is a reason why firms often use schooling as a personnel pre-selection criterion. As already argued above, the most important ‘social’ capacities selected with this criterion are proneness to collaboration and organisational involvement and therefore the willingness to accept company ideologies. Internal socialisation mechanisms can be induced or spontaneous. The former are those consciously prompted by the management. They work through instruments of inducement aimed at sustaining organisational involvement. They are not incentives to productivity, but compensation aimed at favouring psychological processes through which the workers learn to deceive themselves about the role they cover in the organisation and the reasons that justify it. Spontaneous mechanisms, have a non-deliberate character, and automatically emerge from social life in the factory, from interaction among the workers, from participation in collective work. Some of the most important reward instruments in induced socialisation mechanisms are profit sharing, involvement in decision-making, employment stability, seniority-based wage differentials, career. Spontaneous socialisation mechanisms are sustained by some psycho-social processes, like imitation, conformism, and the approval of bosses and colleagues, which in practice work as instruments of inducement. Profit sharing helps the workers to convince themselves that the surplus they produce is not seized and controlled by a class of exploiters, but returns, even if only partially, into the workers’ pockets. As to the part that does not return, dominating ideologies in general and company ideologies in particular, try to convince the workers that it is a payment necessary to obtain the use of capital and to remunerate entrepreneurial activity. The profit share that returns to the workers through profit sharing schemes is normally rather small in large firms. Therefore this kind of remuneration plays not so much the role of a monetary incentive to labour effort as that of an instrument of ideological induction. In fact it serves to diffuse among the workers the illusion that income distribution in the firm is based not on class conflict but on community interests. Besides illusion, share profit schemes also play on ignorance, at least to the extent the workers are not informed on the criteria of distribution and use of the profit share that does not return to them. The strength of this illusion can also be sustained by arrangements for the workers’ involvement in decision-making at all levels of the organisation. Workers may be stimulated to participate in the definition of their labour activity in teams

Production governance structures 213 and departments. They may be allowed to elect Union officials who collaborate with the management in fixing pay, labour times and methods, safety measures etc. Their representatives may even be involved in some managerial decisions, as in the German Mitbestimmung system. Normally these arrangements leave the great strategic decisions on finance, investment, innovation and productive organisation out of the workers’ reach. They might however give the impression of participation in the internal decision-making process, thus contributing to alleviate the frustration caused by labour discipline. This instrument too plays on illusion and ignorance. A modern example of this kind of manipulation is offered by the so-called ‘self managed teams’. These are groups of co-operating workers in a firm who have been entrusted with decision-making responsibility in the organisation and coordination of their work, in the choice of methods to improve product quality, in the selection of criteria to share overload and in many other things. Usually the set-up of self-managed teams goes hand in hand with a shortening of the hierarchical ladder, for the simple reason that they reduce the need for control. In fact workers do not obey officers or bosses but follow ‘natural leaders’. Yet what they have to produce is not decided by the team but by the top management. They only have to decide how to produce it. They may be convinced that they enjoy real freedom, receive merit recognition and even obtain satisfaction from their productive activity. Similar convictions may be induced by other organisational devices which also play on illusion and ignorance, as for instance, in the practices of ‘job enlargement’ and ‘job enrichment’. Occupational stability plays instead on focusing. Workers engaged on longterm or open-term contracts can legitimately expect job security, especially if they are endowed with a ‘human capital’ the firm considers valuable. Focusing, in this case, induces the worker to privilege a positive aspect of his labour condition, low unemployment risk, against some negative ones, like discipline, low income etc. And, since the privileged aspect is the one whose existence most depends on company survival, the worker may be easily induced to invest some affection in the company itself, in other words to accept organisational involvement. In certain circumstances, a worker like this might even be convinced to accept wage cuts to ensure the survival of the firm. Associated with occupational stability is another inducement instrument which is widely used in big modern companies, that of seniority-based wage differentials. I argued above that this instrument works as a means of pre-selection. It serves to select workers endowed with capacities that come to light in the long run. These capacities consist basically in the willingness to endure a long socialisation process terminating in complete organisational involvement. But seniority pay schemes do not work merely as a pre-selection mechanism. In reality they serve to induce the illusion and focusing devices through which workers convince themselves that loyalty is an asset for themselves because it is an asset for the firm. For themselves, as it is conducive to increasing pay; for the firm, since it is a source of increasing productivity. Seniority pay schemes, though, give only the illusion of being associated by productivity growth. In practice, by paying wages that grow

214 Production governance structures more than productivity in old age and less than productivity in young age, they mainly reward loyalty. Organisational involvement requires a long time to be accomplished, as it consists in learning unwritten rules, values and beliefs which are specific to the firm. Therefore it is strongly favoured by remuneration schemes that reward seniority. Career is another inducement instrument that combines the effects of illusion and focusing. Internal career prospects are normally associated with job security, and therefore contribute to induce the workers’ preferences to focus on stability of employment rather than on other labour conditions. But career perspectives also promise improvements in income, labour conditions and various kinds of perquisites. As such, they encourage zeal and dedication. And since a substantial part of the dedication demanded of a career-minded worker involves identifying with the firm and assimilating its objectives, this instrument well serves the pursuit of organisational involvement. However, most career expectations are doomed to be frustrated for the majority of workers. Since internal hierarchies take on a pyramidal form, the number of persons aspiring to promotion is always greater than the number of posts open for promotion. Therefore for most workers career expectations serve merely to nourish an illusion. And it is only toward the end of their careers that they may realise they are destined to retire with an income and a status which fall short of the aspirations of their youth. Illusion works well in convincing a worker to invest his life in the good of an organisation which only ex post and too late reveals its inability to maintain all its promises. Besides instruments of inducement, which sustain self-deception devices and reinforce correct behaviour, there are also instruments of punishment, which aim at inhibiting incorrect behaviour. These, however, are not so important as the instruments of inducement. Since organisational involvement is a psychological process requiring a constructive attitude on the part of the workers, it must be encouraged and rewarded. Punishments, serve more to discourage wrong attitudes than to encourage the right ones. As matter of fact, the most effective punishment is the denial of the rewards. Dismissals, for instance, truncate careers and cancel the benefits and perquisites acquired through seniority, status and fidelity premiums. From the point of view of socialisation, however, it is not so much the punishment in itself that really counts, as its very possibility – a possibility that has an effective power of inhibition in those cases where organisational involvement is a necessary condition for career advancement and accumulation of seniority. Demotion, or just the possibility of being demoted, acts as a reinforcement to the socialisation mechanism based on career prospects, since the fear of losing a position which was conquered with great effort multiplies the value of the organisational involvement through which the conquest is made. Transfers to other offices work in the same way as demotion when they imply a loss of status or income. Imitation is the first process of ideology induction which sustains spontaneous socialisation. This is a fundamental instrument for learning labour practices and routines and therefore is strongly encouraged during training and apprenticeship periods. But it does not just concern the acquisition of technical abilities. It also engenders the capacity to assimilate common values and beliefs. At a deeper level

Production governance structures 215 it serves as a means to learn from others how to prompt self-deception devices. Since these lead to the assimilation of false convictions, an isolated individual has some difficulty in triggering them. It is easier for a worker to accept a colleague’s shared conviction than to produce a new opinion of his own. The beliefs shared by others provide a ready-made solution to the problem of constructing a false consciousness. They economise on the effort of inventing new beliefs. Imitation is strongly sustained by conformism. When a conviction is shared by a group of colleagues, it assumes the force of collective wisdom. And since in conformism people find some defence from uncertainty, each time they have to cope with frustration through a self-deception device they first try to get information on how a reference group has already worked the problem out. A conformist perceives his problem as an individual problem only so long as he fails to comply with mass moods. As soon as he accepts compliance, though, his problem is resolved. Whilst imitation serves to provide low cost production of the beliefs required to trigger self-deception, conformism serves to give it the strength of a consolidated truth. And this strength is more important than the truth in itself. In fact, since the opinions that sustain self-deception are false, they might be entertained with indecision and little conviction. But, as soon as a person realises that they are shared by many others too, all doubts vanish. After all, if truth is invariably impregnated with subjectivity, inter-subjective consensus may give the value of an objective truth to a lie. This is how a false belief can become an ideology. A further reinforcement to the acceptance of widespread opinions can be obtained by the approval of others. When a person takes on an opinion sustaining self-deception, he may find some relief for his frustration. When he realises that that opinion is shared by many others, he soothes the doubts raised by his critical conscience. But when he perceives that that acceptance is approved by the others, he has no escape. At that point, any attempt to keep his critical conscience alive becomes antisocial behaviour, which in itself is a source of frustration. Approval from the bosses is appreciated, especially if some career-based instruments of inducement are at work. But much more powerful is the approval of peers. If ideologies look like objective truths when they have the strength of social diffusion, or even useful truths when they are approved by superiors, the approval of peers gives them the strength of neutrality. This kind of consent does not seem to be associated with any particular interest or intention. A person who prompts this kind of self-deception device, suffers double social pressure: that sustained by his own predisposition to imitate the others and conform to widespread opinions, and that produced by the reinforcement action of social approval.

The market Who makes the prices? Production co-ordination in the market mainly takes place through the price system. Given technology, prices determine the costs of production and establish the degree of profitability in producing commodities. They provide inducements to buy and

216 Production governance structures sell, and regulate the division of labour, income and power among firms. Any study of the market system as a production governance structure must therefore focus on the processes of price fixing and price changes. This is the subject of the present part of the chapter. The other fundamental aspect of the market process, the evolution of knowledge and technology, will be dealt with in the next part. The prices of goods are fixed by the economic agents and not by an auctioneer or the ‘market’. The problems are: which agents fix them, the sellers or the buyers, on the basis of which criteria and under what conditions? Let my begin with a simple and very abstract model – the case of an exchange between a consumer and a producer – and let me assume, for the moment, that both actors are endowed with complete information and Olympic rationality. The producer’s reservation prices are determined at a level that ensures the firm’s survival in the short run. They coincide with average variable costs, whose curve is assumed to have a saucer form. The producer aims to appropriate the whole consumer rent, besides the producer rent. Therefore his aspiration prices, that is, the maximum selling prices, will coincide with the demand curve. The consumer has Marshallian demand prices as his reservation threshold. His aspiration prices aim at securing the whole consumer and producer rents, and coincide with variable costs. Lower prices are not aimed at, because the consumer knows that no firm can produce anything if prices are below prime costs. The case is represented in Figure 5.1, where D is an individual’s notional demand curve and C a variable costs curve. The demand curve, in the stretch between pl and ps, coincides with consumer reservation prices and producer aspiration prices, while the cost curve coincides with consumer aspiration prices and producer reservation prices. Therefore the surface bounded by the demand and cost curves between pl and ps represents the area of price discretion. The price remains undetermined if no other conditions are specified. If, for instance, a negotiation process is set in motion, the final exchange price will depend on the path by which it is reached, on the strength relations between the parties and on other circumstances which I shall not go into here. p, C pl D C pm ps

0

Figure 5.1

q

Production governance structures 217 Now, to take a first step towards reality, consider the case in which many consumers deal with one monopolist. Here D is the market demand curve. There are two good reasons for the price being fixed by the monopolist in this case. The first concerns the degree of exit freedom and the consequent effectiveness of exit threat of the various agents. No single consumer can compel a monopolist, by threatening exit, to sell at a price which is lower than the consumer reservation price, for the loss of a customer does not take a great bite out of the monopolist’s profits when there are many customers. A monopolist, by threatening exit, can easily compel a consumer to pay the producer aspiration price, for consumers have no alternative but to buy from that seller. Thus prices are fixed by the monopolist, who, moreover, to secure the whole consumer rent, sells goods to different consumers at different prices. The entire area of price discretion coincides with profits (gross of depreciation allowances). Since the degree of exit freedom is a measure of the monopolist’s threatening capacity, the first conclusion to be drawn is that prices depend on the effectiveness of exit threat and express the strength of market power. The second reason why prices are fixed by the monopolist can be understood if one releases the hypothesis of complete information. In a market where there are many consumers with differing tastes, it is difficult for a producer to estimate the demand of each with any degree of accuracy. If the monopolist were to bargain each single price with each single consumer, he would incur enormous transaction costs in seeking out the consumer, compelling him to reveal his reservation price, informing him of sales conditions and undertaking negotiations. Under such circumstances the seller might well decide to forgo the consumer rent. If transaction costs are higher than this rent, the producer might fix a single price, pm, ensuring maximisation of monopoly profits. While the monopolist saves on transaction costs, the consumers receive their own rent in exchange. Thus the second reason for the price being fixed by the producer is the transaction costs generated by the lack of information. This second reason alone, however, is not sufficient. In fact, if the information search initiative were taken by the consumers, transaction costs would be quite low. Since there is only one producer, search costs would be practically nil, while, after a first transaction, bargaining costs too, as well as those for the revelation of product qualities, would be quite low, if the consumers had the ability to trade at the producer’s reservation prices. Therefore it is necessary to maintain that consumers do not have this ability. The first reason for the price being fixed by the producer – monopoly power – is necessary and sufficient, whilst the second, high transaction costs, is just subsidiary. In reality this is only necessary to account for the fixing of a single price. Similar reasoning applies to the case of many producers and one buyer. The cost curve in Figure 5.1 is now interpreted as an aggregate curve under the provisional hypothesis that all producers use the same technology. The buyer, levering on his high exit freedom and monopsony power, can impose on the producers a price, ps, that coincides with their reservation threshold, thus securing the entire consumer rent.3

218 Production governance structures Goods markets are of the fix-price type even for most agricultural products. Normally in many markets for agricultural commodities there are many small producers who have to cope with a few big merchants or, as I assume for the moment, one monopsonist. And this is the strong party who fixes the prices. The usual view that agricultural commodities markets are flex-price is often borne out by the observation that prices are highly variable. Yet this phenomenon is easily explainable without resorting to the hypothesis that prices are fixed by an auctioneer or the ‘market’. The fact is that agricultural production is in itself highly mutable due to the variability of weather conditions. Normally farmers define their reservation incomes, that is, incomes that ensure survival of the household or firm. Therefore, when harvests are abundant, the reservation prices implicit in reservation incomes tend to shrink, which is tantamount to saying that the farmer’s supply curve is decreasing. This is why the monopsonist tends to fix prices that vary inversely with quantities produced. Three observations remain to be made. First, the alleged variability of agricultural commodity prices is not so great as microeconomic textbooks claim. Prices do not change daily, but only seasonally or yearly, for they are affected by seasonal or yearly weather conditions. The impression one gets by observing the greengrocer’s counter and the tendency of price tags to change frequently is deceptive. Agricultural products, even those of the same species, are not perfect substitutes, so that the week-to-week increase in the price of oranges is often determined by changes in quality rather than in the quantity demanded. Second, wholesalers perform a stabilising function as they tend to compensate the low profits they earn in bad years with the high profits of good years. The reason is that, in reality, wholesale markets are not so much monopsonistic as oligopsonistic. As I show in the next section, oligopolistic or oligopsonistic structures contribute to price stabilisation. Third, retailers perform a stabilising function too, for not even retail markets are perfectly competitive or perfectly monopolistic. Oligopolistic competition, price policies and the demand curve Monopoly and monopsony are extreme cases. The real world lies between the two. But it is not a world of perfect competition. One might be tempted to resort to the theory of imperfect or monopolistic competition to account for the process of price formation, but it is precisely in this field that the theories of Robinson and Chamberlin fail more noticeably. The theoreticians of imperfect competition assume that market demand curves are decreasing, as in monopoly, and consider each company’s strategies as independent of those of their competitors. Firms are assumed to be price-takers in the long run, just as if they traded in competitive markets. Yet, if the products of different firms are not perfect substitutes and if the number of competitors is not virtually infinite, why should the price and product strategies of each company not take into account those of their competitors? Thus it is to oligopoly that one must look, and on uncertainty that one must concentrate attention. Indeed, the uncertainty of estimates in an oligopoly is exalted by the fact that the demand faced by a firm depends on the other firms’ strategies.

Production governance structures 219 Each seller influences his competitors’ demand curves in the attempt to influence his own, for each firm in fact tries to form its own demand curve rather than to discover it. If the degree of product substitutability and the number of competitors are not too low, it is difficult for each firm to interact strategically with all the other oligopolists by simply observing their actions. In reality what characterises an oligopolistic market with many firms is precisely the difficulty in directly observing all the individual competitors’ strategies. This is typical of competition rather than of an oligopoly. The right concept to use, then, is ‘oligopolistic competition’. It defines a market in which each firm interacts strategically, not with any other single firm, but with all the others taken together. Oligopolistic competition is a portmanteau concept. It applies to many kinds of market structure, from an oligopoly with a few companies, like the European or the American car industry, to the case of a myriad of quasi-atomistic firms, like a retail trade in a big city; from a solar system pattern with a big price-leader company, to a constellation pattern of many dimensionally homogeneous firms without price-leadership. Oligopolistic competition is like an oligopoly because each firm takes into account the possible reactions of competitors to its price, quantity and quality decisions and because profits are not cancelled either in the short or in the long run. It is like competition because prices are strongly constrained by the market structure and normally, once fixed, they remain steady long enough to be treated as parameters in the short run. In reality competition hinges on the quality and scope of products rather than on their prices. In oligopolistic competition the market strategy of each firm aims to create demand in order to become less dependent on competitors’ strategies, to anticipate the latter’s reactions and to protect its own market from them. There is a wide range of instruments that can be used to form the demand, from advertising to marketing, from pseudo-innovation to price policies. Price strategies, in particular, deserve some attention, for they make us understand the basic reasons why the demand curve is perceived as decreasing. It is not so much a question of decreasing marginal utility as the possibility of stealing customers from competitors. Given the imperfect substitutability of goods, information incompleteness is a problem of paramount importance. A firm’s potential customers just do not know its products, or do not know them well enough. To steal a competitor’s customers, a firm must convince them that its goods are different and better than those they already consume. If a good has to be perceived as different, then a new need has to be created. To this end, the firm can use various selling cost policies sustaining advertising and marketing. But it can also use policies based on selling prices, that is, strategies based on price cuts, discounts, credit facilities, warranties, special services associated with the product sold and any other attraction device the firm equates to a price cut. The aim of these policies is not so much that of compensating for the lower marginal utility associated with an increase in consumption, as that of inducing the potential customer to approach the new good, to try it, consume it and, in this way, acquire new information,

220 Production governance structures preferences and habits. Once the new customer has been won over, it will be easier to keep him, for any product market is always partially protected by the imperfect substitutability of goods, the consolidation of special tastes and the availability of idiosyncratic information. But this also means that it is difficult to steal customers from competitors. In other words, each firm perceives its own demand as being not very elastic to price cuts. Behind the so-called kinked demand curve there is a situation of information incompleteness and endogeneity of tastes. Why prices are sticky Now let me make a step backwards to get rid of a simplification. I initially defined a notional demand curve by assuming perfect knowledge, maximising behaviour and exogeneity of tastes. The demand curve was derived from a representative individual’s utility curve. Thus it was possible to determine the monopoly price on the basis of the notional demand curve. In an oligopolistic competitive market the situation is different. First, since there is competition among several producers, each one’s degree of exit freedom and monopoly power is reduced with respect to a monopoly situation, and it is difficult to compel the buyers to trade precisely at their reservation prices. Each firm perceives a demand curve which is lower than the notional one, and the latter becomes irrelevant. Second, since there is imperfect information, demand estimates are always impaired by uncertainty. Thus each effective demand curve is located inside the price discretion area but firms do not know exactly where. Third, each firm knows that its demand curve is not independent of those of other firms, and this, while contributing to increase uncertainty, at the same time, however, facilitates the task of price fixing. In fact information on the ability to raise prices can be acquired, albeit imperfectly, from observing the competitors’ price changes. Price decisions are easy in an oligopoly where there is price-leadership. When there is no price leader, information is conveyed by the prices of the closest competitor (closest, particularly in terms of product substitutability). In general and in theory, the price structure remains undetermined. The area of indeterminacy can be reduced by specifying the ‘industry’ organisation in terms of the number and size of competitors, the normal profit rates of different firms, the rates of interest etc., but from a theoretical point of view these specifications should be considered ad hoc hypotheses. Thus it is better to accept indeterminacy as a general characteristic of oligopoly. But one should not feel too disheartened, for the real theoretical problem is not so much that of determining the level of prices, as that of explaining why, once fixed, prices tend to be stable. A demand curve is perceived by a firm as more elastic the less a reaction to price decisions is expected from competitors. The azc curve in Figure 5.2 is highly elastic, presupposing that, when a price is changed, competitors do not follow suit. The inelastic bzd curve, presupposes a prompt reaction on the part of competitors. It is to be expected that a price cut along the azc curve will not be imitated by competitors and will therefore increase revenues, whilst a price increase

Production governance structures 221 p b

a z

c

d 0

q

Figure 5.2

along the same curve will reduce revenues. Conversely, a decrease or an increase in prices along the inelastic curve will produce opposite effects: reductions in revenue following price cuts, because it is expected that competitors will lower their prices too; increases in revenue following price increases, because competitors are expected to augment their prices. To say that the demand curve is badly specified, as I have done above, does not just mean maintaining that it is located somewhere inside the price discretion area, as azc and bzd are presumed to be. It also means positing that the effective demand curve is located somewhere between the azc and bzd curves in Figure 5.2. The theory of the kinked demand curve alleges that the real demand curve is perceived by a firm as close to azd. This is the result of a combination of hypotheses deriving from a pessimistic view of human nature. A firm surmises that a decision to cut prices will be interpreted by competitors as an act of hostility and that a hostile reaction is to be expected. Conversely, a decision to augment prices is expected to be perceived as a sign of weakness or stupidity so that competitors will try to take advantage from it. Thus, the best policy is not to change prices. Looking at the problem from the point of view of information collection, it can be observed that, not only do competitors’ prices convey information for a firm’s price fixing, but they also provide strong incentives not to modify that information. Oligopolistic competition stabilises information. Changes in market structure are not normally brought about by price changes – although these may reflect them – but by other causes, as I will show below. Thus there is another reason why prices are stable. If a given price structure supplies information, a change in prices destroys it. A firm that cuts its prices confuses its competitors and creates conditions for destabilising the market structure. If it decides to raise them, it confuses its customers. In both cases the destroyed information has to be reconstructed, and this costs money. The theory of small menu costs highlights the problem. Firms that decide to change their prices must alter their price-lists and bear the expense

222 Production governance structures of informing customers and convincing them that the changes are justified. They have to tackle the dilemma of choosing between losing customers because of a bad information offer and losing profits because of menu costs. Often this kind of cost is not so ‘small’. If the increase in costs for menu changing is higher than the expected revenue from the price changes, no change will be made – a situation that normally occurs in firms facing a kinked demand curve, for which the expected revenue from price changes is quite low. It also occurs frequently in firms that produce a wide range of products, for menu costs are quite high in these cases. A firm may have good reason to change prices when a change in customers’ tastes leads to an increase in demand. But tastes are influenced by the firm itself. Thus, apart from the case of generalised cost increases, it is only when innovations or pseudo-innovations are introduced that it is worth changing prices. Apart from innovation proper, firms compete mostly through marketing, packaging, advertising, by trying to attract customers with ostensible promises. In these cases it may be worth adopting aggressive price policies, trusting in the fact that competitors cannot reduce their prices unless they change their product image. Now consider price changes in relation to cost changes. If the latter are idiosyncratic, there is no inducement to adjust prices. For instance, an increase in wages caused by a company strike will almost certainly be considered by managers as the cause of a profit squeeze, whether prices are changed or left unaltered, for the demand curve is perceived as highly elastic to price increases. It is preferable to avoid losing market shares and wait for better times. But a cost reduction caused by a process innovation may induce price cuts, although not necessarily so. It depends particularly on the speed of imitation expected from competitors. If this is not considered very high, an aggressive price policy might be tempting. Often, however, managers decide to use the innovation to swell profits, possibly in anticipation of using the higher cash flow to buy out a competitor firm or to enter a new market. Finally, if an increase in costs is perceived as general, as occurs after a collective wage agreement or an oil shock, then every company management knows that all its competitors are exposed to the same pressure and can legitimately expect that an increase in price will not be interpreted as an irresponsible action. In these cases price increases tend to be widespread. Thus it is understandable that inflationary pressures are stronger in boom phases, notwithstanding the great supply elasticity of industrial goods. It is not the increase in aggregate demand that inflates prices, but rather the increase in costs, especially labour costs, caused by the growth in employment. The above considerations would appear to be open to criticism of redundancy: if the supply curve is horizontal, what need is there to assume a kinked demand curve or the bite of small menu costs to justify price stickiness? The real problem is: how is a horizontal supply curve justified? Neither a full cost and normal pricing theory nor a constant prime costs assumption provide a satisfactory justification. Post-Keynesians often argue that firms build reserve excess capacity aiming at earning normal profits with normal utilisation and that they fix prices by using a mark up rule. But why should companies not take advantage of an increase in demand by raising prices during boom phases?

Production governance structures 223 p, C

C

a

0

b

S

q

Figure 5.3

In reality it is the oligopolistic structure of markets that accounts for price stickiness and therefore justifies a horizontal supply curve. To understand why supply curves tend to have infinite elasticity it is necessary to specify the kinked demand curve theory in dynamic terms. This is done in Figure 5.3, where I represent a firm’s reaction to a demand shift from point a to point b. Line S is the ensuing supply curve. Starting from point a, the firm that observes a demand increase has to cope with the dilemma whether it is more profitable to raise the price or the quantity. If the demand increase is idiosyncratic, for instance because it was brought about by effective marketing strategy, it is better not to imperil the outcomes of the product policy with a price increase, for one can reasonably expect competitors to profit from such an increase by recovering market shares. Conversely, should a firm observe a shift in demand from b to a and realise it was caused by its competitors’ aggressive product policy, it would not react by cutting prices. It would prefer to avoid triggering a price war in a period of internal weakness, a period, that is, in which it undergoes effective market aggression by competitors. Firms perceive their demand curves as kinked precisely when they have to take decisions occasioned by changes in the conditions of market competition. And this is why their supply curves remain flat when demand changes in time. Labour exploitation in the market Recent neoclassical and neo-institutional literature on ‘make or buy’ is based on the fanciful vision of a market economy in which an invisible hand ensures efficient allocation of resources, except in some particular cases of failure that require solutions other than market competition. Then the problem of the choice between making or buying goods is tackled in terms of the reasons why social actors are induced to forgo the market and resort to hierarchical organisations like the firm or the state: when is it necessary to give up buying and choose making?

224 Production governance structures This view is the fruit of an ideology that tries to disguise capitalism as a Walrasian Utopia. But if you really want to focus analysis on a capitalist economy, rather than on a generic ‘market economy’, you must begin by presupposing that organisation is fundamental and primitive. Thus the correct approach to the problem becomes: when is it necessary to give up making and resort to buying? If it is true that the capitalist firm and the employment contract are required to extract surplus value from production, why should a capitalist renounce these institutions and make use of exchange rather than of authority relations? And is it possible that labour is capitalistically exploited in the exchange process instead of through the productive process? In the rest of this section I tackle the latter problem. But the conclusions I reach do not only apply to the employer-employee relationship. They can easily be extended to various forms of market relation between capitalist firms, including many cases of contracting, subcontracting, sales contracts and contracts for services, by means of which a strong actor takes advantage of market power to extract a surplus from a weak actor’s activity. The theorem of maximisation of total value, developed in neo-institutional economics, endeavours to demonstrate the following miracle: if a transaction between two or more actors is likely to generate a surplus, then it will be stipulated regardless of the way the surplus is distributed. The actors will freely choose to enter into it because all of them may get an advantage from it. The theorem depends on the crucial hypothesis that there are no wealth effects. Another hypothesis, which is equally crucial but rarely made explicit, is that there is perfect liberty for all the actors. The no wealth effects hypothesis states: (1) given a couple of alternative transactions, Y0 and YT , there is a sum sufficient to compensate each actor for passing from one to the other; (2) this sum is independent of the actors’ wealth; (3) each actor has sufficient wealth to pay for the compensation. The perfect liberty hypothesis states: (1) there are non objective impediments to an actor’s freedom of choice; (2) no actor is endowed with power that enables him to condition the choices of others. Now, consider the case of a firm that has to decide whether it is worth firing a worker and then establishing an exchange relation with him in which he will supply the output in the form of a commodity. The Y0 transaction is based on an employment relation whereby worker L produces the output under the command of employer B by performing a labour activity in the company controlled by the latter. With transaction YT , L performs labour activity as a self-employed worker and then sells his produce to B. In the Y0 situation the employer bears control costs, or more generally organisation costs, which reduce the net surplus he earns. In situation YT there are no control costs because, by assumption, the self-employed worker controls himself without sacrifice. The worker’s effort is assumed identical in both situations, as is the output, gross of control costs. This and other interesting cases are illustrated in Figure 5.4, where Yl is the worker’s income, Yb that of the employer. Total income, net of control costs, is Yn = 1,000 dollars, while Yg = 1,400 dollars is gross income. Therefore Yg – Yn = 400 dollars is the cost of control. Line Yn Yn defines all the possible income distributions between B and L. Vector Y0

Production governance structures 225 Yb

Yg YTS YTB

Yn

Y0

YTL

YTD 0

Ys

Yn

Yg

Yl

Figure 5.4

= (500, 500) represents a given distribution in the capitalist firm, assuming that net value added is divided fifty-fifty between profits and wages. Now suppose B decides to fire L. Then he rents him the means of production at a price of 900 dollars, and buys from him, at a price of 1,400 dollars, the output L produces as a self-employed worker. Subsequently B can resell the output at the market price of 1,400 dollars. We pass from transaction Y0 to transaction YTB. In the new situation B collects 900 dollars, that is, he earns the surplus generated by the elimination of control costs. L can freely agree to enter this new transaction because he does not lose out on it. In fact he earns 1,400 – 900 = 500 dollars. Alternatively, L could resign from his job in B’s firm, hire the means of production at a rent of 500 dollars, and then sell on the market, at 1,400 dollars, the output he produces autonomously. In this new situation, YTL, all surplus generated by the elimination of control costs is secured by L, and B would have no reasons to object. All allocations between YTL and YTB are optimal, while those included in the Y0YTLYTB area are however better than Y0. The allocations outside this area are not feasible so long as all the actors preserve full freedom, for none of them would autonomously agree to enter a transaction in which he would be worse off. What happens if some of the basic hypotheses are removed? Suppose that, when L proposes to B to pass to transaction YTL, in which he would hire the means of production at 500 dollars, the employer refuses. And suppose the latter makes a counter-proposal: to sell L his means of production instead of renting them. But suppose that B is not wealthy enough to buy the capital assets and has no access to credit facilities. Therefore he will refuse B’s proposal. At this point B could propose

226 Production governance structures to L to rent him the means of production at a price of 900 dollars, and L will have no reason to refuse it, for he has nothing to lose in situation YTB. In this example I have removed the third condition of the hypothesis of no wealth effects: L has insufficient wealth to accept B’s first proposal. I have, however, maintained the hypothesis that B has sufficient wealth to refuse L’s proposal. This is a new assumption. Let me call it the ‘asymmetric wealth effect hypothesis’. It states: not all the actors in a transaction have sufficient wealth to be able to induce the other party to accept a preferred allocation. Its theoretical implication is significant. A transaction can be undertaken not only and not necessarily because it maximises total value, but also and especially because it modifies income distribution in favour of the strong party. The weak party might draw some advantage from it, but the rate of exploitation might increase, as is the case, for instance, in all the transactions located to the left of the bisecting line in Figure 5.4. But things could be worse. Suppose B proposes to L a transaction similar to YTB, but he fixes the rent for the use of means of production at more than 900 dollars. This solution would be located to the left of YTB. L would not prefer it to Y0, because he would earn less from it than if he were an employed worker. However B could threaten to dismiss him anyway if he did not accept the proposal. If there is unemployment and if L has insufficient wealth to survive when unemployed, he would be compelled to accept the over-exploitation imposed by B. He would no longer make a really free decision. The command prerogative renounced by the employer when he rescinded the employment contract, would be more than compensated by the influence that an asymmetric distribution of wealth enables him to exert on exchange conditions. A self-employed worker would end up being exploited more than a wage worker. Note that modification of the no wealth effect hypothesis, and its substitution with that of asymmetric wealth effects, has also led to a change in the perfect liberty hypothesis. Now freedom of choice is also distributed asymmetrically. This is because wealth, as shown in Chapter 4, is a source of power. Its asymmetric distribution causes some social actors to be endowed with limited and others with more extensive exit freedom. Wealth is one of the conditions for freedom of choice. The worker who is unable to reject his employer’s indecent proposals, due to his incapacity to live on unearned income when he is unemployed, is not completely free; and certainly not as free as the employer who has sufficient wealth to reject the worker’s proposals.4 The fact that misdistribution of wealth is a condition of misdistribution of freedom perhaps explains why the perfect liberty hypothesis is rarely made explicit by those who assume no wealth effects, but does not justify it. Freedom is limited by other conditions besides poverty. I shall not list them all here, but recall just a few. One has already been mentioned, and that is the existence of mass unemployment.5 The more extensive is mass unemployment, or the more rapidly it grows, the more effective is the capitalists’ blackmail of workers. Thus the ‘industrial reserve army’ serves as an instrument that limits freedom of choice. Another condition pertains to the normative institutions that limit the workers’ self-organisation capacity. The anti-combination laws that regulated the English

Production governance structures 227 labour ‘market’ during the industrial revolution by making it more ‘competitive’ are a good example of normative institutions aimed at reinforcing the asymmetry in the distribution of freedom. In fact they forbade the workers’ coalitions, not the capitalists’. Another condition is created by class-biased credit rationing. If the workers have more limited and costly access to finance than the capitalists, this is tantamount to having less wealth than them. A particular condition of freedom limitation is often generated by the specificity of the workers’ ‘human capital’. Many studies on the ‘hold up’ problem have brought to light the risk of inefficiency linked to asset specificity, and have suggested that an expedient to cope with this problem is to assign ownership of complementary capital goods to the actors who made specific investments. Only in this way – goes the story – can these actors be induced to invest without the fear of seeing their capital values impaired by the other parties’ opportunism. But things rarely work this way in the real capitalist world. Many firms which use advanced technologies also use workers endowed with specific capacities without transforming them into the firm’s owners. A different solution is more frequently adopted in these cases. Through apprenticeship and other kinds of atypical employment contracts, a firm partakes of the worker’s investment in ‘human capital’. Then, once labour capacities have been formed, the firm may offer the worker a normal employment contract. But asset specificity limits the exit freedom of the worker, who would see his capital devalued in case of dismissal. Who could prevent the firm from using this freedom asymmetry opportunistically? A commonly exploited practice, in these cases, and especially in present day capitalism, is to fire the worker after completion of apprenticeship and then set up an exchange relation with him whereby he sells the firm the products or services produced in conditions of formal self-employment. More generally one could say that, in a capitalist system, most of the norms regulating labour uses, labour ‘markets’, labour organisations, labour ideologies and finance markets, among other things, tend to produce limitations in the workers’ freedom of choice. The conditions of freedom asymmetry depend much more on the systemic effects of the power structure than on the individual endowments of the capitalists who take advantage of them. But ‘most’ does not mean ‘all’. Institution and culture production apparatuses are a class battlefield. Organised workers also fight to check and limit the systemic effects of the capitalist power structure, and not just to conquer improvements in the immediate conditions of production and distribution. Whenever they obtain laws that reduce labour time and improve their access to health, education and the political arena, for instance, they succeed in widening their freedom field by squeezing that of the capitalists. Which brings me to the question: if the employment relationship is a fundamental condition for capitalist exploitation in production, what are the reasons that sometimes induce the capitalists to prefer buying instead of making? So far I have argued by reducing these conditions to control costs and by assuming that labour productivity does not vary in the transition from capitalist control to selfcontrol. In reality, organised production generates various co-operative benefits,

228 Production governance structures for instance in the form of scale, scope, space and time economies, which market exchange is unable to produce. Labour productivity in a factory may, however, be impaired by the workers’ action, especially when this is protected by labour laws. Thus, the first general conclusion is: a capitalist’s preference for buying instead of making arises in situations where the workers’ counter-power is so extended that control costs tend to outweigh the benefits of organised co-operation. There are then two reasons for that preference, which may occur together or separately. The first is political in nature, and consists in the existence of a strong worker counter-power. This accounts for the tendency to decentralisation and ‘marketisation’ of labour that often takes place in periods following important victories of social movements, as it happened in the capitalist world in the 1980s, after the advance of worker movements in the 1970s. The second is technological, and concerns the existence of labour processes that can be decentralised without substantial loss in labour productivity. This explains why recourse to market rather than organised exploitation is more widespread in two kinds of activity, besides the cases of asset specificity already referred to: in those labour services where no great inventive dynamism is required; and in those whose performances are sensitive to creative effort but do not require strong complementarity of ‘human capital’. An example of the former case is seen in the use of contracting and subcontracting for decentralising standardised and routinised labour services. Just think of the worker co-operatives that work hard as cleaning contractors for big organisations. The co-operative workers do the same things they would do if they were employees of the big organisation, but they earn profits instead of wages. They do not have to put up with the bosses’ harassment and can even consider themselves capitalists. Yet their profits are often lower than the wages they would earn under a collectively agreed employment contract, while the safety, social security and labour conditions are worse. The kind of transaction prevailing in these cases can be traced back to the contract for services model. Similar to this is the case of decentralisation and ‘marketisation’ of components and semi-finished goods. Many big firms buy on order from small suppliers who may be either co-operatives or capitalist firms, but who remain tied to the former in conditions of more or less informal subordination. The transaction model used to mobilise and exploit labour in these cases is the sale contract. An example of the other case is given by some forms of agreement used to exploit the services of consultants. These are not hired as employees and preserve a condition of professional autonomy. Just think of the appointments assigned to lawyers, designers, advertising agents etc. The services produced by these experts are too sensitive to creative ability for them to be effectively motivated and controlled within a power hierarchy. Their production does not require big investments in team production structures and complementary capital assets. The transaction model referred to in these cases is the agency relation. The principal assigns goals to the agent, controls his performance and fixes ex ante his remuneration pattern, but leaves him substantial freedom to choose the best means to achieve the prefixed ends.

Production governance structures 229 Exploitation of labour in the market takes place through price fixing. When an employee is transformed into a self-employed worker by an employer’s autonomous decision, the new transaction normally involves a weak and a strong party. The reservation price of the weak party, the worker, is determined by alternative employment opportunities outside the relationship. If there is unemployment, difficult access to credit, asymmetric wealth distribution and so on, the alternative opportunities are not very good. Therefore the workers’ reservation prices are quite low. If, moreover, the weak party has made specific investments in personal skills, his reservation price discounts the loss of asset value in alternative jobs where specific capital is not required. Since the strong party has good information on the weak party’s abilities (for the latter was his employee), and since he has a higher exit freedom, he can fix a price near the worker’s reservation threshold. He secures most of the rents produced by the latter, who thus turns out to be overexploited. Labour exploitation in the market is made possible by the oligopsonistic power and the ensuing ability of the capitalists to fix prices near the workers’ reservation prices. The institutions governing market relations, contrary to the claims of some orthodox Marxists, are not intrinsically capitalistic. Nor is exploitation in the sphere of exchanges intrinsically capitalistic. The market existed in many pre-capitalist societies and in all probability will survive in post-capitalist society. Thus, it is even possible to conceive socialist exploitation of labour through exchange, as would occur, for instance, when a democratically managed co-operative firm uses market power to exert influence on another co-operative firm. Market institutions are not typical of capitalism. The notion of a ‘mercantile economy’ elaborated in Marxian theory is rather abstract and general. It defines a set of possible systems in which social and economic relations are established in the market. Capitalism is a subset of a ‘mercantile economy’ – a subset that preserves market transaction institutions and transforms them into institutions for sustaining capitalist exploitation and accumulation. This is why it is legitimate to speak of capitalist subsumption of the market or capitalist exploitation in the market, but not of an intrinsically capitalist nature of the market. Value as a social relation Since in oligopolistic competition prices tend to be rigid, there are valid counterarguments to those who claim that prices furnish good information for efficient allocation of resources. Obviously all theories of Pareto efficiency, market clearing and general equilibrium are thoroughly impaired, as is to be expected in the light of the hypotheses of non-perfect competition, non-tâtonnement and non complete information. But also weakened are those theories which appeal to Hayek’s argument that prices contribute to reveal dispersed information on individual tastes and natural scarcity and that the market process pushes an economy toward a spontaneous ‘order’, whatever this means. The information dispersed in the consumers’ minds is put there by companies and other strong social forces, not by nature, and nobody has yet proved that the

230 Production governance structures firms’ production and selling decisions eventually turn out to be efficient or wellordered from a social point of view. Prices do not help the consumers to choose their consumption patterns efficiently. Moreover, prices are too sticky to reflect changes in natural scarcity and technical conditions of production with any precision and promptness. So, we know pretty well what prices are not: they are not indices of scarcity and instruments of efficient allocation. What do we know of what they are? First, as far as horizontal competition is concerned, that is, competition among sellers, prices are observed by each firm to collect information on its competitors’ market strategies. On the ground of this information firms decide their own strategies. Since prices are sticky in oligopolistic competition, one could say that they are used to stabilise the power structure in each industry. Prices determined in an oligopolistic context are, first and foremost, indices of market power and reflect the relative profitability of firms more than the conditions of natural scarcity. Production co-ordination in the market is ensured by the governance of power relations among producers. It is an ongoing process through which each actor disciplines the others by trying to overwhelm them. Second, in vertical competition, that is, in the income distribution struggle between sellers and buyers, prices are used to transform power into exploitation. In this context they are class inequalityproducing instruments. Their stickiness serves to stabilise social relations and to blunt the harshness of economic conflict. Thus rigid prices also play the role of inequality-preserving devices. This twofold role of prices, their action in producing and preserving inequality, is most apparent and effective in the labour ‘market’ and in the market for labour services, where one can always see it at work in changing and stabilising the power relations between social classes. Wages and the prices of labour services change when class conflict heightens and are sticky during social truces. More generally one could say that prices are the signs through which transaction institutions materialise. And since transaction institutions serve to produce, modify and stabilise social relations, one cannot but agree with Marx when he says that value is a social relation. But in a capitalist economy values reflect more the social division of power than the ‘social division of labour’.

The governance of knowledge production Epistemic uncertainty and informational complexity Information and knowledge are concepts economists tend to use as synonyms. But they are not. Information is about signs, knowledge about theories. A piece of information is a sign referring to an aspect of an object, which can be either an event or a state of the world. An information set concerning an object is a collection of signs. But signs have meanings which are defined by the subject who interprets them. Therefore the pieces of information and information sets are never purely objective tokens: they are subjective interpretations of data. What is the ancient desk I have under my arms in this moment? If I look at it from the point of view of its utility to me, it is just a desk, and to define it in this

Production governance structures 231 way I have to collect information on its capacity to satisfy my writing needs, the width of the plane, the number of drawers etc. But a carpenter would look at it from another angle. To him it is a wooden table, and the information he needs to define it concerns the quality of wood, the dimension and form of the pieces that compound it, the kind of joints uniting them etc. An antiquarian would look at it from still another point of view. He would need information on the epoch of its making, the style, the degree of wear etc. Interpretation requires a theory, that is, an organisation of information pieces into an image of the object, a mental simulacrum capable of representing it in its entirety. The image gives knowledge of the object. But note that an image is created, not found. Information is collected, knowledge is constructed. The two processes, of course, information collection and knowledge construction, are not independent of each other, for you cannot build a reliable theory of an object without information on it, whilst you cannot collect, select and interpret information without a theory. Theories are consistent, simplified and incomplete models of reality. They are models, i.e. artifacts working not just as descriptions of objects, but as idealised representations of them. They are incomplete models, in that normally not all necessary information about the objects is available. They are simplified models, in that not all available information is taken into account, theories being constructed through a selection of the available information, by skipping the information pieces considered irrelevant and elaborating the relevant ones. Finally they are consistent models, because information is organised in the form of a logically coherent scheme. The construction of knowledge raises two kinds of problems, that of uncertainty and that of complexity. In the presence of uncertainty and complexity social actors react in two ways, depending on whether they are dominated by conservative or innovative moods. In the former case they let themselves be guided by institutions and habitual behaviour, thus giving up any opportunity offered by the possibility of change. In production, this attitude is manifested by compliance to the routines and techniques in use. When innovative moods prevail social actors try to take advantage of the possibility of change, which they can do by producing new knowledge. The construction of knowledge is like the quest for the Grail in a unknown land. The knight does not know the exact nature of the object of his aspirations, where the object is hidden and where to start from to begin his quest. He must mark out a route before leaving, but has no reliable information to do so. He must start out on the venture anyway, risk leaving from the wrong base and heading for the wrong destination. He must expose himself to the disappointment of eventually finding an object that does not come up to his expectations. This is the problem of epistemic uncertainty: lack of knowledge ex ante – ex ante, because an image on future events has to be formed before taking a decision. Epistemic uncertainty is linked to the essentially hermeneutic character of knowledge. Different knights will set out the quest by leaving from different places, that is, they will adopt different points of view. They will have different ideas on the nature of the object they are looking for and the place where they might ‘find’ it. Therefore they will mark out different routes, that is, they will formulate altern-

232 Production governance structures ative hypotheses depending on different intentions, convictions and knowledge interests of an essentially subjective nature. Knowledge has a constitutionally interpretative character, and this is the basic reason why uncertainty is structural quite independently of the randomness of the objective word changes. It is not possible to verify the truth of a theory objectively if it is impossible to collect information by abstracting from subjectivity. But knowledge also has a constitutionally constructive character, for knowing a new object implies producing it – producing it, first as a simulacrum, and then, possibly, as a real object. A new good is by definition a thing which was unknown and not existent beforehand. Inventing it is an act of knowledge and creation. This is why genuine uncertainty is always radical: it pertains to a world not only not yet known, but just not yet existing. Once the knight has set out, most information has to be collected during the quest, for new problems will continually arise in the search process. And each time a new problem is worked out, it may become necessary to redesign the route. The knight will have to cross deserts and mountains, avoid getting lost in woods and oceans, weigh up the advice of wise masters and false prophets, in other words, permanently play a problem-solving activity in which the data and the variables incessantly change. And problems of collection, assessment and selection of information continually have to be tackled. Alas, all has to be done with the help of inexorably limited reason. This is the problem of informational complexity: excess of information ex post – ex post, because most information is made available after a decision has been taken, i.e. after the construction of a particular theory has started. If uncertainty is tackled with the formulation of hypotheses and assumptions on unknown aspects of the object, complexity is coped with through the selection and elaboration of existing information. Many pieces of information, both hypothetical and real, are then organised into a theory of the object with the aim of reconstructing an organic vision of it. Market pluralism and the evolution of knowledge Any firm competing in an oligopolistic market is continually exposed to the risk of losing market shares because of its competitors’ innovations. This kind of uncertainty is always present, and is the stronger the more intense is competition, the higher is commodity substitutability, the more dynamic is research activity and scientific development in the industry and in society. A firm must always cope with this form of systemic uncertainty – systemic, because it pertains to states of the world that do not depend on the firm’s decisions. Besides this, there is an idiosyncratic uncertainty which is generated by the firm’s managers, especially in connection with invention activity.6 Uncertainty generated by the introduction of a new product or a new process reflects the difficulty in producing the expected outcomes – an increase in sales, a reduction in costs, a growth in market shares, in profits etc. Thus idiosyncratic uncertainty is

Production governance structures 233 better defined in negative terms, as the possibility of invention decisions producing losses instead of profits, regression and failure instead of growth. It is interesting to note that the two kinds of uncertainty, systemic and idiosyncratic, are in a complex relation with each other. In fact a firm’s introduction of new goods and processes is stimulated by competitive pressure, and is decided in anticipation of possible innovative decisions by competitors. This is why many economists interpret knowledge production as an instrument for coping with uncertainty. The fact is that those who do not innovate, do not reduce the risk of failure, for new goods and processes can be introduced by competitors. Thus, the higher the idiosyncratic risk linked to decisions to innovate, the lower the systemic risk expected from the competitors’ innovative decisions. The innovator entrepreneur tackles the latter kind of uncertainty by trying to anticipate the market. Therefore when calculating the overall risk of an innovative decision, he should take into account the risk inherent in not deciding, and measure the overall risk as the difference between the two. A sensible conclusion is that the overall risk of an innovative investment might be rather low! This is the idea to which many entrepreneurs refer when they say that the introduction of a certain innovation was necessary – necessary for the survival of the firm, necessary for reducing the risk of failure. After all, this idea matches the common view that the production of new knowledge, insofar as it is knowledge of the world, serves to reduce human ignorance and, to the extent that it contributes to change the world, serves to augment human mastery of it. Through competition the market generates the systemic uncertainty that stimulates innovative decisions. This is why the market is an effective institutional apparatus for enlivening the evolution of knowledge. And here the notion of ‘market’ is intended in a broad sense, to the point of using it also to embrace network institutions working as forums. It includes, for instance, markets for ideas ‘exchanged’ in academic environments. The vagueness of this definition will be removed in another section. But the market also performs other important functions in the process of knowledge evolution. When I referred to epistemic uncertainty in the previous section, I argued that an ‘objective’ verification of a theory is impossible because of the hermeneutic character of human knowledge. A theory is always based on conjectures and, until it is put into practice, it remains a conjectural theory. What is impossible is to prove that a theory is ‘objectively’ true. But this does not mean that all theories are equally valid, nor that all can resist with the same effectiveness to the proof of facts. Another aspect of the epistemological conception mentioned in the previous section concerns the constructive character of theories. To start with, theories furnish knowledge by conceptually re-constructing the object; they then manifest their practical effectiveness by contributing to re-construct the world. The verification of a theory, according to this view, can only be pragmatic. A theory ceases to be just conjectural when it is put into practice, that is, when it becomes a substantiated

234 Production governance structures theory, when it is verified. It is realisation that verifies the theory, that makes it real and true.7 Not all conjectural theories succeed in becoming substantiated. Some fail, often together with the firms which launched and tested them. And it is the market that decrees the success or the failure of a theory. Market competition plays an essential role in the struggle for survival of ideas: it performs a selection function, allowing realisation only of the most valid conjectures. But ‘valid’ in which sense? To answer this question it might be useful to specify the institutional and social context of knowledge evolution, which I will do by returning to the main subject of this part of the chapter: capitalist competition through knowledge production. Then it will be easy to understand that the success and the validity of new ideas are not determined by exogenous consumer preferences. Individual preferences, in fact, are never so exogenous with respect to economic progress in general and the productive process in particular. The firms who produce new ideas, precisely because these are novelties, also have to produce a need for them. There is no doubt that the construction of a new product invariably implies the construction of its image and of a desire for it in the consumer’s mind. Even from a purely commercial viewpoint verification-realisation of an invention is never purely ‘objective’, for it never depends solely on social and economic processes which are external to the subjective choice of the firm who launches it. To some extent the forecasts sustaining innovative choices are self-realising expectations. Then, what are the selection criteria regulating the competitive process in the evolution of knowledge? In a capitalist economy, that is, an economy in which survival depends on accumulation, the prevailing selection criterion is growth. Only those ideas which make a better contribution to the growth of firms catch on, and independently of their capacity to satisfy ‘objective’ and ‘genuine’ human needs. The real and fundamental need satisfied by innovations is the need of accumulation. The market rewards with growth the ideas that have been better elaborated with a view to capital accumulation and punishes with regression and failure ideas which do not fuel accumulation. The apparently tautological nature of this proposition should not give cause for surprise in the light of the partially self-realising nature of expectations upholding innovative investments. Human evolution does not just take place through selection of right decisions and elimination of mistaken choices. It also works through learning and imitation. Successful ideas establish themselves when the prizes awarded by the market encourage people to insist on them. At the same time the punishment of unsuccessful ideas induces people, at least those who strive to survive, to modify their choices and decisions. Moreover the diffusion of good ideas is sustained by the attempts of latecomers to imitate innovators. The market plays its disciplinary function through rewards and punishments, thus inducing people to consolidate the right choices and to abandon the wrong ones. Finally the market plays a financial function in the process of knowledge evolution. The problem here is that the realisation of a new idea requires the investment of financial resources which are often quite large. Where do these

Production governance structures 235 resources come from? Well, it is precisely through the credit and capital markets that firms who want to pass from theory to practice obtain finance. Most important is finance that accrues ex post. Successful innovations bring about profits and cash, together with accumulation. Success manifests itself in the ability to repay debts. Ex post finance is also important because it enables existing firms to selffinance further development. Moreover, since self-finance and re-payment of debt are judged by the market as a sign of success, they also furnish inducement for the accrual of other finance. This inducement arises both because the success of a firm’s innovation reduces the perception of idiosyncratic risk attached to any attempt to launch new ideas and because an abundant cash flow and a high market valuation enable a successful firm to advance collaterals for new debt. Thus a sort of virtuous circle is set in motion that links the success of an idea to the possibility of launching new ones, which is how the market plays its financial role in the process of knowledge evolution. There is, however, a more general reason why a network type of governance structure, is conducive to the evolution of knowledge. To understand this let me recall the Grail quest metaphor. Why is it better, from the social point of view, that many knights, instead of just one, set out on the quest? One might think the answer is that you cannot trust in the ability of each individual to mark out the right route, so that the multiplication of routes increases the probability of hitting on the good one. But the problem is that the Grail, the Good, does not exist. It is not an existing, though concealed, object that has to be found out. Since every knight has his own idea of what the Good is, many of these can be found or, rather, created. The greater the number of knights who set out on the quest, the greater the number of beliefs, intentions, hypotheses and theories that are tested, and the higher the number of new Goods produced. Now, to get away from the metaphor, just substitute that capital letter with a small one. In the economic field, theoretical pluralism is fostered by freedom of enterprise and the market process. Firms compete with each other by searching for the goods each in its own way. The higher the number of firms, the higher the probability of producing many new goods. Competition fosters the evolution of knowledge not because it selects the best good among a given potential variety, but because it widens the variety. The expression ‘growth of knowledge’, which is often used by historians of science, is deceptive. There is no such thing as the knowledge, which grows by simple accretion, for there is no such thing as the best idea. Since there are many researchers, inventors, entrepreneurs, with many different ideas, conjectures and interests, who use many different methods, procedures and information sets, there are always many different ideas in any one field, and many of them can get validation on the ‘market’. The right expression therefore would be ‘the evolution of knowledges’ if the plural of the word were possible. The evolution of knowledge brings growth of human riches. Through continual innovation, it fosters an increase in the quantity of goods and their multiplication in number. But it does more than that. It promotes the multiplication of human aspirations with the capacity to satisfy them, at least potentially. And it favours a proliferation of ideas and values, as well as an increase in the heterogeneity of

236 Production governance structures people, which can only enrich humankind. But let the above be said without attaching any positive value to the word ‘riches’. I state this not because it has yet to be proved that the growth of such wealth leads to an increase in human happiness, but because there is no proof that the direction in which people build their spiritual enrichment is always chosen by themselves. Firms, hierarchies and the construction of knowledge A theory is a conceptual artifact which organises information into a hierarchical structure. Various pieces of information are transformed into concepts and these are linked to each other so as to form an organic whole. In this way a unitary vision of the object is constructed. Some concepts, normally the most abstract and general, are considered fundamental and constitute the theory core. Then secondary, more specific and less abstract, concepts are organised around the core. They enrich and articulate the model and make it more concrete. Then other, still less abstract and more specific, concepts are organised around the secondary concepts, and so on and so forth. A firm does not just need to know the nature of an object. It also needs to know how to produce it. This implies that any conceptual construct must be accompanied by practical instructions defining the way to apply it. Thus, on the way to be substantiated, a conjectural theory may become quite complex indeed, spanning from a few highly general and abstract concepts down to a wide variety of particular and practical dispositions. In the presence of complex productive problems, no single human mind can master all the information required to work them out, define the relevant and effective concepts and organise them in an organic whole. This is where organisation comes into it. Since a theory is a hierarchical conceptual structure, the organisation aiming at producing knowledge tends itself to take on a hierarchical structure. To understand how this structure works, it might be advantageous to distinguish between three kinds of flow in the knowledge-production process, top-down, bottom-up and horizontal. These flows are strictly linked to each other so that in practice it is impossible to separate them. But, for the sake of clarity, it might be useful to reconstruct them separately. The top-down flow is started by a top manager, who takes strategic decisions and poses the problem of their realisation. To do this he elaborates a plan which takes the form of a theory core of a complex model. The theory core is a partial solution to a general problem. On its basis the problem is structured, i.e. it is broken down into residual under-problems, and into application instructions. Underproblems and instructions are transmitted to the various branches in the lower layer of the hierarchy. Then each second layer branch takes up its under-problem, elaborates a partial solution to suit its tasks, structures the unresolved parts in the form of other under-problems, and transmits these to the third hierarchical layer together with instructions for application. The process is then repeated by the third layer and goes on gradually involving the whole organisation. When it reaches the base of the pyramid, a worker solves the final problem of the right number of turns to give a screw, and no residual problems remain, in principle.

Production governance structures 237 Summing up, each branch of each hierarchical level undertakes a knowledgeproducing activity consisting in: (1) finding a partial solution to a problem transmitted from the higher layer, (2) structuring residual under-problems, (3) transmitting these to the lower level branches together with application instructions. The top management has the task of defining the general problem with a strategic decision. The lowest hierarchical level leaves no residual problems to be solved. Now consider the bottom-up flow. Problem-solving activity presupposes the collection and elaboration of information. Since individual ability in mastering information is bounded, relevant information must be selected and reduced in quantity. Moreover it is necessary for each actor to transmit information to his superiors. In other words a process of selection, elaboration and transmission of information is set in motion which goes from the bottom to the top of the organisation. Each individual uses his information to work out his problem. He elaborates it by transforming it into a theory. With this elaboration, various pieces of information are chunked into a model which, in the form of an informative block, gives rise to new information. This, in turn, is transmitted to the higher hierarchical layers as a single piece of information. Each office at the higher level receives various blocks of chunked information from its various under-branches. It selects and elaborates them, and uses them to work out its problem, which it does by building a theory. In this way information is again chunked to give rise to a new information block which will be transmitted to the higher layer as new information. Each branch in each hierarchical layer undertakes a complex information processing activity consisting in: (1) collecting and selecting the information at hand, (2) chunking various pieces of information in the form of block-information, (3) transmitting this new information to the higher hierarchical layer. The role played by this process in overcoming the individual calculation limits should be clear. A simple example will help to clarify it. Assume that no individual is able to elaborate efficiently more than seven pieces of information simultaneously, and suppose that a certain organisation has to cope with 100 pieces of information, only 49 of which are really relevant. In this case seven workers are sufficient to start up the knowledge-production process. Each of them selects information by choosing the seven pieces he considers relevant to solve his problem. He then organises them into an information block which he transmits to the boss. The latter will thus receive seven pieces of block-information from his seven subordinate workers, and can use them to work out his problem. Eight persons, located on two hierarchical layers, are sufficient to manage 107 pieces of information (the 100 original ones, only 49 of which are selected as relevant, plus 7 new pieces of block-information), even if none of them can master more than 7 pieces. As already observed, the two knowledge flows are inseparable, since the solution of a problem requires both the acquisition and the theoretical elaboration of information. But acquisition presupposes selection, and this presupposes model elaboration. Each actor then undertakes four functions simultaneously: (a) he collects and selects information, possibly receiving part of it from a lower hierarchical level, (b) he elaborates it by producing a partial theory to work out a problem (problem-solving), (c) he transmits it to the higher hierarchical layers in

238 Production governance structures the form new block-information, (d) he structures residual problems and transmits them to the lower hierarchical layer together with application instructions (problemframing). Although the problem-solving and problem-framing work is undertaken by single individuals, it is the organisation as a whole that produces knowledge. And it does it through a process of task division using the hierarchical ladder to effect both synthesis and analysis work. The synthesis activity, consisting in collecting, filtering, elaborating and chunking information, develops mainly through the bottom-up flow. The analysis activity, consisting in structuring, decomposing and assigning problems to productive branches, develops in the top-down flow. The point of encounter of the two elaboration processes is the theoretical activity. But this is not the whole story, nor perhaps the most interesting part of it. The crucial questions are: how does an individual carry out his theoretical job? does he work it out by himself? and does he find the solutions in a book of blueprints? How does a worker realise that he has finally discovered the right number of turns to give a screw? The point is that nobody works and thinks on his own. A worker continually exchanges information with his colleagues, he asks for and makes suggestions, he observes other workers who have similar or complementary jobs to his own, he requests the advice of experts, he imitates the practices of his successful peers. This is the horizontal process of knowledge production. Theories, problems and information continually travel through the junctions of a more or less informal network involving individuals located at the same layer of a hierarchy or those in its immediate proximity, or even outside the organisation. Such networks take the form of expert forums and constitute what is sometimes called a ‘community of practices’. A worker decides to give 10 turns to his screw; then he observes that a colleague gives it 12, another 9, another 8, still another 14; then he looks at the reactions of the workers further down the assembly line and discovers that most of them are quite satisfied with 10 turns, other with 11, still other with 12; so he decides to settle for 11. If all workers behave in the same way, they eventually discover the right number of turns is 11. This kind of scientific activity gives rise to the acquisition of know-how which remains partly in the form of tacit knowledge. But the ‘knowledge socialisation’ process, insofar as it takes place through conversation, also helps individuals to make know-how explicit. Through communities of practices individuals carry out various functions of knowledge conversion: socialisation of tacit knowledge, through observation and imitation; transformation of tacit knowledge into explicit knowledge, through conversation; combination of different pieces of knowledge, by chunking the information they convey; transformation of new explicit knowledge into new tacit know-how, by putting it into practice. Problem-solving and problem-framing are theoretical activities undertaken by individuals, which however take place through social interaction and practical action. These are the two characteristics of the communities of practices that make

Production governance structures 239 them so helpful in knowledge production: the social and practical dimensions of the information flows which circulate in them. The horizontal process of knowledge production often creates information redundancy, in that individuals normally acquire more information than is strictly necessary to work out their specific problems. Redundancy may prove to be a hindrance, as in the case of information overload, but may also produce valuable benefits, for it enriches the information sets of individuals thus enabling them to learn by intrusion, that is, by freely appropriating the wisdom of others. Workplace scientific forums are networks in which individuals play the role of junctions combining both knowledge and hard work, that is, both theoretical and experimental activity. Through them, knowledge is pooled, thus enhancing the capacity to collect, select, chunk and elaborate information. Then, when a problem has been solved and framed in a community of practices, it is transmitted as new information to the upper layers of the organisational hierarchy and as new problems to the lower layers. But most of the work has been carried out within a community of peers. Knowledge is a collective wealth, and is produced by the organisation. An organisation performing this task can be considered as a social device practising the division of knowledge. Although each specific piece of information is possessed by single individuals, and although the specific pieces of theory are elaborated by single individuals, no person is able to acquire general knowledge. Nor can it be said that the knowledge available to the upper layers of the hierarchy is more valuable than that available to the lower. Of course upper level theories are more general, yet they are not more complete. They are certainly more abstract, but, taken separately, they do not count for much. Knowledge is an essentially and structurally social fact, for the simple reason that no single individual has Olympic rationality. The innovation gap and decreasing returns to hierarchy From the point of view of knowledge production, an organisation can find itself in one of two states: stasis or exertion. Stasis occurs when all production problems have been solved and no new relevant information is taken into consideration, which happens when no conspicuous change, either exogenous or endogenous, takes place. In a stasis situation knowledge flows are blocked. Each branch of the organisation has resolved its specific problem and has identified the best way of carrying out its productive task. Therefore it has no need to elaborate new theories, nor transmit new problems to its subordinate branches. No new relevant information is taken into consideration, and the activity of selecting and elaborating information is no longer required. All information considered relevant is already incorporated in the routines and techniques in use. Thus the flow of information transmission is also blocked. When the firm is in this state, most of knowledge has become tacit and praxis has been routinised. Some problem-solving activity may still take place, of course, but only minor problems are dealt with in this state, problems which are

240 Production governance structures mainly caused by modest stochastic variability, and the solution of which can be confined to the branch and the layer where they arise. A state of exertion may be triggered off by an internal change. In fact there is always the possibility that new ideas spring from the inventive capabilities involved in the problem-solving activities. Any branch at any hierarchical level may discover at any time a new way of working out a problem which is not compatible with the theory elaborated at a higher hierarchical level or with existing routines and techniques in use. Such a discovery may take the form of new information to be transmitted to the upper layers of the organisation, and from which, as a consequence, a decision change is called for. In turn, the upper layer branch involved, when trying to elaborate the new information, may discover it is dealing with a new fact requiring a change in the structuring of problems and theories at an even higher branch level. Since a firm hierarchy, as an instrument of knowledge production, reflects the hierarchical structure of theories, the degree of radicalism of an invention can be measured by the number of hierarchical layers and branches involved in the process of restructuring problems and re-elaborating theories. When an organisation is in this state of adaptation to novelty, new routines have to be produced and old ones modified. Often inventions remain trapped for a long period. This occurs when a branch has attained an invention, either from its own praxis or from information supplied by lower layer branches, but has decided it is not necessary to transmit it upward the hierarchy as new information and downward as new instructions. It might believe that it is not worthwhile changing existing routines and techniques in use, for instance, because present production is still highly profitable, is less risky than the innovation, is constrained by sunk costs, or because the innovation requires huge investments or costly restructuring of organisation. Often several such paralysing conditions occur simultaneously. In these cases the new knowledge survives in a latent state, imprisoned in the minds of some individuals or in the electronic files of some offices and awaiting better times for implementation. It is this tendency to delay innovations that renders productive routines relatively stable. In fact change and the production of new knowledge is potentially incessant. If every alteration in conventional wisdom gave rise to a change in productive practices, the latter would always be in a state of permanent revolution. In reality firms operate with relatively stable organisational structures and production techniques, and change, especially radical change, occurs catastrophically rather than continuously, and intermittently rather than continually. When such a gap between new ideas and old practices occurs, a separation between the theoretical and the pragmatic dimensions of knowledge takes place. In the account of the knowledgeproduction process presented here, these two dimensions have been kept united, on the presupposition that a theory is always at the same time a model of the word and a bet on its change. Yet, when an invention is inert it happens that the knowledge power of a theory is not translated into practical power, ‘know what’ is not transformed into ‘know how’, conjectures are not put to the test. This is the innovation gap. The firm living in this state possesses a wealth of potential innovation, but goes on producing by using obsolete knowledge. Innovation consists in the

Production governance structures 241 revolutionary process that transforms new theoretical knowledge into practical knowledge. An innovation gap may be caused by a sluggish management. But it can also arise at the lower layers of the firm hierarchy. The formulation of an idea and the transmission of ensuing instructions and under-problems to lower hierarchical layers is an innovation decision. Knowledge production always implies decisionmaking. But the ability to take decisions is not uniformly distributed in a hierarchical organisation. The top management has complete decision autonomy, in that its ability to launch new ideas is not internally constrained. Actors working in lower layers of an organisation have a more or less constrained ability to take decisions, the more constrained the lower the layers. Constraints arise because the problems to work out have been framed by a higher layer decisor. There is a difficulty which is typical of all authoritarian organisations. Actors who have limited autonomy, that is, a constrained ability to take decisions, do not normally have much personal incentive to innovate. It is not just the fact that the economic benefits of inventive activity are mostly appropriated by actors working at the higher or top layers of the organisation. Even more important is the fact that, lacking a real decision-making faculty, people are unable to obtain satisfaction from creative activity, to find self-realisation in their work. Furthermore, since the identification of a personal contribution to an invention tends to fade as new ideas are assimilated and spread among the upper layers of a hierarchy, people are also frustrated by the lack of recognition of their merits. A worker deprived of such incentives has no reason to exert his inventive ability, his willingness to catch new opportunities, his theoretical skills. This seems to substantiate Boulding’s thesis on the existence of something like a decreasing return to hierarchy in knowledge production. The wider is a hierarchical ladder and the more authoritarian its governance relations, the lower will be the personal efforts of employees in producing innovations and the less efficient will be innovative activity. Confirmation of Boulding’s thesis is offered today by the tendency of the most knowledge intensive and innovation prone companies to flatten their learning hierarchies and to count more on horizontal flows of information than on vertical ones – a tendency facilitated by the diffusion of computer-based information technologies and electronic networks. Appropriation of knowledge and capital fetishism A theory remains the exclusive dominion of its creator until it is published, after which it becomes public knowledge. This is the case with a mathematical theorem. Anyone can use it when it has been made known to the scientific community. Similar to this, in principle, is the ‘publication’ of the inventions of new economic goods. The inventor is able to keep his theory to himself so long as it remains conjectural. But as soon as he tries to realise it by introducing the new product on the market, everybody gets to know about it. This possibility of making known and publicising ideas during their realisation is what makes knowledge a public good, a good that can be used without rivalry and without excludability. In fact

242 Production governance structures nobody reduces the stock of knowledge by using it and nobody can be excluded by others from using public knowledge, in principle. Knowledge is also an indivisible good. A theory must be used en bloc and cannot be split into small parts. This creates the conditions for prompting increasing returns to scale. A new idea is like a new machine: its dimension does not change when you change the quantity of goods produced with it. And not only the use, but also the production of knowledge is encumbered by indivisibilities, for research and development activities normally require huge investments in private and public capital. Moreover, strong external economies may emerge in the use and production of knowledge, as the theories developed by a certain social actor propitiate the investigation and invention capacities of others. This makes knowledge a merit good. The more abundant is the knowledge available to a certain society the higher is social welfare. And since knowledge is scattered in individual minds, social welfare rises as education spreads and increases. Knowledge is often a gratifying good in itself, as it gives satisfaction to the individuals who have it quite independently of the material goods it enables them to produce. Therefore all communities assign it a positive value also from the point of view of the individuals’ welfare. Unfortunately a serious valuation problem arises with knowledge which is typical of merit goods: those who do not have it are unable to value its utility. Ignorant people tend to underestimate the importance of the struggle against ignorance and therefore tend to under-invest in knowledge production. Finally, new knowledge is typically a non insurable good, or at least one that is not fully insurable. Since new ideas normally take on the form of inventions rather than discoveries, it is impossible to define probability distributions for events which might be provoked by them. This is why it is so difficult for an inventor to reduce idiosyncratic risk by paying an insurance premium, although it may be possible to share it among the many actors who are prepared to finance an innovative investment. Having clarified the characteristics of knowledge that make it a very special good, I can now argue about the effects of the institutional setting on its production and use, on the growth of its quantity and quality and on its contribution to social welfare. In particular I wish to focus on the effects produced by capitalist institutions. To this end it might be useful to resort to an analytic device adopted by Kenneth Arrow, that of taking an ‘ideal socialist system’ as a benchmark. The specification furnished by the term ‘ideal’ is important: it is a model defined precisely as one conducive to the best use and the highest production of knowledge. The ensuing reasoning is thus an exercise with counter-factuals and not a comparative analysis of alternative economic systems. But before going ahead I must substitute the term ‘socialist’ with ‘communist’. The reason for this change is that a public good is allocated efficiently, even in a market system, only when it is allocated on the basis of the principle ‘from each according to his ability, to each according to his needs’. For instance, everyone is free to use a mathematical theorem, even though its production has been financed by all the citizens through graduated tax-rates and carried out by a scientist by means of his personal ability.

Production governance structures 243 In an ideal communist system everyone is free to make use of knowledge. Thus knowledge is not exclusively appropriable, is socially diffused and produces maximum social welfare. At which Mr Alchian might ask: if one cannot appropriate his own ideas and cash in on their fruits, why should he waste time and money producing them? This question is more than legitimate, and indeed it gives us an insight into the most important problem, and the paradox, of knowledge intended as an economic good: an intrinsic contradiction seems to exist between the conditions of its production and those of its use; if everybody can use it, nobody would want to produce it, or at least so it would appear. As is well known, the solution adopted in capitalist systems is counterfeit appropriation, that is, the creation of a system of property rights on the product of the human mind – a system based on trade marks, copyrights and patents. Why is it counterfeit? Because the products of the human mind cannot be naturally appropriated as in the case of most real private goods. In fact their use by the inventor discloses their contents and makes public use possible. The point is that the system of counterfeit appropriation, which seems to solve the problem of production incentives, produces forms of inefficiency which are not present in an ideal communist system. First, it leads to the sub-optimal use of knowledge by introducing elements of excludability. The quantity of a new good which can be produced by a firm that owns the patent is normally lower than the quantity that would be produced if all firms could use the invention. If the patent owner were prepared to assign its use only in exchange for royalties, production would certainly increase, but then so would production costs, so that the social benefit would anyhow be lower than could be achieved without patent ownership. Furthermore, any limitation of access to knowledge also hinders activation of all those forms of scale economies which are usually associated with cultural diffusion. Not to mention the limitations on the process of variety multiplication. An innovation is always susceptible to improvements, if it is available to the public; and the greater the number of firms that produce the new good, the wider the range of possible improvements. Private appropriation of knowledge brings about a weakening in precisely the most positive effects of market pluralism. But there is another problem. By virtue of the strong indivisibilities associated with the production and productive use of knowledge, private appropriation of profits from innovation enables the entrepreneurs to use their privilege to expand market shares and acquire monopolistic power. Monopoly gives rise to an immediate social loss, because it enables the sellers to exploit the buyers and because the consumers’ freedom of choice is greatly impaired. To this must be added the intrinsic loss that derives from the limitation of pluralism and the quantities produced. As is well known, even if it is true that monopoly may enable huge investments to be made in research and development, it is likewise true that it blunts the competitive spur to use knowledge productively. An oligopolistic company with a big market share could be prepared to invest in research activities in an attempt to

244 Production governance structures prevent the entry of potential competitors or aggression from extant competitors. But it could also have no inducement to compete with itself, at least so long as existing plants were profitable. One of the effects of the market power of big companies is to delay the economic use of inventions. But not only is there an effect of inhibition of the knowledge diffusion process, there is also one of deformation of development direction. In fact capitalist firms aim only instrumentally to satisfy the consumers’ aspirations. Their real aim is capital accumulation. As a consequence investments orient research and development activities towards sectors that promise high profits and fast growth, rather than social welfare. Just think of the huge quantity of capital and energy spent in pseudo-innovations, planned obsolescence, and the production of useless, absurd and noxious commodities. And the deformation of development directions causes deformation of human personality. Together with new goods, especially if they are not really useful, firms also have to produce a need for them, and these are often fictitious, futile and harmful needs. This aspect of the pernicious consequences of capitalist appropriation of inventions can be traced back to the merit good nature of knowledge. The fact that the struggle against ignorance is considered good from the social point of view implies that some collective subject has the prerogative of deciding what is good for people. No church, dictatorial state or big party should claim this right. But then, by the same token, nor should Coca Cola or Phillip & Morris. The diffusion of knowledge is a merit good only if citizens are the real decisors of the validity of inventions, only if the final test of conjectural theories takes place in markets where consumers are really sovereign and in forums where people are really free, and only if the collective decisor, when necessary, is entrusted democratically. And this is not the case in a system in which private appropriation of knowledge transforms this good into an instrument of power and capital accumulation. Now consider the non-insurable nature of innovative investments. The main consequence of this feature is under-investment. As I have shown above, in highly competitive systems this difficulty is contrasted by the systemic uncertainty attached to non-investment, as firms are compelled to bear some risk for fear that their competitors will get ahead of them. But in a system in which the private appropriation of knowledge has produced monopolistic powers this fear is reduced. It may seem a paradoxical conclusion, but that is just what it is: private appropriation, which was conceived as an incentive to research, produces the opposite effect in the presence of strong idiosyncratic and low systemic uncertainty. To be sure, the possibility of funding investments with debts and shares helps to ease the trouble by pooling the risk, but this form of ‘insurance’ contributes to worsen the deformation of development direction caused by the subordination of knowledge evolution to capital accumulation. To sum up: there is both low investment and bad investment in research activities, when these are rewarded through private appropriation of knowledge. Let me indulge in a final blow. A system that favours private appropriation of knowledge has ideological implications which contribute to a profound deformation of social ethics. The conviction that everyone is the owner of the product of his

Production governance structures 245 own scientific research refers back to the idea that everyone is the owner of his own ‘human capital’. A man’s perception of himself comes out devastatingly deformed. People are really conceived as the summation of their fortunes, as subjects who have a value only as capital. A form of social ethics is associated with this anthropology in which any kind of altruistic or fair behaviour is interpreted as being useless and therefore irrational, if not downright stupid. Yet, as I argued in chapter 1, man is not a something on which one can settle property rights, at least in a social system based on the recognition of inalienable rights of freedom. A legal system that does not admit slavery and forbids the buying and selling of people, cannot allow the institution of real property rights (including the right to sell) on an individual, or even the right to own (and sell) himself. Now, ownership of flows of ‘human capital’ (as the products of inventive activity are considered to be) are justified by ownership of stock. But, if the latter is arbitrary, then so too is the former. This is another reason why property rights on knowledge are counterfeit and quite arbitrary. Moreover, an invention produced by an individual is never produced by the ‘human capital’ of that individual alone. Accumulated knowledge is not only a public good, that is, a good accessible to everyone without rivalry, it is also a collective good, that is, a thing produced by the co-operative effort of many social actors. This is true for every single new idea, as I argued when I illustrated the role played by hierarchical organisations in coping with informational complexity. It is also true in a wider context, for external economies of scale are always associated with the use of knowledge. It follows that individual ownership of the products of research is not only counterfeit, it is also unjust. It implies exclusive appropriation by a few people of goods which they themselves have not wholly produced. The individual profits ensured by a patent or a copyright are always the result of theft to the detriment of the community, the appropriation of an economic surplus produced by scale economies generated by social co-operation. If knowledge is power, its evolution is thwarted Now consider the power granted by the exclusive control of knowledge. Quite apart from the negative value judgements normally associated with power, there is another serious reason for condemning the capitalist tendency to use science as a source of power: such use limits the evolution of knowledge. I am not referring here to power as the technical capacity to do things. I rather refer to the usual notion of power as a social relation, to the capacity of an actor to determine the actions of other actors and make them serve goals they did not choose. What happens in a capitalist system is that power in the technical sense is reduced to a form of capital and then, given that it is distributed asymmetrically among the social classes, transformed into a source of social power. So, let me now look from another angle at some of the points I have already touched on in previous sections. First of all, a firm that has the exclusive control of a new idea tends to use it to build monopolistic positions through which it exploits its customers. This form of

246 Production governance structures power is made possible by limiting the consumers’ exit freedom. Since the product is new and is not produced by other firms, the consumer is unable to coerce the seller with an exit threat. An inventor of a new good must try to create a demand for it, which he will do by spreading information aimed at moulding the consumers’ aspirations. He will tend to produce deforming information – deforming, in the sense that it changes the form of individual preferences. Aspirations are moulded so as to make them serve the seller’s goals. The consumers are deprived of the capacity to define their aspirations autonomously. Thus, they will have reduced exit freedom, not only because of the innovator’s monopolistic power, but also because of the urgency of the need created by the seller. They cannot even choose to abstain from consumption. In reality, the monopolistic power produced by an invention is rarely unbounded, for in most cases there is some substitutability between a new good and those already existing. Therefore some degree of oligopolistic competition is normally preserved. But the exclusive control of a new idea secures its owner some power over his competitors. The fixing of the price of a new good inevitably conditions the price of its old, partial substitutes. The change in the information set caused by the innovation also conditions the information associated with the old goods. When consumers’ tastes and market shares change, the firm who does not innovate may be compelled to cut prices or bear higher selling costs to make up for the information destruction caused by the innovation. The innovator firm thus tends to assume a position of price leader enabling it to acquire power over its competitors. The monopolistic power of an innovator who has the exclusive control of a new idea may be long lasting, for the start of a research trajectory conditions the future evolution of research. Consider the activity required to perfect the initial idea, commercialise and standardise it; and the activity aimed at realising imitations and improvements, pseudo-innovations and pseudo-improvements. Moreover, if the theory that produced the new trajectory is strongly innovative, it may give rise to a new technological paradigm which will inevitably condition research in industries that produce complementary goods. Obviously the firm who innovated first enjoys a strong competitive advantage and conditioning power over other firms, whether these collaborate or compete with it. And the more exclusive is control of the new idea, the higher are the advantage and the power. Finally, consider the power acquired by the management of the innovating firm over its employees. The articulation of research in a hierarchical organisation implies that all employees take some innovative decisions: selection, elaboration and aggregation of information, formulation of pieces of theory. But decisionmaking power assigned to the lower layers of the hierarchical ladder is limited and conditioned by that exerted at the top of it, if for no other reason because the strategic decisions taken at core theory level determine the research line followed by the whole hierarchy. This means that, whilst the power of the innovator managers is exerted through free decisions, the delegated and residual power of employees is exerted through decisions which are strongly constrained by the upper layers of the firm’s hierarchy. The employees’ research activity is not a free activity, not

Production governance structures 247 only because their goals are decided by others, but also because their choice of means is conditioned. The appropriation of knowledge should be considered harmful quite independently of the effects of unjust income redistribution and limitation of human freedom. It has, among other things, the effect of limiting and conditioning further research growth. Since science is produced through science, the maximum social benefit from scientific evolution is obtained when the public good nature of this product is preserved. But it is precisely this aspect of knowledge that is impaired by private appropriation and power accumulation. This consequence can be seen in the market for final goods, where the capacity to condition the formation of consumers’ aspirations, their consumption decisions and their income expenditure deviates and hinders people in the search for their genuine aspirations. Commercial conditioning limits a person’s capacity to know himself, which is the most ominous kind of limitation of knowledge growth. It can also be seen when an innovator exerts oligopolistic power over competitors and suppliers and conditions their lines of research. In this case appropriation of knowledge reduces market pluralism and therefore hinders the production of variety and the multiplication of ideas. Knowledge appropriation weakens precisely those aspects of the market process that would better favour scientific growth. Finally, it can be seen in the limitation of the autonomy of employees in an organisation where the managers’ control of knowledge discourages the research effort of all the workers who feel their creative potential is conditioned and their decision-making ability limited. And the workers are further discouraged by expropriation of the economic fruits of their research activity. The production and use of knowledge in Nowhere Since the contradiction between production and use is intrinsic in knowledge intended as an economic good, no solution which does not surpass the purely economic dimension of this good can be an ideal solution. Perhaps this idea inspired Kenneth Arrow when he said that an efficient allocation of knowledge presupposes that ‘motives of a non-economic character must be playing a major role’ (1994: 13) or, better still, that ‘in an ideal socialist economy, the reward for invention would be completely separated from any charge to the users’ (1971: 149). I have already shown the positive effects a market system has on the evolution of knowledge. I have also shown the negative ones produced by capitalist appropriation. This distinction is important to dismantle the ideological constructions based on a surreptitious identification of the market system with capitalism. It is important also for understanding why Arrow’s reference to the Soviet system as a possible embodiment of his ‘ideal socialist economy’ is wrong. In fact it is now well known that the failure of the USSR in its competition with Western capitalism was caused, among other things, by a scarce propensity to innovation. The point is that the world is too varied for us to be able to reduce comparison just to market capitalism and state ‘socialism’. Why not also consider a system of state capitalism or one of market socialism or communism? If the Soviet system is

248 Production governance structures interpreted as a form of state capitalism, then the historical example put forward by Arrow does not disprove the conclusions of his analysis. State capitalism may have some merits, for it might reduce the obstacles that hinder the use of knowledge. In fact: it might provide generalised education and culture diffusion better than a system in which the merit nature of knowledge is only partially and imperfectly recognised; it might guarantee public and free use of innovations; it might avoid the allocative disadvantages of monopoly. It may also have some drawbacks, for it might hinder the production of knowledge. It tends to deform the path of knowledge evolution by privileging directions preferred by the political class; it might limit the freedom of research; it might reduce the growth of knowledge variety by discouraging pluralism; it might reduce individual incentives to innovation; it might blunt the spur of competition and systemic uncertainty to the spreading of innovative moods among entrepreneurs. In short, state capitalism may facilitate the use of knowledge but may hinder its production. A market capitalism system has the opposite merits and drawbacks. And obviously it is better to produce in abundance a good which is used ineffectively than the other way round. What can be said of a hypothetical ‘market communism’ system? A democratic community would take care, through public intervention, of creating the conditions for an efficient use of knowledge as a public and merit good. Both general education and access to inventions, for instance, should be governed by the principle ‘to each according to his needs’, thus preserving some of the merits of state capitalism. The problem appears to be: how to ensure that production of knowledge is regulated by the principle ‘from each according to his ability’, while eliminating the drawbacks of state capitalism? The market – one made up of workers’ managed firms and strictly regulated by effective anti-trust laws – would furnish the competitive spur to innovative moods. In a system like this, by the way, all the workers of a firm would bear the risks not insured by the wider community, so that an entrepreneurial attitude would spread through all the strata of a productive organisation. This would defuse the decreasing returns to hierarchy caused, in a capitalist system, by the scarce participation of workers in decision-making and in profit sharing from innovation. The widening of decision-making capacities to include all members of the firm would stimulate their inventive attitudes. Moreover, through the elimination of oppression and exploitation, most of the workers’ grounds for shirking and opportunism would also be eliminated. I do not think all these advantages would be lost due to the impossibility of a full private appropriation of inventions. An innovative firm would have a competitive advantage anyway so long as its competitors fail to imitate it. This would certainly be a monopolistic privilege, but only a temporary one in a system where nobody has the right to block access to knowledge or to claim royalties for the use of patents. And a limited and temporary monopoly profit is a reward a community should be prepared to pay to spur inventive activity. However, individual innovative efforts are also triggered by other motivations. The nature of these other motivations has been highlighted by William Morris in

Production governance structures 249 his splendid Utopian novel News from Nowhere. If, like Morris, you assume that all people, to a greater or lesser extent, wish to give some sense to their labour activity and to seek personal gratification and self-fulfilment in it, you realise that there are two particular non-economic motivations for creative activity: the desire for self-realisation, and the desire for merits recognition. They are both egocentric, but not anti-social, sentiments. Self-realisation consists in the pleasure you receive from transfusing your taste and fantasy, your personality, into the object of your creation, by giving life to something you consider beautiful and useful, by generating with your mind a theory or a project that you are able to realise. In realising it, you realise yourself. Morris gives examples that may appear innocent, given the rural and artisan nature, as well as the prevalently individual dimension, of the creative activity undertaken in Nowhere. But why should one not be gratified by creative activity undertaken in an industrial team? Just think of the proverbial alacrity and creative involvement of the workers who contribute to the production of such a national legend as the Ferrari car. The desire for recognition is no less important. A creative individual, in his boundless egotism, is not content with just admiring himself and seeing himself reflected in his work: he wants to be admired and recognised by others. He wants the applause, he wants the product of his work to be a market success. The desire for recognition is important because it contributes to bend creative gratification to the service of others, to ensure that each person uses his inventiveness in a socially useful way. This desire must not be confused with the search for economic rents. Dante Alighieri and Vincent Van Gogh would not have produced better masterpieces had they been able to collect royalties. And note that, even with the desire for recognition, the motivation behind the effort is not eliminated by the collective nature of the prize. The applause goes to all the Ferrari personnel not just to the designer, to the entire orchestra not just to the conductor. These two kinds of motivation are present in all people to a greater or lesser degree. They are partially present in some capitalist firms too, as the Ferrari example shows. But there is no doubt that their effectiveness would be exalted in a system where labour exploitation and capital fetishism have been abolished. Proof of this opinion is provided by those islands of imperfect communism which thrive in the capitalist world, and precisely in those activities where a great deal of creativity is required, and which society remunerates largely with non-economic rewards, for instance in scientific communities. Appendix: the magnificent and progressive destiny of the ‘learning economy’ The aggressive market competition of present times has pushed the most dynamic companies of global capitalism to invest in permanent innovation. An increasing number of firms in the so called ‘core sector’ is now able to survive only by continually producing innovations and pseudo-innovations, thus permanently abiding in a state of learning exertion. This implies they have to obtain high creative effort

250 Production governance structures from all employees. One of the most important organisational consequences is that internal hierarchical ladders are flattened, thus keeping decreasing returns to hierarchy at bay. Authoritarian control on labour activity is considerably reduced, while the scope of the decision-making ability of all workers is widened. To make up for the effects of control loosening and to trigger organisational involvement, firms invest a lot of money and intelligence in implementing incentive pay schemes and company ideologies. Hierarchy flattening is favoured by the spreading of new information technologies and by the diffusion of computer-based networks. Core companies tend to supplement the hierarchical structures, which are mainly aimed at the control of hard production, with the forum-like structures aimed at knowledge production. A technological revolution seems to be currently in motion capable of dramatically re-moulding social relations and the organisational fabric of capitalism. And just like all revolutions, the present one too is awakening fantastic hopes of regeneration. Professional extrapolators do not appear to have to make much effort to ‘prove’ that humankind has eventually arrived at the threshold of its paradise, and that most of the well-known bad causes of human disease are nearing their end. A new koiné has pervaded the media and the minds in global networks – a koiné by which fantasies like the end of labour, exploitation, authoritarianism, Fordism, industrialism, capitalism, even the end of history, are becoming common sense. Thus, some notes of warning are necessary. To start with, beware of extrapolators. It is not true that the contemporary knowledge-intensive economy is qualitatively different from the old-fashioned hardware economy of Taylor’s times. Hard work has always gone hand in hand with knowledge work in capitalism. The ingenuous example I took from Chaplin’s Modern Times aimed to convey just this truth. A worker has to invest some inventive effort even in deciding on the right number of turns to give a screw if he wants to survive in the capitalist firm. Certainly there has recently been a great leap forward in the intensity of brainpower use in production. But this is a quantitative change, not a transformation in substance – as can easily be seen by looking at the percentages of highly skilled workers in the labour force. In the USA, for instance, this percentage jumped from 26 in 1972 to 31 in 1996. In Germany it jumped from 18 to 25 (Cirino 1999). Is this a secular trend? I doubt it. It is understandable that the intensity of knowledge work tends to rise during big technological revolutions. And this tendency has occurred other times in the history of capitalism. But there have also been periods in which, on the contrary, capital ‘tends to replace skilled workers by less skilled, mature labour-power by immature’ (Marx, 1990 I: 788) – a tendency that had already been noted by Adam Smith. A similar change also occurred after the establishment of Fordism, and was pointed out by Braverman (1974). It seems that a long wave motion, rather than a secular trend, is at work here (Gordon, Edwards and Reich, 1982). And one can even envisage a scenario for the near future in which ‘the growth of knowledge is thwarted’, so that

Production governance structures 251 in the place of humans, many production processes would be administered largely by artificially intelligent machines. Technology would be used extensively, not to enhance human creative powers, but by as much as possible to replace them. Some economic growth occurs but it would not result from substantial human learning or innovation. It would largely emanate from a growing output of physical goods and automated services. (Hodgson, 1999: 189) If there is no end of history inscribed in the human soul, why should one be inscribed in the minds of engineers and entrepreneurs? Second, beware of generalizers. 31 per cent is still much less than 69 per cent and, moreover, concerns only the US labour force. A few other advanced capitalist countries follow suit. But what about the rest of the world? Third, beware of simplifiers. Flattening a company hierarchy does not mean thoroughly eliminating it. There are still top and ‘middle up-down’ managers; intranet newsgroups are regulated by ‘knowledge managers’; and ‘self-managed work teams’ can do without a boss but are anyway steered by ‘natural leaders’ who have a commitment to accountability to the company ‘mission’. Hierarchical control in knowledge production shrinks, but empowerment of the rank a file with greater decision-making ability is mainly limited to how rather than what to do, to the choice of means rather than to that of goals. In addition, the reduction of authoritarian control is often more than compensated by an increase in ideological intrusion. Finally, one should also beware of wishful dreamers. Is ‘the learning economy’ really a step toward the supersession of capitalism? The emergence of knowledge work as an important dimension of all labour activity would probably create a further condition for worker-managed firms being able to show greater efficiency than capitalist firms – if they were given a chance. But technological changes alone do not modify power relations. And capitalism is demonstrating that it is able to survive this technological revolution too. Possibly some technical boundaries of authority will loosen as ‘cognitive capitalism’ develops, but there will be employers and employees anyway. Probably base pay will be supplemented by incentive pay for a higher number of workers, but it is however still a wage. Presumably high-tech companies will engage a higher number of self-employed workers, but most of them will supply hetero-directed and exploited work nonetheless. Not only ‘cognitive capitalism’ is not in itself the end of history, but nor is it even the end of capitalism. On sources In writing this chapter I have taken account of the so-called competence approach to the theory of the firm. This approach was started by Alfred Marshall (1925) and Edith Penrose (1959). It was then developed in various directions by authors like G. B. Richardson (1972), S. Lippmann and R. P. Rumelt (1982), R. Nelson and

252 Production governance structures S. Winter (1982), B. Wernerfelt (1984), R. N. Langlois (1992), D. J. Teece, G. Pisano and A. Shuen (1990). More recent contributions can be found in G. Hamel and A. Heene (1994), N. J. Foss and C. Knudsen (1996) and N. J. Foss (1997). The competence approach tends to explain the existence of the firm on the ground of the specific resources at its disposal, resources which are intended both in terms of the particular technical endowments that give it a dynamic competitive advantage and in terms of the labour capacity endowments of the actors who take part in production activity. When emphasis is placed on the latter aspect, as I have done in this chapter, it becomes necessary to focus on the governance structures that enable the firm to extract the greatest surplus possible from human resources. The central role of governance structures in organising productive co-operation was investigated in depth by O. E. Williamson (1980, 1982, 1985, 1986). P. Bolton and J. Farrell (1990) studied with acumen the problem of control costs in centralised control systems; P. Milgrom and J. Roberts (1990, 1992, chapter 4) the technical co-ordination function of productive organisations. On the use of incentive pay to cope with information asymmetries, vast literature is now available. The works from which I have drawn with the greatest profit in this field are J. Stiglitz (1974, 1975), J. Mirrlees (1976), M. Harris and A. Raviv (1979), B. Holmstrom (1979, 1982, 1990), J.-J. Laffont and J. Tirole (1986), M. Riordan and D. Sappington (1987). There is copious literature on efficiency wages. Fundamental contributions have been made by G. Calvo (1979), S. Salop (1979), C. Shapiro and J. Stiglitz (1984), J. J. Yellen (1984), J. Stiglitz (1986), L. Katz (1986), J. B. Bulow and S. Summers (1986). I found the work of S. Bowles and R. Boyer (1988) particularly congenial. My personal version of the efficiency wage theory was presented in E. Screpanti (2000). Finally, on the analysis of ‘firm cultures’ as factors of organisational involvement, the essays by J. W. Meyer and B. Rowan (1977), W. G. Ouchi and J. B. Johnson (1978), W. G. Ouchi (1980), J. Van Maanen and E. H. Schein (1978) are particularly stimulating. My sources for price theory are the institutionalist and post-Keynesian approaches. To avoid entanglement in the boundless literature produced in this field, I will limit my references to the two comprehensive treatments of M. R. Tool (1995) and P. Arestis (1992). I will just add a mention of P. M. Sweezy (1939) and N. G. Mankiw (1985), to whom we owe the theories of the kinked demand curve and of small menu costs, and of a few other classic works, R. L. Hall and C. J. Hitch (1939), M. Kalecki (1939), G. J. Stigler (1978), A. M. Okun (1981). The main inspirer of the third part of this chapter is Brian J. Loasby (1976, 1984, 1986, 1996, 1998, 1999), an economist who made fundamental contributions to the definition of a new research programme in which the market dynamics is seen as a process of knowledge growth and the entrepreneur as a knowledge producer. Important contributions to this line of research have also been made by Harper (1989, 1996).

Production governance structures 253 What I like less in this approach is the idea of a ‘growth of knowledge’. Whether it is based on the epistemology of Popper, Lakatos or other post-positivist philosophers, this idea conveys the opinion that knowledge is something which accumulates through scientific ‘discoveries’. These seem to consist in a sort of unveiling of reality which the scientific subject only has to apprehend. Then the competitive process leads to the selection of the best ideas, that is, those that better reflect objective reality. To be truthful, Loasby does not entirely go along with such an approach, mainly by virtue of the influence that Shackle (1972, 1979) exerted on his thought, but nor is he completely immune from it, and certainly not to the point of accepting a constructivist or hermeneutic or pragmatist approach to epistemology. More consistent with the Popperian position is Harper, who, although recognising the impossibility of truth verification, is convinced that the market can provide some sort of corroboration of entrepreneurial truths through falsification. An unacceptable corollary of this conviction is that the objective test bench of entrepreneurs’ conjectures is represented by exogenous consumer tastes. K. E. Boulding (1956, 1971) is more consistent with a pragmatist and constructivist approach, having proposed the famous theory of the ‘image’ as a device aimed at the knowledge and determination of future events, as well as the theory of decision-making as an instrument of knowledge production. The idea that a hierarchical organisation also functions as a knowledgeproduction apparatus is relatively new and as yet there is no extensive literature on the subject. Loasby put forward the concept of a production organisation as a ‘visible college of scientists’ (1986: 56) and clarified the role of the organisational structure of firms in implementing the creation and testing of new knowledge. Pioneer work in this field has been done by M. Egidi (1992, 1996, 1997), who also has the merit of attempting some experiments. E. Jaques (1990) and N. M. Kay (1997) made the first attempts at a general theoretical elaboration. D. W. Conrath (1973) put forward the notion of a hierarchy as a system of nested ‘clusters’. The theory of firms as organisations that create knowledge through various kinds of transformation involving tacit and explicit know-how has been developed by I. Nonaka and H. Takeuchi (1995). F. Cillario and R. Finelli (1998) and G. Cirino (1999) have much to say about ‘cognitive capitalism’. G. M. Hodgson (1999) explores some interesting implications of the ‘learning economy’ for the possible futures of capitalism. G. A. Miller (1956) investigated chunking as a device to reduce information complexity. The notion of a theory as a hierarchy of conjectures has been developed by D. H. Harper (1996), who uses it to account for entrepreneurial activity. But the idea of a hierarchical structure of theories is much more ancient. G. A. Kelly (1955, 1963) investigated the psychological rationale of hierarchical theorising. Among the economists’ contributions on this subject, noteworthy are those of H. A. Simon (1962), H. Margenau (1966) and P. E. Earl (1984). On the characteristics of knowledge as a commodity, the investigation through which K. J. Arrow (1971) brought to light the properties of indivisibility and non-

254 Production governance structures insurability of this good was path-breaking. In the same article Arrow also singled out the public good nature of knowledge, which he subsequently defined in greater detail in Arrow (1994). It is this very nature that causes partial inefficiency of the institutions responsible for private appropriation of knowledge. On the merits and drawbacks of the patent system, the essays by A. Silberston (1971) and P. Dasgupta and P. A. David (1985, 1988) are illuminating. As to the deep motives of creative work, J. O’Neill (1998) investigated into the desire for recognition as a better motivation than the lust for appropriation, while U. Pagano (1985) focused on the potential gratification of labour activity and brought to light its organisational implications. A. Maslow (1970) was the first to discover the paramount importance of the needs for recognition and self-realisation in motivating labour activity in organisations.

Notes 1 This conclusion seems to be contradicted by the practice of job evaluation. With this practice, the capacities required for a certain job (technical knowledge, experience, responsibility etc.), are listed, measured and weighted. Jobs are classified by using the scores and weights attributed to various capacities. Then wages are differentiated in relation to job classes. Job evaluation, however, is not very widespread in modern industry. Indeed it has been much criticised, for instance, on the grounds that it is exposed to the arbitrariness of score and weight attribution, it is based on past experience and does not take into account any potential variation or innovation in the labour process, it tends to hinder flexibility in the use of workers. 2 The notion of ‘socialisation’ is used here in the sense proposed by Talcott Parsons (1951), that is, as referred to the processes of adaptation to social roles that invest the whole life of an individual. 3 If the producers used different technologies, the monopsonist could attempt to collect the producers’ rent by fixing differential prices. In this case too, however, the high transaction costs caused by the existence of many different sellers, would produce a strong incentive to fix a single price. 4 To understand the role played by asymmetric distribution of freedom, observe the two limit cases represented by YTD and YTS in Figure 5.4. The former describes a situation in which, say, the workers expropriate the employer, or deprive him of authority, and then use the means of production to pursue the goals of a democratic firm they control. In this situation the capitalist’s freedom is completely abolished. The other case is represented by allocation YTS, which could be described as follows. The employer dismisses some workers, and than rents the use of some slaves from a slave dealer. Presumably these workers would be paid no more than a subsistence income, YS, for their exit freedom is practically nil. Some modern kinds of temporary work hire resemble this model – a refined and liberal form of part time locatio servi. 5 With unskilled workers it is perhaps the level of unemployment that acts as a whip for the unruly. With skilled workers who enjoy efficiency wages, it is the change in employment that plays that role, as I have shown in the second part of this chapter. 6 In this chapter I use the terms ‘invention’ and ‘innovation’ as synonyms, as I am mainly interested in highlighting the consequences of new knowledge production, independently of the stage of development of the goods incorporating it. For the same reason I will avoid, as far as possible, distinguishing between research and development activities. 7 The English word verify, the French vérifier, the Italian verificare derive from the Medieval Latin expression verus ficare, where verus has both the meanings of ‘true’ and ‘real’ and ficare means ‘to be made’.

Different forms of capitalism 255

6

Different forms of capitalism

The need for a general theory of capitalism Today there seems to be a vast series of different organisational arrangements capable of regulating the functioning and evolution of production activities and, at a first glance, it seems impossible to reduce all of them to a common institutional setting. Is it reasonable to speak of ‘capitalism’ when referring to such diverse phenomena as public companies, state firms, financial markets, the managerial revolution, the pension funds revolution and the like? What hampers our capacity to understand all these phenomena as particular forms of a unique and general mode of production is the widespread acceptance of a traditional notion of capitalism whereby ownership of the means of production unites the claims on residual control and residual income. This view, even if it has been rediscovered and refined in recent research, corresponds to a common sense doctrine, going back to the classical economists and Marx, according to which capitalism is a mode of production based on private ownership of capital assets. It is a theory elaborated on the ground of observation of the economic system prevailing in Great Britain in the late-eighteenth and early-nineteenth centuries. It certainly does not correctly describe the capitalist forms prevailing in modern industrialised economies, and a number of economists, starting with the pathbreaking work of Berle and Means have attempted to reformulate the notion of ‘capitalism’ to cope with a reality that Smith could not have envisaged. But it must not be forgotten that the first economist who investigated the separation between ownership and control was Karl Marx. The fact is that processes like the growth of mega-corporations, the diffusion of financial wealth among vast strata of the population and the development of financial institutions specialised in risk pooling, are continually driving towards an ever increasing separation of the claims on residual control from those on residual income. Is there a sense in using the notion of ‘capitalism’ in investigating such a world? Any affirmative answer to this question has to face a decisive theoretical problem: how to account for the fact that modern economies are propelled by the motor of capital accumulation, if the distribution of surplus value does not legally warrant its use for backing the growth of firms and if the decisions to invest in real assets are not necessarily motivated by the profit goal? What we need, to tackle

256 Different forms of capitalism this problem, is a theory of capitalism in which the function of property rights in allocating surplus value is separated from the function of governance structures in disciplining investment activity. We need a theory which is general enough to encompass all possible forms of capitalism, but, at the same time, flexible and complex enough to accommodate many idiosyncrasies. In other words, we need a general theory of capitalism articulated into a series of models of some specific forms.

Institutional systems Institutional systems, modes of production and institutional forms An institutional system is a set of consistent institutions sustaining the structure of a society and regulating its evolution. It is difficult to observe an institutional system in the real world; indeed it is impossible if the principle of institutional plurality holds. Since a society is made up of heterogeneous social actors, i.e. subjects endowed with different and contrasting aspirations and behavioural attitudes, the institutions which regulate it will be heterogeneous, and hardly can be fully consistent with each other. They may govern behaviour in different and alternative ways and may therefore generate chaos and conflict rather than order and harmony. However, it is also difficult for a society to survive for a long time if it is completely chaotic. Actually, when we think of a society we envisage a system of social actors and social relations enjoying a certain stability and some form of structural consistency. And, since a social structure is sustained by institutions, we must assume that at least part of these are consistent with each other. Thus, let me start again: an institutional system is a set of institutions which tends to self-consistency, with at least one well-structured subset which I will call a core subset. The institutions that belong to it will be called dominating or core institutions. To say they are self-consistent means to posit that each of them is in a relation of complementarity with the other elements of the subset.1 They sustain and justify each other and form an organic whole, so that it is not possible to change or modify one of them without modifying the whole core subset. I will call peripheral institutions those which are extraneous to the core, and which can be changed without altering the structure. They may play ancillary or auxiliary functions with respect to the core institutions, but they can even contrast them. Since I posited a ‘tendency to consistency’, one might think that, in the evolution of a society, the weight of its core institutions tends to grow as against that of the peripheral. Fortunately this is not so. Society is the environment where the struggle for existence and dominance of human beings takes place. In this environment the relations of power, exploitation and co-operation change incessantly. The social actors themselves are continually changing. Therefore the institutions will also change. There will be conservative forces, which tend to consolidate the institutional system by eliminating ineffective and destabilising institutions, absorbing some peripheral institutions in the core, isolating conflicting ones. But there will always be innovative and revolutionary forces too, which produce alternative institutions,

Different forms of capitalism 257 weaken the dominating ones and undermine the core consistency. The innovation process contributes to create that institutional variety which is a fundamental condition for evolutionary change. Thus, even if intrinsic forces exist which tend to reinforce the core subset, this tendency will never get the upper hand of history. There may be long periods in which the processes of social stabilisation and institutional consolidation prevail, but not even asymptotically can an institutional system reach perfection. Yet, what I will do in this chapter is to assume precisely this perfection. If what is at stake, in studying society, is not description, but understanding, then some theoretical straining is necessary. This I will do by means of an operation of amputation: I will define some ideal-type institutional systems merely in terms of core institutions and by assuming that no impurity, no peripheralness, infect them. Candidly following Karl Marx, I will call ‘mode of production’ an institutional system composed only of core institutions. To define the structure of a mode of production it is necessary to structure it, which I will do by dividing its institutions into four subsystems.2 The first is the labour utilisation subsystem, which regulates the extraction of surplus from production and therefore defines the fundamental class relationship. In the ‘ancient’ mode of production, for instance, this subsystem is centred on the institutions regulating slavery; in the capitalist mode of production it is based on the employment contract. The second is the distribution-allocation subsystem. It regulates the surplus distribution among the social actors and the exercise of its uses. It consists of the institutions defining and sanctioning property rights, the transfer of wealth, the determination of its yields, the practices and customs for its uses. The third is the legal-political subsystem. It determines the state form and its relation with civil society. Thus it defines the rights and duties of social actors; the prerogative and functions of the political class; the ways of action and organisation of political actors and the rules of law formation. The normative and government action of the state is bounded by this subsystem. The products of the action itself, however, i.e. the laws and political measures, are not necessarily part of the legalpolitical subsystem. In fact, apart from the constitutional rules and regulations governing the life of the state apparatus, most of those rules and measures contribute to define the other three subsystems. Finally, the fourth is the ideological subsystem.3 Perhaps this is the most complex of all, certainly the most difficult to define, if for no other reason than beliefs are among the most fleeting and evanescent attributes of individuals’ behaviour, but especially because many ideological beliefs are entertained by people in an implicit, informal, often even unmindful way. It should be possible to single out at least the essential characters of the ideology dominating in a mode of production. An ideological subsystem furnishes the ethical and theoretical justifications for social and political action and the cognitive bases of decision-making. Therefore it does not just have a rationalisation function, it also has a cognitive and a pragmatic one. The fact that a mode of production is characterised by a stable institutional system does not mean that it cannot change. The natural and technical environment,

258 Different forms of capitalism and the social actors themselves, continually change, Thus an institutional system will be the more stable the higher its capacity to adapt institutions to historical changes. A theoretical problem immediately arises: since a mode of production has been defined on the ground of its core institutions alone and since these constitute an organic, consistent and stable whole, how is it possible to speak of the adaptive change of a mode of production rather than of the mutation from one mode of production to another? Is it possible to speak of internal modifications in a mode of production, i.e. modifications which leave its fundamental characteristics unaltered? The answer to this question is affirmative and is implicit in the way the question has been raised. It is necessary to distinguish, in a mode of production, the fundamental or general characteristics of its institutions or of its subsystems from what could be defined as its special characteristics. Only a change in the former implies the transition from one mode of production to another. The latter can be modified without altering the former. However, since they are core institutions anyway, so that they cannot be changed independently of each other, modifications of special characteristics will consist in structural changes. It is possible for a mode of production to remain unaltered in its fundamental characteristics, and yet take different specific forms. Therefore the notion of a ‘mode of production’ has to be complemented with that of ‘specific institutional forms of a mode of production’, for brevity, ‘institutional forms’. The fundamental characteristics of capitalism I defined capitalism as an economic system in which surplus value is extracted from the production process by using wage labour and is utilised in the circulation process for sustaining capital accumulation. This simple and synthetic definition calls for a more detailed description which can be done by singling out the fundamental characteristics of a capitalist mode of production. The fundamental institution of capitalism is the employment contract. This is the nucleus of the labour utilisation subsystem. It can take different specific forms: it can be an individual or collective contract; a fixed-term or open-term contract; it can be defined implicitly in different measures; it can be combined with other kinds of labour utilisation transactions (contract for services, sharecropping contract etc.); it can give rise to different forms of payment (piecework and overtime pay, bonuses etc.); it can be stipulated by different categories of employers (private owners, public companies, state firms etc.) What cannot be changed is its fundamental characteristic, that is, its ability to generate the workers’ obligation to obedience and the employers’ prerogative to command. In the employment relationship, both contracting parties are endowed with freedom of contract. This gives us a clue as to the fundamental characteristics of the legal-political subsystem. Freedom of contract presupposes that individuals are also endowed with a vast series of universal rights of freedom – rights that must be attributed to citizens as such, that is, irrespective of any difference in

Different forms of capitalism 259 class, gender, race etc. Freedom of contract can be effective only if it is accompanied by rights of association, speech, assembly, thought, ownership and so on. Universal rights cannot be completely generated by transaction institutions, because these produce effects which only involve the contracting parties. Therefore it is necessary for a normative sphere to be constituted which founds, and at the same time is founded on, the citizens rights. This is the constitutional state.4 The state, as a subject of normative action guaranteeing the citizens’ rights, cannot emanate from a private will (like an absolute king or a Machiavellian prince), nor can it boast a legitimacy which is over the law (like God’s grace or papal investiture). In a constitutional state the citizens are subject to the law without functional distinction, so that the political class itself, as well as the legislative, executive and judicial bodies, must be subject to the law. And this implies the existence of constitutional laws. The state must draw its legitimacy from the will of free citizens. If these are recognised as individuals endowed with freedom and inalienable rights, the state cannot rule against their will. Yet human rights are not a gift of nature. They originate from the law and therefore are constituted with the state constitution. Modern constitutions define the fundamental laws of the state together with the fundamental rights of citizens. Thence they found these rights. This may seem a circular reasoning: the state is founded on the citizens’ rights, and these are founded by the state. Precisely so: since neither an original social contract nor a system of natural rights ever existed, only the state can found itself, which it does by founding the source of its legitimacy, i.e. the citizens’ rights and duties. This philosophical circularity necessarily causes some arbitrariness in determining the rule of law: everything depends on how the fundamental rights of citizens are defined. Thus different institutional forms may structure a constitutional state. There may be different electoral franchises, electoral systems, divisions of powers, different forms of monarchy or republic and so on and so forth. But all of them will have a general characteristic in common: sanction of the freedom of contract. There can be no capitalism without freedom of contract and a state which guarantees it.5 Perhaps, even more variegated than the legal-political subsystem, is the ideological one. In history different countries developed different ideological supports for capitalist development. Just think of the role played by the Protestant and Catholic religions in the establishment of capitalism in the North and South of Europe. Even a ‘Marxist-Leninist’ ideology has been used to sustain capitalist development. Moreover, different ideologies may prevail in the same country at different periods. What must all these ideological systems have in common for us to be able to speak of a general characteristic which is typical of the capitalist mode of production? I think the answer to this question is simple: individualism. All the capitalist ideologies have this idea in common: that the foundation of the social being lies in the individual subject. In the Middle Ages there was the idea of a community as a ‘mystic body’, a collective being, transcendentally constituted, which gives sense to the individuals’ lives only to the extent these functionally participate in the life

260 Different forms of capitalism of the social organism. With the birth of modern capitalism the social being is constituted immanently and is embodied in the individual, an egoistic, rational and free social atom (I am speaking of ideology). Social life, political structure and motivation to action, are thought to originate from individual wills. After all, what other idea, better than this, can justify the constitution of economic agents as autonomous contracting parties in a transaction? As to the distributive-allocative subsystem of a capitalist mode of production, it governs the allocation of economic effort and the attribution of its yield so as to nurture capital accumulation. Also in this case the property regimes and governance structures can be the most disparate, from private property to state ownership of the means of production, from concentrated to diffused wealth, from the owners’ to the managers’ control, from the goods to the companies markets. But a principle remains firm in all institutional forms: the principle of individual competition. Economic and political decisions are taken by individuals in competition with other individuals. The consequences of decisions fall on the decisors themselves, which implies individual responsibility. The principle of individual competition holds independently of the organisational forms within which action takes place. Whatever these forms, markets, small or big businesses, bureaucratic apparatuses, they must function so as to make the individuals responsible for their performances and recipient of their results, which is accomplished by making people compete with each other. Property right regimes Now it is time to consider some specific characteristics of the subsystems. And, to prepare the ground for a classification of different institutional forms of capitalism, it is convenient to focus on the distribution-allocation subsystem. The allocation and distribution aspects can be distinguished and separated by means of the notions of ‘property right regimes’ and ‘accumulation governance structures’. A property rights regime is a system of normative, transaction and behavioural institutions that regulate the distribution of wealth and surplus value. In particular it establishes who are the social actors entitled to acquire property rights, as well as the rules governing their transfer. Of great importance are the norms which determine the attribution of residual incomes, and thence the personal distribution of wealth. Some of these norms may be defined by law, others by custom, still others may be ingrained in the common values and beliefs of a society. Observation of the history of modern capitalism brings out three main types of property rights regimes. Let me call them ‘concentrated private property’, ‘diffused private property’ and ‘state property’ regimes. In a regime of concentrated private property wealth is amassed in the hands of individuals belonging to a specific class, the bourgeoisie, while individuals belonging to the working classes are propertyless. In this regime it is the distribution of wealth that creates the conditions for exploitation. The lack of wealth deprives the workers of any control of their life conditions and therefore compels them to work as wage workers under the command of the owners of capital. Since the workers

Different forms of capitalism 261 do not possess any wealth, they can earn their living only by furnishing labour activity. Diffused private property prevails in contemporary capitalism, especially in advanced countries. In this regime, whilst capital is still highly concentrated in statistical terms, a vast and growing mass of people have obtained access to wealth and capital ownership, both directly, as savers possessing shares in joint stock companies, and indirectly, as investors or beneficiaries of various kinds of financial institutions, investments funds, pension funds, employee stock ownership plans etc. Furthermore many forms of financial assets, like bonds and other types of credit, including bank deposits, qualify the investors to receive part of the surplus value produced by firms, even if they have no title to ownership. It seems that modern capitalism tends to transform any citizen into a wealth owner. State property takes ownership diffusion to the extreme, since publicly owned wealth is legally the property of all citizens. No matter what the ideology sustaining or justifying state property is, nor even the mechanisms ensuring actual control of production and the prerogative to take accumulation decisions, the formal owners of state firms are the citizens, who possess national wealth collectively. Legally they are all shareholders of equal importance, since in this system the ‘one head one vote’ principle formally overrides Mises’s ‘one penny one vote’ rule holding in private property regimes. This kind of property regime is not necessarily socialist, if a socialist system is defined as one in which the workers actually control production decisions and the use of value added. If control is out of the workers’ reach and decisions are taken in view of capital accumulation, then this regime entails a capitalist system. Note that what really counts, in the attribution of surplus value in private property regimes, is not so much control of the means of production, as the possession of wealth in general, that is, of any kind of rent-bearing assets, both real and financial, as well as credit and possessory titles. A property right, in this sense, is none other than a claim on the future yields of an asset. An important aspect for which a state property regime differs from private property regimes is the severing it brings about between the title to wealth and the claim on its yields. If all citizens were formally peer shareholders of national wealth, the ‘national dividend’ should be divided and distributed democratically, and the political class should have no discretionary control over these payments. In reality this did not occur, at least not in the historically known systems, where the national dividend is formally attributed to the state and materially controlled by the ruling class. This means that the expression ‘public wealth’ is only a euphemism. The proper word is ‘state wealth’. Accumulation governance structures What really matters is capital accumulation. Whatever the property right regime, an economic system can be defined ‘capitalist’ if surplus value is used to sustain capital accumulation. The systems of institutions regulating capital accumulation are called ‘accumulation governance structures’. Note that the difference between a property rights regime and an accumulation governance structure is that the

262 Different forms of capitalism former governs the allotment of surplus value ownership to the income recipients, whilst the latter governs the attribution of its uses to the investment decisors. An accumulation governance structure plays three functions: disciplinary, selective and financial. In its disciplinary function it regulates the distribution of rewards and punishments in relation to the decisors’ performances, so that control activities enhancing accumulation are encouraged and those hampering it are prevented or dissuaded. In its selective function it regulates the allocation of actors to roles and positions so as to assign the fittest person to each office and to dismiss the unfit. In its financial function it regulates financial flows so as to make them accrue to the decisors who are better able to use them in view of accumulation. In Chapter 4 I showed that the same functions are performed in production governance structures. In that chapter I also argued that there are two fundamentally different kinds of governance structures: networks and hierarchies. Economic networks are institutional systems that regulate markets. All the social actors are independent subjects endowed with freedom of contract, and interact with each other through exchange relations. The disciplinary, selective and financial mechanisms work through market competition. Economic hierarchies are institutional systems regulating economic organisations. The social actors are linked to each other by formal bonds of subordination6 and interact through command relations. The disciplinary, selective and financial mechanisms work through organisational competition. In particular, there are goods markets and companies markets. In the former individual consumers and firms appear as subjects, the objects of transaction being real inputs and outputs, money and credit. In companies markets, or markets for corporate control, the firms themselves are treated as things and made the object of transaction. There are, on the other hand, internal hierarchies and external hierarchies. The former are structures of power relations binding the members of an organisation, the latter structures of power and subordination relations among organisations. Internal hierarchies An internal hierarchy is a necessary condition for the extraction of surplus value in the production process in any form of capitalism. Therefore, not only it is present in all of them, but is a general characteristic of the core institutions of capitalism, and is essential both in the labour utilisation and in the distribution-allocation subsystems. As argued in Chapters 4 and 5, in the former subsystem it functions as a production governance structure. In the latter subsystem, as I am going to show in the present chapter, it works as an accumulation governance structure. This does not involve the whole hierarchy of a firm, but only its upper layers that organise cadres and executives. The command hierarchy of a firm is normally structured in the form of an internal labour ‘market’. Officers are enrolled in the lower layers of the hierarchy, and are engaged on long-term or open-term contracts. Salaries are not determined by market conditions, but by efficiency considerations in view of the duties and skills required by the position in the hierarchy. A different

Different forms of capitalism 263 class of salary is associated with each layer. Although incentive pay may entitle officers to differential salaries within each layer, substantial pay increases are mainly achieved through promotion. The amount of power assigned to any one officer is greater the higher his position on the hierarchical ladder. And power is desired by officers and executives, who are thus strongly motivated to climb the organisational structure of the firm. There may be some inter-firm mobility at the bottom and top layers of the hierarchy. But for the great bulk of officers it is intra-firm, vertical mobility that motivates choices and effort. Hierarchies work as competitive environments. Since there are always many candidates for promotion to any position, competition is strong and takes the form of tournaments. And the road to promotion is good performance. All the three functions of an accumulation governance structure are performed in a hierarchy. The financial function is fulfilled by allotting internal cash flows to the divisions and departments exhibiting better profits and growth opportunities. These allotments furnish resources for future success. At the same time they are perceived by the division or department managers as rewards for their actual performance and a spur for future achievements. But often the allocation of funds to the intermediate officers of a division or a department takes place in an implicit way. The possibility of implementing a decision implies the possibility of using funds. The overall funds of a division are procured by means of good performance. If there is team production, the decisions of the single officers or cadres affect the team performance and its assignment of funds. Therefore it is in the interests of every officer to control the effort of his subordinate workers and clerks: effective control leads to good performance and, as a consequence, procures the funds required to take decisions and extend power. The disciplinary function is fulfilled by granting rewards (in the form of incentive pay and higher salaries and prestige for higher positions on the ladder) and punishments (in the form of fines and income setbacks caused by demotions and dismissals). The selection function too is fulfilled through the promotion and dismissal system. Since organisational competition is won by ability and good performance, the winner of any position normally turns out to be the fittest person for it. Inefficient officers are prevented from reaching high positions. A tournament, that is, the process through which organisational competition takes place, tends to divide the players into two classes: the winners and the losers. Victory is, therefore, a positional good: the winners get it, the losers get defeat; the former secure a good, the latter an evil, i.e. a negative amount of the good at stake. This is why competition is so strong. Moreover, when the prize at stake is an authority role in a hierarchy, victory is associated with the acquisition of power, defeat with the acceptance of a subordinate position. The power structure of a hierarchy transforms the temporary good called ‘victory’ into the permanent one called ‘authority’. This means that, when Russell axiom on the desirability of power applies, the power structure of a firm’s hierarchy gives sufficient inducement to organisational competition, quite independently of the salary structure associated with it.

264 Different forms of capitalism In what sense is an internal hierarchy an accumulation governance structure? Good performance for an intermediate director or officer normally boils down to having the ability to exert control over production activity in order to foster productivity growth and extract surplus value. But it often also entails the ability to increment output and market shares, that is, the size of the firm. And growth in size means growth of capital. This is especially true for directors of divisions in multi-division companies. Success in managing a division or a department attracts internal funds and therefore contributes to allocating surplus value for uses which better enhance accumulation. Power is an instrument for enhancing accumulation, while accumulation performance is an instrument for increasing power. This is the fundamental reason why the internal hierarchy of a firm is an accumulation governance structure: it selects, disciplines, finances and motivates individuals to serve capital accumulation even independently of their money rewards. The problem now is: when executives have reached the top of the ladder and have no further promotion prospects, what motivates their effort and how is their performance disciplined and rewarded? Executives, directors and officers are normally not motivated by profit, but by power, prestige and pay. Typically these three variables are isomorphic, as they are all increasing functions of the position on a hierarchical ladder. And since the number of total subordinates to a position is greater the higher the position, pay, power and prestige also increase with the number of subordinates – a principle which, starting with Lydall (1959), has been demonstrated empirically in a number of researches on the personal distribution of executives’ pay.7 Thus the best way for top managers to pursue their goals is to increase the number of subordinate employees and enhance company growth, that is, capital accumulation. Since they are interested not so much in transitory pay increases as in the accretion of their permanent income, power and prestige, what they actually try to further is long-run company growth. An internal hierarchy works as an accumulation governance structure in its disciplinary function also because the width of the hierarchy in itself is a value for the managers. Its growth is a reward for the ability to make it grow, its shrinking is a punishment. External hierarchies The disciplinary mechanism based on organisational competition works better still when a firm is part of an external hierarchy of business concerns. This is the accumulation governance structure prevailing in the Japanese Keiretsu and in the German system of bank control. The firms belonging to a corporate group are organised in hierarchical structures in which minority shareholdings, debt fetters, technological interdependencies, legal and customary rules, and many other kinds of institutional arrangements, are used to enable big concerns to supervise, monitor and discipline small ones. These corporate groups are normally led or assisted by a ‘main bank’ (or Hausbank), which possesses and controls, directly or indirectly, small but effective minority stakes in all the other firms. At the same time a main bank can be partially controlled by other banks and/or by the big industrial companies of the group. A

Different forms of capitalism 265 main bank also ensures flows of finance and entertains customer relations with members of the group. Finally officers of the main bank are members of the board of directors (or the Aufsichtsrat) of the supervised companies. In this system the ‘external’ controllers of a firm can no longer be considered outsiders, since they possess relevant internal information, they can monitor and supervise the managers’ activity, they can finance their investment decisions, and they can promote or remove the chief executives of controlled firms. The latter, however, preserve extensive decision-making autonomy. Discipline in this accumulation governance structure works through the growth of firms and by virtue of the correlation existing between rapid growth and efficient management. Inefficient officers are not only punished by the slowdown of their companies, but may also be demoted or sidelined or even fired. The financial function is fulfilled through the financial policy of the main banks, which handle credit facilities and assist and sustain new share and bond issues with the aim of favouring long run growth of the most promising companies. Companies markets The accumulation governance structure prevailing in the USA is rather different, especially in the form that gave its ‘best’8 in the 1980s. Big concerns, in this system, have widely dispersed shareholdings and are actually controlled by their managers. Financial capitalists are normally not interested in controlling insiders, nor have the necessary competence to do it. Yet financial markets do exert some control, for managers who outrageously disregard the shareholders’ interests, may be compelled to respond to the raiders. A strategic variable in this accumulation governance structure is the value of the firm. Since the managers’ performances are reflected in the market valuation of their companies, this is the lever through which the financial, disciplinary and selective functions are performed. When the shareholders lose control of the firm, the difference between title documents and negotiable instruments, between shares and bonds, tends to vanish, at least on economic grounds. In both the cases of shares and bonds the owner is deprived, not only of any effective control of the firm, but also of the control on residual income. In both cases the market values and yields of the two kinds of financial assets are determined by a combination of managers’ decisions and market valuations, the owner being unable to affect either. The only real difference boils down to the degree of riskiness of the two kinds of assets. External finance accrues easily and cheaply to high value firms. These firms can successfully resort to new issues of share capital and, at the same time, they can obtain cheap credit without needing to raise leverage. Not so with companies whose performances are badly valued by the market. The financial function of this accumulation governance structure tends to reinforce the positive effects profitable growth has on cash flow and internal finance, as well as the negative ones brought about by low profitability. Market valuation also works as a disciplinary and selection device. First, it gives signals as to the ability of managers, which contributes to their reputation

266 Different forms of capitalism and increases their probability of finding new, more prestigious and better paid jobs in other firms.9 Second, there is a widespread use of incentive pay in this system, especially of schemes linking rewards to the value of the firm. Thus good managers are paid better than bad ones, and pay differences tend to reflect the differences in market valuations. Third, managers are constantly under threat of hostile take-overs. This threat may be kept at bay by a high market value, but it is likely to become effective on unsuccessful managers, i.e. those who are unable to make the value of their firm come up to its potential. Goods markets The most widespread accumulation governance structure is based on the working of goods markets. It operates in all modern economies, and even in state capitalist systems, where it regulates accumulation in the small business, retail trade and service sectors. It was the dominant structure in nineteenth-century capitalism and is still dominant in many backward economies. Goods markets are here defined as encompassing both real commodities and credit, but excluding transactions in company shares. The financial function in goods markets is performed through the allocation of internal finance and bank credit. Firms with good performances in goods markets also have high profits and cash flows. Therefore they can self-finance their growth. Their profitability ensures easy and cheap access to credit facilities, as well as the ability to repay debts. Badly performing firms, on the contrary, have to face financial distress and growth regression. The disciplinary function is based on the allotment of profits and losses. Efficient entrepreneurs enjoy profits and growth, and therefore increases in income, prestige and power. Inefficient ones are punished with losses, economic slowdown and social setbacks. Finally the selection function is based on the working of goods markets as a competitive environment, i.e. a milieu where the struggle for survival is regulated by the jungle law reinforced by bankruptcy law. Structurally inefficient firms face permanent losses and sooner or later are expelled from the market. Thus inept entrepreneurs are doomed to failure, whilst the able ones enjoy success. This mechanism ensures that capital accumulation is correlated to entrepreneurial ability, but also that unfit entrepreneurs are eliminated from the market. Institutional forms and the problem of institutional stability Table 6.1 delineates the notions of a ‘property rights regime’ and an ‘accumulation governance structure’. By focusing on some of their possible combinations, it provides a classification of a few ideal capitalist forms. Of course the historical examples to which this classification refers do not completely fit the scheme. The ideal-types do so very well, although they do not completely fit history. Note that the particular accumulation governance structure characterising each capitalist form in Table 6.1 is not exclusive. Indeed all the other governance structures are also present to some degree in each form. Neither classical capitalism nor corporate

Different forms of capitalism 267 Table 6.1 Institutional forms of capitalism

Property Rights Regimes Concentrated private property Goods Markets Accumulation

Diffused private property

State property

Classical capitalism

Companies Markets

Market-oriented corporate capitalism

External hierarchies

Bank-oriented corporate capitalism

Governance Structures

Internal hierarchies

Decentralised state capitalism Centralised state capitalism

and state capitalism can do without goods markets, for instance. Likewise, internal hierarchies are present in all forms, if anything, because a power hierarchy is a necessary requisite of capitalist production. But ideal-types are characterised solely by the accumulation governance structure which is considered dominating. The principle of impurity suggests that no real economy ever coincides precisely with a typical form. Yet the principle of dominance grants that some institutions are organised into systems able to regulate the great bulk of transactions. Impurities cannot be ignored when accounting for historical evolution, since they produce the variability through which competitive evolution drives institutional change. Not even the principle of dominance can be ignored, for history evolves through change, of course, but also through the ‘long durée’. There are periods of structural instability, revolution and catastrophic change, in which chaotic movements prevail, institutional forms are in disarray and the distinction between dominating and peripheral institutions is temporarily blurred. There are long periods of rest, during which change is slow and gradual and dominating institutions are organised in structurally stable forms capable of enduring institutional evolution by smoothly assimilating innovations and external shocks. Thus history can be read both as a process of recurrent structural change, and as a succession of stages in which stable forms of social and economic organisation persist undisturbed for a long time. As I am mainly interested in model classification, I feel justified in freeing some historical cases of all impurities, thus defining a few ideal institutional forms of capitalism which vaguely resemble reality. In the rest of this chapter I try to reconstruct the institutional systems of four capitalist forms: classical capitalism, market-oriented corporate capitalism, bankoriented corporate capitalism and state capitalism. They are only partial attempts

268 Different forms of capitalism at reconstruction, since I will mainly focus on the distribution-allocation subsystems. To reconstruct an institutional form implies more than analysing its structure. An institutional form, as an ideal-type, is defined as being made up solely of core institutions organised into a self-consistent system. Institutions are not logical assertions or mathematical equations. Their consistency is not a formal property. Institutions pose constraints on social action. To say they are consistent with each other means that the constraints posed by each of them does not contrast with those posed by the others. In an institutional system which is self-consistent in this sense different institutions are complementary. They sustain each other and thus contribute to produce stability in social relations. It is not sufficient to describe the structure of the various subsystems. It is also necessary to show how the influence on social action exerted by any one of them upholds that exerted by the others. When this condition holds, the social relations determined by the decisions of social and political actors are stable. Since institutions are produced by social and political action, the stability of relations and behaviour bears on the stability of institutions. The institutional stability of a capitalist form consists of the capacity to resist institutional invasions, innovations and shocks in such a way as to maintain its structure substantially unaltered. Institutions are consistent with each other if they are structured in a system endowed with this property. Note that the notion of ‘institutional stability’ is slightly different from that of ‘evolutional stability’. The latter has been developed especially in game theory, where it defines a property of repeated games whereby a ‘strategy’ (or ‘rule’ or ‘institution’) governing social interactions is capable of resisting the emergence of a new strategy and avoiding its invasion of the social system. The stable rule or strategy affects the behaviour of most players, and this contributes to diffuse and reinforce the grip of the rule itself. The rule is evolutionally stable if it remains dominant when a new rule arises. ‘Institutional stability’ differs from ‘evolutional stability’ in three ways. First, it refers to systems of institutions and not to a single or a few institutions. Second, instead of focusing on the process of behavioural change and rule substitution it stresses the structural-functional relations of the institutions making up a system, trying to show the capacity of the various institutions or institutional subsystems to sustain and reinforce each other. Third, since an institutional system cannot be changed gradually and piecemeal, but can only be modified in block, it conveys a vision of social change as morphogenetic rather than smooth and gradual. The institutional system is the focus of institutional analysis, which looks at the interactions between the structure of social relations and that of institutions. It is not the only way of studying the conditions of social stability. There is another, more traditional one, which focuses directly on social relations. It is the method prevailing in Marxist theory, in which the study of stability conditions takes the form of an analysis of social reproduction. A social system is socially stable if it is able to reproduce in time the social relations on which it is built. In the preceding chapters I showed that there are four sets of conditions for social reproduction:

Different forms of capitalism 269 technological, financial, political and ideological. Now I can add that these conditions are ensured by the labour utilisation, the distribution-allocation, the legal-political and the ideological subsystems of institutional structure. In the rest of this chapter the various forms of capitalism are scrutinised via the method of institutional analysis, i.e. by investigating the structure of their institutional systems and the conditions of their stability. The method of social analysis, however, will not be completely ignored, and something will be said also on the conditions of social reproduction.

Forms of capitalism Classical capitalism This institutional form is based on the association between a concentrated private property regime and a goods markets governance structure. The control of labour in the production process is ensured by a relational contract whereby a personal employer buys command over an employee in exchange for a wage. Since operation of the labour process requires the use of some means of production, a condition for workers to agree to enter an employment contract is that wealth is out of their reach and is concentrated in the hands of a class which uses it to produce profits. Thus control of the production process is secured by the ownership of the means of production. The capitalist firm is a thing, an asset owned by a person who bears unlimited responsibility for its activity. Firms are small in size and internal hierarchies are relatively tiny and simple. Society is divided into two groups: the capitalist class, or bourgeoisie, which includes all the owners of means of production; and the working class, or proletariat, which comprises all the propertyless employees. The political class responds to the bourgeoisie. This relation is legally warranted by a limited franchise, privileging wealth owners, which keeps the right of vote out of reach of the working class.10 The state sanctions and protects property rights and the ensuing rights of freedom. In particular it guarantees freedom of contract in all transactions, not only de jure, but also de facto. And since free competition is considered one of the basic conditions of freedom, the economic action of the state aims, above all, at limiting or abolishing monopolistic practices. This action is particularly strong in the so-called ‘labour market’, where state intervention boils down to promoting labour mobility and flexibility and forbidding and punishing workers’ coalitions. As Adam Smith had already realised, it is this kind of ‘competition’ in the labour market, whereby the capitalists are organised in the state and the workers are disorganised by the state, that ensures the conditions for wealth concentration and for the real wages to settle down to a subsistence level. Thus property rights secure the microeconomic and microsocial conditions for controlling the labour process and the state; while the state guarantees the macroeconomic and macrosocial conditions for exercising property rights. The dominating ideology in classical capitalism is based on liberalism and laissez faire doctrine, a complex philosophical construct plunging its roots in

270 Different forms of capitalism seventeenth- and eighteenth-century enlightenment. This ideology evolved in different ways in various countries and epochs. It is useless and impossible to summarise its basic traits here, let alone do justice to the richness of its varieties. What I want to do, instead, is to bring to light the role it played in linking the political action of the ruling classes to the fundamental interests of the bourgeoisie. There is in fact a double link. The first provides philosophical guidelines for politics. Even if the ruling class were fully and really autonomous from the capitalists, the laissez faire doctrines would have induced it to work in their interests, e.g. by establishing the rule of law, rights of freedom and freedom of contract; by enacting free trade rules and fighting monopolistic practices; by preventing and repressing workers’ organisations that could threaten freedom and competition in the labour market and the political arena; by defending the natural rights of men, the most important of which being property rights. The second link provides philosophical justifications for a restricted franchise privileging wealth owners and discriminating workers, women and race minorities. This link ensures that the ruling class is not fully autonomous from the capitalists and is, also de jure, the ‘business committee’ of the bourgeoisie. The philosophical justifications for a limited franchise may be different, but all of them must have a principle in common, namely, that the right of ownership is the most important human right and the one deserving the attribution of complete political rights.11 The legal-political and ideological subsystems ensure the political and cultural conditions for social reproduction. The legal-political subsystem takes care of the political stability of the class system, the ideological provides the knowledge and belief framework required to shape and justify political and social action. But the basic conditions of social reproduction are warranted by the working of the labour utilisation and distribution-allocation subsystems. Control of the production process is effected to produce surplus value and this legally accrues to the firm’s owners. The capitalists’ goal is valorisation of capital, and surplus value is mostly reinvested in the firm. Since the workers’ pay is determined in ‘competitive’ markets that tend to fix it at a subsistence level, wages are unrelated to labour productivity. Thus wealth concentration is stabilised. Social reproduction works through a distribution mechanism whereby the wealth-owners increase their wealth and the proletarians replenish the labour force. Wealth concentration also contributes to raise the average saving propensity and to boost capital accumulation: workers’ consumption is very low and their saving is nil; capitalists, save most of their income. Financial markets reinforce this condition. Banks supply finance to profitable investment projects. Capitalists who reinvest less than they earn supply savings; those who invest more than they earn, demand them. Banks make profits by transforming the savers’ credit into the investors’ debt. They supply credit by selecting borrowers on the ground of profitability and riskiness of their investments, as well as the availability of wealth assets to be offered as collateral. As a consequence, financial capital accrues to fast-growing firms, thus sustaining capital accumulation, while it cannot accrue to the workers, thus reinforcing social reproduction.

Different forms of capitalism 271 Finally, goods markets take care of the selection and discipline of capitalists. Efficient entrepreneurs expand their markets and are rewarded with high profits and increasing power. The laws of competition foster capital accumulation in efficient firms. Inefficient entrepreneurs are punished with losses, slow growth and bankruptcy. The laws of competition hinder accumulation in inefficient firms and expel incompetent entrepreneurs. There is also a technical condition of social reproduction. Entrepreneurial competence is a firm-specific asset. It is acquired through long experience and consists of a set of particular abilities developed both through family education and through managerial practice in the family company. The choice of techniques is a prerogative of entrepreneurs, who exert it to extract surplus value and nurture accumulation. Historical experience shows that, after the industrial revolution, the mechanisation processes tended to de-skill manual work and impoverish the workers’ technical abilities. Employees tend to be deprived of any firm-specific ability. In these conditions, it is efficient to assign control and ownership to the entrepreneurs. Whilst the market-based accumulation governance structure tends to attribute control to efficient owners, the private property regime enables the latter to choose the techniques that justify and sustain their control. Classical capitalism is institutionally fairly stable because its property rights regime and accumulation governance structure reinforce each other. Actually historical instances of classical capitalism did exhibit institutional stability for a long time. Yet this is not an absolute stability, for this type of capitalism contains the conditions for its own transformation. History has proved it to be unstable in the very long run. Its final collapse was brought about by three main trends: the growth of personal wealth, the growth of the size of firms and the growth of the workers’ movement. The great fortunes accumulated during century-long development induced capitalists to diversify risk. Wealth owners tended to invest their riches in many firms. Thus a class of financial capitalists was formed who specialised in the practice of portfolio management. Moreover, the growth of firms, which took place through extensive processes of concentration and centralisation, led to the expansion of internal hierarchies. Family education was no longer sufficient for training entrepreneurs. The old kind of selection was exposed to hereditary taints and family disputes. Entrepreneurial ability became more and more firm-specific with the growth of firms and the development of firms’ capabilities. It could be acquired only through long experience within the internal company hierarchies. This led to the formation of a class of entrepreneurs who specialised in the management of organisations, production activities and real investment decisions. Eventually the public company proved to be the most effective form of ownership structure in big concerns. It was the most efficient to cope with the problem of manager formation, selection and discipline. It partially freed the managers from systematic external monitoring on the part of shareholders, thus enabling them to pursue capital accumulation without obeying a strong urge to maximise profits in the short run. It reduced external control to various forms of market or hierarchical discipline contingent to episodes of performance setbacks.

272 Different forms of capitalism Lastly the growth of the workers’ movement progressively undermined the old form of labour market ‘competition’. Union activity gradually led to legalisation of many worker organisations. Moreover, political activity, as organised, for example, in the Chartist movement in the United Kingdom and in various kinds of democratic and socialist parties in continental Europe and the USA, led to extension of the franchise, development of workers’ protection laws and growth of labour strength. Working time was gradually reduced and work conditions were improved. Wage rates, which were increasingly fixed through collective bargaining, tended to rise beyond subsistence level for a substantial part of the working class. Mutual aid and pension schemes developed, and the working classes became able to supply a share of national saving. Corporate capitalism In this institutional setting the great corporation supersedes the small family company. Its ownership is structured in the form of a joint stock company with a dispersed shareholding, while internal control is implemented through huge and complex hierarchies of command. The firm is no longer a thing that belongs to a person or a group of persons endowed with residual control. Rather, it is itself a person, that is, a social entity with legal personality, able to enter into contracts and transactions with all the other social actors. The entrepreneurs are managers who formally act both as agents of the shareholders and functionaries of the company. Their objective, in the former role, is the value of the firm, in the latter, company growth. In principle there is no contrast between the two goals, since the growth variable is an argument of the firm’s valuation function.12 However, as I will show below, such a contrast might arise as a consequence of the way the accumulation governance structure disciplines managers. Workers, cadres and functionaries are directed through internal hierarchies. Wages are high enough to enable workers to save and accumulate wealth. Many workers become rentiers and, directly or indirectly, the owners of part of the companies’ shares. However, to the extent that owners are deprived of all residual control, the workers remain employees under the command of the managers and, as such, can be exploited. Social reproduction in corporate capitalism is mainly granted by three conditions. The first, which is the same holding in classical capitalism, consists in the non-neutrality of techniques. Technical change is controlled by the entrepreneurs and is class-biased: the labour process is moulded in such a way as to render efficient the assignment of control to the entrepreneurs themselves. This mechanism, though, is exalted in corporate capitalism, where it is much more powerful than in classical capitalism. The innovation process and even invention activity are internalised in the great corporations. Research is mostly performed in R&D departments and is oriented and controlled by managers. The second condition consists in a particularly strong form of class-biased credit rationing. Since banks and financial markets lend money to earn profits, while minimising risk, they tend to privilege borrowers who: are endowed with substantial

Different forms of capitalism 273 wealth to offer as collateral; are proponents of profitable and relatively safe investment projects; have a reputation for managerial ability. The managers of existing firms normally exhibit these three qualities; not so the workers. Thus the managers have easier and cheaper access to credit than the workers. This contributes to stabilise class structure. It might seem strange that class-biased credit rationing is stronger in corporate capitalism than in classical capitalism, for a wage rate that is higher than subsistence enables workers to save and accumulate wealth so that at least self-finance should become an easier way to vertical class mobility. However, first, entry and exit financial barriers have become huge in corporate capitalism, with the consequence that, given the relatively low magnitude of workers’ savings, no worker controlled firm can hope to set up as a big corporation. Second, workers can enter small business, but big business has monopolistic and monopsonistic power in industrial markets. This kind of market power is normally used to establish sub-contracting relations and to externalise and localise some components of production, especially the components in which labour activity is not easy to control in a big internal hierarchy. Therefore the workers who organise themselves into small firms and co-operatives normally work for big business and often turn out to be even more intensely exploited than the employed workers. Third, not only do workers have scarce managerial expertise, but they have no familiarity with financial markets either. Thus they tend to invest their wealth in financial institutions like banks, investment trusts and pension funds, so renouncing any possibility to use it to acquire control over the means of production. Finally, the financial institutions who collect the workers’ savings concentrate their efforts on portfolio management, aiming at long-run financial returns. They diversify investments and practise shortrun trading. The result is that financial markets and institutions work, among other things, at collecting wealth from the common people to finance big corporations, certainly not at fostering the workers’ control over the managers of their companies. The third condition of social reproduction consists of a particular interaction between the accumulation governance structure and the property rights regime. Corporate capitalism is institutionally stable mainly because its accumulation governance structure, given a diffused private property regime, assigns control and finance to efficient managers, who act in such a way as to reinforce that property rights regime. They do it by: nurturing the firms’ growth, thus making it increasingly difficult to unify shareholding in the hands of a few owners; sustaining income growth among all classes, thus contributing to spread savings and wealth and induce investment diversification among the rentiers; launching and financing employees’ stock ownership plans, pension funds and the like, thus promoting dispersion of shareholding. Institutional stability hinges on the tendency of the property rights regime and the accumulation governance structure to sustain each other by reinforcing both the technical and the financial conditions of social reproduction. A few comments will suffice about the working of the legal-political and ideological subsystems. The most important ideological change occurring in the transformation from classical to corporate capitalism consists in the transition from liberalism to democracy and, in connection with this, in a modification of

274 Different forms of capitalism the conception of private property. When wealth ceases to be a title to residual control on production, ownership needs no longer be considered a necessary condition for access to political rights. To make sure that the state supports capital accumulation, it is not necessary for it to be controlled by the owners of capital. When wealth possession is within the reach of wide strata of population, it is no longer possible to use ownership to restrict political rights. These are entrusted to all citizens, the state becomes ‘democratic’ and the ruling class formally represents all the people. To become a member of the ‘legal country’ it is sufficient to be a member of the ‘real country’. Together with this change, a subtle but dramatic modification hits the notion of ‘ownership’. Its definition now departs from the Roman law principle – which views it as a ‘right to use, enjoy and abuse’, i.e. as an unrestrained individual control and disposal of personal wealth – to approach the Scholastic principle – which views it as a ‘power to procure and dispense’, that is, as a prerogative assigned by community to individuals in the public interest.13 So widespread becomes the idea that private ownership has a public function, that even the managers of big corporations, especially if they are ‘public’ companies, pretend to be working in the interests of a generic public rather than in those of a few shareholders. Evidently, a state legitimised in such a democratic way and founded on such democratic constitutions can no longer act as a simple ‘business committee’ of one social class. A ‘democratic state’ is the proper structure of the legal-political subsystem of capitalism. Only a democratic state is a capitalist state proper. With respect to it, liberal states of the nineteenth century should be considered as just transitional forms. Now the state can take care of overall social stability and general capital accumulation. Tactically, it can gather and offer support from time to time to one particular social group or another. It can be Keynesian or monetarist, it can act as a welfare or a warfare or a workfare state. But, strategically, it works – as Gramsci observed – as an instrument of rationalisation and acceleration of accumulation, it operates according to a plan, it pushes, spurs, stimulates and punishes. Structurally, it is the general capitalist, the representative of capital in general. As such, it ensures that the technical, political, ideological and financial conditions of social reproduction work well. Market-oriented corporate capitalism There are two types of corporate capitalism, market-oriented and bank-oriented. Since their labour utilisation, legal-political and ideological subsystems do not differ widely, I will confine myself to outlining their distribution-allocation subsystems only, which indeed differ substantially. Market-oriented corporate capitalism combines the diffused private property regime with an accumulation governance structure based on companies markets. The disciplinary, selection and financial functions are fulfilled through competition in financial markets, that is, through the activity of raiders and financial speculators. Speculators specialise in the collection and elaboration of information on company performance, in the evaluation of firms. Financial markets enjoy a high

Different forms of capitalism 275 degree of information-efficiency in Fama’s ‘weak’ sense, so that the speculators’ evaluations are immediately reflected in the market values of firms. As a consequence, well-performing firms, besides benefiting from increasing flows of internal funds, are rewarded by high market values, and thus face no difficulties in collecting external flows of finance. Raiders specialise in evaluating the gap between the potential and the actual profitability of firms, the effectiveness of managers’ performance. When this gap is perceived as positive, i.e. the market value of the firm is considered lower than its latent fundamentals, managers are judged as inefficient. Then a raid may start off. The raiders attempt to re-unite ownership and re-transform the firm into a thing. When they gain control, they try to realise the full potential of the firm’s value by busting up the company or firing and substituting its old management. This process works both as a disciplinary and a selection mechanism. Competition in financial markets is a disciplinary and selective device because the managers of a firm have to continually face take-over threats. If they are efficient, they are rewarded, not only with growth and external finance, but also with increasing power (brought about by growth), good reputation and high prestige (brought about by growth and market valuation), the safety of their job (ensured by a raiderproof market valuation), and high pay (often linked to market valuation). If they are inefficient, the market endangers their position, pay, power, prestige and reputation. The main drawbacks of this governance structure are well known. The market value of a firm is widely affected by the speculators’ expectations. But speculators, who normally have good expertise in portfolio management and trading, are not firms’ insiders and do not have better information than the managers on a company’s prospective profitability. Certainly they do not have better information on longrun growth prospects. Therefore they form their expectations on the grounds of short-run information variables, such as present profitability, balance sheets, priceearning ratios, dividend payments and the like. This phenomenon bears responsibility for the managers’ ‘short-termism’, i.e. their proneness to comply with the speculators’ preferences by privileging short-run targets to the detriment of innovative and risky investments and long-run growth.14 A manager who directs his efforts at long-run growth to the detriment of present profitability could be punished with a low valuation by speculators. This would raise the gap between potential and actual profitability and therefore could prompt a raider’s aggression. Paradoxically, this system is capable of sustained and stable growth only because managers have developed instruments of defence, such as poison pills, greenmail, majority rule restrictions for the board of directors etc., which serve to raise the cost of raiding. Bank-oriented corporate capitalism Bank-oriented corporate capitalism differs from the market-oriented type in that it is based on an external hierarchy governance structure. Minority shareholding is used to form pyramidal coalitions of firms led by big concerns, often big banks.

276 Different forms of capitalism A big bank dominates a coalition by using a variegated set of instruments, besides shareholding: proxy votes of their customers, the conditioning power ensured by their loans and their assistance in new share issues, the influence of their members on the boards of directors, the borrowing of votes (stimmenleihe) from other banks, and the like. They concede wide autonomy to the managers of controlled companies, yet not unlimited autonomy. Banks entertain customer, monitoring and supervision relations with controlled firms, and are able to collect reliable information on the managers’ performances, investment opportunities and profit and growth prospects. They are insiders, and are better equipped than speculators and raiders to evaluate the fundamentals. They have effective control power, to the point of being able to dismiss and substitute the managers of controlled companies. Thus control on relevant investment decisions is exerted ex ante, through bank monitoring, rather than ex post, through speculation. Bad investments are discouraged or just barred in bank-oriented systems. As already observed, an external hierarchy control structure can perform all three functions of an accumulation governance structure. The financial function is fulfilled through the selected assignment of credit and assistance in new share issues. Banks are interested in the long-run profitability of their investments, therefore they use inside information for directing flows of finance to efficient, innovative and fast-growing firms, while they tend to ration the inefficient, high risk and low profit ones. Finance flows to the channels that better feed accumulation. Bank-oriented corporate capitalism has a series of advantages over the marketoriented type, the most important of which is that it protects managers from the conditioning of myopic speculators, thus eliminating short-termism and enabling the managers themselves to concentrate their efforts on long-run growth. Managers have a higher confidence to preserve their position, and this encourages them to face long-term projects. Another advantage of bank-oriented systems is that they keep the firms’ financial structure under control, at the same time assigning selected flows of finance. In this way financial fragility is reduced. Since financial availability is not biased or conditioned by the vagaries of speculation, dynamic instability is also reduced. Institutional stability is reinforced because leading banks also take care of the ownership structure of controlled firms, especially shareholding dispersion, with the aim of preventing loss of control, hostile takeovers and the like. Recent developments in the European and American financial systems, especially the increasing role played in the latter by institutional investors, seem to reveal a tendency to their convergence. If this is a real possibility, one could conceive of a new capitalist form that combines some of the merits of the bank-oriented and the market-oriented forms. State capitalism In a state capitalist system private ownership is abolished or reduced to a peripheral institution. Productive capital is publicly owned. Since the state ‘expresses the will and the interests of all people’,15 ownership of the means of production takes

Different forms of capitalism 277 ‘the form of state property’. ‘State property is the common wealth of all people’. Since the state possesses (almost) all national wealth, it is the dominant employer. Therefore (almost) all citizens are employed workers. They have the ‘right to a job … and the right to choose the kind of employment and work in accordance with their skills and education’. But they also have ‘the duty to a rigorous observance of labour discipline’. The employer, on his part, ‘exerts control on the quantity of labour’ and, by combining ‘material and moral incentives, takes care of the increase in labour productivity and production efficiency’. Wages are not fixed through the market or through negotiation, but are determined by ‘the state on the grounds of labour productivity increases’. Evidently, a first condition to its being a capitalist system is ensured: the employment contract is the fundamental institution for labour mobilisation. But a second condition is also ensured: freedom of contract is recognised for all citizens, who also enjoy a vast series of ‘social, economic, political and personal rights’. However, for it to be a proper capitalist system another important condition is required: command over labour in the production process must aim at the extraction of surplus value and the decisions to invest surplus value must aim at capital accumulation. Accordingly, ‘the state takes care of the increase in labour productivity’ by virtue of the authority it has in the labour process. It also has ‘the direction of the national economy’. This is ‘exerted on the basis of state plans’ and by ‘combining central management with the economic autonomy and enterprise of firms … To this end the state uses economic calculations which take into account profits, costs of production and other economic levers and spurs’. The main goal of the state’s economic action is ‘the increase of social wealth’ or, to be precise, ‘the dynamic, planned and proportional growth of the national economy’, in other words, balanced capital accumulation. The accumulation governance structure is based on internal and external hierarchies which can be mixed up in different ways and degrees. At one extreme there is the possibility of a wide autonomy of the firms’ managers, who can be endowed with extended control prerogatives in production activity and investment decisions. They can fix the prices of their products and trade with each other in goods markets. The state takes care of overall growth and efficiency by monitoring the managers, assigning them price and quantity norms or guidelines, sustaining them with financial assistance, restraining them with budget constraints and orienting them with indicative or negotiated planning. Firms may be linked to each other and to the state apparatus through external hierarchies. This is what I call ‘decentralised state capitalism’. At the other extreme there is the possibility of a complete internalisation of hierarchy, with the national economy reduced to a single big company and the local managers confined to the position of department or division directors deprived of any strategic decision-making autonomy. In this case only final goods are exchanged in goods markets, whilst investment goods are traded at administered prices in internal pseudo-markets. Finance flows through hierarchical channels in implementing central planning decisions.16 This I call ‘centralised state capitalism’. Both extreme cases can give rise to an accumulation governance structure capable of playing disciplinary and selective functions in a way which is similar to that of bank-oriented corporate capitalism.

278 Different forms of capitalism Summing up, as far as the labour utilisation and the distribution-allocation subsystems are concerned, state capitalism is a fully-fledged capitalist form. What about the legal-political and ideological subsystems? Here is where contradictions and inconsistencies, as well as some conditions for institutional instability, can emerge. Some students believe that this system ceases to be capitalist and becomes socialist as soon as the conditions for a proper democracy are established. By this many people mean a Western-like form of procedural democracy: it would be sufficient for citizens to be allowed to vote for candidates to parliamentary elections who are pre-selected by two parties instead of just one. Since the means of production are owned by the state – goes the story – a formal democratic control of the state on the part of workers-citizens would make it a socialist state, i.e. a state through which ‘the producers control production’. I have some doubts. To start with, if a basic division of labour persists between a working class and a political class specialised in administration activity, and if some basic inequalities are perpetuated in the distribution of information, knowledge, political competence and income, one can hardly believe the workers are able effectively to control the political class, even if the latter is organised into 20 parties instead of one or two. The ruling class can even declare its pursuit of ‘the maximum satisfaction of the growing material and spiritual needs of the people’. But, if class ignorance persists, who prevents the politicians from interpreting this goal as coinciding with the ‘increase of national wealth’, i.e. with capital accumulation? The basic contradiction, however, goes deeper, and resides in the fact that, in a formally democratic state-capitalist system, the workers would relate themselves to the state in a twofold relationship: on the one hand, they are the citizens from whom the ruling class obtains legitimacy, on the other, they are state employees obliged to a ‘rigorous observance of labour discipline’. How can the producers effectively control production if they have to obey the state officers in the labour process? Obviously the workers’ control over the state must be ineffective if the state officers are committed to managing the labour process to increase its efficiency in order to sustain accumulation. The practical consequences of this contradiction bear on the institutional stability of the system. If there were many political parties, no political class would be confident of preserving power. The workers will continue to feel exploited and oppressed so long as they are submitted to a labour discipline oriented to capital accumulation, and will naturally identify their main enemy with the ruling employer. In such a system there could be recurrent élite changes until eventually a political group decides, in order to stabilise power, to abandon its role of single employer, privatise the means of production and take the position of a socially neutral political actor. No political class is at ease in the role of the main enemy of the people who grant legitimacy. The only way of dissolving, if not resolving, this contradiction is to stick to a one-party system, which precludes, however, the possibility of legitimising power through decent procedures of formal democracy. As a consequence, legitimisation must be ensured by a strong metaphysical and/or meta-historical ideology. The

Different forms of capitalism 279 ruling party must be considered as empowered, not by the people, but by God or History or some other transcendental entity. And the stronger the ideological grip of the legitimisation principle, the safer and more long-lasting is the political grip of the ruling class.17 Either way, whether it is governed through a multi-party or a one-party system, state capitalism is institutionally less stable than classical or corporate capitalism, the main reason being that a basic inconsistency exists between its legal-political and its labour utilisation subsystems.18 However, state capitalism can well survive by going hand in hand with corporate capitalism, thus giving rise to mixed forms of capitalism. A mixed corporate and state capitalist form could combine some advantages of the bank-oriented system, especially in terms of the ideological and legal-political conditions of legitimisation, with some virtues of state-owned firms that are particularly important in all the activities in which market failures may occur. There is ground to believe that this combination could reinforce the institutional stability of the system. Perhaps there is another reason for intrinsic weakness of state capitalism – one that was brought to light by the recent debacle of Soviet-type economies. It is well known that a field in which this kind of economy has failed more dramatically in the competition with corporate capitalist systems is that of innovative capacity. The reason for this failure is easily explainable if one accepts the view that those economies were forms of state capitalism rather than socialism. As was shown by Sutton (1968, 1971, 1973), the Soviet Union succeeded in growing fast so long as it could take advantage of imported technologies to overcome part of the technical gap with advanced economies. But, as soon as it became necessary to count on endogenous growth, many insurmountable difficulties emerged as a consequence of a certain systemic weakness in innovative capacity. The explanation of this weakness is now common knowledge. In historical experiments of state capitalism building most of the spurs and incentives to innovation associated with market competition and private appropriation were eliminated. The attempts to activate the non-economic stimuli associated with forum-like networks of knowledge communication were highly unsuccessful. Moreover, because of the strong bureaucratic and authoritarian climate of productive apparatuses, the decreasing returns to hierarchy in knowledge production have been exalted in those systems. However, I am not sure this is a built-in and fundamental ‘contradiction’ of state capitalism, because historical experience is still in progress (for instance in China), but also because the failure of one or a few experiments is unable to prove or disprove very much. The capitalist class in general An objection often raised against the notion of ‘state capitalism’ is that the abolition of private property brings about the abolition of the bourgeoisie. How is it then possible to speak of capitalist exploitation if the very actor of exploitation has disappeared? The objection is easily extended to the idea that managers are the real capitalists in a system where ownership is separated from control.

280 Different forms of capitalism The rationale for this criticism stems from a definition of the capitalist which is appropriate to classical capitalism. The bourgeoisie is the class of people who own the means of production. They are interested in the valorisation of capital and therefore control and use the means of production with the aim of extracting a surplus which will be reinvested to increase the capital stock. In this way a link is posited between individual aspirations to enrichment and the general process of capital valorisation. By virtue of the coupling of ownership and control, individual interests motivate a kind of behaviour that contributes to sustain overall capital accumulation. By contrast, in a system where ownership and control are separate, managers would act as employees of the (private or public) owners, and should be likened, in class analysis, to the other employees rather than to the owners themselves. Moreover, since they are remunerated with a salary, and not with profits, they would have no individual interest in profit maximisation. Any link between personal motivation to enrichment and capital accumulation would then be severed. The managerial class could not be defined as bourgeoisie, and an economic system dominated by the managers could not be considered a capitalist system proper. Thus several bizarre notions and theories have been elaborated by various authors, from the ‘new class’ to the ‘nomenklatura’, from ‘bureaucratic collectivism’ to ‘degenerate workers’ state’, which, far from working out the theoretical problem, in reality contributed to complicating it. The solution proposed by the anarchists, on the ground of Proudhon’s suggestion to reinterpret ownership in terms of possession or effective control, has recently been revived even by some non-socialist thinkers, but does not do away with all perplexities. In particular it does not account for the reasons why an economic actor who is deprived of any property title on the means of production, and therefore of any claim on residual income, should be interested in profit and accumulation. The idea to pose the lust of power at the base of the managers’ interests and motivations grasps an essential aspect of reality, but is not sufficient to work out the theoretical problem. This idea sheds no light on the characteristics by which the capitalist class differs from the political class; nor does it enable us to come to grips with the intrinsically capitalist nature of the dominating classes. It might lead to the construction of a general theory of political power, but it tends to blur the differences existing between capitalism, feudalism, oriental despotism and other forms of domination. I think that a step ahead on the road to a satisfying solution could be made by resorting to Bakunin (1911), in particular to his analysis of the role played by the social division of labour in determining the power and exploitation structure of capitalism. Bakunin focused on the isomorphism linking the ‘intellectual-practical abilities’ dichotomy to that of ‘command-execution functions’. After that, only one more step remains to be taken to find a solution to the problem: to recognise that intellectual and command capacities are considered capital, namely a form of ‘human capital’. Capitalist firms in general, but especially the modern mega-corporations, require managerial personnel endowed with a set of capacities which are rather complex and difficult to acquire, capacities that can be achieved only with years and years of study and experience, that is, through massive investments in ‘human capital’.

Different forms of capitalism 281 Part of this capital is generic, and is mostly produced by study outside the firm. Another part is specific, and consists in the knowledge attained from experience and career within a particular firm. Still another part, consisting of the knowledge of markets and trades, as well as the acquisition of a good managerial reputation, is predominantly generic, or is only trade-specific, but can be obtained through the activity undertaken within a firm. For the sake of synthesis, let me call the whole set of the managers’ ‘human capital’ assets a ‘control capacity’. The people who are endowed with it aspire to high returns from it, just like the owners of real and financial assets. From the standpoint of managers of capitalist firms, the profits distributed to shareholders are not a gain, but a sort of faux frais. If managers have effective control of production and utilisation of surplus value, they also control the distribution of profits and tend to handle this tool with the aim of balancing the firm’s financial structure so as to take care of external pressures exerted through financial markets. Normally they pay a minimum dividend and try to keep it stable. Their real interest is not maximisation of profits but valorisation of their ‘human capital’. The return on this kind of asset is a composite and rather complex magnitude, and includes, besides the monetary income directly earned by managers, also that received indirectly in the form of entertainment expenses, fringe benefits, various perquisites and, especially, in the form of satisfaction produced by prestige, status and power. This latter good in particular, as argued in Chapter 4, represents the fundamental motivation of managerial actions. Now it can be treated as the principal component of the returns on the managers’ ‘human capital’. All these components, at any rate, tend to grow as a function of the firm’s size. Thus managers are induced to promote expansion of the firm size, not of its shortrun profits. Not only the returns, but also the worth of managerial capacities grows with the size of the firm. Finally, the firm’s success nurtures the reputation of managers. The instrumental variable in strategic decisions is the growth of the firm’s overall capital. This may seem strange, since the managers’ ‘human capital’ is only part of the firm’s assets, and one might think that managers should aim only at the return on their own assets rather than that on overall capital or, worse still, that they should pursue their own objectives to the detriment of the other components of the firm’s assets. But there are strong complementarities among all the capital components, so that it is impossible to make them grow separately. Those who have an interest in the growth of one of them (the one I called the ‘control capacity’) and, by virtue of it, hold power over all the others, will tend to use this power to serve that interest. But to make the managers’ capital grow they have to make all the other assets grow. The workers too have some ‘human capital’. They however signed an employment contract whereby they gave up their faculty to decide autonomously their own activity in the production process. Therefore they cannot exert any substantial control on the use of their labour capacities. Capitalists, having acquired a command prerogative, can indeed exert that control. One could say that the managers, by virtue of their control capacity, i.e. of a characteristic of their own ‘human capital’, also acquire control over the use of the workers’ capacities. And it is through this control that they can perpetrate exploitation. They will use workers

282 Different forms of capitalism to pursue the goals of the firm and their own goals. Managers appropriate residual income, the net returns on all ‘human capital’ used in the firm, their own and that of the workers, and can do so because they have control. The workers, notwithstanding their endowment of technical capacities, cannot avoid exploitation. From this point of view, ‘human capital’ is no different from any other form of wealth: ownership is not sufficient to avoid exploitation, because it does not warrant control over the uses of complementary goods. The workers’ ownership of their own ‘human capital’ does not warrant control over the use of managerial capacities. In the first chapter of this book I proposed a general definition of the working class. Now, at the end of the book, I am able to present a general definition of the capitalist class: this is the group of individuals who are endowed with that particular kind of ‘capital’ called ‘capacity of control’, and who therefore can exert command on the use of the firm’s assets and the worker’s ‘human capital’. On sources The notion of ‘mode of production’ used here comes from Marx, but does not precisely coincide with that used in Marxist theory. In the Introduction to A Contribution to a Critique of Political Economy, a mode of production is defined on the grounds of an ‘economic base’ and a legal-political and ideological ‘superstructure’, the economic base being articulated into sets of ‘productive forces’ and ‘social relations’. My approach differs from this mainly because, instead of focusing on social and material relations proper, I focus on the institutions that regulate them. Moreover, instead of the notions of ‘economic base’, ‘forces of production’ and ‘social relations’, I use those of ‘labour utilisation subsystem’ and ‘distribution-allocation subsystem’, the latter being successively split into those of ‘property rights regimes’ and ‘accumulation governance structures’. These differences are less substantial than might seem at first glance. They boil down in fact to a differentiation in method: in the present chapter I practise institutional analysis rather than social analysis proper. A more relevant divergence stems from the tendency of Marx and most Marxists to consider the economic base more important than the legal-political and ideological superstructure, even worse, often to reason as if there were a one-way causal link between the former and the latter. On this problem I took advantage of a Gramsci’s teaching: that the ideological and political spheres are no less important than the economic one in determining the structure and dynamics of capitalism, and that the causal links between the different spheres are complex, multifaceted and certainly not one-sided. Yet there is an aspect in which the labour utilisation subsystem is more important than the others: it is the sphere where the most fundamental institution of capitalism is defined. The notion of ‘institutional system’ introduced in Chapter 3 and expounded here has been developed and used by many contemporary students. Its revival has perhaps been stimulated by the emergence of social system theory, on which see S. Deakin and J. Michie (1997), but the extensive literature produced so far is not confined to this approach. Interesting work has also recently been done on the

Different forms of capitalism 283 analysis of ‘sub-systems’. See P. Hall (1984, 1986), R. Whitley (1992) and C. Lane (1994) for applications to business systems and industrial relation orders, G. Teubner (1993) for the theory of legal systems, M. L. Tushman and L. Rosenkopf (1992) for the notion of ‘technological core sub-system’. M. McKelvey (1991) offers a perceptive survey of the research on national systems of innovation. The theory of ‘capitalist forms’ or ‘institutional forms of capitalism’ freely draws from some streams of thought that flourished in the 1980s, particularly the so-called ‘Regulation school’ and ‘Social Structure of Accumulation school’. Literature in this field is so extensive now, that I hope to be excused if I cite only some essential works: M. A. Aglietta (1979), A. Lipietz (1987), G. D. De Bernis (1990), D. M. Gordon (1980), D. M. Gordon, R. Edwards and M. Reich (1982), J. O’Connor (1984), S. Bowles, D. M. Gordon and T. E. Weisskopf (1986), H. Gulolp (1989), D. M. Kotz (1990), R. Boyer (1999). W. Samuels (1971) investigated the interaction between the legal and economic subsystems as part of a social stabilisation process whereby the self-sustaining structure of property rights and power relations are endogenously determined. The idea that institutional stability hinges on the interaction between the technological structure and the property rights regime of a capitalist form was developed by U. Pagano (1991a, 1991b) and U. Pagano and R. Rowthorn (1994), by combining contributions of some neo-institutional economists, such as S. Grossman and O. Hart (1986) and O. Hart and J. Moore (1990), and some radical economists, such as H. Braverman (1974), R. Edwards (1979), S. A. Marglin (1974), D. M. Gordon, R. Edwards and M. Reich (1982). G. M. Hodgson (1984, 1988, 1993) developed the idea that no institutional system can ever be pure and that institutional variety plays an essential role in sustaining institutional evolution. On classical capitalism I limit myself to refer to three classics: Adam Smith’s Wealth of Nations, Karl Marx’s Capital, and Karl Polanyi’s The Great Transformation. The idea that class exploitation is based on the social division of labour between the people playing intellectual-and-command activities, on the one hand, and those compelled to do practical-and-execution work, on the other, was developed by the anarchists. Particularly enlightening are the contributions of M. Bakunin (1911), C. Berneri (1964) and J. W. Makhaïski (1979). Contemporary literature on modern capitalism and the role banks and the markets for corporate control play in governing accumulation is boundless. Let me therefore refer to just two comprehensive general treatments, namely, P. Milgrom and J. Roberts (1992) and F. Barca (1994). Some more specific contributions I have taken advantage of are: M. Bianco (1994), T. Bresnahan, P. Milgrom and J. Paul (1991), M. Burkart, D. Gromb and F. Panunzi (1997), K. Iwai (1999), J. Stein (1989), J. F. Weston, K. Chung and S. Hoah (1990) and R. Zeckhauser and J. Pound (1990). The literature on state capitalism mainly developed in the form of a Marxist debate on the social nature of the Soviet Union. I will limit myself to referring to just three classic works originally published in the 1940s, A. Bordiga (1976), T. Cliff (1974), and R. Dunayevskaya (1992) and to two more recent contributions I found particularly enlightening, P. Chattopadhyay (1987) and P. Zarembka (1992).

284 Different forms of capitalism Finally, on the role played by hierarchy and pay in disciplining and motivating the functionaries of capital, the following contributions are instructive: H. F. Lydall (1959), P. B. Doeringer and M. J. Piore (1971), H. A. Simon (1979, 1982), E. Lazear and S. Rosen (1981), R. A. Lambert, D. F. Larcker and K. Weiglet (1989), J. M. Abowd (1990), C. Jensen and K. J. Murphy (1990) and S. Rosen (1990). In classical political economy and in most of Marx’s and Engels’ works the capitalist was identified with the owner of the means of production. This is the result of a misperception caused by some ‘imperfections’ and historical contingencies of the British form of classical capitalism – a form in which the managerentrepreneur of a capitalist firm tended to coincide with the owner. Marx himself, however, elaborated the categories required to criticise that theory. He did it in two fundamental yet little known chapters of Capital, III, namely the 23rd and the 27th, as well as in the Addenda of the Theories of Surplus Value, III. He anticipated A. A. Berle and G. C. Means (1933) when he distinguished the capitalist as the ‘pure owner of capital’ from the manager as a ‘functioning’ or ‘industrial’ or ‘active’ capitalist, and when stated that ‘the mere manager, who does not possess capital under any title, neither by loan nor in any other way, takes care of all real functions that fall to the functioning capitalist as such’ (1990, III: 512). I endeavoured to elaborate these categories in Screpanti (1994, 1998). This chapter develops materials I already dealt with in three essays (Screpanti, 1992, 1999a, 1999b). The second of these essays defines the notion of an ‘institutional system’, while the first elaborates the idea that the different types of capitalism observed in history are transitional forms tending to ‘pure capitalism’. The reconstruction of different institutional forms of capitalism on the ground of the concepts of ‘property rights regimes’ and ‘accumulation governance structures’ is attempted in the third essay.

Notes 1 For the moment this definition of self-consistency of institutions will suffice. I will return to the argument in another section in which I delve into the question of institutional stability. 2 Note that a core subset is an element of a set which is not fully organised into a system, while a subsystem is an element of a system and part of a core subset. 3 It can be considered as a system of institutions (rather than beliefs) if one accepts the extension of the concept suggested in note 4 of chapter 2. 4 This terms aims at abstracting from the different traditions of thought existing in modern capitalist systems. It broadly translates notions like ‘Rechtsstaat’, ‘stato di diritto’, ‘limited government’ or ‘rule of law’. 5 Note that a constitutional state so defined need not be a democratic state. The rule of law is compatible with a liberal state with limited franchise as well as with a totalitarian ‘people’s republic’, not to speak of the modern ‘mediocracies’ (the rule of the mediocre by means of the media). As a matter of fact, as Jurgen Habermas (1973) observed, the modern constitutional state has always been in contradiction with the democratic principle. 6 These bonds of subordination are ‘formal’, not personal as in a feudal system. They are explicitly defined by transaction and normative institutions and pertain to roles not to the persons holding them. 7 The distribution of executive salaries fits Pareto law rather well. This fact can be explained on the grounds of two assumptions, namely, that the number of direct subordinates of an executive does not vary greatly and that the ratio of an executive’s salary to his direct subordinates’ salary

Different forms of capitalism 285

8

9 10

11

is fairly constant (Simon 1979, 1982). The two assumptions imply that both the number of total subordinates of an executive and the executives’ pay grow exponentially with the position on the hierarchical ladder and with firm’s size. This is not the right place to make efficiency comparisons among real economic systems, nor to focus on the changes they are presently undergoing. After all, I am only interested in model building, here. Therefore I will ignore both the debate on the effectiveness of corporate control markets and the institutional transformations the American system of external control has been enduring in the 1990s. As is well known, inter-firm mobility of top managers is higher in the American system than in the German and Japanese. In the advanced capitalist countries of the nineteenth century the electoral franchise was restricted on many grounds, gender, race, literacy, wealth, income. These restriction bases could be combined in various ways in different countries, but they all conspired against an extended form of democracy. In France a revolution was needed in 1830 to bring the number of electors to 240,000. In Great Britain the Reform Act of 1832 increased the number of electors from 500,000 to 813,000. Around mid-century in this country there were 1 million electors in a population of 27.5 millions. Then the Reform Acts of 1867 and 1883 brought the number of electors to 29 per cent of the male adult (over 20 years old) population. A brief detour on eighteenth- and nineteenth-century capitalist ideologies may be useful to bring to light the nature of some of these justifications. One view was based on natural law philosophy. It had been put forward in its most authoritative form by John Locke (1960). Property rights are natural rights deriving from the ownership any individual has on his person, his labour and the products and instruments of his labour activity. Indeed many enlightened philosophers intended personal freedom and the citizens’ rights as grounded on self-ownership. Listen to Genovesi (1765, I, pp. 12-13): ‘What are a person’s primitive rights? I call a right the moral faculty to make use of what one owns […] Nature endows us with [a mind and a body] and, although these can each be modified in various ways, they cannot be separated from us. Now, whatever is part of my nature, and cannot be separated from me, is mine by nature […] thus it is my natural ownership; and therefore my natural right.’ It follows that property rights, as natural rights of freedom, must be defended by any state whose rule of law is based on natural rights. Since the parliamentary representatives’ authority is delegated by the people, and since public administrators have to defend citizens’ rights, only wealth owners must have the faculty to give up a part of their freedom by delegating it to Parliament. The weakness of this argument is apparent: if a natural right to property is derived from a natural right on labour capacity, why should political rights be conceded only to the owners of accumulated labour and not also to the owners of living labour? This was in fact a major argument of the democratic and socialist parties that fought in the early nineteenth century to overcome the gap between ‘the legal country and the real country’. A more compelling and theoretically valid justification for restricted franchise was based on the view that freedom implies independence. The right to elect political representatives should be assigned only to persons who can enjoy real freedom. This implied not only that the possession of wealth, as a base of economic independence, was a condition of political freedom, but also that women, slaves, servants, indentured and waged workers could not be recognised any political right. These people were dependent on their masters, employers, fathers and husbands. They had no autonomy of interests and thought, so that their will could effectively be represented by their superiors. There is a simple truth in this theory: that real freedom is not just a formal property of legal systems, and that legal freedom disjoined from the material bases of individual autonomy is a fake. A slave gave up his decisionmaking faculty with an ‘original slavery covenant’, a woman with marriage, a wage-earner with an employment contract. The lives of these persons are marked by an obligation to obedience. They have lost any real freedom by virtue of a transaction institution that cannot be contrasted by the normative institutions produced by a state based on ‘limited government’. The trick, of course, is to assume that obligations to obedience have been undertaken willingly and autonomously. With these undertakings slaves, servants, wage workers and women ‘freely’ gave up their freedom and decision-making capacity. The right to elect a political delegate

286 Different forms of capitalism implies the faculty to renounce a part of personal decision-making capacity and empower an agent with it. Which part of personal decision-making ability can be delegated by those who have none? To this must be added the observation that a person who spends most of his life at work, and who earns a subsistence wage or no personal income at all, cannot acquire the culture and the competence required to exert political rights. 12 Assuming expected dividends (and profits) to grow at a constant rate, gd, equal to the growth rate of capital assets, g, minus the growth rate of new share issues, s, the market value of a firm is ∞

K = (1 − ν )π

∑ (1 + g ) / (1 + r )  = (1 − ν )π (1 + g − s ) / ( r − g + s ) t =1

13

14

15 16

17

18

t

d

where n represents the retention ratio, p the present profit and r the rate of interest. The market value of the firm is an increasing function of present profits and the growth rate of capital, and a decreasing function of the retention ratio, the growth rate of new share issues and the rate of interest. Most contemporary constitutions, while protecting property rights, attribute them a public function and reserve the state the right to limit and expropriate private ownership in view of public welfare. Since the speculators have a short-term horizon, they are not even interested in forming longterm expectations, and often focus their attention on guessing the other speculators’ expectations rather than evaluating the fundamentals. This injects volatility into market values. As a consequence, market-oriented corporate capitalism, although institutionally stable, may be financially fragile and highly unstable from a dynamic point of view. Speculative bubbles and financial crashes often spill over the stock exchange and hit the real economy, thus amplifying the business cycle. The quotations in this section are taken from the Soviet constitution of 1977. If it is reasonable to conceive of a capitalist form in which all wealth is state-owned and all production activity is organised into one big state company, a proper label for it would be ‘hyperstate capitalism’. A good historical example is offered by Vatican City with its Church hierarchy and its huge global company, one of the most efficient, long lasting and fast growing organisations in human history. Lenin once said that the Communist Party ruled in the Soviet Union on the strength of a ‘historical right’. He seemed to have a teleological view of history of the Hegelian kind which enabled him to conceive the leading group of the Communist Party as the embodiment of Zeitgeist. The communist vanguard understood the sense of history better than the other politicians (as its ability to conquer power proved) and, despite the fact that that sense consisted in a conception of history as a process of proletarian and human conscience awakening, better than the proletarians themselves (who were in fact not acquainted with the developments of Western philosophy). Therefore, in their interests, the proletarians should have been directed and educated by the Communist Party. History has proved that this kind of legitimisation principle was rather shaky and that the institutional system founded on it was institutionally unstable. It is interesting to note that another experiment of state ‘socialism’ was more long-lived: the Jesuit republic of Paraguay ruled over more than 150,000 souls, lasted almost one century and broke down by an external shock rather than by internal contradictions. Evidently God’s Providence has a stronger grip on the people’s minds than Dialectical Materialism. Thus state capitalism could turn out to be useful in some special historical situations of economic under-development, where a strong and rapid industrial take off is needed. State capitalism, by virtue of its totalitarian organisation, can help in strengthening national efforts, while it can help in keeping it transitional by virtue of its institutional instability.

Conclusions 287

Conclusions

Towards the autonomy of capital What do all the capitalist forms presented in Chapter 6 have in common that makes them particular cases of a unique mode of production called ‘capitalism’? The question was already posed and answered at the beginning of Chapter 1. Now, at the end of the book, I would like to reformulate both the question and the answer in an historical vein. This reformulation is facilitated by an evolutionary reading of the classification pattern reconstructed in the foregoing sections – a reading that looks at the different capitalist forms as stages of a development process and puts them in perspective. If it is true that the anatomy of man also explains the anatomy of the monkey, then that reading should give us a clue as to the essence of capitalism in general. And it seems that the evolution of modern economies tends toward a pure capitalist form in which the ‘autonomy of capital’ settles and consolidates. One of the basic features of capitalism was brought to light by Marx. It consists in its tendency to destroy all forms of human relations based on personal bondage. Marx focused his attention on the institutions regulating the servant-master relationship: slave or feudal bonds were considered incompatible with capitalism. But also other societal structures, like the family, were believed to be undergoing a de-structuring transformation process in which the ‘natural’ cement was being increasingly replaced by legal and formal instruments of aggregation and constitution. Since the basic cell of the capitalist mode of production is commodity, all social relations tend to be reduced to exchange relations. Social exchange is mediated by the market, and this implies that all social actors must be endowed with freedom of contract and, as a consequence, with a vast series of rights of freedom which make the individuals unfettered by any personal bondage. Yet, in classical capitalism, the very essence of capital, i.e. control of the production process in the firm, was still based on a sort of personal relation. Living labour is dominated by ‘dead labour’, but this exists as an active subject only in as much as it is privately owned by a person. The capitalist firm is not a person, but a thing belonging to a particular individual. The entrepreneur obtains control of the labour process only because he is the owner of the means of production. As a consequence, the selection and replacement of entrepreneurs, as regulated by family relations, the family law and the law of inheritance, poses an institutional constraint

288

Conclusions

on the efficient allocation of control. In fact, although the family could work as an educational apparatus for the transmission of entrepreneurial competence, there is no guarantee that hereditary transmission of ability works efficiently. This problem becomes more and more serious as accumulation progresses and firms grow in size, because the accumulation of firm-specific entrepreneurial competence and ability is increasingly acquired through internal careers and massive investments in ‘human capital’ that the family as such is unable to provide, while the growth in personal wealth induces risk diversification attitudes that tend to weaken the owners’ affection to a particular firm. The latter effect is reinforced by the tendency of family wealth to be dispersed among many heirs. The solution to this problem is coming forth in a process of institutional evolution that is breaking all linkage between the claimants on residual control and the claimants on residual income. With the development of the joint stock company, the firm becomes a legal person endowed with rights and duties which do not coincide with those of the shareholders, while, with the dispersion of shareholding, the owners become rentiers who are de facto indistinguishable from the bondholders and other kinds of creditors. By the same process that tends to transform the firm into a person, capital becomes a thing at the firm’s disposal. This process approaches completion in the Japanese Keiretsu and the German-type corporate capitalism. The Keiretsu consists of a company coalition held together, not only by personal ties among managers, but also by complex structures of cross-shareholding linkages. A group of 12 top companies, constituting the core of a Keiretsu, can become completely autonomous from external shareholders if each company possesses 5 per cent of shares in the others: the majority of shares of each member of the group is possessed collectively by the other members. As a consequence the managers, without personally possessing a single share, can legally control the entire group if they act in concert. But they can of course control a wider set of companies if the core group has relevant stakes in other, subordinate members of the Keiretsu. The German system is even more revolutionary. Manufacturing companies are organised in hierarchical structures sustained by financial fetters and shareholding blocks led by banks. Each bank has command over at least one of these structures, so that personal shareholders are deprived of any residual control. The interesting point is that each bank is controlled by no shareholder other than the other big companies (mainly banks), so that the bank managers, even if they do not own shares, can collectively control the entire system without being obliged to account to outsiders. The banks’ capital is collectively self-owned.1 This poses three crucial theoretical problems: who is the formal subject of the accumulation and production processes, if it is not a personal owner? who is the material subject, if the formal one is not a physical person? which institution is at the basis of labour exploitation, if it is not the private ownership of the means of production? These problems had already been implicitly worked out by Marx, who did so by playing with a Hegelian dialectic reasoning according to which the totality of

Conclusions 289 capital, through the laws of competition and accumulation, compels the individual owners to serve its own goals, i.e. the goals of a universal subject. Thus the capitalists are properly, although still in a vague sense, ‘functionaries of capital’. Marx also clarified that capital is a social relation and not just a thing, meaning that capital constitutes itself as a self-valorising subject only when it establishes a social relation with wage labour. The problem with these ideas is that they reduce capital to an hypostatised and unreal subject. You can surmise how it operates, but you cannot see it. And you cannot see it because the firm, in classical capitalism, is not a person but a thing belonging to a person. How can a thing be a subject? And how can a person be an instrument of a thing, let alone of the totality of things? Yet the evolution of capitalism has confirmed Marx’s arguments, in the twofold sense that it has proved their validity and has given them a real content. The formal subject of the accumulation and production processes in pure capitalism is the capitalist firm or, rather, the set of capitalist firms which are collectively self-owned. These firms are legally the owners of their means of production and not things belonging to individual owners. Therefore they are not hypostatised subjects, but real social entities endowed with contractual capacity: from a formal point of view they are able to enter transaction and contractual relations with any other subject. This works out the first theoretical problem. The second too has been worked out by the evolution of capitalism. The capitalists, i.e. the material subjects of capital accumulation, are the ‘functionaries of capital’, the officers of the capitalist firms. They are no longer the owners of capital, nor the shareholders’ agents. But, to the extent that firms are self-owned personal entities, they are properly the functionaries of the firms. An organisation may be the formal subject; it cannot be the material subject, though, because an organisation as such does not have a mind and cannot take decisions. Its mind is in reality the mind of its managers. Its decisions are its managers’ decisions. Its goals are the managers’ goals. And since these, by virtue of a personal interest in power, pay and prestige, coincide with the firms’ growth, we can say that the goal of capitalist firms is capital accumulation. Moreover the personal interests of the managers cannot be arbitrary, since their activity is regulated by accumulation governance structures that discipline, finance and select the individuals in such a way that they have to serve capital accumulation: only those managers who correctly define their personal goals and efficiently pursue them are selected and rewarded. The correct personal goals are those coinciding with the growth of the firm. Finally, the evolution of capitalism also contributed to give real content to the view that capital is a ‘social relation’. Marx gave a twofold meaning to this concept. He intended it as the authority relationship constituted through the exchange of labour power; and as the functional relation by which dead labour subsumes living labour in its self-valorisation process. But the authority relationship established in the employment contract, in classical capitalism, is still a relationship between a personal employer and a worker. Worse still, it is not an authority relationship proper, i.e. a power relationship formally constituted as the function of a role and

290

Conclusions

a position in an organisation hierarchy. Furthermore, insofar as ‘dead labour’ as such does not enter as a party in the employment contract, one cannot understand how it can subsume anything under itself. All residuals of dialectic mist are dissipated when we realise that dead labour is none other than the capitalist firm. It really can subsume living labour in its self-valorisation process if it is a legal subject. It does this by means of the employment contract, which is now to be interpreted, not as a transaction between two individual persons, but as an institution constituting the social relations in which the firm consists. In entering into the employment contract a worker undertakes the obligation to work under the command of the other party. The other party pays a wage in exchange for the prerogative of commanding labour activity. But it is an organisation. It is structured in the form of a hierarchy of functional positions endowed with authority and constrained by obligations. This very structure is the structure of capitalist social relations. The firm is constituted by the employment contracts into which it enters and from which its internal authority structure originates. In fact it cannot work, it cannot even exist without such relations. The capitalist firm is a nexus of contracts of employment. Thus the relation linking the workers to capital is not a relation with a person. The firms’ officers who exert command over labour, do not do it in the form of personal power, but by virtue of their role in the firm’s hierarchy, i.e. in a form legitimised by an authority position. Their power is legally delegated by the firm, and ‘dead labour’ really and formally consists in the organisation. Now it is possible to understand the deep sense in which capital is a social relation: it is the system of obligations, prerogatives and constraints that regulate labour activity in the capitalist firm.

Notes 1 Another case of self-ownership of capital is that of a firm who buys the majority of its own shares. If self-owned shares have a right of vote at the shareholders’ meeting, then that firm is considered as legally self-owned and, in fact, legally controlled by its managers. Self-owned shares, though, are not entitled to a right of vote in many modern legal systems. Yet this is a residue of a traditional conception of the firm as an entity possessed by external shareholders. It is a convention and, as such, may be changed or circumvented. Indeed, this legal constraint to control through self-ownership can be circumvented with the practice of stimmenleihe (bank A lends its votes in bank B’s shareholder meeting to bank B’s managers, who normally return the favour in due time).

References 291

References

Abowd J. M. 1990, ‘Does Performance Based Managerial Compensation Affect Corporate Performance?’ Industrial and Labour Relations Review. Aglietta M. A. 1979, A Theory of Capitalist Regulation: The U.S. Experience, London. Alchian A. and Demsetz H. 1972, ‘The Property Rights Paradigm’, Journal of Economic History. Alford R. R. and Friedland R. 1985, Powers of Theory: Capitalism, the State, and Democracy, Cambridge. Altvater E. 1973, ‘Notes on Some Problems of State Intervention’, Kapitalistate. Amariglio J. L. and Ruccio D. F. 1994, ‘Postmodernism, Marxism, and the Critique of Modern Economic Thought’, Rethinking Marxism. Amariglio J. L. and Callari T. 1993, ‘Marxian Value Theory and the Problem of the Subject: The Role of Commodity Fetishism’, in E. Apter and W. Pietz (1993). Antonelli G. and De Liso N. 1997 (eds), Economics of Structural and Technological Change, London. Apter E. and Pietz W. 1993 (eds), Fetishism as Cultural Discourse, Ithaca, NY. Arestis P. 1992, The Post-Keynesian Approach to Economics: An Alternative Analysis of Economic Theory and Policy, Aldershot. Arrow K. J. 1971, ‘Economic Welfare and the Allocation of Resources for Invention’, in NBER, The Rate and Direction of Incentive Activity: Economic and Social Factors, Princeton, 1962; reprinted in D. M. Lamberton (1971). Arrow K. J. 1994, ‘The Production and Distribution of Knowledge’, in G. Silverberg and L. Soete (1994). Arrow K. J. 1996 (ed.), The Rational Foundations of Economic Behaviour, New York. Ashton D. and Green F. 1996, Education, Training and the Global Economy, Cheltenham. Bacharach P. and Baratz M. S. 1970, Power and Poverty, New York. Bakunin M. 1911, ‘Les endormeurs’, in Oeuvres, vol. V, Paris. Bakunin M. 1963, Archives Bakounine, vol. I. (A. Lehning, ed.), Leiden. Barca F. 1994, Imprese in cerca di padrone: Proprietà e controllo nel capitalismo italiano, Rome. Barrow C. W. 1993, Critical Theories of the State: Marxist, Neo-Marxist, Post-Marxist, Madison. Barry B. 1989, Democracy, Power and Justice, Oxford. Bartlett R. 1989, Economics and Power: An Enquiry into Human Relations and Markets, Cambridge. Bellanca N. and Giocoli N. 1998, Maffeo Pantaleoni: Il principe degli economisti italiani, Florence.

292

References

Bellofiore R. 1998 (ed.), Marxian Economics: A Reappraisal, 2 vols, London. Berle A. A. 1967, Power, New York. Berle A. A. and Means G. C. 1933, The Modern Corporation and Private Property, New York. Berneri C. 1964, ‘Lo stato e le classi’, in Scritti scelti: Pietrogrado 1917 – Barcellona 1937 (C. Masini and A. Sordi, eds), Milan. Bianchi M. 1998 (ed.), The Active Consumer: Novelty and Surprise in Consumer Choice, London. Bianco M. 1994, ‘Il controllo della public company’, in Banca d’Italia, Contributi all’analisi economica, numero speciale: Il mercato della proprietà e del controllo delle imprese: aspetti teorici e istituzionali, Rome. Block F. 1977, ‘The Ruling Class Does Not Rule: Notes on the Marxist Theory of the State’, Socialist Revolution. Block F. 1980, ‘Beyond Relative Autonomy: the State Managers as Historical Subjects’, Socialist Register. Bolton P. and Farrell J. 1990, ‘Decentralization, Duplication and Delay’, Journal of Political Economy. Borcherding T. E. 1983, ‘Toward a Positive Theory of Public Sector Supply Arrangements’, in J. Roberts and S. Prichard (1983). Bordiga A. 1976, Struttura economica e sociale della Russia d’oggi, Milan. Bornstein S., Held D. and Krieger J. 1984 (eds), The State in Capitalist Society, London. Boulding K. E. 1956, The Image: Knowledge in Life and Society, Ann Arbor. Boulding K. E. 1971, ‘The Economics of Knowledge and the Knowledge of Economics’, American Economic Review, 1966; reprinted in D. M. Lamberton (1971). Bowen H. R. 1943, ‘The Interpretation of Voting in the Allocation of Economic Resourses’, Quarterly Journal of Economics. Bowles S. and Boyer R. 1988, ‘Labor Discipline and Aggregate Demand’, American Economic Review: Papers and Proceedings. Bowles S. and Gintis H. 1983, ‘The Power of Capitalism: On the Inadequacy of the Conception of the Capitalist Firm as Private’, Philosophical Forum. Bowles S. and Gintis H. 1988, ‘Contested Exchange: Political Economy and Modern Economic Theory’, American Economic Review. Bowles S. and Gintis H. 1992, ‘Power and Wealth in a Competitive Capitalist Economy’, Philosophy and Public Affair. Bowles S. and Gintis H. 1993, ‘The Revenge of Homo Economicus: Contested Exchange and the Revival of Political Economy’, Journal of Economic Perspectives. Bowles S. and Gintis H. 1996, ‘The Distribution of Wealth and the Viability of the Democratic Firm’, in U. Pagano and R. Rowthorn (1996). Bowles S., Franzini M. and Pagano U. 1999 (eds), The Politics and Economics of Power, London. Bowles S., Gordon D. M. and Weisskopf T. E. 1986, ‘Power and Profits: the Social Structure of Accumulation and the Profitability of the Post-War U.S. Economy’, Review of Radical Political Economics. Boyer R. 1999, ‘The Variety and Dynamics of Capitalism’, in J. Groenewegen and J. Vromen (1999). Braverman H. 1974, Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century, New York. Bresnahan T., Milgrom P. and Paul J. 1991, ‘The Real Output of the Stock Exchange’, in Z. Griliches (1991).

References 293 Bridges A. B. 1973, ‘Nicos Poulantzas and the Marxist Theory of the State’, Politics and Society. Bulow J. B. and Summers S. 1986, ‘A Theory of Dual Labor Markets with Applications to Industrial Policy, Discrimination and Keynesian Unemployment’, Journal of Labour Economics. Burkart M., Gromb D. and Panunzi F. 1997, ‘Large Scale Shareholders, Monitoring, and the Value of the Firm’, The Quarterly Journal of Economics. Butkiewicz J. L. 1986 (ed.), Keynes’ Economic Legacy: Contemporary Economic Theories, New York. Calvo G. 1979, ‘Quasi Walrasian Theories of Unemployment’, American Economic Review. Cannoni A. 1999, La partecipazione politica in Europa: Un’analisi comparativa dei dati Eurobarometro, Degree dissertation, University of Siena. Carver T. 1998, The Postmodern Marx, Manchester. Cazzaniga M. S. 1995 (ed.), The Cognitive Neurosciences, Cambridge, MA. Chattopadhyay P. 1987, ‘On the Marxian Category of “Competition of Capitals” and Its Relevance for “Postrevolutionary Economy”’, Research in Political Economy. Cicero M. T. 1962, De officiis (C. F. W. Müller, ed.), Rome. Cillario L. and Finelli R. 1998, Capitalismo e conoscenza: L’astrazione del lavoro nell’era telematica, Rome. Cirino G. 1999, ‘Il lavoro “cognitivo” nella fase dell’accumulazione flessibile: uno schema interpretativo del “fenomeno” dei cosiddetti “lavoratori della conoscenza” ’, Proteo. Clarke S. 1983, ‘State, Class Struggle, and the Reproduction of Capital’, Capitalistate. Coase R. H. 1937, ‘The Nature of the Firm’, Economica. Coase R. H. 1960, ‘The Problem of Social Costs’, Journal of Law and Economics. Cohen G. A. 1995, Self-Ownership, Freedom, and Equality, Cambridge. Coleman J. S. 1986, Individual Interests and Collective Action: Selected Essays, Cambridge. Commons J. R. 1924, Legal Foundations of Capitalism, New York. Conrath D. W. 1973, ‘Communication Environment and its Relationship to Organisational Structure’, Management Science. Conte R., Hegselemann R. and Terna P. 1997 (eds), Lecture Notes in Economics and Mathematical Systems, Berlin. Cullis J. and Jones P. 1987, Microeconomics and the Public Economy: A Defence of Leviathan, Oxford. Dahl R. A. 1957, ‘The Concept of Power’, Behavioral Science. Dallago B., Brezinski H. and Andreff W. 1992 (eds), Convergence and System Change: the Convergence Hypothesis in the Light of the Transition in Eastern Europe, Aldershot. Damasio A. B. 1994, Descartes’ Error, New York. Dasgupta P. and David P. A. 1985, ‘Information Disclosure and the Economics of Science and Technology’, Discussion Paper, Centre for Economic Policy Research, Cambridge. Dasgupta P. and David P. A. 1988, ‘Priority, Secrecy, Patents and the Socio-Economics of Science and Technology’, Discussion Paper, Centre for Economic Policy Research, Cambridge. De Bernis G. D. 1990, ‘On a Marxist Theory of Regulation’, Monthly Review. De Jouvenel B. 1962, On Power: Its Nature and the History of its Growth, Boston. de Sousa R. 1987, The Rationality of Emotions, Cambridge, MA. Deakin S. and Michie J. 1997, ‘The Theory and Practice of Contracting’, in S. Deakin and J. Michie (eds), Contracts, Co-operation, and Competition: Studies in Economics, Management, and Law, Oxford.

294

References

Doeringer P. B. and Piore M. J. 1971, Internal Labor Markets and Manpower Analysis, Lexington, MA. Doise W. 1993, Logiques sociales dans le raisonnement, Neuchâtel. Downs A. 1957, An Economic Theory of Democracy, New York. Dunayevskaya R. 1992, The Marxist-Humanist Theory of State Capitalism, Chicago. Earl P. E. 1983, The Economic Imagination: Toward a Behavioural Analysis of Choice, Brighton. Earl P. E. 1984, The Corporate Imagination: How Big Companies Make Mistakes, Brighton. Easton D. 1953, The Political System, New York. Edwards R. 1979, Contested Terrain: the Transformation of the Workplace in the Twentieth Century, New York. Egidi M. 1992, ‘Organisational Learning, Problem Solving and the Division of Labour’, in M. Egidi and R. Marris (1992). Egidi M. 1996, ‘Routines, Hierarchies of Problems, Procedural Behaviour: Some Evidence from Experiments’, in K. J. Arrow (1996). Egidi M. 1997, ‘Technological and Organisational Innovations as Problem Solving Activities’, in G. Antonelli and N. De Liso (1997). Egidi M. and Marengo L. 1995, ‘Division of Labour and Social Co-ordination Models: A Simulation Model’, in G. Nigel and R. Conte (1995). Egidi M. and Marris R. 1992 (eds), Economics, Bounded Rationality and the Cognitive Revolution, Aldershot. Egidi M. and Messori M. 1995, ‘Teoria economica e teoria dell’organizzazione’, in G. Rodano (1995). Egidi M. and Ricottilli M. 1997, ‘Co-ordination and Specialisation as Co-evolutionary Processes’, in R. Conte, R. Hegselemann and P. Terna (1997). Ellerman D. P. 1992, Property and Contract in Economics: The Case for Economic Democracy, Oxford. Elster J. 1979, Ulysses and the Sirens, Cambridge. Elster J. 1983, Sour Grapes: Studies in the Subversion of Rationality, Cambridge. Elster J. 1986 (ed.), Multiple Self, Cambridge. Elster J. 1998, ‘Emotions and Economic Theory’, Journal of Economic Literature. Engels F. 1955, Anti-Dühring: Herr Eugen Dühring’s Revolution in Science, London. Engels F. 1972, The Origin of the Family, Private Property and the State, London. Engels F. 1978, ‘Introduction’ to K. Marx, The Civil War in France, in K. Marx and F. Engels (1978). Erasmus R. 1965, The Praise of Folie (Englisshed by Sir Thomas Chaloner) (C. H. Miller, ed.), Oxford. Evans P., Rueschemeyer D. and Skocpol T. 1985 (eds), Bringing the State Back in, Cambridge. Festinger L. 1957, A Theory of Cognitive Dissonance, Stanford. Foss N. J. 1997 (ed.), Resources, Firms and Strategies, Oxford. Foss N. J. and Knudsen C. 1996 (eds), Towards a Competence Theory of the Firm, London. Foucault M. 1979, Discipline and Punish, New York. Frankfurt H. G. 1971, ‘Freedom of the Will and the Concept of a Person’, Journal of Philosophy. Friederich C. J. 1963, Man and His Government, New York. Fukuyama F. 1992, The End of History and the Last Man, London. Galbraith J. K. 1983, The Anatomy of Power, London. Genovesi A. 1765, Lezioni di commercio o sia di economia civile, 2 vols, (reprint: Varese 1977).

References 295 Giddens A. 1968, ‘ “Power” in the Recent Writings of Talcot Parsons’, Sociology. Gordon D. M. 1980, ‘Stages of Accumulation and Long Economic Cycles’, in T. Hopkins and I. Wallerstein (1980). Gordon D. M., Edwards R. and Reich M. 1982, Segmented Work, Divided Workers: The Historical Transformation of Labor in the United States, Cambridge. Gough I. 1979, The Political Economy of the Welfare State, London. Gramsci A. 1975, Quaderni del carcere, 4 vols, Turin. Graziani A. 1989, The Theory of the Monetary Circuit, Thames Papers in Political Economy, London. Griliches Z. 1991 (ed.), Output Measurement in the Services Sector, Chicago. Groenewegen J. and Vromen J. 1999 (eds), Institutions and the Evolution of Capitalism: Implications of Evolutionary Economics, Cheltenham. Groenewegen J., Pitelis C. and Sjöstrand S.-E. 1995 (eds), On Economic Institutions: Theory and Applications, Aldershot. Grossman S. and Hart O. 1986, ‘The Costs and Benefits of Ownership: a Theory of Vertical and Lateral Integration’, Journal of Political Economy. Gulolp H. 1989, ‘The Stages and Long Cycles of Capitalist Development’, Review of Radical Political Economics. Gustafsson B. 1991 (ed.), Power and Economic Institutions: Reinterpretations in Economic History, Aldershot. Habermas J. 1973, Legitimationsprobleme im Spätkapitalismus, Frankfurt a. M. Hall P. 1984, ‘Patterns of Economic Policy: An Organizational Approach’, in S. Bornstein, S. Held and J. Krieger (1984). Hall P. 1986, Governing the Economy, Oxford. Hall R. L. and Hitch C. J. 1939, ‘Price Theory and Business Behaviour’, Oxford Economic Papers. Hamel G. and Heene A. 1994, Competence-based Competition, New York. Hargreaves Heap S. 1898, Rationality in Economics, Oxford. Harper D. A. 1989, ‘Entrepreneurship and the Market Process’, Discussion Paper in Economics Series A, University of Reading. Harper D. A. 1996, Entrepreneurship and the Market Process: an Enquiry into the Growth of Knowledge, London. Harris M. and Raviv M. 1979, ‘Optimal Incentive Contracts with Imperfect Information’, Journal of Economic Theory. Hart O. and Moore J. 1990, ‘Property Rights and the Nature of the Firm’, Journal of Political Economy. Hegel G. W. F. 1977, Phenomenology of Spirit, Oxford. Heilbroner R. 1985, The Nature and Logic of Capitalism, New York. Heiner R. A. 1983, ‘The Origin of Predictable Behaviour’, American Economic Review. Held D. and Krieger J. 1984, ‘Theories of the State: Some Competing Claims’, in S. Bornstein, D. Held and J. Krieger (1984). Helmstädter E. and Perlman M. 1996 (eds), Behavioral Norms, Technological Progress, and Economic Dynamics, Ann Arbor. Hertje J. E. 1989 (ed.), The Economic Role of the State, Oxford. Higgins E. T. and Sorrentino R. M. 1990 (eds), Handbook of Motivation and Cognition: Foundations of Social Behaviour, 2 vols, New York. Hindess B. 1987, Freedom, Equality and the Market, London. Hirch J. 1977, Appareil d’état et reproduction du capital, Paris. Hirsch F. 1977, Social Limits to Growth, London.

296

References

Hirschman A. O. 1970, Exit, Voice and Loyalty: Responses to Decline in Firms, Organizations, and States, Cambridge, MA. Hirschman A. O. 1982, Shifting Involvements: Private Interest and Public Action, Princeton, NJ. Hirschman A. O. 1993, ‘Exit, Voice and the Fate of the German Democratic Republic: an Essay in Conceptual History’, Word Politics. Hobbes T. 1931, Leviathan, Indianapolis. Hodgson G. M. 1984, The Democratic Economy: a New Look at Planning, Markets and Power, Harmondsworth. Hodgson G. M. 1988, Economics and Institutions: a Manifesto for Modern Institutional Economics, Cambridge. Hodgson G. M. 1993, Economics and Evolution: Bringing Life Back into Economics, Cambridge. Hodgson G. M. 1999, Economics and Utopia: why the Learning Economy is not the End of History, London. Hodgson G. M. and Screpanti E. 1991 (eds), Rethinking Economics: Markets, Technology and Economic Evolution, Aldershot. Holloway J. and Picciotto S. 1987, ‘Toward a Materialist Theory of the State’, Introduction to J. Holloway and S. Picciotto, State and Capital: a Marxist Debate, Austin. Holmstrom B. 1979, ‘Moral Hazard and Observability’, Bell Journal of Economics. Holmstrom B. 1982, ‘Moral Hazard in Teams’, Bell Journal of Economics. Holmstrom B. 1990, ‘Multi-Task Principal-Agent Analysis: Incentive Contracts, Asset Ownership and Job Design’, Site Working Paper 6, Stanford University. Hopkins T. and Wallerstein I. 1980 (eds), Processes of the World System, Beverly Hills. Hotelling H. 1929, ‘Stability in Competition’, Economic Journal. Hubbard R. G. 1990 (ed.), Asymmetric Information, Corporate Finance, and Investment, Chicago. Hume D. 1888, A Treatise of Human Nature, Oxford. Hyman R. and Ferner A. 1994 (eds), New Frontiers in European Industrial Relations, Oxford. Iwai K. 1999, ‘Persons, Things and Corporations: The Corporate Personality Controversy and Comparative Corporate Governance’, American Journal of Comparative Law. Jaques E. 1990, ‘In Praise of Hierarchy’, Harvard Business Review. Jeffrey R. 1974, ‘Preferences among Preferences’, Journal of Philosophy. Jensen C. and Murphy K. J. 1990, ‘Performance Pay and Top-Management Incentives’, Journal of Political Economy. Jessop B. 1982, The Capitalist State: Marxist Theories and Methods, New York. Jessop B. 1990, State Theory: Putting Capitalist States in Their Place, Cambridge. Johnson M. K. and Sherman S. J. 1990, ‘Constructing and Reconstructing the Past and the Future in the Present’, in E. T. Higgins and R. M. Sorrentino (1990). Johnson-Laird P. N. and Oatley K. 1992, ‘Basic Emotions, Rationality, and Folk Theorem’, Cognition and Emotion. Jossa B. 1998, Mercato, socialismo e autogestione, Rome. Jossa B. and Cuomo G. 1997, The Economic Theory of Socialism and the Labour-Managed Firm, Cheltenham. Kalecki M. 1939, Essays in the Theory of Economic Fluctuations, London. Katz L. 1986, ‘Efficiency Wage Theories: A Partial Evaluation’, NBER Macroeconomic Annual. Kay G. and Mott J. 1982, Political Order and the Law of Labour, London. Kay N. M. 1997, Pattern in Corporate Evolution, Oxford.

References 297 Kelly G. A. 1955, The Psychology of Personal Construct, New York. Kelly G. A. 1963, A Theory of Personality: the Psychology of Personal Construct, New York. Keynes J. M. 1921, A Treatise on Probability, London. Keynes J. M. 1936, The General Theory of Employment, Interest and Money, London. Kolakowski L. 1978, Main Currents of Marxism, 3 vols, Oxford. Körner S. 1974 (ed.), Practical Reason, Oxford. Kotz D. M. 1990, ‘A Comparative Analysis of the Theory of Regulation and the Social Structure of Accumulation Theory’, Science and Society. Krupp S. R. 1966 (ed.), The Structure of Economic Science, Engelwood Cliffs, NJ. La Grassa G, 1993, Dal capitalismo al capitalismo, Rome. La Grassa G. 1996, Scritti di metodo e dintorni, Milan. La Grassa G. and Bonzio M. 1990, Il capitalismo lavorativo e la sua ri-mondializzazione, Milan. Laffont J.-J. and Tirole J. 1986, ‘Using Cost Observation to Regulate Firms’, Journal of Political Economy. Lambert R. A., Larcker D. F. and Weigelt K. 1989, ‘Tournaments and the Structure of Organisational Incentives’, The Wharton School, University of Pennsylvania. Lamberton D. M. 1971 (ed.), Economics of Information and Knowledge, Harmondsworth. Landes W. M. and Posner R. A. 1975, ‘The Independent Judiciary in an Interest-Group Perspective’, Journal of Law and Economics. Lane C. 1994, ‘Industrial Order and Transformation of Industrial Relations: Britain, Germany and France Compared’, in R. Hyman and A. Ferner (1994). Langlois R. N. 1986 (ed.), Economics as a Process: Essays in the New Institutional Economics, Cambridge. Langlois R. N. 1992, ‘Transaction Cost Economics in Real Time’, Industrial and Corporate Change. Lasswell H. D. 1930, Psychopathology and Politics, Chicago. Lasswell H. D. 1936, Politics: Who gets What, When, How, New York. Lasswell H. D. and Kaplan A. 1950, Power and Society, New Haven. Lazear E. and Rosen S. 1981, ‘Rank Order Tournaments as Optimal Labour Contracts’, Journal of Political Economy. LeDoux J. H. 1995, ‘In Search of an Emotional System in the Brain’, in M. S. Cazzaniga (1995) . LeDoux J. H. 1996, The Emotional Brain, New York. Legrenzi P., Girotto V. and Johnson-Laird P. N. 1993, ‘Focusing in Reasoning and Decision Making’, Cognition. Lindberg L. 1975 (ed.), Stress and Contradiction in Modern Capitalism, Lexington, MA. Lindblom C. E. 1982, ‘The Market as Prison’, Journal of Politics. Lipietz A. 1987, Mirages and Miracles: The Crisis of Global Fordism, London. Lippmann S. and Rumelt R. P. 1982, ‘Uncertain Imitability: An Analysis of Interfirm Differences in Efficiency under Competition’, Bell Journal of Economics. Loasby B. J. 1984, ‘Entrepreneurs and Organisation’, Journal of Economic Studies. Loasby B. J. 1986, ‘Organisation, Competition, and the Growth of Knowledge’, in R. N. Langlois (1986). Loasby B. J. 1996, ‘The Imagined, Deemed Possible’, in E. Helmstädter and M. Perlman (1996). Loasby B. J. 1998, ‘Cognition and Innovation’, in M. Bianchi (1998). Loasby B. L. 1976, Choice, Complexity and Ignorance: an Inquiry into Economic Theory and Practice of Decision Making, Cambridge.

298

References

Loasby B. L. 1999, Knowledge, Institutions and Evolution in Economics, London. Locke J. 1960, Two Treatises on Government, New York. Lukes S. 1974, Power: A Radical View, London. Lydall H. F. 1959, ‘The Distribution of Employment Income’, Econometrica. Machiavelli N. 1833, Opere complete, Florence. Makhaïski J. W. 1979, ‘L’évolution de la social-démocratie’, in A. Skirda (1979). Mankiw N. G. 1985, ‘Small Menu Costs and Large Busines Cycles: a Macroeconomic Model of Monopoly’, Quarterly Journal of Economics. Margenau H. 1966, ‘What is a Theory?’, in S. R. Krupp (1966). Marglin S. A. 1974, ‘What do Bosses do? The Origins and Functions of Hierarchy in Capitalist Production’, Review of Radical Political Economics. Marglin S. A. 1984, ‘Knowledge and Power’, in F. Stephen (1984). Marglin S. A. 1991, ‘Understanding Capitalism: Control versus Efficiency’, in Gustafsson B. (1991). Marshall A. 1925, Principles of Economics, London. Martini R. 1958, ‘Mercennarius’: Contributo allo studio dei rapporti di lavoro nel diritto romano, Milan. Marx K. 1969, Theories of Surplus Value, 3 vols, London. Marx K. 1970, A Contribution to a Critique of Political Economy, London. Marx K. 1978a, Wage Labour and Capital, in K. Marx and F. Engels (1978). Marx K. 1978b, Economic and Philosophical Manuscripts of 1844, in K. Marx and F. Engels (1978). Marx K. 1978c, Contribution to the Critique of Hegel’s Philosophy of Right, In K. Marx and F. Engels (1978). Marx K. 1990, Capital, 3 vols, London. Marx K. and Engels F. 1978, The Marx-Engels Reader (R. C. Tucker, ed.) New York. Maslow A. H. 1970, Motivation and Personality, 2nd edn, New York. Maugham W. S. 1949, The Moon and Sixpence, Rome. McClelland D. C. 1978, Human Motivation, Cambridge. McKelvey M. 1991, ‘How do National Systems of Innovation Differ? A critical Analysis of Porter, Freeman, Lundval and Nelson’, in G. M. Hodgson and E. Screpanti (1991). Meyer J. W. and Rowan B. 1977, ‘Institutionalized Organizations: Formal Structure as Myth and Ceremony’, American Journal of Sociology. Michels R. 1925, Zur Soziologie der Parteiwesens in der Modernen Demokratie, 2nd edn, Leipzig. Milgrom P. and Roberts J. 1990, ‘The Economics of Modern Manufacturing: Technology, Strategy and Organization’, American Economic Review. Milgrom P. and Roberts J. 1992, Economics, Organization and Management, Engelwood Cliffs, NJ. Miller G. A. 1956, ‘The Magical Number Seven, plus or minus Two: Some Limits on Our Capacity for Processing Information’, Psychological Review. Miller, J. B. 1990, ‘Minimal Power Structures in a Weighted Organization’, Mathematical Social Sciences. Mirrlees J. 1976, ‘The Optimal Structure of Incentives and Authority within an Organization’, Bell Journal of Economics. Morris W. 1915, News from Nowhere, in The Collected Works of W. Morris (M. Morris and S. Cockerell, eds), London. Morriss P. 1987, Power: A Philosophical Analysis, Manchester. Mosca G. 1884, Sulla teorica dei governi e sul governo parlamentare, Turin. Mosca G. 1923, Elementi di scienza politica, Turin.

References 299 Musil R. 1978, Der Mann ohne Eigenschaften, in Gesammelte Werke, vol. I, (A. Frise, ed.), Hamburg. Nelson R. R. and Winter S. G. 1982, An Evolutionary Theory of Economic Change, Cambridge, MA. Nigel G. and Conte R. 1995 (eds), Artificial Societies: The Computer Simulation of Social Life, London. Niskanen W. A. Jr. 1971, Burocracy and Representative Government, Chicago. Niskanen W. A. Jr. 1975, ‘Burocrats and Politicians’, Journal of Law and Economics. Nonaka I. and Takeuchi H. 1995, The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation, Oxford. North D. C. and Thomas R. P. 1973, The Rise of the Western World: A New Economic History, Cambridge. O’Connor J. 1984, Accumulation Crisis, New York. O’Neill J. 1998, The Market: Ethics, Knowledge and Politics, London. Offe C. 1975, ‘The Theory of the Capitalist State and the Problem of Policy Formation’, in L. Lindberg (1975). Offe C. 1984, Contradictions of the Welfare State, Cambridge, MA. Okun A. M. 1981, Prices and Quantities: A Macroeconomic Analysis, Oxford. Ouchi W. G. 1980, ‘Markets, Bureaucracies, and Clans’, Administrative Science Quarterly. Ouchi W. G. and Johnson J. B. 1978, ‘Types of Organizational Control and their Relationship to Emotional Well-Being’, Administrative Science Quarterly. Pagano U. 1985, Work and Welfare in Economic Theory, Oxford. Pagano U. 1991a, ‘Property Rights Equilibria and Institutional Stability’, Economic Notes. Pagano U. 1991b, ‘Property Rights, Asset Specificity, and the Division of Labour under Alternative Capitalist Relations’, Cambridge Journal of Economics. Pagano U. 1999, ‘Is Power an Economic Good? Notes on Social Scarcity and the Economics of Positional Goods’, in S. Bowles, M. Franzini and U. Pagano (1999). Pagano U. and Rowthorn R. E. 1994, ‘Ownership, Technology and Institutional Stability’, Structural Change and Economic Dynamics. Pagano U. and Rowthorn R. E. 1996 (eds), Democracy and Efficiency in the Economic Enterprise, London. Paoletti G. 1994 (ed.), La cassetta degli attrezzi di Marx, Milan. Pareto V. 1916, Trattato di sociologia generale, Florence. Pareto V. 1989, Oeuvres Complètes, vol. XXIX (G. Busino, ed.), Geneva. Parsons T. 1952, The Social System, Glencoe, Ill. Pashukanis E. B. 1978, Law and Marxism: A General Theory, London. Penrose E. 1959, The Theory of the Growth of the Firm, Oxford. Perroux F. 1950, ‘The Domination Effect and Modern Economic Theory’, Social Research. Perroux F. 1974, Pouvoir et économie, 2nd edn, Paris. Pietri P. 1973, Puerto Rican Obituary, New York. Pitelis C. 1991, Market and Non-Market Hierarchies, Oxford. Poggi G. 1990, The State: its Nature, Development and Prospects, Stanford. Polanyi K. 1957, The Great Transformation: The Political and Economic Origins of Our Times, Boston. Poulantzas N. 1969, Political Power and Social Classes, London. Poulantzas N. 1975, Classes in Contemporary Capitalism, London. Powell W. W. and DiMaggio P. J. 1991 (eds), The New Institutionalism in Organizational Analysis, Chicago. Reati A. 1968, ‘Une classification des formes de marché d’après l’effet de domination’, Économie appliquée.

300

References

Renner K. 1976, The Institutions of Private Law and their Social Functions (O. KahnFreund, ed.), London. Richardson G. B. 1972, ‘The Organization of Industry’, Economic Journal. Riordan M. and Sappington D. 1987, ‘Information, Incentives and Organizational Mode’, Quarterly Journal of Economics. Roberts J. and Prichard S. 1983 (eds), Crown Corporations in Canada: The Calculus of Instrument Choice, Toronto. Rodano G. 1995 (ed.), Relazioni Pericolose: L’avventura dell’economia nella cultura contemporanea, Rome. Rorty A. 1980, ‘Self-Deception, Akrasia, and Irrationality’, Social Science Information. Rosen S. 1990, ‘Contracts and the Market for Executives’, NBER working papers. Ruccio D. F. 1986, ‘Essentialism and Socialist Economic Planning: a Methodological Critique of Optimal Planning Theory’, Research in the History of Economic Thought and Methodology. Ruccio D. F. 1993, ‘Marx’s Critical/Dialectical Procedure: A Review Essay’, Research in the History of Economic Thought and Methodology. Russel B. 1938, Power, Woking. Sabine G. H. 1951, A History of Political Theory, New York. Salop S. 1979, ‘A Model of the Natural Rate of Unemployment’, American Economic Review. Samuels W. 1971, ‘Interactions between Legal and Economic Processes’, Journal of Law and Economics. Schumpeter J. 1950, Capitalism, Socialism and Democracy, 3rd edn, London. Scott W. R. 1994, ‘Institutions and Organizations: Towards a Theoretical Synthesis’, in W. R. Scott and J. W. Meyer (1994). Scott W. R. and Meyer, J. W. 1994 (eds), Institutional Environments and Organizations, London. Screpanti E. 1992, ‘The Advent of the Capitalist Utopia: Transition and Convergence’, in B. Dallago, H. Brezinski and W. Andreff (1992). Screpanti E. 1994, ‘La teoria del capitalismo in Marx: da Smith a Mrs Thatcher’, in G. Paoletti (1994). Screpanti E. 1995, ‘Relative Rationality, Institutions and Precautionary Behaviour’, in J. Groenewegen, C. Pitelis and S.-E. Sjöstrand (1995). Screpanti E. 1997, ‘Banks, Increasing Risk, and the Endogenous Money Supply’, Economic Notes. Screpanti E. 1998, ‘Toward a General Theory of Capitalism: Suggestions from Chapters 23 and 27’, in R. Bellofiore (1998). Screpanti E. 1999a, ‘Discontinuous Changes in Institutional Systems’, in J. Groenewegen and J. Vromen (1999). Screpanti E. 1999b, ‘Capitalist Forms and the Essence of Capitalism’, Review of International Political Economy. Screpanti E. 2000, ‘Wages, Employment and Militancy: A Simple Model and Some Empirical Tests’, Review of Radical Political Economics. Sen A. 1974, ‘Choice, Ordering and Morality’, in S. Körner (1974). Sen A. 1983, ‘The Profit Motive’, Lloyds Bank Review. Shackle G. L. S. 1966, ‘Policy, Poetry and Success’, The Economic Journal. Shackle G. L. S. 1969, Decisions, Order and Time in Human Affairs, 2nd edn, Cambridge. Shackle G. L. S. 1972, Epistemics and Economics, Cambridge. Shackle G. L. S. 1979, Imagination and the Nature of Choice, Edimburgh. Shafir E., Simonson I. and Tversky A. 1993, ‘Reason-based Choice’, Cognition.

References 301 Shapiro C. and Stiglitz J. 1984, ‘Equilibrium Unemployment as a Discipline Device’, American Economic Review. Shaw B. and Cummings L. L. 1992 (eds), Research in Organizational Behaviour, London. Silberston A. 1971, ‘The Patent System’, Lloyds Bank Review, 1967; reprinted in D. M. Lamberton (1971). Silverberg G. and Soete L. 1994 (eds), The Economics of Growth and Technical Change, Aldershot. Simon H. A. 1951, ‘A Formal Theory of the Employment Relationship’, Econometrica. Simon H. A. 1953, ‘Notes on the Observation and Measurement of Political Power’, The Journal of Politics. Simon H. A. 1955, ‘A Behavioural Model of Rational Choice’, Econometrica. Simon H. A. 1957, Models of Man: Social and Rational, New York. Simon H. A. 1959, ‘Theories of Decision-Making in Economics and Behavioural Sciences’, American Economic Review. Simon H. A. 1962, ‘The Architecture of Complexity’, Proceeding of the American Philosophical Society. Simon H. A. 1979, ‘On Parsimonious Explanations of Production Relations’, Scandinavian Journal of Economics. Simon H. A. 1982, Models of Bounded Rationality, 2 vols, Cambridge, MA. Simon H. A. 1990, ‘A Mechanism for Social Selection and Successful Altruism’, Science. Skirda A. 1979 (ed.), Le socialisme des intellectuels, Paris. Skocpol T. 1979, State and Social Revolution, Cambridge. Skocpol T. 1985, ‘Bringing the State Back In: Strategies of Analysis in Current Research’, in P. Evans, D. Rueschemeyer and T. Skocpol (1985). Smith A. 1970, The Wealth of Nations, Harmondsworth. Smith A. 1978, Lectures on Jurisprudence (R. L. Meek, D. D. Raphael and P. G. Stein, eds), Oxford. Smithies F. 1941, ‘Optimum Location in Spatial Competition’, Journal of Political Economy. Spence A. M. 1973, Market Signalling: Information Transfer in Hiring and Related Processes, Cambridge, MA. Steedman I. and Krause U. 1986, ‘Goethe’s Faust: Arrow’s Impossibility Theorem and the Individual Decisor’, in J. Elster (1986). Stein J. 1989, ‘Efficient Capital Markets, Inefficient Firms: a Model of Myopic Corporate Behaviour’, Quarterly Journal of Economics. Stephen F. 1984 (ed.), The Economics of Organisation and Labour, London. Stigler G. J. 1978, ‘The Literature of Economics: the Case of the Kinked Oligopoly Demand Curve’, Journal of Political Economy. Stiglitz J. E. 1974, ‘Incentives and Risk Sharing in Sharecropping’, Review of Economic Studies. Stiglitz J. E. 1975, ‘Incentives, Risk and Information: Notes Toward a Theory of Hierarchy’, Bell Journal of Law, Economics and Organization. Stiglitz J. E. 1986, ‘Theories of Wage Rigidity’, in J. L. Butkiewicz (1986). Stiglitz J. E. 1989, ‘On the Economic Role of the State’, in A. Hertje (1989). Sudgen R. 1997, ‘Measuring Opportunity: Towards a Contractarian Measure of Individual Interest’, unpublished manuscript, University of East Anglia. Sutton A. C. 1968, Western Technology and Soviet Economic Development: 1917 to 1930, Stanford. Sutton A. C. 1971, Western Technology and Soviet Economic Development: 1930 to 1945, Stanford.

302

References

Sutton A. C. 1973, Western Technology and Soviet Economic Development: 1945 to 1965, Stanford. Sweezy P. M. 1939, ‘Demand under Conditions of Oligopoly’, Journal of Political Economy. Teece D. J. , Pisano G. and Shuen A. 1990, ‘Firm Capabilities, Resources and the Concept of Strategy’, University of California, Berkeley. Teubner G. 1993, Law as an Autopoietic System, Oxford. Tool M. R. 1995, Pricing, Valuation and Systems: Essays in Neoinstitutional Economics, Aldershot. Tushman M. L. and Rosenkopf L. 1992, ‘Organizational Determinants of Technological Change: Toward a Sociology of Technological Evolution’, in B. Shaw and L. L. Cummings (1992). Tversky A. and Shafir E. 1992, ‘Choice under Conflict: The Dynamics of Deferred Decision’, Psychological Science. Van Maanen J. and Schein E. H. 1978, ‘Toward a Theory of Organizational Socialization’, Sloan School of Industrial Administration, Massachusetts Institute of Technology. Vanberg V. J. 1994, Rules and Choice in Economics, London. Vannucci S. 1999, ‘A Lattice-Theoretical Classification of Coalitional Power Correspondences’, Quaderni del Dipartimento di Economia Politica, Siena. Veblen T. 1919, The Place of Science in Modern Civilization and Other Essays, New York. Weale A. 1976, ‘Power Inequalities’, Theory and Decision. Weber M. 1922, Wirtschaft und Gesellschaft, Tübingen. Wernerfelt B. 1984, ‘A Resource-based Theory of the Firm’, Strategic Management Journal. Weston J. F., Chung K. and Hoah S. 1990, Mergers, Restructuring and Corporate Control, Engelwood Cliffs. Whitley R. 1992, ‘Societies, Firms and Markets: The Social Structuring of Business Systems’, in R. Whitley (ed.) European Business Systems, London. Williamson O. E. 1980, ‘The Organization of Work: A Comparative Institutional Assessment’, Journal of Economic Behaviour and Organization. Williamson O. E. 1982, ‘Efficient Labour Organization’, CSOI Discussion Paper, University of Pennsylvania. Williamson O. E. 1985, The Economic Institutions of Capitalism, New York. Williamson O. E. 1985, The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting, London. Williamson O. E. 1986, ‘The Economics of Governance: Framework and Implications’, in R. N. Langlois (1986). Yellen J. J. 1984, ‘Efficiency Wage Models of Unemployment’, American Economic Review. Young D. 1995, ‘The Meaning and Role of Power in Economic Theories’, in J. Groenewegen, C. Pitelis and S.-E. Sjöstrand (1995). Zarembka P. 1992, ‘The Development of State Capitalism in the Soviet System’, Research in Political Economy. Zeckhauser R. and Pound J. 1990, ‘Are Large Shareholders Effective Monitors? An Investigation of Share Ownership and Corporate Performance’, in Hubbard R. G. (1990).

Index of names 303

Index of names

Abowd J. M. 284 Aglietta M. A. 283 Alchian A. 243 Alford R. R. 144 Alighieri D. 249 Altvater E. 143 Amariglio J. L. 108 Arestis P. 252 Arrow K. J. 242, 247, 248, 253, 254 Ashton D. 20 Bacharach P. 183 Bakunin M. 181, 280, 283 Baratz M. S. 183 Barca F. 283 Barrow C. W. 143 Barry B. 183 Bartlett R. 183 Beccaria C. 190 Bellanca N. 109 Bellofiore R. 5 Bentham J. 53 Berle A. A. 183, 255, 284 Berneri C. 283 Bhandari R. 5 Bianco M. 283 Block F. 143 Bolton P. 252 Bonzio M. 40 Borcherding T. E. 141 Bordiga A. 283 Boulding K. E. 241, 253 Bowen H. R. 141 Bowles S. D. 5, 41, 182, 183, 252, 283 Boyer R. 252, 283 Braverman H. 41, 250, 283 Bresnahan T. 283 Bridges A. B. 143

Bulow J. B. 252 Burkart M. 283 Callari T. 108 Calvo G. 252 Cannoni A. 144 Carchedi G. 5 Carver T. 108 Chamberlin E. 218 Chattopadhyay P. 283 Chung K. 283 Cicero M. T. 18, 19 Cillario L. 253 Cipolla F. P. 5 Cirino G. 250, 253 Clarke S. 141 Cliff T. 283 Coase R. H. 41 Cockshott P. 5 Cohen G. A. 2, 181 Coleman J. S. 41 Commons J. R. 40, 42 Conrath D. W. 253 Cullis J. 141 Cuomo G. 40 Dahl R. A. 170, 182 Damasio A. B. 109 Darwin C. 146 Dasgupta P. 254 David P. A. 254 De Bernis G. D. 283 De Francesco M. 5 De Jouvenel B. 183 de Sousa R. 109 Deakin S. 282 Demsetz H. 41 DiMaggio P. J. 108

304 Index of names Dimitri N. 5 Doeringer P. B. 284 Doise W. 109 Downs A. 126, 141, 142 Dunayevskaya R. 283 Earl P. E. 108, 253 Easton D. 182 Edwards R. 41, 250, 283 Egidi M. 181, 253 Ellerman D. P. 39, 40 Elster J. 91, 108, 109, 110 Engels F. 91, 141, 142, 284 Erasmus R. 50 Fama E. 275 Farrell J. 252 Festinger L. 109 Feuerbach L. 105 Finelli R. 253 Fiorito L. 5, 111 Foss N. J. 252 Foucault M. 169 Frankfurt H. G. 109 Friederich C. J. 159, 182 Friedland R. 144 Fukuyama F. 108

Heilbroner R. 182 Heiner R. A. 108 Held D. 144 Hindess B. 108 Hirch J. 143 Hirsch F. 183 Hirschman A. O. 109, 173, 182 Hitch C. J. 252 Hoah S. 283 Hobbes T. 111, 161 Hodgson G. M. 5, 40, 41, 108, 251, 253, 283 Holloway J. 143 Holmstrom B. 252 Hotelling H. 141 Hume D. 149 Iwai K. 283 Jaques E. 253 Jeffrey R. 109 Jensen C. 284 Jessop B. 144 Johnson J. B. 252 Johnson M. K. 109 Johnson-Laird P. N. 109 Jones P. 141 Jossa B. 5, 40

Galbraith J. K. 174, 183 Genovesi A. 285 Giddens A. 183 Gintis H. 41, 182, 183 Giocoli N. 109 Girotto V. 109 Gordon D. M. 250, 283 Gough I. 143 Gramsci A. 109, 110, 142, 274, 282 Graziani A. 41 Green F. 20 Gromb D. 283 Grossman S. 41 Gulolp H. 283

Kalecki M. 33, 252 Kaplan A. 182 Katz L. 252 Kay G. 41 Kay N. M. 253 Kelly G. A. 253 Keynes J. M. 106, 107, 167 Khang K. 5 Knudsen C. 252 Kolakowski L. 91 Kotz D. M. 283 Krause U. 109 Krieger J. 144

Habermas J. 284 Hahn F. H. 5 Hall P. 283 Hall R. L. 252 Hamel G. 252 Hargreaves Heap S. 108 Harper D. A. 252, 253 Harris M. 252 Hart O. 41, 283 Heene A. 252 Hegel G. W. F. 102, 104, 105

La Grassa G. 40 Laffont J.-J. 252 Lambert R. A. 284 Landes W. M. 141 Lane C. 283 Langlois R. N. 252 Larcker D. F. 284 Lasswell H. D. 121, 182 Lazear E. 284 LeDoux J. H. 109 Legrenzi P. 109

Index of names 305 Lenin V. I. 2, 176, 286 Levy G. 5 Lindblom C. E. 143 Lipietz A. 283 Lippmann S. 251 Loasby B. L. 108, 252, 253 Locke J. 111, 284 Louça F. 5 Lukes S. 183 Lydall H. F. 264, 284 Machiavelli N. 121, 122, 140 Makhaïski J. W. 283 Mankiw N. G. 252 Marengo L. 181 Margenau H. 253 Marglin S. A. 41, 182, 283 Marshall A. 251 Martini R. 40 Marx K. 1, 2, 3, 4, 18, 19, 20, 21, 39, 41, 42, 91, 105, 106, 141, 142, 147, 180, 181, 182, 191, 192, 250, 255, 257, 282, 283, 284, 287, 289 Maslow A. H. 108, 254 Maugham W. S. 70 McClelland D. C. 49, 108 McKelvey M. 283 Means G. C. 255, 284 Messori M. 181 Meyer J. W. 252 Michels R. 121, 177, 181, 182 Michie J. 282 Miconi B. 5 Milgrom P. 181, 197, 252, 283 Miller G. A. 253 Miller J. B. 183 Mirrlees J. 252 Moore J. 283 Morris W. 248, 249 Morriss P. 183 Morroni M. 5 Mosca G. 121, 140, 141, 182 Mott J. 41 Murphy K. J. 284 Musil R. 72 Nell E. 5 Nelson R. R. 108, 251 Niskanen W. A. Jr. 141 Nonaka I. 253 North D. C. 41 O’Connor J. 283 O’Neill J. 108, 254

Oatley K. 109 Offe C. 143 Okun A. M. 252 Ouchi W. G. 252 Pagano U. 5, 41, 108, 183, 254, 283 Pantaleoni M. 109 Panunzi F. 283 Pareto V. 52, 121, 141, 182 Parsons T. 182, 254 Pashukanis E. B. 136, 142 Paul J. 283 Penrose E. 186, 251 Perroux F. 166, 182 Pessoa F. 71 Picciotto S. 143 Pietri P. 104 Piore M. J. 284 Pisano G. 252 Pitelis C. 141 Poggi G. 143 Polanyi K. 15, 101, 283 Posner R. A. 141 Poulantzas N. 143 Pound J. 283 Powell W. W. 108 Proudhon P.-J. 38, 191 Raviv M. 252 Reati A. 5, 182 Reich M. 250, 283 Renner K. 40 Ricardo D. 39 Richardson G. B. 251 Ricottilli M. 181 Riordan M. 252 Roberts J. 181, 197, 252, 283 Robinson J. V. 218 Rorty A. 109 Rosen S. 284 Rosenkopf L. 283 Rowan B. 252 Rowthorn R. E. 41, 283 Ruccio D. F. 108 Rumelt R. P. 251 Russel B. 146, 166, 181 Sabine G. H. 140 Salop S. 252 Samuels W. 283 Sappington D. 252 Schein E. H. 252 Schumpeter J. 141, 142 Scott W. R. 108

306 Index of names Sen A. 109, 183 Shackle G. L. S. 107, 253 Shafir E. 109 Shapiro C. 252 Sherman S. J. 109 Shuen A. 252 Silberston A. 254 Simon H. A. 40, 106, 107, 183, 253, 284, 285 Simonson I. 109 Skillman G. 5 Skocpol T. 143 Smith A. 39, 41, 42, 49, 190, 250, 269, 283 Smithies F. 141 Spence A. M. 197 Spencer H. 106 Steedman I. 109 Stein J. 283 Stigler G. J. 252 Stiglitz J. E. 141, 252 Sudgen R. 53 Summers S. 252 Sutton A. C. 279 Sweezy P. M. 252 Takeuchi H. 253 Teece D. J. 252 Teubner G. 283 Thomas R. P. 41

Tirole J. 252 Tool M. R. 252 Tortia E. 5 Tushman M. L. 283 Tversky A. 109 Tylecote A. 5 Van Gogh V. 249 Van Maanen J. 252 Vanberg V. J. 41 Vannucci S. 5, 183 Veblen T. 106, 169 von Hayek F. A. 229 Weale A. 183 Weber M. 182 Weigelt K. 284 Weisskopf T. E. 283 Wernerfelt B. 252 Weston, J. F. 283 Whitley R. 283 Williamson O. E. 41, 252 Winter S. G. 108, 252 Wright Mills C. 121 Yellen J. J. 252 Young D. 182 Zarembka P. 4, 5, 283 Zeckhauser R. 283

Index of names 307

Subject index

abilities 13–14, 193, 195–7 accumulation governance structures 185, 261–5, 273, 282, 284, 289 accumulation of capital 12, 22–3, 32, 91, 102, 133–4, 139, 147, 166, 168–71, 187, 234, 277–8 agency relationship (or contract) 5, 8–9, 11, 42, 159, 195, 228 alienation 18–19, 21, 41–2, 85, 103, 125, 211 altruism 45–9, 74, 82, 107, 121, 149, 153–4, 171, 175, 181 a-rational actions 64, 87, 89, 102 aspirations 54–6, 59–69, 76, 79, 81, 86–7, 94, 97, 103–4, 131–2, 145, 168, 194, 211, 235, 246 asset specificity 8, 164 associations (see representative organisations) authority 12–15, 158, 162–5, 263, 289–90 autonomy of capital 287–90 autonomy of choices 82, 84–6, 101 behavioural institutions 55–9, 63, 86, 106–7, 110, 112–14 bounded rationality 8, 56, 57, 74, 76, 100–1, 106–7, 237 capital fetishism 20–1 capitalism 2, 4, 29–31, 39, 48, 72, 73, 91, 102, 117, 123–5, 128, 130, 134–5, 137, 142, 154–5, 160, 176, 224, 229, 234, 255, 258–60, 287–90 capitalist class 2, 7, 30, 91, 102, 125, 130, 138, 279–82, 284 capitalist firm 10–12, 25, 27–8, 41, 152, 156, 161, 170, 187–8, 191–2, 224, 272, 280–1, 287

capitalist ideological hegemony 96–8, 126, 269–70 capitalist mode of production (see capitalism) capitalist system (see capitalism) capitalists (see capitalist class) class-biased control 37–8, 41, 172 class-biased credit rationing 33, 37–8, 41, 97, 137–8, 172, 227, 272–3 classical capitalism 267, 269–72, 280, 284, 287, 289 collective actors (see social actors) collective movements (see social movements) command 10, 12, 13, 16, 18, 134, 137, 144, 156–60, 162, 178, 199 commodity fetishism 18–21, 41–2 companies markets 262, 265–6, 274 company ideologies 104, 205, 210–15 competencies 13, 193, 195–6 conditional improvement 57, 62 construction of knowledge (see knowledge production) control costs (see organisation costs) co-operation 147–51, 153, 169, 171, 173, 175, 181, 186, 188, 204 co-operative firm 37–8, 151, 161, 171, 173, 191, 229 core institutions 29, 120, 131, 256–8 corporate actors (see corporate organisations) corporate capitalism 272–4, 279 corporate capitalism (bank-oriented) 267, 275–6 corporate capitalism (market-oriented) 267, 274–5, 286 corporate organisations 26–7, 41, 187–8 counter-power 158, 160, 174–5, 192, 199, 228

308 Subject Index ofindex names decreasing returns to hierarchy 241, 248, 279 delegated power 158, 178–80, 246 democracy 123–4, 126, 128, 136, 140–1, 160, 278 democratic organisations (see representative organisations) desire for power 49–51, 80–1, 146–7, 153, 166–7, 182 desire for recognition 49–51, 65, 85, 96, 102, 153, 249, 254 desire for self-realisation 49–51, 102, 249, 254 dissatisfaction 55, 56, 59, 60, 62, 73, 77, 80 dominating class ideological hegemony (see capitalist ideological hegemony) domination 135–6, 140–1, 160–1, 168 efficiency wages 205–10, 252 egoism 45–9, 74, 82, 112, 117, 145, 149, 153–4, 171 emotions 47, 52–4, 109–10 employer 10, 13, 18–19, 21, 30, 102, 135, 162–3, 201, 269, 289 employment contract 5, 8–12, 13, 15–19, 29–31, 39–42, 91, 102, 105, 113, 120, 133, 135, 157–8, 162, 183, 191–3, 195, 224, 228, 258, 269, 277, 285–6, 289–90 employment relationship (see employment contract) entrepreneurs (see managers) evolutionary process 27, 61, 62, 100, 106, 145–6 exit faculty (see exit freedom) exit freedom 155, 163–4, 168, 173, 181, 183, 217, 220, 227, 229 exploitation 1–3, 5, 10, 12, 15, 17, 22, 23, 27–8, 30, 134, 136–7, 147, 155–6, 159, 167–8, 171, 181, 199–201, 211, 224, 226, 228–30 external hierarchies 262, 264–5, 275–7 false consciousness 19, 76, 78, 96, 101, 211 focusing 76–8, 83, 94, 213–14 forms of capitalism (see institutional forms) forums 153–4, 239, 250 freedom of choice 66, 67, 69, 73, 79, 82–5, 102, 105, 147, 224, 226–7 freedom of contract 123, 143, 258–9 frustration 76, 77, 78, 80, 94, 96, 101, 111, 129–33, 211

goods markets 262, 266, 271, 277 governance relations 150–3, 156, 171, 183 habituation 83–4, 86 hierarchy 11–12, 14, 149–50, 156, 158, 185, 189–91, 214, 236–8, 250–1, 262, 284 homo economicus 44, 79, 90, 101, 106 homo relativus 99–102 human capital 2–3, 7, 18, 20, 80, 97, 196–8, 207, 213, 227, 245, 280–2, 288 idealisation 76, 80–3, 103, 121 ideological hegemony 96–8, 142 ideology 18–19, 48, 86–98, 101, 105, 109–10, 122–3, 125, 131–2, 170, 188, 195–6, 259–60, 270, 279 ignorance 76–8, 83, 94, 124–7, 136, 141, 212–3, 278 illusion 76–8, 83, 94, 103, 212–14 incentive pay 9, 14, 195–6, 199–204, 209, 252, 263, 266 incentive wage (see incentive pay) influence 159–60, 168, 226 information asymmetries 14–15, 25, 35, 119, 151, 153–6, 172, 192–5, 207 innovations 56, 59, 60, 61, 62, 67, 79, 222, 233–5, 240–1, 243, 254 institutional compliance 57, 59, 60, 62 institutional forms (of capitalism) 258, 266–8, 283 institutional incompleteness 62, 69, 112–19, 123, 134–5, 137 institutional plurality (principle of) 28–9, 41, 67, 69, 73, 131 institutional revision 58, 61, 62, 131 institutional systems 119–20, 131, 256–8, 268, 282 intentional actions 56, 62–3, 86–7, 89, 100, 102 interest rates 33–4, 42, 137–8, 144 internal hierarchies 262–4, 271–2, 277 inventions 60, 61, 232, 240, 242, 254 irrational actions 87–9, 94, 106 kinked demand curve 221–3, 252 knowledge evolution 232–5, 244, 248 knowledge production 231–2, 236–42, 248, 250, 254 labour action 15–16, 39 labour activity 10, 14–16, 18, 25, 102, 199 labour capacities 12–15, 186, 193, 197, 199

Index Subject of names index 309 labour commanded (see purchasable labour) labour process 14, 21, 40, 103, 105 labour services 10–12, 15–16, 195, 230 labour time 13–14, 22, 42–3, 102 laws 117–20, 124, 259 managers 34–6, 43, 97, 161, 166–7, 187, 211, 264, 271–2, 275, 280–2, 288–9 manipulation 83–4, 87, 168 markets for corporate control (see companies markets) mass force 147–8, 152, 156, 175, 188–9 median voter 111, 127–9, 141 mental feelings 45–51, 54, 66, 68, 93, 96, 107, 145 modal voter 127–9, 141 mode of production 29, 31, 36, 257–8, 282 multiple self 53, 66–7, 69–74, 79–80, 82, 87–8, 101, 109, 131, 211 neoinstitutionalism (see new institutional economics) networks 149–50, 153, 185, 262 new institutional economics 4–5, 8, 113, 135 normative institutions 25, 110, 114–16, 123, 137, 170, 193, 226 obedience 10, 12, 39, 83–6, 103–5, 135, 163, 170 oligopolistic competition 218–19, 230, 232–5, 246 opportunism 8, 101, 104, 148–9, 152, 155, 192–3, 203, 227 organisation costs, 8, 190, 194–5, 199, 204, 207, 224–5, 227 organisational involvement 65–6, 70, 103, 177, 205, 210–15, 252 organisations 26–7, 65–6, 116, 146, 157, 175, 177, 181, 190, 224, 236–9, 289–90 partnership contract 8, 10 performance pay (see incentive pay) peripheral institutions 120, 256 political actors 98–9, 121–4, 127–8, 136 political class 98, 121, 123, 125–7, 129–30, 136, 138–9, 141, 143, 278 positional goods 166, 169, 173–5, 183, 263 poverty 83, 86, 105, 155, 226

power 13–14, 18, 21–3, 28, 36, 48, 81–2, 121–2, 134–7, 141, 153, 155, 157–62, 165, 169–70, 172–4, 177–80, 182–3, 230, 245–6, 263–4, 281 power bases 161–4, 167–8, 170 power instruments 164–5, 169–70 power seeking (see desire for power) power sources 159, 163–4, 167, 226 price fixing 216–20, 229 price stickiness 220–23, 229–30 principle of equivalence 1, 6–8, 10–12, 16–17, 42, 145, 159 principle of improvement 61, 62, 100 principle of impurity 29, 31, 41, 267 private property (concentrated) 260 private property (diffused) 261, 274 procedural democracy (see democracy) production governance structures 149–50, 185, 187, 262 profits 6–7, 10–11, 22 property right regimes 260–1, 282–4 property rights 4–5, 24–5, 113, 117, 135, 137, 143, 256, 269, 283, 285 purchasable labour 22–3, 41–2 purposeful actions (see intentional actions) rational actions 56, 57, 58, 59, 61–3, 87–9, 92, 100, 102, 106, 109 rational behaviour (see rational actions) representative organisations 26–7, 111, 151, 154, 161 reservation prices (see reservation thresholds) reservation thresholds 55, 56, 58, 59, 62, 107, 132, 216–18, 229 residual power 158, 178–80, 193, 246 routines 56, 59, 62, 63, 111, 156, 239–40 satisfaction 55, 56, 69, 78, 107, 130–2, 166 self-deception devices 76–79, 80, 83, 93–4, 96, 101, 103, 109, 121, 212, 214–15 self-managed firm (see co-operative firm) self-ownership 2–3, 20 services (contract for) 5, 8–9, 11–12, 16–18, 39, 41–2, 159–60, 192, 195, 228 sharecropping contract 7–9, 160, 195 small menu costs 221–2, 252 social actors 23–7, 29, 44, 90, 112, 114–15, 119, 121–5, 132, 136, 145–6, 157, 186–7

310 Subject Index ofindex names social classes 2, 29–31, 90, 130, 138, 157 social conventions 107, 114–16, 134–5, 137 social movements 26–7, 115–16, 152–4, 157, 175 social reproduction (the cultural conditions of) 96–9, 270, 273–4 social reproduction (the financial conditions of) 31–4, 272–3 social reproduction (the political conditions of) 99, 136–8, 143, 270, 273–4 social reproduction (the production conditions of) 34–9, 271–3 Soviet Union 2, 247, 279, 283, 286 specialisation 147–8, 153, 156, 175, 188–9 state 117–20, 122–3, 131–9, 141–3, 259, 269, 274 state autonomy 138–9, 141–3 state capitalism 2, 144, 248, 267, 276–9, 283, 286

surplus value 7, 10, 12, 22, 42, 136, 147, 155, 160, 277 systemic effect 169–71, 183 team production 148, 190, 196, 198–9, 202–5 tolerance thresholds (see reservation thresholds) transaction institutions 23–6, 28, 40, 110, 112, 115, 117, 135, 137, 148, 165, 230 undervaluation 76–9, 83, 94 unintentional actions 64, 86–7, 89, 102, 110 wage 7, 10, 16, 22, 80, 102, 160, 195, 200, 206–10, 230 workers (see working class) working class 2, 30–1, 91, 95, 102, 105, 132–3, 137–8, 173, 176, 201, 278, 281–2