The Active Consumer: Novelty and Surprise in Consumer Choice (Routledge Frontiers of Political Economy)

  • 75 19 1
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

The Active Consumer: Novelty and Surprise in Consumer Choice (Routledge Frontiers of Political Economy)

THE ACTIVE CONSUMER The consumer is also a maker. Just like a producer, consumers inevitably and constantly find themse

1,591 343 1MB

Pages 279 Page size 396 x 612 pts Year 2001

Report DMCA / Copyright


Recommend Papers

File loading please wait...
Citation preview


The consumer is also a maker. Just like a producer, consumers inevitably and constantly find themselves transforming time, material resources, and market goods into something that is good for them. But what is good for them is not inscribed in the goods already available; it has to be discovered. Options and opportunities must be detected and exploited. The consumer therefore is also an entrepreneur, looking for ways to create and add value as well as differential advantage for him or herself. Novelty is an outcome of this process; it is also a fundamental motivation for this activity. Consumers seem to delight in trying new solutions, exploring new combinatory possibilities. Many of these facets of consumer behaviour are barely touched on in traditional economic theory, where the consumer’s motivation is subsumed under an unanalysed preference set, the consumer simply responds passively to exogenous changes, and he or she never engages in producing change. Other literatures have taken up some of these themes, but not as a central concern, rather as insights from their separate disciplinary perspectives. The active consumer makes these themes central. The book aims to provide an economic–theoretical understanding of the many ways in which innovation can structure consumer choice. The authors show from different points of view how central novelty can be in consumer behaviour, how it is related to technical change, how new consumer capabilities are developed and organized, or may involve the consumer in costly errors, how novelty dictates the multiple lives of products, and how the desire for it can induce change in labour supply patterns. Conceptual and linguistic shifts needed to discuss consumption in this new light are also dealt with. Marina Bianchi teaches microeconomics and industrial organization at the University of Cassino, Italy. She has published in international journals on expectations and learning, on institutional evolution in a game-theoretic framework and on innovative strategies in firm behaviour.


EQUILIBRIUM VERSUS UNDERSTANDING Towards the Rehumanization of Economics within Social Theory Mark Addleson EVOLUTION, ORDER AND COMPLEXITY Edited by Elias L. Khalil and Kenneth E. Boulding INTERACTIONS IN POLITICAL ECONOMY Malvern After Ten Years Edited by Steven Pressman THE END OF ECONOMICS Michael Perelman PROBABILITY IN ECONOMICS Omar F. Hamouda and Robin Rowley CAPITAL CONTROVERSY, POST KEYNESIAN ECONOMICS AND THE HISTORY OF ECONOMIC THEORY Essays in Honour of Geoff Harcourt, Volume One Edited by Philip Arestis, Gabriel Palma and Malcolm Sawyer MARKETS, UNEMPLOYMENT AND ECONOMIC POLICY Essays in Honour of Geoff Harcourt, Volume Two Edited by Philip Arestis, Gabriel Palma and Malcolm Sawyer SOCIAL ECONOMY The Logic of Capitalist Development Clark Everling NEW KEYNESIAN ECONOMICS/POST KEYNESIAN ALTERNATIVES Edited by Roy J. Rotheim THE REPRESENTATIVE AGENT IN MACROECONOMICS James E. Hartley

BORDERLANDS OF ECONOMICS Essays in Honour of Daniel R. Fusfeld Edited by Nahid Aslanbeigui and Young Back Choi VALUE DISTRIBUTION AND CAPITAL Edited by Gary Mongiovi and Fabio Petri THE ECONOMICS OF SCIENCE James R. Wible COMPETITIVENESS, LOCALISED LEARNING AND REGIONAL DEVELOPMENT Specialization and Prosperity in Small Open Economies Peter Maskell, Heikki Eskelinen, Ingjaldur Hannibalsson, Anders Malmberg and Eirik Vatne LABOUR MARKET THEORY A Constructive Reassessment Ben Fine WOMEN AND EUROPEAN EMPLOYMENT Jill Rubery, Mark Smith, Damian Grimshaw EXPLORATIONS IN ECONOMIC METHODOLOGY From Lakatos to Empirical Philosophy of Science Roger Backhouse SUBJECTIVITY IN POLITICAL ECONOMY Essays on Wanting and Choosing David P. Levine THE POLITICAL ECONOMY OF MIDDLE EAST PEACE The Impact of Competing Arab and Israeli Trade Edited by J. W. Wright, Jnr THE ACTIVE CONSUMER Novelty and Surprise in Consumer Choice Edited by Marina Bianchi

THE ACTIVE CONSUMER Novelty and Surprise in Consumer Choice

Edited by Marina Bianchi

London and New York

First published 1998 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 This edition published in the Taylor & Francis e-Library, 2001. © 1998 Marina Bianchi – editorial matter and selection; contributors – individual chapters 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 The active consumer: novelty and suprise in consumer choice/[edited by] Marina Bianchi. p. cm. Includes bibliographical references and index. 1. Consumption (Economics) 2. Consumer behaviour. I. Bianchi, Marina. HB801.A275 1995 658.8´342—dc21 97-32108 ISBN 0–415–17190–3 (Print Edition) ISBN 0-203-02291-2 Master e-book ISBN ISBN 0-203-17214-0 (Glassbook Format)

To two early navigators of these and other uncharted waters, Tibor Scitovsky and Israel Kirzner


List of figures List of tables List of contributors

xi xii xiii

1 Introduction




The hedonics of taste: utility, novelty and change 2 Choice without utility? Some reflections on the loose foundations of standard consumer theory




3 Economic change, choice and innovation in consumption



4 Taste for novelty and novel tastes: the role of human agency in consumption




Consumers as producers and problem solvers: consumption capabilities 5 Cognition and innovation

87 89


6 The organization of consumption





7 Consumer goals as journeys into the unknown




Adoption and diffusion of new goods: illustrative analyses 8 Work and the sirens of consumption in eighteenth-century London




9 Silk purses out of sows’ ears: mass rarefaction of consumption and the emerging consumer–collector



10 Novelty, imitation and habit formation in a Scitovskian model of consumption




Consumption and communication


11 Consumption in postmodernity: social structuration and the construction of the self



12 On the consumption of signs







2.1 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 8.1 8.2 8.3 8.4 8.5 9.1 9.2 10.1 10.2 12.1

Pareto’s drawing of Buridan’s ass Selecting the choice set Local efficiency gains Local utility gains Local efficiency and utility gains Local efficiency gains with no utility gains Local utility gains with no efficiency gains Local efficiency gains and utility losses Local utility gains and efficiency losses Time of starting and stopping work, 1750 and 1800 Changes in the odds ratio: Monday and weekdays Changes in the odds ratio: old holy days Real wages in London, 1749–1805 Time allocation in the Becker model Rarefaction processes in the firm–consumer relationships Types of rarefaction processes Time paths of consumption stocks sj, j = 1, 2, 3 Estimated invariant distribution, consumption stock of good 1 Arrow


29 77 77 78 79 80 80 81 81 146 147 148 154 163 181 187 207 208 239


8.1 8.2 8.3 8.4

Changes in labour input, 1750–1800 Incidence of paid work, 1749–1803 Price of clothing, selected benchmark years 1750–1800 Incidence of paid work, 1749–1803


149 155 161 165


Marina Bianchi is Associate Professor of Economics at the University of Cassino, Italy. Metin M. Cosgel is an Associate Professor of Economics at the University of Connecticut, Storrs. Peter E. Earl is Professor of Economics at Lincoln University in New Zealand. Davide Gualerzi is currently a Researcher at the Department of Economic Sciences, University of Pisa, Italy. Guido Guerzoni is Assistant Professor of Economic History at the Luigi Bocconi University, Milan, where he also teaches Cultural Economics. Michael Hutter is Professor of Economic Theory at Witten/Herdecke University, Witten, Germany. Richard N. Langlois is Professor of Economics at the University of Connecticut, Storrs. Brian J. Loasby was Professor of Management Economics at the University of Stirling where he remains as Honorary and Emeritus Professor. Fabio Ranchetti is Associate Professor of Economics at the University of Pavia and the Politecnico di Milano, Italy. S. Abu Turab Rizvi is an Associate Professor in the Department of Economics at the University of Vermont. Rajiv Sethi is Associate Professor of Economics at Barnard College of Columbia University. Gabriele Troilo is Assistant Professor of Marketing at the Luigi Bocconi University, Milan. Liisa Uusitalo is Professor of Marketing Communications at the Helsinki School of Economics, Finland. Hans-Joachim Voth is Visiting Professor at the economics department, Stanford University.


1 INTRODUCTION Marina Bianchi

When the first tulips appeared in the Netherlands is not known, though a shipment of them was recorded in Antwerp in 1562. By the first half of the seventeenth century, however, a craze for tulips had swept Holland – a country renowned for its moralizing about vigilance and moderation – like a disease. By that time several hundred varieties existed, with pretentious or at least tantalizing names – Miracle, Paragon of Perfections, General of Generals, Prince, Semper Augustus – and with what seemed infnitely fine variations of colour, shading and colour combinations. During the craze prices reached vertiginous heights and many fortunes were quickly made and unmade. Still, the fortune of tulips lasted long and well beyond the speculative period of the early 1630s. What made the appeal of this flower irresistible was its ever-varying colour possibilities. A virus tended to affect bulbs with the result that plain colours could ‘break’ into splendid yet always unpredictable flamed variants. No wonder that the tulip was thought to possess miraculous properties and to generate spontaneously new mutations. It is not difficult to relegate this, along with many other examples of consumer ‘manias’, to those forms of passionate behaviour which lie beyond the realm of economic enquiry. The contagiousness of the ‘fever’, as well as the futility of the effort spent, reinforces the sense that this sort of behaviour is irrational and exceptional. Still, the tulip example, like many others, shows something that is not at all bizarre, namely the pleasure that consumers seem to take in novelty and change. There is more to learn from novelty and surprise in capturing consumers’ desires and choices than we are accustomed to consider.



NOVELTY AND SURPRISE Novelty may be defined as a mismatch between present and past experience. The degree of this mismatch can vary widely. It might be very high, as when something is new relative to everything that has been experienced before, or low, as when we encounter new combinations of already experienced events. The time dimension of novelty is also important, as when mismatch is relative to a more or less recent set of experiences. Surprise, on the other hand, is something contrary to our existing set of expectations. Novelty too can be surprising when it is not expected, and vice versa, but novelty and surprise are distinct. Something surprising (unexpected) may not be new, and correspondingly, something new can be expected.1 These distinctions and characterizations of novelty and surprise are to be found in the writings of D.L. Berlyne (1960, 1971), a Canadian experimental psychobiologist who studied the hedonic components of human behaviour. Novelty and surprise, in Berlyne’s view, are among the variables which, by generating changes in the stimulus potential of a specific situation, are also responsible for how pleasurable, interesting and rewarding the situation is perceived to be. Berlyne found, in experiments, that a novel situation, or one that is uncertain, complex, surprising, seems to have a double effect on its pleasantness. A high degree of novelty – one which is signalled by the presence of a sharp disjunction between actual and already experienced events – is perceived as distressing and threatening, and therefore as unpleasant. Correspondingly, a low degree of novelty, being repetitive and unchallenging, is also perceived as unpleasant. Both situations trigger actions by subjects that tend to increase pleasure, either by reducing novelty or by increasing it. Solving a difficult puzzle, understanding a complex mathematical model, mastering a new language with laborious application, are all pleasureinducing activities insofar as they reduce complexity, novelty and uncertainty. On the other hand, even deciding to start these activities when one is not forced to, or to follow more complex paths in solving problems, are part of a desire to increase pleasure, in as much as they add challenge and novelty to a situation which was previously perceived as already routinized, hence dull and boring.2 It is not difficult to show that pleasure-inducing processes, considered in this light, have no ending or, if one prefers, no equilibrium point.3 Success in solving a problem or in mastering uncertainty inevitably reduces its stimulus or novelty potential and the pleasure associated with it. The same is true for those activities which tend to increase novelty, where it is difficult to find a balance between falling back on reliable routine(s) or risking too much. Therefore, even if a point could be found where pleasure is at its maximum, and the degree of novelty is neither too much nor too little – a situation for which, in Berlyne’s terminology,



the arousal potential is optimal – one cannot, even if one would want to, protract this situation through time. Instead, new actions would be taken that introduce variation and change. The result of this dynamic is that, even in the absence of any exogenous shock or, better, because of this absence, change would be produced endogenously in the form of exploratory and search behaviour. By uncovering the motivational importance of novelty, complexity, uncertainty, and the impact they have on behaviour and choice, Berlyne’s approach throws new light on the way cognitive and learning processes work.4 If cognitive theories have acquainted us with the procedures and organization structures of learning, we also know that learning, search, trial and error procedures are a self-perpetuating activity.5 Problems are generated not only by environmental uncertainties, but by the fact that any problem solution is also the condition for new problems to arise. Learning in this way is a discovery process, endogenously innovative. But, invoking Berlyne, we can now also see not only that learning and novelty go together, but also that learning, novelty and pleasure are connected in quite basic ways. People delight in producing change, in creating new options or in controlling ones which are complex.6 In brief, the process of discovery is pleasant, and actively looked for. But what is discovery, and what is its role in production and consumption activities?

DISCOVERY IN PRODUCTION AND CONSUMPTION In Israel Kirzner’s book Discovery, Capitalism and Distributive Justice (1989), the brilliant chapter entitled ‘The meaning of discovery’ contrasts the pure act of discovery with that of production. Production involves the execution of a coherent plan where well-defined goals are achieved with a minimum expenditure of effort, time and material resources. Discovery, on the contrary, is the creation ex nihilo of options which did not exist before. As a simple illustration of the difference, imagine an explorer whose way is blocked by an unexpected river and no bridge to cross it. In the vicinity, however, there is a tree which, for its strength and length, might respond to the need of reaching the opposite side. In building a bridge with the tree and in doing it well, such an explorer is obeying the rules of efficient production. However, that he or she could act in the way supposed, that this new option was opened, as Kirzner never tires of pointing out, is a pure act of discovery. Noticing a tree, and innovatively connecting it to the possibility of bridging the river, was not given in advance but created. Chance and search play a role in processes of this sort; there might have been no trees, or no tree noticed, had the explorer not looked for some way to cross the river. But,



Kirzner correctly stresses, discovery is not reducible to chance or search. Discovery is wholly due to the explorer’s ability to take advantage of existing opportunities in ways that are not yet explored, to his or her alertness to detect new and gainful options. Therefore, if search aims at filling known gaps in knowledge, discovery deals with unknown gaps, with ‘errors’ and as yet ungrasped opportunities. Conversely, if chance may provide these unexpected, unforeseen opportunities, they will remain idle without the ability to recognize their value and to exploit it. Discovery and production are logically distinct, yet in real life there is an inescapable element of discovery in every act of deliberate production (Kirzner 1989: 38). That is Kirzner’s novel insight. In a world of genuine uncertainty and open-ended ignorance (see Loasby 1991), without discovery, without the creation of the conditions for it to happen, no productive plan would be possible. Production, in other words, is not separable from, in fact is, entrepreneurship: the ability to innovate, to discover gainful opportunities where nobody has yet detected them. Correspondingly, though Kirzner does not emphasize this point, production helps discovery. Accumulated knowledge and skills, expertise and capabilities, favour alertness and innovation. An expert explorer, though taken by surprise by the river, still travels with tools versatile and flexible enough to cope with different scenarios, and is highly skilled in the art of using them innovatively and profitably. Search, therefore, even when filling known gaps, may reveal unknown ones. In the same way, production, though deliberate, will tend to involve the emergence of new and unplanned possibilities to explore, and some of these will be recognized because of a learned capability. Novelty, in short, is not simply exogenous but also inherent to the production process. The upshot of this discussion is that if discovery, and therefore also novelty and surprise, is ingrained in any act of production, it represents a central and unavoidable, not a peripheral, element of our decision processes and ‘cannot be imagined away’ (Kirzner 1989: 66). It is not difficult to recast both this discussion and the previous one on the motivational aspects of behaviour in terms of the consumer’s decision problems. There are two ways in which novelty is relevant for the consumer. As has been emphasized already by economists, in consuming the individual also produces his or her own commodity set, either indirectly, by earning spendable cash, or directly, by transforming market goods and own time and effort into the desired consumption commodities. But the consumer as producer is also engaged in a process of discovery, in being able to detect as well as to create new and favourable options. Though market goods may already exist as inputs in consumption, still the output (or outcome) is not inscribed in them once for all, but is the result of multiple, as yet unknown, combinatory consumption possibilities. How to grasp them and how to exploit them effectively has yet to be discovered. Consuming therefore implies also the production of novelty, the



creation of opportunities for consumption to happen. On the other hand, novelty is also a stimulant for action. Both the change which is produced by reducing too high a level of uncertainty and novelty and the change which comes by increasing too low a level, are enjoyable and actively pursued. Examples of both come readily to mind in connection with consumption activities. This dual role of novelty – to have both a ‘hedonic’ and a ‘productive’ value – changes the way we are accustomed to thinking about consumer behaviour and choices and opens new research paths. Before starting to address some of them more closely, I want to reflect briefly on the incentive structure that motivates consumer decisions in our inherited economic theory of consumption.

OUR INHERITED THEORY: FROM UTILITY TO INDIFFERENCE The pursuit of pleasure and the avoidance of pain have been the motivational ingredients of consumer behaviour since its first systematic analysis by Gossen and especially Jevons. Utility theory translated these motives into a model of choice based on constrained maximization whose power in describing rational behaviour has been very strong. Yet the discussions that from the beginning accompanied utility maximization explanations have raised several problems, which still remain unsolved. The very word utility had in it the seed of an original sin which various attempts to replace it with a better term – desirability, ophelimity, wantability – tried in vain to eradicate. The problem was that ‘utility’ meant both the objective qualities in a good that responded to individual wants and desires (usefulness) and the desire itself, the subjective happiness anticipated (see Georgescu-Roegen, 1966, 1988). Utility seemed to rest both in the object and in the mind. Which one was to be maximized? The first led to considerations of quality, and of the characteristics and efficiency of the good in satisfying wants, but also introduced unwelcome normative distinctions between what is useful and what pernicious, between ‘legitimate’ and ‘illegitimate’ wants. The second, the feeling of the mind, introduced into the theory an element of subjectivity that seemed to defeat all attempts to measure desire and to undermine the empirical grounding of the theory. Bentham’s unsuccessful struggle to elaborate a coherent felicific calculus is testimony to the difficulty (Bentham 1970 [1789]: Ch. VI: 3; see also Mitchell 1968 [1918]). As is known, economic theory chose to focus on the feeling of the mind, leaving aside considerations of objective usefulness. The law of decreasing marginal utility, or of satiable



wants,7 seemed to restrict the variability of desire within the requirements laid on the theory (to explain downward-sloping demand curves). Still, as the critics did not fail to notice, the psychological grounds of this law, where desires are pure quantitative reflections of utilityyielding objects, were very feeble (Viner 1968 [1925] for a discussion of these criticisms) or, worse, unnecessary (Cassel 1934: one can use demand functions directly with no reference to utility). The introduction of the indifference curve as a step towards ordering consumer preferences and choices meant a radical change of the whole debate and offered a logical way of ending what were thought to be the psychological and philosophical sophistries connected with the reference to utility.8 By simply assuming a complete ordering of preferences where indifference is some sort of imaginary divide between what is preferred and what is not, we have all that is needed (aside from satiation) for a theory of equilibrium choice. As Pareto put it (1971 [1909]: 119n. and 113), the entire theory could now rest on pure experience, the empirically given curves of indifference; there was no need for considerations of pleasure and its measurement. Slutsky’s contribution, Samuelson’s Revealed Preference theory, and the subsequent axiomatization of choice, represent successive refinements of the abandonment of ‘utility’.9 Though utility is still the name economists give to whatever it is consumers are supposed to maximize, utility is in fact superfluous to the standard mode of explaining behaviour – a conviction challenged by Ranchetti in Chapter 2. There has been a clear loss in this evolution: consumer motivation or the incentive to action. Since preferences, as revealed by market choices, are the ultimate, indecomposable, and given elements of actions, what is maximized is no longer known, knowable, or even relevant. How to proceed from the known (of choice) to the unknown (of ophelimity and pleasure) – Pareto’s aim when he introduced the indifference curve (see his letter to Maffeo Pantaleoni 1899, quoted by Georgescu-Roegen 1988) – has in fact been sidestepped as a problem.10

THE ROLE OF NOVELTY IN CONSUMPTION By losing sight of what consumers’ goals and motivations are, economic theory has introduced a strange asymmetry between production and consumption choices (though the logic of decision remains the same). While in the case of production firms face a well-defined objective function, the profit function, quantitatively assessable and calculable, in the case of the consumer, the objective function, though internally consistent, is goal empty. Ends coherently match the means, but ends themselves are not rationally chosen. Had we to apply the same



rule to the firm – pure consistency of choice independent of any specification of the ends – even profit-minimizing firms would be rational. Hypothetically consumers, but not firms, may never – or never be found to – err, since goals are redefinable in terms consistent with consistency. Economists may tacitly assume that this asymmetry lies in reality, not in the theory. But is that really so? What is worse, is that this asymmetry and our knowledge about it has recently grown sharper. Thanks to the institutional theory of the firm much progress has been made in the understanding of the organization structure, goals and incentives of the ‘firm’ (the literature on the topic is increasingly vast; see March 1988; Cyert and March 1992; Williamson 1993). There is no corresponding stream of research on the consumer side. Much insight and empirical understanding is embedded in the marketing literature, in anthropological studies of consumption, and in historical and material culture enquires, but little of this has been taken into economic thinking. This book arises from the recognition of this disparity. It explores several possible responses to the relative poverty of economic discourse about consumption; and it uses as point of entry a novel reconsideration of the incentive structure of consumer choice. I have already mentioned the first step in this direction: that of considering consumers also as active producers of their consumption set. Far from belonging to distinct and distant worlds of experience, producers and consumers face decisional problems of the same kind. To acknowledge this implies that considerations of utility in terms of ‘usefulness’, abandoned though they have been by economic theory as dispensible normative intrusions into free choice, must become important again. If market goods are in fact inputs for the production of individual welfare, these inputs can be more or less efficient and consumers may be expected to decide about that too.11 As I have said, this step has already been taken in the economic literature, mainly through the contributions of Becker’s model of household economics and Lancaster’s characteristics approach. On both, there is more in this book (especially the chapters by Loasby, Gualerzi, Bianchi). Nonetheless, two things are completely missing from the analyses of Becker and Lancaster which are worth noting. The first is that consumers, though represented as active producers, completely lack the entrepreneurial dimension which, as Kirzner has shown, inevitably characterizes production.12 Consumers are never engaged in any act of discovery. How inputs of goods are transformable into consumption outputs is always given in advance in the theory and is assumed perfectly known. There are no unexpected rivers to cross for the consumer, and all rivers invariably have bridges. Efficiency thus comes to be confined to (known) cost reduction, nothing more. This leads to the second element missing from the analyses of consumers as producers, their motivational structure. Motivation either remains quite unanalysed, as in the case of Lancaster, or it is completely reduced to the efficiency dimension as just defined. Opportunity



cost advantages are deemed the only motives behind choice; and any act of choice, it is assumed, can be translated into some corresponding cost representation. But how costs are assigned to the various choice options in the first place remains unknown. Furthermore, considerations of possible pleasure advantages are completely eliminated. No consumer is thought to enjoy exploring, for example; rather, the emphasis is entirely on the costs associated with activities such as search. Trying to recover and address these two missing elements requires taking several steps along new paths. One is to recognize that in the production of consumption possibilities there are constant new gainful opportunities to grasp and new losses to prevent. As does the firm, so also must the consumer create ‘profit’ opportunities and efficiency gains. As I have stressed, these profit advantages are not openly displayed, but are the result of multiple and changing combinatory options: for example, one can multiply dress opportunities using cheaper separate components, or one can custom-design one’s own multi-media centre out of selected components.13 Exploring how consumers discover, organize and exploit efficiency gains in consumption is a first step towards bridging the analytical gap between firms’ and consumers’ behaviour. A second step is recovering the motivational structure of consumer decision processes, which means recovering that aspect of ‘utility’ which is related to subjective pleasure and happiness. Though introducing motivations, tastes, desires, into the analysis might seem to broaden scope at the expense of explanatory content, we have seen that this does not have to be the case. Desire for novelty and change seems to follow recognizable strategies of action and these are associated with empirically detectable exploratory behaviour and learning. They provide the incentives both for ordering rules and recognizability and for innovation and change. Also the social dimension of the consumer – which includes the communication side of individual choices – must recover a more refined and articulated place in the motivational structure of consumer decision procedures. In brief, consumption strategies, as we have seen, divide roughly into those associated with creating ‘efficiency’ gains and those linked to ‘pleasure’ gains. The workings of these two may be difficult to distinguish operationally: they may overlap and reinforce each other – the discovery of a new and unexpected consumption possibility may be a source of efficiency gain, though it is also a source of pleasure from the simple fact of its being new and surprising – or they may work in opposition to each other. Think, for example, of microwaving prepackaged food but losing the pleasure of cooking or, conversely, of tanning for two minutes under a tanning machine versus enjoying three hours at the beach; the end products may be quite similar, though distinct options are available for reaching the end. In both cases they may be difficult to distinguish and separate. Still, analytically, they are decisive for an



understanding of the multi-dimensionality of consumer choices. They tell us where the potential for innovations lies, as well as the sources of possible conflicts. In introducing this distinction between efficiency and pleasure gains or losses, I have anticipated many of the topics treated in the book. Let me now identify them more closely.

PART I THE HEDONICS OF TASTE I made the assertion above that there has been a succession of evasions in economic theory as to what the utility function really represents. In Chapter 2 Ranchetti traces some of these through an historical and critical study of the linkages between choice and utility. He argues in particular, that the term utility is used by contemporary economists purely as a qualifier in the name given to the mathematical function which represents individual preference orderings. Yet these same economists often justify their recourse to the term because of its hedonic resonance, because it gives content to what they claim is our strong intuition that some measure is needed of consumer satisfaction. That these two uses of the term utility are unproblematically lumped together is surprising, for Ranchetti. They belong to two different, indeed antithetical, traditions of thought about choice and consumption. One goes back to Pareto and his attempt to free the ordering of preferences from metaphysical entanglements; an empirical science demanded that entities such as pleasure and happiness be dismissed. The other has its roots in utilitarianism where, on the contrary, the term utility is strongly ethically biased (value laden). Edgeworth is probably the best economic representative of this tradition. He introduced indifference curves and the mathematical apparatus of contracting still in use, yet, in his view, ethics should have priority over economics. This was not just his personal preference. The individual utility calculus he thought should be subordinated to the social utilitarian calculus since the first does not guarantee that social happiness will be maximized. The fact that considerations of the measurement of pleasure continue to surface in the current pure theory of choice seems to suggest that they are still needed, despite the longstanding empiricist desire to oust them as impure elements. Yet of this philosophical conflict and its historical roots, modern theorists show no awareness. If it is not clear on the basis of received economic analysis just what agents are supposed to maximize, it is even less clear what happens to this process of maximization when novel goods or characteristics appear. Product innovation is taken up by Gualerzi in Chapter 3. He introduces a critical literature on the consumer which shows that, differences in approach notwithstanding, there exists a body of work that looks beyond the givenness of preferences which dominate the traditional approach. This critical literature does not amount to a new,



unified frame of reference for an alternative theory of consumption, as is evident if we focus on two of the crucial points which, for Gualerzi, must characterize an adequate alternative. One of these is: how technical change and product innovation translates into consumers’ decision procedures; the other, how consumers subjectively channel these changes so as to modify their tastes and preferences. Lancaster and Ironmonger, for example, correctly address the problem of innovation and quality changes in consumption and link it to the hierarchical nature of wants, yet leave quite unanalysed the way consumers evaluate and adapt new commodities to their individual wants. ‘Tastes’ continue to be exogenous. Conversely, adaptive models which endogenize preferences have variously stressed the dependence of preferences on habit formation, on production specifications or, as in Duesenberry, on emulation. Still, in (properly) removing the exogeneity of preferences, these models seem to remove also any active role for the consumer. Gualerzi outlines a theoretical path that, if explored, could avoid the reductionist approaches just mentioned and provide for a theory of the consumer which takes full account of changes in production and consumers’ own active responses. The problem of how new goods enter into and alter consumers’ decision procedures is also taken up in Chapter 4 by Bianchi. Goods do not exist, nor are they chosen, in isolation. They are part of a complex and changeable network of interrelations with other goods and characteristics. It is this combinable potential of goods which allows for variation and change in consumption. New goods, in order to become ‘goods’ have to play on recognizability, exploiting known similarities and rivalries among goods. But to be new they have to introduce characteristics and functions which alter the existing order and timings in the consumption set. Often in this process small changes activate large ones. To identify some features of this process of cumulative change, the chapter uses a specific historical example, the Indian painted calicoes which appeared in European markets in the latter part of the seventeenth century. The example shows that if a new good, besides being novel (within recognizable bounds), is flexible and unspecified enough as to function so as to combine in multiple ways with existing goods, then a constant stream of novel combinations may be produced. In seeking out and effecting these combinations the consumer plays the active role.

PART II CONSUMERS AS PRODUCERS AND PROBLEM SOLVERS The three chapters in Part II explicitly adopt the strategy of recasting consumers’ actions as production activities. Each of the authors has contributed already to understanding firm



organization and behaviour as capabilities and their chapters here take up the challenge of extending the new insights to consumption. Following the initiative of George Richardson, Loasby was one of the first economists to draw attention to the decision structure and strategy of firms which, in a world of uncertainty, must anticipate the unexpected and develop appropriate response capabilities (see Loasby 1991). Since the problem of the consumer too, though not addressed in the standard theory of choice, is how to deal with and to produce change, Loasby seeks to localize the difficulty by exploiting Gilbert Ryle’s distinction between knowledge that – here knowledge of the opportunity set – and knowledge how, which involves the very process of choosing (Ryle 1949). Choosing, however, is costly, time consuming and often suboptimal. Knowledge, values, and preferences cannot be assumed at the outset, but emerge and are developed incrementally as choices are made. Like Kirzner, Loasby stresses that a more effective choice entails an increase in the skills of decision-making, which range from evaluating options already available to the discovery of how to introduce new ones. To focus on this process of evolving capabilities requires an analysis of the conditions which make such evolution possible. The recognition of markets as institutions shows their importance in supplying a stable framework against which perceptions and actions can be identified and distinguished – a role recognizable especially when markets are missing. But variation is also needed for learning to take place. Individual differences will occur as consumers build their own patterns of connections and construct their own consumption strategies. This inevitably will generate variety and increased opportunties for innovation. Loasby’s contribution is complemented by that of Langlois and Cosgel in Chapter 6. They ask: how are consumers to acquire the alertness and other capabilities needed to discover consumption opportunities and take advantage of them? Drawing on earlier work by Langlois (1990) and Langlois and Robertson (1993, 1995) for the firm, they seek to adapt those findings to the consumer. Consumers can delegate their problem to the market, as when consultant services are bought. But they can also internalize such functions, as occurred, for example, with the generation of do-it-yourselfers who played such an important role in the early stages of the microcomputer industry. Routines seem to be as important to the consumer as they are to the firm and, as with the firm, these have strong implications for communication. Routines must be shared interpersonally. Moreover, a coordination problem arises when producers and consumers have different sets of routines; solving it favours the emergence of standards, of ways of fixing connections among the various subsystems in order for change to be channelled and created autonomously. In the case of consumer-generated innovations, which are most effective in assemblages and recombinations of existing possibilities, modular systems are decisive for economizing on capabilities and favouring the generation of novelty. In the case of producer-generated innovations, however, the creation of brands provides



standardized meanings and this allows consumers to do away with specialized, external consultancy for ascertaining uses and quality. Within this framework, the boundaries between producers and consumers are permeable. They will of course shift, reflecting the changing nature and distributions of capabilities in the economy as it undergoes its own changes. The decision implications for consumers who have to acquire mastery in a highly uncertain and never fully known world (like walking in a dense, pathless wood), are given an important twist in Chapter 7 by Peter Earl, who has written imaginatively elsewhere on both the firm and the consumer. He argues here that consumers, like firms, are systematically prone to making mistakes. Mistakes are related to the very nature of being a consumer and are not confined to the simple information failures familiar from economic analysis. Consumers face decisions that are not marginal, that require standards to be met and are binary in nature; they pursue goals that are sequential. Even successful outcomes, such as that of ‘arriving’ at a high income position, may be sources of costly errors. Having arrived, one might then inappropriately stay, based on biased estimates of one’s continued success, perhaps leading to opportunities needlessly sacrificed. Having arrived, one also may begin to display an impaired ability to develop new tastes. Such responses make it difficult or impossible to expand, or adjust, the feasibility set: no axiom of completeness can be applied. The sources of possible mistakes are numerous. They include difficulties (or resistances) to rendering the unfamiliar familiar – think of the well-known degree of under-exploitation of the use opportunities in such available market goods as VCR machines; misplaced assumptions; and difficulties in framing and perceiving one’s own aspirations.

PART III THE ADOPTION AND DIFFUSION OF NEW GOODS The focus in Part III shifts from the theoretical to the empirical or, in one case, to simulation study of some of the issues now familiar. How consumers pursue new goods and adjust their working decisions in response to new consumption alternatives – the lacuna in available theories particularly stressed by Davide Gualerzi – is taken up in an historical context by Hans-Joachim Voth in Chapter 8. During the eighteenth century (though starting earlier), as many economic historians have noted, a flood of new goods appeared on the English market. The problem for economic historians has been to show where the buying power for these goods came from. Voth, exploiting a new data source, argues that it came from decisions by would-be consumers to lengthen their own working week. Using witnesses’ accounts from the Old Bailey Sessions Papers during the



second half of the eighteenth century, he is able to show that while time use during the day remained unchanged, ‘Saint Monday’ disappeared, as did many political and religious days off. More work – up to 50 percent more – came to be done in an average week. Voth challenges the adequacy of the reasons usually given for this increase in labour supply, such as the need to keep costly machines in use as the capital/labour ratio rose, declining nutritional constraints, or a response to declining wages along a backward-bending labour supply curve. On the contrary, what seems to be of major relevance is the increased availability of new and cheaper consumer durables. In terms of Becker’s model of time allocation, one can imagine that there was a substitution effect in favour of these goods, partly because they were new and gave greater pleasure, and partly because they were cheaper and generated increases in consumption productivity. The productivity gains and cost savings also issued in a positive income effect. The logic of the model requires that if the income effect caused demand to rise for other goods as well, namely, high monetary cost goods which added to total consumption costs, then, with given labour productivity, consumers had no option but to extend their working week. In Chapter 9, Guerzoni and Troilo give a modern version of the siren song of consumption described by Voth. The personalized jeans recently offered in selected outlets by Levi’s, for example, and the emergence of owners’ clubs among possessors of the classic Volkswagen ‘Beetle’, are expressions of an apparently contradictory phenomenon: the simultaneous presence in modern consumption patterns of high-tech mass customization and an antitech customization which privileges goods and services that are anachronistic, pre-modern. Such cases, the authors argue, reflect a change in the relationship between producers and their customers, whereby standardized goods are yielding ground to more individualized variants. These last, through customization and the passage of time, tend to become unique; as Guerzoni and Troilo put it, they are subject to a process of ‘rarefaction’ – using this word according to its Latin root. This process can be activated by firms in search of strategic advantages, but, significantly, also by consumers in search of differentiated and new goods. This chapter is rich in examples which uncover the novel aspects of the phenomenon. They show too that there are close analogies with the activity of collecting. These individualized goods are scarce and special and they form a series – they receive their meaning and significance from being part of a set. How to enlarge the series with new ‘specimens’ becomes a new strategic behaviour on the part of the consumer. The consumer behaves as a collector, a serial consumer. Functionality, in this behaviour, tends to dissolve, being replaced by the pleasure of the search, the stimulus of discovery and the gratification of possession. If constant newness, partly consumer induced, is in fact the normal situation that needs to be explained, how do we move from the micro-inducement of the single consumer to that ‘shared’ behaviour which creates and alters whole markets? This was the great analytical gap



that Jevons could not bridge. Recent work on the diffusion of innovation, facing a similar gap, has tended to be ‘epidemiological’ and probabilistic, rather than price theoretical in nature – as in diffusion models. In Chapter 10 Rizvi and Sethi offer a simulation model complement to such work. What is new and unique is that they expressly try to model the Berlyne theory, as re-expressed for economists by Scitovsky. Not only that, but this is the first attempt to explore Scitovsky’s insights on a population basis, rather than for individual laboratory subjects. Their model identifies three sources of motivation in consumer behaviour: the pursuit of novelty and associated stimuli; the (possibly conflicting) sticking with habits; and the desire to imitate and conform to the behaviour of others. When habit formation has eroded the stimulating power of novel situations, one seeks variation by drawing on the alternative behaviours of other consumers. The interplay of these forces, not being deterministic, generates time paths of consumption stocks which are unpredictable in the short run. However, regularities are discernible in the long run and take the form of invariant frequency distributions. These results therefore challenge the traditional understanding of a consumer who predictably, though passively, responds to price–income stimuli. Here the very fact that the consumer is active breaks the certitude of the responses. Consequently, what for the actor may be a deliberative and purposeful choice takes on a random character for the observer.

PART IV CONSUMPTION AND COMMUNICATION Pierce insisted that meanings are not attached to goods, but are social constructs – relational in character, therefore, like signals exchanged between individuals. This perspective, applied to consumption, is explored in Chapters 11 and 12 by Liisa Uusitalo and Michael Hutter. Uusitalo draws on a very wide range of literature covering recent psychological, sociological, and cultural studies of consumption, to show how the theoretical representations of consumption have changed over time within each of these disciplines. Changes in the way consumption has been understood reflect, she suggests, the post-World War II growth experience and corresponding changes in the vision of society. In what may be called the postmodern phase of consumption – characterized by a disruption of ideological and social bonds and by a ‘de-differentiation’ of society into loose networks of relations – two different features of consumption are emphasized in the three literatures she analyses. On the one hand, consumption is seen as providing the social and integrating force which was previously performed, in the main, by work activities. In a postmodern world, it is around consumption,



whether in the form of cultural preferences, life style patterns or even purely imaginary experiences and emotions, that social bonds and communities are formed (and destroyed). On the other hand, consumption is represented as a process of self-production and selfidentification. Though echoing an old tradition, according to which consumers strove for social status and distinction, this view also tries to explain new and different phenomena, from taste differentiation to upward social mobility. These can now be viewed as indicating the way consumers actively use signs in self-expression. Having surveyed the way consumption appears in postmodern society and its literature, Uusitalo introduces a sceptical note. She observes that postmodern tendencies in consumption have been inferred more or less exclusively from the study of signs, not from the way consumers see consumption, nor from their behaviour. This leaves open the possibility that postmodern readings, which stress ‘imaginary consumption, affective communities and consumers’ self-construction’, are ‘just artefacts’. Hutter presents a very different case for attending to signs. Eschewing the postmodernist penchant for ‘reading’ consumption through its signs, he argues that there is an economics of signs behind what appears to be a self-contained economics of material exchange and production; and not just ‘behind’ the appearances, but actually driving them. For signs enter directly into consumption in the form of pictures, sounds, electronic signals, while many economic activities exist and are performed uniquely for their signifying qualities. Search, negotiation and control, which are involved in transactions, also include a whole chain of signs and signal procedures relating to trust and credibility, competence and commitment. Reputation-related and self-identity types of activities make use of signs. Monetary exchange, to take a more obvious example, makes use of a medium which has both a signifying and a signified value. The use of signs in economic relationships is thus much more widespread than usually perceived. Drawing on semiotics, signs are treated in this chapter as the relational link between a class of signifiers and a world of signified objects. But they are also forms of distinction, which actively mark the boundaries between signifiers and objects, and help the observer to give meaning to the outside world. For their inclusive–exclusive quality, therefore, signs and their use constantly create and recreate the plays which constitute the social world. Hutter spells out several consequences of this new set of distinctions that are relevant for the economic consumption of signs: how consumers themselves use signs, how signs are produced and reproduced, and how signs are priced. To conclude, the chapters collected here are first attempts to address several specified lacks in the traditional economic theory of consumption. They do so in almost every instance by altering the terms of reference for understanding consumption. Novelty is made central; acquiring capabilities to perceive and exploit the unexpected is considered as unavoidable.



The first element poses measurement problems for standard models – the new and the old are not comparable goods – and forces us to re-examine the accepted link between equilibrium as a resting place and optimality. The second also makes necessary a rethinking of economic rationality, since wise decisions come to involve learning ways of dealing with the unexpected. At two different levels, we also see from these chapters that the traditional sharp distinction between production and consumption only serves restricted goals. Novelty and discovery are integral to production, even as traditionally considered, as well as to consumption. Moreover, if we view production and consumption phenomena as being encompassed within a complex of sign exchanges, the distinction loses much of its force. In short, these chapters invite reconsiderations and outline something of an agenda for those who enjoy exploring new paths – precisely the character here ascribed to the active consumer.

ACKNOWLEDGEMENT The financial support of the Italian CNR and MURST 40 per cent is gratefully acknowledged.

NOTES 1. This distinction is clearly made by Henry Home (Lord Kames) (1762), to whom Berlyne (1971: 142) refers. 2. A large body of experimental evidence is provided in Berlyne and Masden (1973). Reference to the psychobiology of arousal and mood is in Thayer (1989). 3. As Scitovsky, the first to discover Berlyne’s relevance to economics, did not fail to notice. 4. The motivational components of behaviour and search is an aspect often neglected even in the psychological literature, as, for example, Amsel (1989) remarks. 5. On the endogenous and innovative character of learning see Bianchi (1992), and, for learning applied to the firm’s behaviour, see Bianchi (1995). 6. The fact that the search for novelty is activated by both ‘pleasure’ and ‘problem-solving’ considerations is noticed and discussed by Witt (1994). He also offers some insights on the environmental conditions which might favour or prevent the changes induced by novelty. 7. As Marshall says, wants are of an endless variety, but each separate want is limited (1961 [1890]: 93). 8. For example, are some people better pleasure machines than others? And should they then be better rewarded? 9. For a still useful discussion of the early history, see Stigler (1950); for an overall critical analysis


INTRODUCTION and of the assumption of indifference in particular, see Georgescu-Roegen (1966, 1968) and Ranchetti in this volume. The problems arising from purely ordinal preferences when welfare considerations are at stake are analysed by Roemer (1996). 10. This lack of empirical content as to what enters into preferences has been aptly remarked by Ulrich Witt (1991: 562). 11. See, however, Chapter 9 by Guerzoni and Troilo for a revealing analysis of how anachronistic, premodern, ‘Jurassic’ goods continue to encounter the consumer’s favour. 12. This has been treated earlier by Earl (1986). 13. For the role of assemblage see Chapter 6, Langlois and Cosgel.

BIBLIOGRAPHY Amsel, Abram (1989) Behaviorism, Neobehaviorism, and Cognitivism in Learning Theory. Historical and Contemporary Perspectives, Hillsdale, NJ: Lawrence Erlbaum Associates. Bentham, Jeremy (1970 [1789]) An Introduction to the Principles of Morals and Legislation, J.H. Burns and H.L.A. Hart (eds), London: Athlone Press. Berlyne, D.E. (1960) Conflict, Arousal, and Curiosity, New York: McGraw-Hill. —— (1971) Aesthetics and Psychobiology, New York: Appleton-Century-Crofts. Berlyne, D.E. and Masden, K.B. (eds) (1973) Pleasure, Reward, Preference, New York: Academic Press. Becker, Gary S. (1965) ‘A theory of the allocation of time’, Economic Journal 75: 493–517. Bianchi, Marina (1992) ‘Knowledge as expected surprise. A framework for introducing learning in economic choice’, Research in the History of Economic Thought and Methodology 10: 43–58. —— (1995) ‘Markets and firms. Transaction costs versus strategic interaction’, Journal of Economic Behavior and Organization 28: 183–202. Cassel, Gustav (1934) The Theory of Social Economy, New York: Brace and Company. Cyert, R.M. and March, James (1992) A Behavioral Theory of the Firm, 2nd edn, Oxford: Blackwell. Earl, Peter (1986) Lifestyle Economics. Consumer Behavior in a Turbulent World, New York: St. Martin’s Press. Georgescu-Roegen, Nicholas (1966) Analytical Economics. Issues and Problems, Cambridge, Mass: Harvard University Press. —— (1968) ‘Utility’, in David L. Sills (ed.) International Encyclopedia of the Social Sciences, New York: Macmillan and Free Press, vol. 16: 236–67. —— (1988) ‘Ophelimity’, in J. Eatwell, M. Milgate and P. Newman (eds) The New Palgrave. A Dictionary of Economics, London: Macmillan, vol. 3: 716–18. Home, Henry (Lord Kames) (1762) Elements of Criticism, London: Routledge, Thoemmas Press (1993) vol. I. Kirzner, Israel M. (1989) Discovery, Capitalism, and Distributive Justice, Oxford: Basil Blackwell. Lancaster, Kelvin (1991) Modern Consumer Theory, Aldershot: Edward Elgar.


INTRODUCTION Langlois, Richard N. (1990) ‘Creating external capabilities: innovation and vertical disintegration in the microcomputer industry’, Business and Economic History, second series, 19: 93–102. Langlois, Richard N. and Robertson, P.L. (1993) ‘Business organization as a coordination problem’, Business and Economic Theory 1: 1–11. —— (1995) Firms, Markets and Economic Change, London: Routledge. Loasby, Brian (1991) Equilibrium and Evolution. An Exploration of Connecting Principles in Economics, Manchester: Manchester University Press. March, James G. (1988) Decisions and Organizations, Oxford: Basil Blackwell. Marshall, Alfred (1961 [1890]), Principles of Economics, London: Macmillan. Mitchell, Wesley C. (1968 [1918)] ‘Bentham’s Felicific Calculus’, in Alfred N. Page (ed.) Utility Theory: A Book of Readings, New York: John Wiley: 30–48. Page, Alfred N. (ed.) (1968) Utility Theory: A Book of Readings, New York: John Wiley. Pareto, Vilfredo (1971 [1901]) Manual of Political Economy, New York: A.M. Kelley. Roemer, John E. (1996) Theories of Distributive Justice, Cambridge, MA: Harvard University Press. Ryle, G. (1949) The Concept of the Mind, London: Hutchinson. Scitovsky, Tibor (1992) The Joyless Economy. The Psychology of Human Satisfaction, rev. edn, Oxford: Oxford University Press. Stigler, George J. (1950) ‘The development of utility theory’, Journal of Political Economy 58, August: 307–27; October: 373–96. Stigler, George J. and Becker, Gary S. (1977) ‘De gustibus non est disputandum’, American Economic Review 67: 76–90. Thayer, Robert E. (1989) The Biopsychology of Mood and Arousal, Oxford: Oxford University Press. Viner, Jacob (1968 [1925]) ‘Utility and the Classical Theory of Demand’, in Alfred N. Page (ed.) Utility Theory: A Book of Readings, New York: John Wiley: 123–38. Williamson, Oliver E. (1993) ‘The evolving science of organization’, Journal of Institutional and Theoretical Economics 149 (1): 36–63. Witt, Ulrich (1991) ‘Economics, sociobiology, and behavioral psychology on preferences’, Journal of Economic Psychology 12: 557–73. —— (1994) ‘Endogenous change. Causes and contingencies’, Advances in Austrian Economics 1: 105–17.


Part I THE HEDONICS OF TASTE Utility, novelty and change

2 CHOICE WITHOUT UTILITY? Some reflections on the loose foundations of standard consumer theory Fabio Ranchetti

In common life, we may observe, that the circumstance of utility is always appealed to. (Hume 1777) Human sciences may be built up of heterogeneous materials; the regular tiers of mathematical construction imposed upon the loose foundations of common-sense. (Edgeworth 1887)

PROLOGUE In contemporary economic analysis, the consumer is dealt with using the tools and the language of utility theory. The problem which the consumer faces is viewed as a problem of choice among some ‘objects’ called commodities (or commodity bundles). The problem of choice is reduced to the constrained maximization of utility or, more precisely, of a utility function. The modern standard definition of the utility function is: ‘a function u: X → ℜ is > if, for all x, y ∈ X, x > y ⇔ = u(x) ≥ (u)y’ a utility function representing preference relation ~ ~ (Mas-Colell et al. 1995: 9). Then, a remark is usually added by the contemporary economist, stating that ‘the ability to represent preferences by a utility function is closely linked to the assumption of rationality’ (ibid.). And that is all: the system of choice of the individual consumer is considered as completely described by this procedure. Usually not a single word is employed to explain, justify, or motivate a procedure which, if to the contemporary economist’s mind seems rigorous and straightforward, to a more naive or perhaps more speculative mind appears quite odd, if not decidedly problematic. In fact, why should the problem of consumer choice be regarded as the problem of maximizing a function? And why



should this be a utility function? Granted he or she is a maximizer of a function of some sort, why could our consumer not maximize a prudence function or an elegance function? (Both prudence and elegance can certainly be considered attractive characteristics, even by modern standard economists.) In the hope of letting more light onto the issue, let us turn to two of the most famous authorities on consumption and choice. In their widely celebrated article on consumers’ tastes, George Stigler and Gary Becker (1977) state that their [then] ‘new’ theory of consumer choice:

transforms the family from a passive maximizer of the utility from market purchases into an active maximizer also engaged in extensive production and investment activities. In the traditional theory, households maximize a utility function of the goods and services bought in the marketplace, whereas in the reformulation they maximize a utility function of objects of choice, called commodities, that they produce with market goods, their own time, their skills, training and other human capital, and other inputs. Stated formally, a household seeks to maximize U = U (Z1, ... Zm) with Zi = fi (X1i , ... Xki , t1i , ... tli , S1, ... Sl , Yi), i = 1 ... m where Zi are the commodity objects of choice entering the utility function, fi is the production function for the ith commodity, Xj i is the quantity of the jth market good or service used in the production of the ith commodity, tj i is the jth person’s own time input, Sj the jth person’s human capital, and Yi represents all other inputs. (Stigler and Becker 1977: 26) And that is all. Again, as in textbook economics, no explanation or justification whatsoever is given for representing the consumer as a utility maximizer; nor do we get more light if we turn to Becker’s most recent work on the subject (1996). Maybe, according to Stigler and Becker (as well as to most contemporary economists), this is a question on which theorists ‘should not dispute’, as they should not dispute on consumers’ tastes. But it is certainly a sign of the state of contemporary economics, that, contrary to normal scientific rules, it is considered unnecessary to justify the basic assumptions and tools of the theory and art they practise.



In the following pages I shall try to provide, with reference to contemporary consumer theory, what the contemporary economist might have been expected to present, namely that critical examination of the basic principles and tools which makes a theory meaningful. In particular, I shall examine the rationale, the meaning, the implications and legitimacy (or not) of the use of the utility function in consumer theory or, in modern jargon, ‘rational choice theory’. My argument turns on my showing how utility is inextricably connected with strong ethical, if not straightforwardly metaphysical, assumptions, that are usually passed over in silence by contemporary economists, but that would be better explicitly stated. Finally, as a result of the analysis of some puzzling questions raised by the introduction of utility in consumer theory (rational choice theory), I conclude by suggesting that the contemporary economist might better try to explore alternative hypotheses on which to build a new and more appropriate theory of the consumer. My treatment falls into two parts. In the first part, which is mainly historical, I shall consider first the introduction of the utility function into economic analysis in the last third of the nineteenth century; second, the efforts to develop a new theory of choice and utility freed from the non-scientific and metaphysical presuppositions that encumbered the economic theory of the old economists. In the second part, which is mainly critical, I shall illustrate some of the main difficulties posed by the notion of utility.1 We start with the questions when, why, and how were utility and the utility function introduced into economic theory.

THE PATENT MARRIAGE AND APPARENT DIVORCE OF CHOICE AND UTILITY Act One: Enter utility/the marriage To understand the reasons for the introduction of the notion of utility into modern economic analysis, it is instructive to consider Edgeworth’s thought. I have chosen Edgeworth for two reasons. First, because, as has been remarked, his general conception of man ‘survives more or less intact in much of modern economic theory’ (Sen 1976–77: 90). Second, because in his thought the economic notions and analytical tools of modern (and contemporary) consumer theory reveal most clearly their necessary link with, and dependence on, non-economic notions and tools. Among contemporary economic theorists, Edgeworth is best known for his contract and indifference curves and for his exceptionally ingenious analysis of the ‘shrinking of the core’



of an economy.2 Much less known is the fact that, like his contemporaries Marshall and Wicksteed (and before them, Adam Smith), Edgeworth arrived at economics from ethics. His first book, New and Old Methods of Ethics: or ‘Physical Ethics’ and Methods of Ethics,3 published in 1877, was in fact a critique and a development of utilitarianism, the doctrine which identifies what is ‘good’ with utility (or happiness, or pleasure), the latter being defined as the satisfaction of rational desire. In Edgeworth’s thought, utilitarianism is essentially connected to the problem of distributive justice, that is, to the problem of distributing a given quantity of means to a determinate set of individuals in such a way as to ensure ‘the greatest quantity of happiness to the greatest number’ (which is, as is well known, the formula Bentham made famous). Edgeworth’s novelty and originality lie in his having applied mathematical analysis, the Calculus par excellence, to the solution of this problem. It is precisely within this ethical context that Edgeworth first introduced a utility function and structured the problem as that of the constrained maximization of the function. Edgeworth’s original utility function was the following:

k{ f (y) – f (β )} where k represents the ‘capacity for pleasure’ of a sentient, y the means of pleasure and β the ‘threshold of pleasure’, i.e. ‘the lowest value of the stimulus for which the sentient has any pleasure at all’ (Edgeworth 1877: 42).4 What is remarkable for our argument, is that in New and Old Methods of Ethics one can find no economic theory whatsoever. Edgeworth’s conception of utility is there absolutely independent of any economic theory, and especially of marginalist theory, which, at any rate (and at that early stage of his thought), he probably ignored. Contrary to modern economics, where the utility function is immediately applied to the economic problem of an individual having to choose between different bundles of commodities, Edgeworth introduced the utility function with the sole aim of defining exactly utilitarianism, which is a moral principle, if not a law, pertaining to distributive justice. Furthermore, it is to be noticed that Edgeworth derives the utility function from a theoretical tradition, experimental psychology, that was quite distant from the economic theory then prevailing. (In fact, Edgeworth was influenced by the experimental work done on the continent by the psychophysics school of E.H. Weber, G. Fechner, Max Wundt, J. Delboeuf, H. von Helmholtz, and others).5 But in 1879, on Hampstead Hill, Edgeworth met Jevons; and his brilliant mind turned from ethics to economics. Within two years, under the influence of Jevons, but reading Marshall and Walras as well, Edgeworth produced his chef d’œuvre, that most extraordinary book – Mathematical Psychics. An Essay on the Application of Mathematics to the Moral Sciences. In this work, the tool of the utility function, first introduced in Edgeworth’s previous book on



ethics, was applied to the economic problem of discovering the best possible contract between individuals mutually exchanging commodities in the marketplace. In relation to this problem, Edgeworth invented the indifference curve and the contract curve, which he defined in terms of a generalized utility function. In fact, contrary to the approach of Jevons and Marshall, and for the first time in economic analysis, Edgeworth regarded utility not as the sum of two (or more) independent functions, one for each different commodity, but as a single function of two (or more) variables (see Edgeworth 1881: 104). Following Edgeworth’s own example and notation, given two individuals, Robinson Crusoe and Friday, ‘whose interest depends on two variable quantities’ (1881: 20), their total utility functions are defined by P = F(x, y), for Crusoe, and ∏ = f (x, y) for Friday (x and y representing physical quantities of the two commodities exchanged in the market). On the basis of those utility functions, Edgeworth then defines the contract curve, the locus of all final settlements, by means of the following equation: ∂∏ /∂x ∂P ∂∏ ∂P ∂∏ ∂P/∂x = (1), or = (1b). ∂x ∂y ∂y ∂x ∂∏ /∂y ∂P/∂y On the basis of the same utility functions, Edgeworth defines the indifference curve of Crusoe, as the set of all contracts with respect to which his total utility is constant, by means of the following equation: ∂P dx + ∂P dy = 0. ∂x ∂y Likewise, on the basis of Friday’s utility function, ∏ = f(x, y), the equation which represents his indifference curves is: ∂∏ ∂∏ dx + dy = 0. ∂x ∂y On the basis of these notions, Edgeworth is able to define exactly the problem of the contract, i.e. the problem of exchange in the market, as a problem of constrained maximization.6 I am not going to dwell here on Edgeworth’s economic analysis, nor will I go into the details of his most celebrated analytical tools, the contract curve and the indifference curve, or discuss their more recent development, especially the notion of the core and the theorem of the limit. These are in fact the best known aspects of Edgeworth’s thought. Instead, I will elaborate on the double-sided nature of the fundamental notion of utility, and therefore on the double nature of the utility function in Edgeworth’s discourse. On the basis of the same notion, utility, and the same tool, the utility function, Edgeworth in fact founds two quite



different sciences: ethics and economics. These sciences are not independent of each other. To see this, it is useful to recall Edgeworth’s important distinction between two species of the Calculus of Pleasure and Pain, or ‘Hedonical Calculus’: the ‘Economical’ and the ‘Utilitarian’, the former corresponding to Economics and the latter to Ethics. Moving from the general conception of man as ‘a pleasure machine’; Edgeworth states:

The Economical Calculus investigates the equilibrium of a system of hedonic forces each tending to maximum individual utility; the Utilitarian Calculus, the equilibrium of a system in which each and all tend to maximum universal utility. (Edgeworth 1881: 15–16)

On Edgeworth’s distinction, two points need to be considered. The first is the formal unity of moral sciences. According to Edgeworth, since the object of both sciences (Economics and Ethics) is a problem of finding a maximum, both are founded upon a unique principle and must follow a unique method, the mathematical method utilized in the calculus de maximis et de minimis, i.e. the differential and the integral calculus. (It is to be noted, however, that this calculus need not necessarily be a numerical calculus.)7 The second point is the difference between the two sciences, and the two corresponding calculi. Whereas the object of economics is the individual agent ‘actuated only by selfinterest’, with no consideration for the happiness or utility of other individuals,8 the object of ethics is society, in which each individual is actuated by sympathy and benevolence towards others, and therefore takes into account the happiness or the utility of other individuals. Consistently with this twofold point of view, in the economical calculus it is the individual who makes the calculation of his or her own pleasure or utility, whereas in the utilitarian calculus it is instead necessary ‘to compare the happiness of one person with the happiness of another, and generally the happiness of groups of different members and different average happiness’ (Edgeworth 1881: 7). Therefore, notwithstanding the unity of social sciences and their common method, one must not confuse Economics with Ethics, nor forget their hierarchical order: the individual good (the object of the Economical Calculus) should be subordinated to the universal good (the object of the Utilitarian Calculus). To Edgeworth, utilitarianism means ‘vivre pour autrui’, and it strongly opposes ‘the condition of that perfect disintegration and unsympathetic isolation abstractedly assumed in Economics’ (ibid.: 12). In Mathematical Psychics, towards the end of the part devoted to the economical calculus, after having expounded his celebrated analysis of ‘the shrinking of the core’, as we say in contemporary jargon, Edgeworth turns to a fierce critique of the ‘reverence’ paid by economists to the results of the market. According to Edgeworth, the results of the market, so well



analysed by the economical calculus, are in fact profoundly deficient, for two orders of reasons. First, if there is no perfect competition, the contract remains indeterminate and the need arises of a ‘superior luminary’ that allows us to find a final settlement that satisfies ‘the sympathy (such as it is) of each with all, the sense of justice and utilitarian equity’ (ibid.: 54). In other words, the economical calculus needs to be supplemented by the utilitarian calculus: ‘competition requires to be supplemented by arbitration, and the basis of arbitration between selfinterested contractors is the greatest possible sum-total utility’ (ibid.: 56; italics in original). Second, and most important for our argument, even if competition were perfect, and hence the contract perfectly determinate, one must equally recur to the utilitarian calculus. For economic equilibrium does not necessarily coincide with the social equilibrium that is ethically desirable. In other words, according to Edgeworth’s analysis, the market solution, even the perfect competitive market solution, does not coincide necessarily with ‘the greatest sum-total of universal happiness or utility’ (Edgeworth, 1889: 508). These are the reasons why the object of the second and last part of Mathematical Psychics is again Ethics. In this part of his economic masterpiece Edgeworth presents what he had already exposed in New and Old Methods of Ethics and again in an article in Mind (1879). Let us sum up what we have been seeing, with a final reflection on Edgeworth’s intellectual double voyage, from ethics to economics, and then back to ethics, and its meaning. Edgeworth’s reflections on distribution and exchange originate from a moral perspective, strongly biased towards utilitarianism. Edgeworth hence developed a very original economic analysis of the market, which concludes with his pointing out the limits, economic and social, of this arrangement of human affairs. These limits provide the necessity and a new motivation to go back to an ethical analysis, in order to find a solid foundation for the distributive justice that is required for the functioning of a civil society. What is particularly interesting for a contemporary economist is the meaning that the notion of utility assumes in this voyage. Edgeworth’s economic analysis, as developed in Mathematical Psychics, is indeed based on the tool of the utility function. However, contrary to most contemporary standard economic theory, it is always clear that the position of the notion of utility, and hence of the utility function, at centre stage of the economic analysis, is not something derived from empirical economic analysis, but is essentially a postulate imposed upon it by the very special ethical theory in which Edgeworth believed. (I return to this in the second part of this chapter.) Furthermore, for Edgeworth, the use of the utility function in economic analysis necessarily has to be supplemented by a superior calculus, the utilitarian calculus. For our argument, the main conclusion is that, in this ‘old’ conceptual framework, the eminence given to utility makes sense, and it is perfectly clear why the theorist has chosen the concept of utility and the utility function to represent consumer choice. But let us now see how and why economic thought tried to reject and abandon the notion of utility, and what



the implications are of this divorce. The best place to go in order to understand this movement away from utility in consumer theory is Pareto’s thought.

Act Two: Exit utility/the divorce In a very important letter to his friend, Maffeo Pantaleoni, who was to become ‘the prince of Italian economists’, on the 28 December of the last year of the last century, Pareto ‘set forth a novel idea that was to transform radically not only economics but also the other disciplines of man’ (Georgescu-Roegen 1987: 717). Pareto writes:

Edgeworth and the others start from the notion of the final degree of utility9 and arrive at the determination of the indifference curves. Now, I leave totally aside the final degree of utility and I start from the indifference curve. In this lies the novelty. (Pareto 1962: 288; original emphasis)

After a few lines, Pareto adds:

Until now economics has been founded upon the final degree of utility, the rareté, the ophelimity etc. Well, that is not necessary. One can start from the indifference curves that are a direct result of experience. (Pareto 1962: 288; original emphasis)

But why does Pareto want to get rid of utility, rareté, ophelimity, etc.? The reason, clearly indicated by Pareto himself, is that he aims at freeing economic theory from metaphysical notions, entities that are not directly and empirically observable, such as utility, rareté, and ophelimity. According to Pareto, economic theory should only register ‘the pure and naked fact of choice’, as the theorist does in the construction of the indifference curve when he starts from the observation of empirical data. To clarify and emphasize his point, Pareto presents a wonderful example and illustration (see Figure 2.1):

I put in two pans cherries and dates; in the first pan I put 10 cherries and 10 dates, in the second 15 cherries and 9 dates. Then I put in the middle of them Buridan’s ass, and I am going to see what happens. (Pareto 1962: 288)



Figure 2.1 Pareto’s drawing of Buridan’s ass.

If the ass chooses the left pan, then that means that he prefers it (and vice versa, should he choose the right pan). If instead he cannot decide on either the left pan or the right one, ‘remaining undecided between the two’, then those two combinations of goods belong to the same indifference curve, of which the observer has thus identified one point. The major implication of Pareto’s new idea and method is therefore the divorce of the economic theory of the consumer from the notion of utility. There are several aspects of Pareto’s analysis that need to be stressed. The first and most important is that the change in the foundation of the theory realized by Pareto is the consequence of his very different philosophical approach. Pareto was not a utilitarist like Edgeworth; he was an empiricist, who could not grant any room to such a metaphysical idea as utility. Furthermore, Pareto adhered to the scientific method summed up in Occam’s razor, the principle that ‘entia non sunt multiplicanda praeter necessitatem’ [entities are not to be multiplied beyond necessity]. Therefore, since one can build indifference curves without knowing consumers’ utilities, the hypothesis and notion of utility are not only meaningless (because metaphysical), but, with respect to the economic theory of consumer choice and demand, they are utterly superfluous. In other words, having severed the essential link with ethics, which characterized its use by Edgeworth, the notion of utility seems to have no role at all to play in economics (at least in that part which is consumer theory). The second aspect is the abandonment not only of the notion of utility, for the reasons just mentioned, but also of the idea that the objects of choice are measurable in terms of some ‘entity’. Third, and connected to this point, is the absence, in Pareto’s economic analysis, of any psychological enquiry into the ‘rational’ mind of the consumer. Pareto (and the modern economist) claims not to be at all interested in the motives of human choice. In fact, according to Pareto, we can define an indifference curve not only with respect to an animal, Buridan’s ass, as in this instance, but also with respect to a machine: ‘for a machine, equal weights give indifference curves’ (Pareto 1899: 290). Pareto’s conception and research programme were then developed and perfected in a number of very well known contributions, among which stand prominent those by Pareto



himself (Pareto 1906), and then by Hicks (Allen and Hicks 1934; Hicks 1939) and Samuelson (1938, 1947). In particular, Samuelson, with his notion of ‘revealed preference’, tried to free economic analysis of ‘any vestigial traces of the utility concept’ (Samuelson 1938: 71), and provide instead ‘operationally meaningful foundations’ (Samuelson 1948: 251) for consumer theory. The story we have presented has usually been told as the progress of economic science from a cardinal to an ordinal notion of utility, the former being metaphysical, the latter purely scientific.10 Utility is cardinal when it is the ‘entity’ from which the economist infers the consumer’s choices, whereas utility is ordinal when it is inferred from the actual, observed choices made by the consumer.11 A distinction can correspondingly be drawn between utility in terms of pleasure (or happiness, well-being, satisfaction of desire, etc.), as it was conceived by the older economists, and utility in terms of preference (or choice), as it is conceived by more contemporary economists. According to the modern and ‘scientific’ definition, ‘utility’ is therefore only the ‘name’ given by contemporary economists to the numbers they select to represent the consumer’s preference ordering. Thus, in contemporary economics: The official meaning [of ‘utility’] is that which represents preferences. This is the meaning given in the major doctrinal texts (e.g., Debreu, Theory of Value, 1959) and in the best textbooks (e.g., Deaton and Muellbauer, Economics and Consumer Behavior, 1980). It is defined with great precision, as a technical term should be. Its use is universal in theoretical economics. There is no alternative term. Economics cannot do without it. (Broome 1991: 10; original emphasis) But is this a true and accurate statement of the position of utility theory in contemporary economics? Is it really true that utility is only the name given to a mathematical function? Let us see, by examining precisely the ‘major doctrinal texts’ of contemporary economics.

THE LATENT LIAISON BETWEEN CHOICE AND UTILITY The ambiguous status of utility in contemporary economics To see things clearly, let us take the most rigorous and celebrated ‘doctrinal text’ of modern economics: The Theory of Value by Gerard Debreu (1959). Here, after having defined the



consumer as an individual whose role in economic theory is to choose a ‘consumption plan to which none is preferred’ (Debreu 1959: 50), and having introduced the notions of preference and indifference, Debreu asks himself (and the reader):

Is it possible to associate with each [indifference] class a real number in such a way that, if a class is preferred to another, the number of the first is greater than the number of the second? In other words, given a set completely preordered by preferences, does there exist an increasing real-valued function on that set? (Debreu 1959: 55–6)

First, it is of interest to note that the answer is negative, unless very restrictive conditions on preferences are assumed. As is well known, these conditions amount to the axioms of consumer preferences: the axioms of completeness (or connectedness), transitivity and continuity, which have been extensively discussed in the literature and which I will not question here.12 If they are assumed, as we concede for the sake of argument, then the sought after ‘thing’ exists and, by a fortuitous coincidence, is named ‘utility function’. The second, more interesting, point to note is the set of reasons given by Debreu for the reintroduction of such an ancient relic (at least in the name). The first reason is an apparently purely technical one: ‘a utility function is a valuable tool in the proof of some results’ (Debreu 1959: 56). If it is really only a technical tool, a fastidious reader might observe, it is very misleading to give it the controversial, if not frankly discredited (as we have seen in the previous pages), name of utility. Words do count. Why then call it utility function? The suspicion that it is not just a purely mathematical device, is in fact immediately reinforced and confirmed by the second reason put forward by Debreu for the introduction of this analytical tool: a utility function, Debreu rather surprisingly remarks, with apparent innocence, ‘also gives a precise content to the intuitive notion of a numerical measure of how satisfied the ith consumer is with xi [a consumption bundle]’ (Debreu 1959: 56; emphasis added). Thus, for ‘technical’ and ‘intuitive’ reasons, back we are with our old friends utility and satisfaction, and to the old idea of measuring individuals’ utilities and satisfactions; even contemporary economic theory it seems cannot do without them. But, are we then really back to square one, or Act One? After a divorce which seemed definitive, is this a second marriage between the same old ‘actors’? If, in order to prove some theorems, a numerical representation of consumer preferences (i.e. a utility function) is convenient, then one can easily see why a mathematical economist might like to introduce it, although the mathematical economist should also be aware of the baggage carried by the technical tools Debreu is using, since no tool is ever neutral. However,



given Debreu’s methodological approach13 and his famous remark that ‘the theory is logically entirely disconnected from its interpretations’ (1959: x),14 one finds it difficult to understand how he could possibly call ‘intuitive’ (whatever meaning one assigns to the rather vague term of ‘intuition’) the notion of a numerical measure of satisfaction. On this point Debreu is indeed wholly silent. On the other hand, there is no doubt that he does here interpret utility in terms of a psychological entity: what else could the ‘satisfaction’ be that an individual consumer obtains from a given bundle of commodities? Consistently, Debreu interprets preference also in a strongly psychological sense: the ‘preferred’ bundle is the ‘desired’ one (see, for example, Debreu 1959: 54). To try to get clarity on this important issue we turn to Debreu’s fellow traveller in the realm of economic theory, Kenneth Arrow, who is usually less esoteric and more outspoken. Arrow starts by describing the process of choice in a very Paretian fashion:

The utility theory of choice states that the choice in any given situation depends on the interaction of the externally given obstacles [Pareto’s ‘ostacoli’] with the tastes [Pareto’s ‘gusti’] of the individual, and that the obstacles and tastes can be thought of as independent variables. The utility theory asserts, more precisely, that the tastes can be represented by an ordering according to preference of all conceivable alternatives. In the usual theory of consumption it is imagined that the individual [like the ass in Pareto’s story] could be asked in advance for his preference between any pair of given bundles. (Arrow 1963: 117–18)

Then, and again in a very Paretian way, Arrow states that the utility function:

is simply a convenient mathematical way of describing an ordinal scale; it does not have any further significance. . . . If we represent preference scales by utility functions, we can say that the demand functions are defined by maximizing the utility function, subject to the constraints implied by the restriction that the total expenditures not exceed total income. (Arrow 1963: 118; emphasis added)

Arrow, again following Pareto’s steps and carrying on Hicks’s and Samuelson’s scientific programme, is very careful to stress the behavioural, as opposed to the metaphysical, content of consumer choice theory:



The only evidence of an individual’s utility function is supplied by his observable behavior, specifically the choices he makes in the course of maximizing the function. But such choices are defined by the preference ordering and must therefore be the same for all utility functions compatible with that ordering. Hence there is no quantitative meaning of utility for an individual. (Arrow 1973: 104; emphasis added)15

Arrow stresses this point many times. For example, in a very clear passage:

Modern economic theory has insisted on the ordinal concept of utility; that is, only orderings can be observed, and therefore no measurement of utility independent of these orderings has any significance. In the field of consumer’s demand theory the ordinalist position turned out to create no problems; cardinal utility had no explanatory power above and beyond ordinal. Leibniz’ Principle of the Identity of the Indiscernibles demanded then the excision of cardinal utility from our thought patterns. (Arrow 1967: 75–6; emphasis added)

On the other hand, and here is the really interesting and quite surprising implication of Arrow’s argument, we are told that ‘this theory of the formation of demand is not a very strong theory, but it is not a tautology either’ (Arrow 1967: 118–19).16 And why is it not tautological? Because, Arrow adds, ‘in many respects it is very close to the hedonistic position in psychology which currently seems to be undergoing a revival’ (ibid.). He concludes, ‘psychological theory is usually much more explicit than pure economic theory about the nature of pleasures and pains’ (ibid.). Hear, hear. Yes, Edgeworth’s Pleasures and Pains. Again, as in Debreu’s case, we are driven back to exactly those entities that, mainly as a consequence of Pareto’s (and Hicks’s, and Samuelson’s, etc.) ‘ordinal revolution’, apparently have long ceased to have any significance and role in economic theory. It is not true then that in contemporary economic theory ‘utility’ is a purely technical tool, whose position is only that of representing individuals’ preferences. Even the ‘sophisticated’ contemporary theory of choice, exactly as the theory put forward by the first ‘naive’ marginalists, cannot escape from a fundamental reference to the old, once rejected, concept of utility. Preference and choice without utility (utility, it has to be stressed, in the old and ‘heavy’ meaning of a calculus of pleasures and pains) seem not to make sense. Correspondingly, the distinction between cardinal and ordinal utility fades away: how can we truly distinguish between the two, if we conceive of utility as a measure of consumer’s



satisfaction? In the end, the divorce between choice and utility shows itself to have been more apparent than real. How can we explain this seeming paradox? Or, how can we explain the persistence of the idea of utility in contemporary economic theory, notwithstanding the strong and clearly expressed opposition to it, and its consequent rejection? To understand this very significant fact we have to go deeper into the complicated relationship between choice, preference, and utility.

Choice as calculus 17 The whole utility theory of choice views choice as a rational calculus, namely the maximization of utility. This holds for both the old cardinalist approach and the modern ordinalist approach, for the old idea that utility consists in individual pleasure, and the modern idea that it only represents individual preferences. A rational calculus presupposes that all different objects of choice be reduced to a common entity, and then be measured by a common standard. Otherwise, the comparison and therefore the choice itself is unfeasible: as Aristotle inferred, ‘there must therefore be one standard by which all commodities are measured’ (Nicomachean Ethics, 1133a).18 Obviously, the selection of the common entity and standard to be used here becomes crucial. For example, we could take weight as the common entity in which different things (e.g. commodities) are compared and evaluated, so that the (rational) individual can choose among them the one ‘preferred’ according to its superior (or inferior) weight; in this case, the individual has only to weigh the different objects, and choose the heaviest (or the lightest). A delightful example of this procedure, in which choosing is equated to weighing, is given by Aristophanes. Towards the end of Aristophanes’ Frogs, in order to be able to choose between the poetic art of Euripides and that of Aeschilus, Dionysus orders a balance to be brought, ‘to prove whose poetry is the weightier’, Euripides’ or Aeschilus’. (It is remarkable that the preferred ‘object’ was thought to be the heaviest. There is in fact no logical reason for preferring this criterion to its opposite, namely choosing the lightest.19 However, this is in line with ordinary language. Usually, we find that people say that one compares two arguments, and then chooses the weightier.) The Chorus comments:

Did ever you hear Such a brilliant idea, So simple and yet so bizarre? Aristophanes’ Frogs, 405



tr. Barrett


(We remind the interested reader that, on the basis of this test, Aeschilus was preferred to Euripides, since his verses, talking about Death and corpses, comprised ‘really ponderous words’, weighed in heavier than those of Euripides, whose talk was about Persuasion and speech, ‘empty words without sense’.) The ‘simple and bizarre’ idea of reducing everything that is the object of choice to an objective quality, here, the weight, in order to make choosing possible, was made less obviously absurd than shown to be in Aristophanes’ play. A respectable method of choice was indeed fashioned by linking the idea of choice with the idea of pleasure, and then by interpreting utility in terms of pleasure. This is exactly what we have seen Edgeworth doing. In fact hedonism, the reduction of the objects of choice to their common feature of being desired objects, which makes possible the calculus (of pleasures) and therefore the choice, has a very long and illustrious tradition, which, as so often, goes back to the Ancient Greeks: ‘our salvation in life lies in the correct choice of pleasure and pain’ and ‘it is in the first place a question of measurement’, we read in Plato’s Protagoras (357a). It is interesting to note that this quotation is the epigraph Edgeworth chose for his less known but very important essay on the technique of measuring pleasures and pains, Metretike (‘metretike’ was in fact the word used by Plato for ‘measurement’). In Metretike, Edgeworth develops and extends his ideas on measuring ‘feeling, doubt and belief’, to ‘the fairest province of Probabilities’, which also falls under ‘the suzerainty of the Principle of Utility’ (Edgeworth 1887: 52). Two aspects of this terse but precious booklet are important for our argument.20 The first regards the nature of utility. According to Edgeworth, utility is a first principle, an arche, derived not from ‘Reason’, but from ‘an instinct and a sentiment’. Therefore, ‘why utility is the standard of good conduct, cannot be proved by syllogism’ (ibid.: 50). The second aspect is Edgeworth’s caution in not extending beyond its limits the doctrine of utility. Edgeworth wants to reassure the old ‘respectable aunts’ who have been alarmed ‘with reckless applications of the Greatest Happiness Principle’ (ibid.: 51) by their young nephews. Notwithstanding his pride in the achievement of the modern philosophy, Edgeworth’s position is in fact quite reticent, even inclining towards an ordinalist point of view:

The fond pretension to arithmetical calculation must be abandoned; but it has been made possible to estimate quantity of belief and quantity of happiness in terms which roughly indicate the more or less. To follow those indications, in preference to custom and tradition, may well be ‘more than half’ the sum of wisdom. It is the crowning height of English Philosophy, and what the ancients, with all their Virtue and all their Wisdom, never attained. (Edgeworth 1887: 51–2)



I do not here intend to examine hedonism. My point was to show that the distinction between a metaphysical and a scientific way of considering utility, between a cardinal and an ordinal utility function, does not stand close inspection at all. Once utility is no longer viewed as the common ‘substance’ of all objects of choice, and it is ‘scientifically’ reduced to a simple indexing, there is no longer any way to give meaning to the maximization of utility: if not pleasure, then what do agents maximize? It is therefore clear why the best theoretical economists (we have quoted Arrow and Debreu), in order to give an economic meaning to their mathematical procedure, have to fall back to the old ‘intuition’ of a calculus of pleasures and pains. In other words, and to conclude this section, the utility function is not a purely technical tool of our economic art, but it is a very impure tool, whose use necessarily commits the economist to a very particular conception of man and human affairs. My impression is that the standard contemporary economist is innocent of these implications. But, if more conversant with them, would he or she be so happily content with the utility function? From the philosophical point of view, ‘the failure of hedonism to provide a rational procedure of choice’ has been forcefully argued in recent times by Rawls (1972). Another very powerful critique has been that by Wittgenstein. In particular, Wittgenstein has shown the ‘innumerable confusions’ that arise from the fact that ‘in science it is usual to make phenomena that allow of exact measurement into defining criteria for an expression; and then one is inclined to think that now the proper meaning has been found’ (Wittgenstein 1967: 76). Wittgenstein’s analysis points out the reductionism and, at the same time, the emptiness that lie behind this view of choice:

The hedonist says: ‘Men desire nothing but pleasure’. Obviously this is no empirical proposition. The hedonist does not find this out by going about asking people what they want. He has no statistics about this. And yet he knows very well that people want all sorts of things. So it isn’t at all like: Everybody wants a motorcar. If someone wants a motorcar, then he wants pleasure, and if he wants to smoke or to write a letter, then he wants pleasure. Pleasure is another word for whatever anyone wants. In other words it is a tautology. Everyone prefers the preferable. So pleasure is the desirable, the preferable. (Wittgenstein 1986: 58, emphasis added; see also 63–4)

As stated, I will not examine hedonism further. Its difficulties as a foundation for a rational method of choice have been clearly exposed, not just from the philosopher’s point of view, but also from the economist’s.21



Conclusions: Buridan’s ass reconsidered In the first part of this chapter, with reference to the case of Buridan’s ass, we saw the theoretical movement to a concept of utility reduced purely to terms of preferences. The agent, here the ass, chooses what it, he or she prefers. If it, he or she does not choose, that indicates indifference between the objects of choice. Preferences are therefore viewed as mental states, of which the observer knows nothing and therefore can say nothing; whereas choices are the agent’s actions observed by the spectator economist. Observed choices are all the economist needs in order to draw indifference curves. Indifference curves, together with the consumer’s budget constraint, are sufficient fully to conceptualize the problem which the consumer faces in the market. In reality, things are much more complicated than that. In fact, in this simple story, which every textbook tells to the innocent student, at least three orders of questions are concealed. The first question is the relationship between preferences on the one side and choices on the other. The modern standard theory seems to try to keep a perfect correspondence between the two, in the sense that preferences are viewed as the determinants of consumer behaviour. But, once modern economists have severed any link with a psychological enquiry into the mind of the consumer, as we have seen with reference to Pareto, preferences should totally disappear from the economic scene, since they are identical with the observed choices. Therefore, one cannot say any more that there exists a causal link between choices and preferences as distinct from choices, or even that choices reflect preferences. In other words, within the framework of modern economics, it becomes impossible to say whether the goods chosen are chosen because they are preferred, or preferred because they are chosen. But, on the other hand, it is generally perfectly possible to conceive of preference without actual choice, as well as of choice without preference. What exactly is choice without preference and how could it be possible? To understand this, which is the second question that arises, it is convenient to go back to our friend, Buridan’s ass. There are two possible interpretations of the dilemma which the ass faces. The first, which seems to be that proposed by Pareto, is that the ass does not choose because it is truly indifferent between the two commodity bundles (that is, it equally prefers them). This is therefore a problem of choice without preference in the sense that the agent, the ass, knows full well that the two bundles are equally preferred: this is the reason why he or it cannot decide which one to take (or ‘to choose’). In other words, decision (choice) is not just a matter of knowledge, but also of some other element or ‘force’. The problem of choice without preference has a long intellectual history, and can be quite easily resolved.22 A solution would be to point out that, in practice, two bundles of commodities are never



identical, and therefore there is always an order of preference between the two (or, conversely, there is never indifference). For example, Montaigne, in his Essais, so defines the problem:

It is a pleasant imagination, to conceive a spirit justly ballanced betweene two equal desires. For, it is not to be doubted, that he shall never be resolved upon any match: Foresomuch as the application and choise brings an inequality of prise: And who should place us betweene a Bottle of Wine and a Gammon of Bacon, with a equal appetite to eat and drinke, doubtlesse there were noe remedy but to die of thirst and hunger. (Montaigne, Essais, II: 14; Florio’s translation)

Then he immediately offers a solution:

In my opinion, it might be said that nothing is presented unto us, wherein there is not some difference, how light so ever it bee: and that either to the sight, or to the feeling, there is ever some choise, which tempteth and drawes us to it, though imperceptible and not to be distinguished. In like manner hee that shall presuppose a twine-thread equally strong all-through, it is impossible by all impossibilities that it breake, for, where could you have the flaw or breaking to beginne? And at once to breake in all places together, it is not in nature. (Montaigne. Essais, II: 14; Florio’s translation)

A more theoretical solution to the same problem would be a random selection of either bundle: ‘random selection is the only reasonable procedure for making choices in the face of symmetric preference’ (Rescher 1959–60: 170). However, one should note that, in this case (a) the reasons for the agent’s choice are independent of any intrinsic qualities of the object of choice; (b) ‘reasonable choice comes to be possible in the absence of preference only by essentially abdicating the right of choice’ (ibid.: 173), since it is delegated to a random process. The second interpretation of the ass’s dilemma, which is the one favoured by contemporary economics, is that the animal does not decide, because he does not know how to order the two bundles, that is, its preference ordering fails to satisfy the axiom of completeness. In other words, the agent does not choose because its, her or his knowledge is not sufficient: this is then a case of choice without (sufficient) knowledge, not a case of choice without preference as previously supposed. If interpreted in this way, Buridan’s ass points to one of the difficulties in the most radical form of the refusal of the older utility version of the consumer theory –



namely the theory of so-called revealed preference. As a sharp critic of this theory remarked: ‘Buridan’s ass died for the cause of revealed preference, though – alas – he was not entirely successful since non-choice leading to starvation would have looked like the chosen alternative, from the point of view of the fundamental assumption of revealed preference’ (Sen 1973: 248–9). In fact, choosing either bundle would have given a ‘wrong signal’ to the observer, since it would have appeared to reveal a view that one bundle was no worse than the other, ‘but this view the ass was unable to subscribe to since it could not decide what its preference should be’ (ibid.). In any case, considering either the first or the second interpretation of the story of Buridan’s ass, there arise some quite strong doubts about the existence of true indifference, and therefore about the existence of proper indifference curves. Pareto’s procedure to find them does not seem to be well founded. It is much safer to state that indifference curves exist only with respect to inanimate objects, as in fact Pareto implied with reference to a machine. But as economists we are not really interested in them. The third and last question I wish to consider concerns the agent’s rationality. Again, a good point of departure to visualize the question is yet another version of the story of Buridan’s ass:

Between two viands, equally removed And tempting, a free man would die of hunger Ere either he could bring unto his teeth. So would a lamb between the ravenings Of two fierce wolves stand fearing both alike; And so would stand a dog between two does. Dante, Paradise, IV, tr. Longfellow

The point illustrated by these verses is that free will and knowledge, even complete knowledge, are not sufficient conditions to make choosing possible. Choosing requires also a criterion of choice, and this cannot coincide with the (complete) preference ordering of the agent. First, this is so for the reasons already given, with reference to the correspondence between choice and preference. Second, it is so because, even granted the separate existence of preferences (which, as we have seen, is difficult to admit within the framework of contemporary economic theory), pure preferences as such do not at all guarantee a rational choice. In fact, even preferences must be evaluated and evaluation requires values. In other words, rationality cannot be reduced, as too hastily thought by some modern economists, to an efficiency relationship between scarce means (the consumer’s budget constraint) and whatever is the agent’s preference ordering as represented by the utility function. Contrary to the simplistic conception of the contemporary economist, the assumption of rationality cannot be made



without a more complex and explicit evaluation of the agent’s ends, or preferences. Unlike Edgeworth (Walras or Wicksteed), the contemporary economist may be quite reluctant and rather badly equipped for such a task; but he or she cannot escape it. Once the necessary and unavoidable link with strongly normative notions has been exposed, only one route remains open to the economist who still wants to stick to the apparatus of the utility theory (in whatever its version, old or modern). That is to adopt explicitly utilitarianism, and rest one’s hopes (as Arrow seems to do)23 on a new sort of Edgeworthian hedonimetry, which could only give a precise content to the utility function, but which has still to be devised, and may never be. In other words, contrary to Schumpeter’s opinion, it is not possible to break the ‘unholy alliance’ between the utility function and the Benthamite philosophy. It is not possible to ‘leave out’ the utilitarianism from pure economic theory ‘without affecting’ its ‘scientific content’ (Schumpeter 1954: 831). In fact, this is the position put forward by Harsanyi (1975, 1976, 1986) in the most forceful and well argued contemporary defence of utility theory: the scientific use of the utility function (and a cardinal one, not just an ordinal one) is integral to the acceptance of utilitarianism. On the other hand, if one does not accept utilitarianism as a method of choice (and, as we showed, there are good reasons for not accepting it), given the logical and analytical problems attached to the axioms of the preference ordering (and therefore to the use of the utility function in economics), an alternative route may look more attractive. The process of human (and asinine) choice might in fact be analysed along different lines, some of which may be traced in several chapters in this volume.

ACKNOWLEDGEMENTS I wish to thank Marina Bianchi for very helpful discussions. Her essay on utility (Bianchi 1985) has been very influential in shaping my own ideas on this complex subject. I am most grateful to Neil De Marchi, whose comments and suggestions on an earlier draft greatly improved my argument. The usual disclaimers apply.

NOTES 1. I will limit my analysis to the notion of utility as it is used in consumer theory. I will therefore not consider welfare economics, nor the theory of expected utility, which raise a quite different set of questions.


CHOICE WITHOUT UTILITY? 2. For recent and detailed expositions of Edgeworth’s economic thought, we refer to Hildenbrand (1993), Newman (1994), and Ranchetti (1996). 3. Physical Ethics and Methods of Ethics are the titles of two books by Alfred Barratt and Henry Sidgwick respectively, which Edgeworth discusses in the first part of his New and Old Methods of Ethics (1877). 4. In modern mathematical terms, assuming, as in the simplest case made by Edgeworth, that ki=k and bi=b for all the i, that is for all the sentients, the problem becomes

∑ {f(y

– f(β)} i i subject to the constraimt that

max k

∑y = Y i


where Y represents the total quantity of the existing means. Under Edgeworth’s hypotheses of f′ positive and f′′ negative, and using the Lagrange multiplier, λ, the solution of the problem follows from: max k

∑ {f( i

∑ yi = Y) y i – f ( β )} – λ(∑ i

5. On this important influence on Edgeworth see Howey (1960: chap. XI) and Mirowski (1994). 6. In fact, in mathematical terms, given two commodities x and y, and the utility function for the two individuals, P = F(x, y) e ∏ = f(x, y), the problem becomes: max P (x, y) subject to ∏ (x, y) = ∏ where ∏ is a given constant which represents the other individual’s utility, and which should not decrease. The mathematical solution of the problem, using the Lagrange multiplier, is: ∂P/∂x ∂P/∂y

∂∏/∂x = ∂∏ ∂y /

This is precisely the equation of the contract curve. The argument exposed is symmetrical: maximizing Friday’s utility, under the constraint that Crusoe’s utility be constant, would yield exactly the same result. It has to be noticed that, as Edgeworth clearly recognized, the argument is easily generalized from a two-individuals case to an n-individuals case. 7. In Ranchetti (1993) I have elaborated on the importance of the epistemological change determined by the introduction of the calculus into economics. 8. It has however to be remarked that Edgeworth is fully aware that if agents took account of the other agents’ interests, they would arrive in the marketplace at a better settlement for all the agents concerned. The debt to Edge-worth’s ideas was in fact acknowledged in the first applications of game theory to economics; see for example Shubik (1959). 9. Final degree of utility is the expression used by Jevons for the degree of utility of ‘the last increment of any commodity secured, or the next increment expected or derived’ (Wicksteed 1925: 59). 10. See, for example, Schumpeter (1954, especially the ‘Note on the theory of utility’: 1053–73) and Stigler (1950).


CHOICE WITHOUT UTILITY? 11. In mathematical terms, the difference could be put in the following manner. An ordinal utility function, u, is a function which is subject to no other restrictions than that given by x ~ > y ⇔ u(x) ≥ u(y). Whereas a cardinal utility function, u, is subject to additional restrictions, such as additivity (u(x , . . . xn) = u1(x1)+ . . . +un(xn)) or an ordered preference-difference representation (x, 1 y) ≥ * (z,w) ⇔ u(x) – u( y) > u(z)-u(w) (that is, the preference for x over y exceeds the intensity of the preference for z over w). 12. See, for example, May (1954), Georgescu-Roegen (1966), Bianchi (1985), Zamagni (1986) and Gay (1987). 13. To understand Debreu’s methodological approach and its connection with the formalist school of mathematics, Debreu (1984) is important. See also Ingrao and Israel (1987) and Napoleoni and Ranchetti (1990). On mathematical intuitionism and formalism, see Menger (1930), Bourbaki (1960), Körner (1960) and Benacerraf and Putnam (1964). 14. It is a feature of mathematical formalism to believe that it is possible to use a technical, namely mathematical, tool irrespective of its ‘original nature’. 15. See also Arrow (1984) Collected Papers, vol. 1, pp. 47–8: ‘The principal merit of the ordinal theory is the operational, behavioristic, pragmatic character of its method. In the domain of social choices it implies the abandonment of the traditional formula of Bentham and Edgeworth: social decisions should maximize the sum of individual utilities. At least, this rule must be regarded as without significance so long as an experimental procedure for measuring the utilities that are to be added is not established.’ See also Collected Papers, vol. 1, pp. 59–60. 16. To this Arrow adds a very telling footnote which is well worth quoting here: ‘I cannot help being struck by the parallelism between the economists’ concepts of tastes and obstacles and Freud’s pleasure principle and reality principle. The former, the phase of the ‘omnipotence of wishes’, seems to correspond to a pure expression of tastes, while the reality principle corresponds to recognition that tastes can only be satisfied insofar as they are compatible with the obstacles presented by the external world, including other individuals. Freud’s use of the term economic in his discussions of metapsychology is remarkably precise. He is referring to the allocation of the scarce resources of the libido among competing uses, just as the individual allocates his scarce income among competing commodities.’ 17. I want to stress the extent of the word ‘calculus’. It does not refer only to mathematical logic and practices, as in the ‘differential calculus’, but also to logic and practices which are not mathematical, as, for example, a game of cards. In ordinary language, one may calculate the alternatives which a consumer faces without necessarily having to refer to mathematics. 18. Aristotle, Metaphysics, Book X, is also fundamental here. 19. In literature, a case for lightness is made by Italo Calvino (1988) in his first Charles Eliot Norton Lecture at Harvard. 20. To my knowledge the most detailed analysis of Metretike is Baccini (1995). Mirowski (1994) is also important.


CHOICE WITHOUT UTILITY? 21. Mainly from the economic perspective, see Sen and Williams (1982), Bianchi (1985), Elster and Roemer (1991). 22. A detailed account of the fascinating history of Buridan’s ass is given by Rescher (1959–60). 23. See, for example, the passage quoted in Note 15 (last sentence).

BIBLIOGRAPHY Allen, R.G.D. and Hicks, J.R. (1934) ‘A reconsideration of the theory of value’, Economica, February and May: 24–48. Aristotle (1955) Ethics, trans. J.A.K. Thomson, Harmondsworth: Penguin. Arrow, K.J. (1963) ‘Utility and expectation in economic behavior’, in S. Koch (ed.) Psychology: A Study of a Science, New York: McGraw-Hill (reprinted in Arrow 1984, vol. 6: 117–46). —— (1967) ‘Values and collective decision making’, in P. Laslett and W.G. Runciman (eds) Philosophy, Politics and Society, Oxford: Blackwell (reprinted in Arrow 1984, vol. 1: 59–77). —— (1973) ‘Some ordinalist-utalitarian notes on Rawls’s theory of justice’, Journal of Philosophy (reprinted in Arrow 1984, vol. 1: 96–114). —— (1984) Collected Papers, London: Blackwell. Baccini, A. (1995) ‘Edgeworth sui fondamenti della probabilità e delle decisioni in condizione di incertezza’, PhD thesis, University of Florence. Benacerraf, P. and Putnam, H. (eds) (1964) Philosophy of Mathematics, Oxford: Blackwell. Becker, G. (1996) Accounting for Tastes, Cambridge, MA: Harvard University Press. Bianchi, M. (1985) ‘Utilità’, in G. Lunghini (ed.) Dizionario di economia politica, Turin: Boringhieri. Bourbaki, N. (1960) Eléments d’histoire des mathématiques, Paris: Hermann. Broome, J. (1991) ‘Utility’, Economics and Philosophy, April: 1–12. Calvino, I. (1988) Six Memos for the Next Millennium, Cambridge, MA: Harvard University Press. Deaton, A. and Muellbauer, J. (1980) Economics and Consumer Behavior, Cambridge: Cambridge University Press. Debreu, G. (1959) Theory of Value, New Haven: Yale University Press. —— (1984) ‘Economic theory in the mathematical mode’, American Economic Review, June: 267–78. Edgeworth, F.Y. (1877) New and Old Methods of Ethics: or ‘Physical Ethics’ and Methods of Ethics, Oxford: Parker. —— (1879) ‘The hedonical calculus’, Mind: July, 394–408. Edgeworth, F.Y. (1881) Mathematical Psychics: An Essay on the Application of Mathematics to the Moral Sciences, London: Kegan. —— (1887) Metretike, or the Method of Measuring Probability, London: Temple. —— (1889) ‘Economic science and statistics’, Nature, September: 496–509. Elster, J. and Roemer, J.E. (eds) (1991) Interpersonal Comparisons of Well-Being, Cambridge: Cambridge University Press.


CHOICE WITHOUT UTILITY? Gay, A. (1987) ‘Introduction’, in A. Gay (ed.) Arrow, Equilibrio, Incertezza, Scelta Sociale, Bologna: Il Mulino. Georgescu-Roegen, N. (1966) Analytical Economics, Cambridge, MA: Harvard University Press. —— (1987) ‘Ophelimity’, in J. Eatwell, P. Newman and M. Milgate (eds) The New Palgrave, A Dictionary of Economics, London: Macmillan, vol. 3: 716–18. Harsanyi, J.C. (1975) ‘Advances in understanding rational behavior’, in R. Butts and J. Hintikka (eds) Logic, Methodology and Philosophy of Science, Dordrecht: Reidel. —— (1976) Essays on Ethics, Social Behavior and Scientific Explanation, Dordrecht: Reidel. —— (1986) ‘Individual utilities and utilitarian ethics’, in A. Diekmann and P. Mitter (eds) Essays in Honor of A. Rapoport, Heidelberg: Physica-Verlag. Hicks, J. (1939) Value and Capital, Oxford: Oxford University Press. Hildenbrand, W. (1993) ‘Francis Ysidro Edgeworth: perfect competition and the core’, European Economic Review, April: 477–90. Howey, R.S. (1960) The Rise of the Marginal Utility School, New York: Columbia University Press. Ingrao, B. and Israel, G. (1987) La mano invisibile, Rome: Laterza. (American edition: The Invisible Hand, Cambridge, (Mass.): MIT). Ingrao, B. and Ranchetti, F. (1996) Il mercato nel pensiero economico, Milan: Hoepli. Körner, S. (1960) The Philosophy of Mathematics, London: Hutchinson. Mas-Colell A., Whinston, M.D. and Green, J.R. (1995) Microeconomic Theory, Oxford: Oxford University Press. May, K.O. (1954) ‘Intransitivity, utility and aggregation in preference patterns’, Econometrica, January: 1–13. Menger, K. (1930) ‘On intuitionism’, in K. Menger, Selected Papers in Logic and Foundations, Didactics, Economics, Dordrecht: Reidel. Mirowski, P. (1994) ‘Marshalling the unruly atoms: understanding Edgeworth’s career’, in P. Mirowski (ed.) Edgeworth on Chance, Economic Hazard, and Statistics, Lanham: Rowman & Littlefield. Napoleoni, C. and Ranchetti, F. (1990) Il pensiero economico del Novecento, Turin: Einaudi. Newman, P. (1994) ‘Edgeworth’s economic calculus’, Metroeconomica, June: 99–126. Pareto, V. (1906) Manuale di economia politica, Milan: Società Editrice Libraria. —— (1962) Lettere a Maffeo Pantaleoni, Rome: Edizioni di Storia e Letteratura. Plato (n.d.) Protagoras 357a. Ranchetti, F. (1993) ‘Dal lavoro all’utilità’, in G. Lunghini (ed.) Valore e Prezzi, Turin: UTET. —— (1996) ‘Francis Ysidro Edgeworth’, in B. Ingrao and F. Ranchetti Il mercato nel sensiero economico, Milan: Hoepli. Rawls, J. (1972) A Theory of Justice, Oxford: Oxford University Press. Rescher, N. (1959–60) ‘Choice without preference’, Kant-Studien, 51(2): 142–75. Samuelson, P. (1938) ‘A note on the pure theory of consumer’s behaviour’, Economica, February: 61– 71. —— (1947) Foundations of Economc Analysis, Cambridge, MA: Harvard University Press. —— (1948) ‘Consumption theory in terms of revealed preference’, Economica, August: 242–51. Schumpeter, J.A. (1954) A History of Economic Analysis, New York: Oxford University Press.


CHOICE WITHOUT UTILITY? Sen, A. (1973) ‘Behaviour and the concept of preference’, Economica, August: 241– 59. —— (1976–77) ‘Rational fools: a critique of the behavioural foundations of economic theory’, Philosophy and Public Affairs, 6 (reprinted in F. Hahn and M. Hollis (eds), 1979 Philosophy and Economic Theory, Oxford: Oxford University Press: 87–109). Sen, A. and Williams, B. (eds) (1982) Utilitarianism and Beyond, Cambridge: Cambridge University Press. Shubik, M. (1959) ‘Edgeworth market games’, in A.W. Tucker and R.D. Luce (eds) Contributions to the Theory of Games IV, Princeton: Princeton University Press. Stigler, G.J. (1950) ‘The development of utility theory’, Journal of Political Economy, August and October (reprinted in G.J. Stigler 1965, Essays in the History of Economics, Chicago: University of Chicago Press. Stigler, G.J. and Becker, G.S. (1977) ‘De gustibus non est disputandum’, American Economic Review, 67(2) (reproduced in Becker 1996, Accounting for Tastes, Cambridge, MA: Harvard University Press). Wicksteed, P.H. (1925) ‘Final degree of utility’, in R.H.I. Palgrave (ed.) Dictionary of Political Economy, London: Macmillan, vol. II: 59–61. Wittgenstein, L. (1953) Philosophical Investigations, Oxford: Blackwell. —— (1967) Zettel, Oxford: Blackwell. —— (1986) Conversations, Indianapolis: Hackett. Zamagni, S. (1986) ‘La teoria del consumatore nell’ultimo quarto di secolo’, Economia Politica, December: 409–66.



INTRODUCTION My aim in this chapter is to discuss the notion of the consumer and indicate a new direction of investigation which can address some of the problems raised in the field of consumption theory. At first sight this issue has been abundantly elaborated. One may recall the central place occupied by consumer behaviour in the micro foundations of economic analysis. However, there is a strange contrast between the central position of the consumer in demand analysis and the peculiar analytical foundations on which it rests. In neoclassical theory consumer behaviour is examined against a background of given, ‘rational’ preferences and the consumer’s role is identified with compliance with the principle of optimization, income and relative prices also given. The result is an optimal choice over goods, hence indirectly the composition of output is determined at the same time. But this has little to do with the analysis of actual consumers. If indeed one purpose of consumer theory is to contribute to an understanding of consumption patterns, consumers’ roles must be specified in a way consistent with the analysis of how consumption goods markets have evolved in advanced industrial economies. This opens up an entirely different research agenda, one in which the consumers are sovereign not in the sense of traditional theory, but because they are active in the process of shaping consumption alternatives. I here try to spell out this new agenda. After a critical examination of traditional consumer theory, I use elements of the criticisms made as a framework for surveying recent alternative views of the consumer’s role. A satisfactory reformulation of consumer choice must start, I suggest, from the notion of economic change and focus on the question of innovation in



consumption. This will direct attention to the dynamism of the production system, but also highlight the fundamental problem which stands in the way of an explicit discussion of the economic role of the consumer. For innovation in production strongly suggests that the items of consumption are defined by firms and their efforts to determine buying patterns. Questioning consumer sovereignty and the exogeneity of preferences then may seem to destroy the very notion of the consumer as an economic agent. The challenge, therefore, is to find a role for the consumer which is consistent with the process of change and at the same time such that individuals actively concur in the way consumption choices are made. It appears difficult for economic theory to consider innovation from this point of view and see through it to a different notion of the consumer. It is nevertheless, as I attempt to show, not only possible, but also quite consistent with the main forces driving the modern consumer economy and long-term trends of structural transformation.

CONSUMER THEORY: SOME ALTERNATIVES TO TRADITION How appropriate are the microeconomic foundations of consumer theory for the purpose of studying consumer choice and consumption patterns? This question is addressed in a survey of consumer theory (Zamagni 1986) which covers about twenty-five years of research in the field.1 Zamagni notes that the refinements of the Paretian ordinal utility paradigm deal with the problem which lies at the core of neoclassical demand theory: the relationship between observed economic variables, such as prices, quantities, income, and preferences, which are not observable. Much effort has been spent in determining the bare minimum requirements which ensure that the principle of rational behaviour is not violated, while yielding demand functions with the desired properties. This research has improved the analytical structure of the theory and in particular its logical consistency and generality. Its usefulness, however, remains that of reasserting maximizing behaviour as the necessary link between consumer behaviour and demand theory. The critical literature instead has argued the existence of distinctively different notions of rationality and criteria of choice from those implied by utility maximization. One series of contributions criticizes the simplified conception of consumer psychology which is at the basis of Paretian ordinal utility, considering explicitly instead the question of value systems and beliefs and incorporating developments in cognitive psychology. Neoclassical consumer theory slights the question of preferences over beliefs, concentrating



exclusively on the question of preferences over goods and states of the world (Bacharach 1986). The principle of cognitive consistency, however, suggests that subjects may suppress information which could indicate that they have made a mistake (Akerlof and Dickens 1982). Beliefs therefore may be as important as preferences for the individual’s description of the world and for defining the problem of choice. Other scholars have tried to improve the rational choice model. Becker’s model of the household production function (1965, 1976), for example, focuses on the household’s allocation of time and resources between work activities, producing income, and consumption activities, yielding utility. The attention paid to the question of time in the consumption process has led to several developments including the distinction between ‘process utility’ and ‘goal utility’ (Winston 1982). The distinction between sources of utility also provides for a richer understanding of the satisfaction connected to work, which otherwise is confined to analysis in the context of the labour – leisure tradeoff (Scitovsky 1976). Some important developments in the direction of an alternative to the Paretian ordinal utility paradigm are based on the recognition of an hierarchy in wants satisfaction, a notion anticipated by Georgescu-Roegen. The fundamental theoretical advance here is that consumption choice is better understood by shifting attention from the satisfaction of a single want, utility, to a series of distinct wants, which are met sequentially and by distinct commodities. This is the fundamental premise of Ironmonger’s (1972) discussion of the role of technical change in determining demand patterns. Hierarchy in wants satisfaction implies discontinuity of preferences, a notion which, Zamagni notes, was lost in the transition from early marginalism to Marshall’s concept of diminishing marginal utility. Preference discontinuity makes it easier to discuss alternative systems of preferences, representing and represented by differentiated life styles (Sen 1977). Zamagni argues that marginalists such as Menger, and even the early Marshall, regarded wants, rather than the notion of utility, as the foundation of demand theory. Lancaster’s ‘characteristics model’ (1971) goes back to this original insight. For this reason, though formally a modification of the rational choice model, it raises a number of questions important for the reformulating of consumer theory and will be examined in some detail later. There is a third important consequence of choosing to focus on wants satisfaction: it leads us to examine the distinctive skills of consumers. Sen has introduced the notion of subject capability (Sen 1985) to argue that what consumers can accomplish with goods for the satisfaction of needs depends on their capabilities. This suggests that there is room for a definition of possible lifestyles, in terms of the structure of production and the ways consumers develop for using and dealing with goods. Adaptive consumption models deploy a notion of rationality alternative to that of neoclassical theory. Consumers do not follow any consistent maximizing behaviour, but



rather an ‘adaptive’ pattern of choice. The characteristics of a behavioural approach to consumer theory, centred on the notion of lifestyle (Earl 1986), will be examined later. However, it must be noted that adaptive consumption models may have diverse roots. In Parrinello’s work (1984), which originates in Sraffa’s analysis of production, adaptive consumption is an implication of endogenous preferences. This notion of endogenous preferences is possibly the most drastic departure from traditional theory. Pollack (1970, 1978) has shown that, assuming preferences depend on past consumption, it is still possible to derive demand from the utility function. However, as observed by Zamagni, the implications of endogenous preferences are much broader. In general, with habit formation and preference interdependence, analysed by Pollack and others, we have examples of a new way of conceptualizing consumer choice without the standard assumption of exogenous tastes. What gives full relevance to the notion of endogenous preferences, however, is the recognition of interdependence between production and taste formation. Simply put, consumer choice is the result of adaptation to new opportunities for consumption created by the market, based on a trial and error decisionmaking process.


The rational choice model and beyond Based on this overview one may conclude that the rational choice model has very serious inadequacies as a way of analysing consumer behaviour. It is quite clear that the alternatives mooted take us well beyond the rational choice model. Nevertheless, as is often the case, there is no single, well-articulated, alternative theory for the analysis of the consumer. In particular, despite advances in the analysis of consumption activities and goals, combined with a better understanding of consumer psychology and decision-making processes, there is no theoretical framework for the analysis of choice in a dynamic context, one in which both preferences and items to satisfy them may change. For this very reason it is hard to conceive a more satisfactory notion of the consumer though, as I have indicated at the outset, such a notion needs to be grounded in the process of change determining consumption patterns. For change appears to be the context most appropriate to investigating a possible active role of consumers in determining economic outcomes. To move us towards that context I will discuss how change in consumption has been taken into account and how it has contributed to redefining the problem of choice.



Wants, technical change and new commodities Lancaster and Ironmonger explicitly introduce the question of technical change, though for somewhat different purposes. Lancaster’s characteristics model confronts the difficulty posed for traditional consumer theory by new products, quality changes and product differentiation. In traditional theory, he points out, ‘Any change in any property of any good implies that we have a new preference pattern for every individual’ (Lancaster 1971: 4). This follows from the fact that utility indexes are associated directly with goods. Consequently, we can proceed as if each new variant product is the same good as before or treat each variant as an entirely new good, throwing out any information concerning demand behaviour with respect to the original variant, and start from scratch’ (ibid.: 8). Goods are the objects of ‘wants’, however, and valuable to consumers, Lancaster insists, because of their ‘characteristics’, i.e. their objective properties. He therefore considers people and things to be related in two steps: first, things are definable by their characteristics (objective and technical); second, people enjoy and select among characteristics (ibid.: 7).2 The data matrix of the ‘consumption technology’ expresses the relationship between the characteristics vector and the goods vector. Hence, the characteristics space can be transformed into the goods space, making of choice a linear programming problem. Given prices, a good will become part of the consumption basket if it lies on the efficiency frontier. Lancaster points out that the ‘operational use of the model requires identification of the relevant characteristics and data on the consumption technology’ (ibid.: 113).3 Characteristics are identified mainly on a priori criteria, while the link between characteristics and things makes it possible to define the consumption technology data. The last step is then to infer individual preferences, the things–people relationship, from the empirical study of markets; ultimately market data should confirm whether consumers do respond to particular characteristics or not (ibid.: 157).4 Lancaster’s objective is thus to put the ‘characteristics model’ at the core of demand theory, and show that ‘product variations and new goods fit easily and naturally’ (ibid.: 10). Traditional analysis would appear as a special case of this more general approach, one in which ‘the number of goods and characteristics is equal and the efficiency surface consists of a single facet’ (ibid.: 50). The superiority of the model, however, hinges crucially on the possibility of identifying relevant characteristics, which should be less in number than products.5 Despite similarities with Lancaster’s model, Ironmonger’s approach (1972) rests not on characteristics but on the separability of wants. The starting point is once again the problems created for traditional consumer theory by new commodities and quality changes. Ironmonger’s revision is centred on a consumption technology, intended to satisfy ‘various



separate wants’, leading to the choice of an ‘optimum budget . . . {which} is found to be a solution of a linear programming problem’ (1972: 12). The maximization of satisfaction is subject in the ordinary way to a budget constraint and to satiation effects. What is new here, however, is that the objects of satisfaction are wants, not some ‘single desire, happiness or utility’. The existence of separate wants is seen by Ironmonger as a way to bring technical change into the consumption process (ibid.: 13). Indeed, the question which comes to light is how new commodities redefine consumer choice. Ironmonger observes that one of the characteristics of a static equilibrium is the assumption of constant tastes. But even without changes in the factors customarily taken to determine taste, such as age, sex, occupation, marital status, the ‘number and nature of commodities are constantly changing’ (ibid.: 12). The stability of tastes, he argues, is far greater than that of the means to satisfy them. Therefore, it is desirable to have a theory which keeps separate the two aspects.6 It follows too that new commodities, in their potential to satisfy separable wants, add a dimension, independent from tastes, income and prices, to the determination of consumer choice. The recognition of a role for technical change in consumption also leads to new possibilities of empirical analysis. A large part of Ironmonger’s own contribution is the study of diffusion paths of new commodities and their effects on market demand.

Changes of taste and endogenous preferences Lancaster succeeded in opening consumer theory to one dimension of change which matters, that of innovation. Shifting attention from the utility function to the separability of wants, Ironmonger established the importance of technical change and therefore innovation from the production side, in determining demand patterns. Technical change transmits dynamism to the domain of consumption, though tastes may well be constant or slowly changing. Neither Lancaster nor Ironmonger, however, go very far into the analysis of taste formation. Interestingly, both seem to think that the exogeneity of tastes is a consequence of the existence of a subjective element in consumer choice. Ironmonger, in particular, argues that the ‘want-satisfying powers’ of commodities depend on their qualities, i.e. their ‘technical nature’, but also on ‘the nature of the wants they serve and the consumer’s valuation of their effectiveness in serving these wants . . . these subjective qualities are the links between commodities and wants’. (1972: 15). Changes in wants are ‘due to the factors determining the make-up of the consumer’s physiology and psychology’ (ibid.: 13). Therefore, even if technical change is a factor determining consumer choice independently from taste, income and prices, there is an element of subjective evaluation by the consumer which follows from the extraeconomic factors which determine ‘the consumer’s fundamental tastes’ (ibid.).



This is a familiar argument. Indeed, it is an often taken for granted presupposition that preferences are exogenous, i.e. determined by extra-economic factors. Its consequences, however, are far reaching. First, it implies that, no matter how inclusive and complex the list of these factors, taste formation remains a matter which cannot be investigated by economic analysis. Second, it suggests that tastes can be taken as a datum, which makes the optimizing procedure both possible and meaningful. Finally, and most importantly, exogenous taste is the foundation of consumer sovereignty. And precisely because of the extraeconomic nature of preferences the economic role of the consumer is magnified, insofar as consumer sovereignty ultimately determines the composition of output. The recognition of the role of technical change does not seem to have taken the consumption technology approach beyond traditional premises. On the other hand, the criticism of exogenous preferences and consumer sovereignty is not new. A wide-ranging marketing literature discusses the manipulative impact of advertising. Within consumer theory, however, argument has centred on the different sources of preference endogeneity: habit formation, preference interdependence and changes of taste due to new goods which become available. Choice, it is suggested, depends on past consumption experience, individuals’ positions in the social structure and firms’ production and marketing strategies. An important though neglected contribution along these lines is Duesenberry’s Income, Saving, and the Theory of Consumer Behavior (1949). Duesenberry’s systematic criticism of consumer theory had little impact, possibly because he addressed himself mainly to the analysis of the macro-economic choice between consumption and saving. Still, Duesenberry develops an articulate discussion of preference interdependence. He argues that what he calls the ‘demonstration effect’ not only sustains consumption spending, but also explains choice. Choice is driven by the search for high quality in consumption and in particular by comparisons with the lifestyles of reference groups, which he considers the next highest income groups. For Duesenberry there can be no theory of consumption choice in the traditional sense, since what is supposed to be fixed, namely, preferences and tastes, are influenced by others’ choices. This insight undermines the economist’s effort to stay away from the social and psychological dimensions of consumers’ decision-making, which is what makes rational choice a paradigm of strong predictions. On the contrary, only enquiry into these issues seems likely to issue in a meaningful theory of consumption spending. Despite the clarity with which the question of taste formation is posed by Duesenberry, the elaboration of an adequate theory of the consumer remains a task largely unaccomplished.



A FRAMEWORK FOR THE ANALYSIS OF CONSUMER CHOICE Consumers: active or passive? We are nevertheless now in a better position to evaluate the implications of the critical perspectives with which we began and to assess their main positive contributions to the analysis of the consumer. The discussion of product innovation has brought to the fore the analysis of changes in the items of consumer choice, but has not led to an analysis of the relationship of change and taste formation; tastes are still, in the contributions of Lancaster and Ironmonger, determined by exogenous preferences. On the other hand, the notion of endogenous preferences creates new problems of its own. If social dynamics and the system of production end up also shaping taste, the question of choice and the role of consumers as economic agents virtually disappears, or remains largely undetermined within an adaptive theory of consumer demand. It might even be concluded that the consumer per se has very little role to play in determining economic outcomes. The criticisms of traditional theory seem then to lead to a dead end: challenges to consumer sovereignty and the basic economic model, where the consumer’s choices determine what goods will be produced, have put nothing in their place. We are left with an unpalatable choice. Either consumer theory cannot account for the dynamics of change originating in the relationship between technical change and taste or, in order to address this, it eliminates any larger role for the consumer. But it is at least now also clearer in which sense traditional consumer theory has neglected the question of change while giving, formally, the greatest relevance to the role of the consumer. The consumer is an active economic agent insofar as he or she behaves as a utility maximizer over known preferences; sovereignty, in turn, depends on the fact that preference formation is an extraeconomic process. The paradox is that when preferences are not extraeconomic, but shaped within the socio-economic structure, the existence of individuals as independent decision-making agents is called into question. If we were to concede that the element of subjectivity involved in choice is a characteristic of individuals as such, but that preferences are endogenous, we would face an apparent contradiction. Reconciling these two aspects is fundamental if we are to retain important roles for the consumer and also discuss technical change in a new perspective.



Choice, individuality and lifestyle On the one hand, I have urged, the analysis of the consumer should be connected to the process of change and capable of identifying its role within that process. On the other, the persistence of a problem of choice, without any reference to utility maximization, requires a redefinition of the goals and motivation of individuals confronting consumer goods markets. A new economically relevant role must be consistent with that motivation. We may recall that, according to Lancaster, goods are valuable insofar as they are object of wants. In turn wants relate to the objective ‘characteristics’ of commodities, but involve a subjective evaluation, reflecting individuals’ preferences. If we were to assume that subjective evaluation really involves the attempt of individuals to pursue a self-image with their consumption choices, we could then conclude that choice must be grounded in some notion of individual identity. A possible direction of investigation would then be to locate motivation, not in the efficient pursuit of ‘utility’, but in a drive to satisfy what have been called ‘identity needs’. Karasek (1980) argues that these are superior needs which reflect the ‘ultimate goal’ of human beings, namely, the development of one’s potential through interaction with the outside world. Karasek’s ‘life style/identity needs’ are analogous to the highest order needs of Maslow’s hierarchy (1968), where self-actualization is at the very top. Identity needs, however, have a more general relevance in the present context. They imply an individual pursuit guiding the consumer and therefore suggest active, directed behaviour in the process of needs satisfaction. Individuals, however, pursue the realization of identity with respect to a constantly changing world of commodities and the evolution of consumption alternatives which that entails. In turn this evolution is fed by the increasing sophistication of wants above and beyond subsistence, which follows from income growth, and by the increasing refinement of the notion of individuality and social identity. Individuals strive for identification within the social structure, bending towards their private aims the system of commodities which at least partly develops along the path of change dictated by industrial transformation. This notion of individuality does not need any extraeconomic, ahistorical notion of preferences; at the same time, it can reconcile the evolving social identity of individuals, which is a social process involving taste formation, with their existence as actors in the sphere of consumption. The effect is to redefine consumers’ choice starting precisely from the fact that consumers are interactive social agents responding to the process of change. Their economic role is consequently defined as inventors of consumption practices which feed back into the production system, validating the final specification of commodities and the selection of technologies of production and distribution.



Such a view fully overcomes the limitations of the notion of exogenous preferences and stresses the fact that it is necessary to link ‘active’ consumption to the process of innovation. Innovation need not be identified purely with the supply side and the development of technology. Consumers are active in relation to a process, that of innovation, which is broader than simply new products. Hence, it would be better to speak of social innovation, the sources of the process we focus on being social, rather than purely technological. The pursuit of individual identity and self-realization along these lines can be fitted into a framework which makes reference to wants as a rather complex system of goals constituting a lifestyle. In turn the antecedent can be considered a collection of consumption choices which gives rise to differentiated patterns of consumption. The notion of lifestyle is central to Earl’s (1986) analysis of consumer choice. Under the heading of the ‘enterprising consumer’ he discusses the strategic decision-making process designed to cope with turbulence, developing what he suggests be called ‘a behavioral approach to household production theory’ (1986: xi). The idea of strategy is elaborated in conjunction with the effort to reconsider the ‘psychological underpinnings of consumer satisfaction’ and redirect attention to ‘what consumers actually feel and are trying to do with their lives.’ (ibid.: 86) To avoid the artificially narrow understanding of choice imposed by the rationality assumption of the traditional theory, Earl draws on personal construct psychology, arriving at the notion of the ‘inquisitive consumer’ which he projects into the domain of choice. The lifestyle idea then emerges as the notion needed to connect economics with ‘the world of the modern consumer’. Earl intends by it ‘“ways of life” as if they are viscous collections of procedures for dealing with fluid situations in which ambiguity is the order of the day’ (ibid.: 4). Though very much in accord with many of the theoretical premises of the lifestyle approach and in particular with the view that ‘there is much more to being a consumer than the reactive activity of making marginal substitutions in a household production system as opportunities change’ (ibid.: 54), the same notion attains a rather different connotation in a framework built on the notion of change in consumption. A lifestyle may indeed be the result of an adaptive pattern of behaviour and of changing personal constructs. However, in my perspective, it depends fundamentally on the interaction of individuals’ attempts to create a self-image, on the one hand, with innovation and technical change, on the other. Lifestyle is both a result and the guiding star of the pursuit of identity and of the invention of consumption practices within an evolving system of commodities. Without the notion of identity it would remain unclear what guides individuals in their adaptation. The concept of an active consumer does not contradict the view that individuals pursue in their choosing a number of goals which are defined within a lifestyle. But it focuses more on the idea that a socially moulded individuality has a role in determining consumer choice and



that it does so by interacting in a creative, active manner with the major source of innovation, i.e. the dynamism of the production system. This does not involve any particular psychological assumptions of behaviour. It only assumes the existence of identity as the ultimate determinant of consumers’ motivation, within a larger process of change in the range of marketable commodities and of consumption alternatives. This notion of individuality therefore suggests a more specific and intelligible economic role for consumers. As a consequence too consumer learning can be grounded in the process which defines consumption alternatives and contributes to their evolution.7

Innovation and consumption choice What I am suggesting is that the search for identity is a very particular type of motivation which complements the social norm implied in the notion of ‘lifestyle’. The collection of goods and consumption practices associated with a lifestyle certainly defines a stratification in the forms of need satisfaction. However, both from the point of view of the development of human needs and from the point of view of the variability of these forms, identity, hence individuals themselves, constitute the element of creative input which can redefine, through choice, consumption alternatives. A so far neglected aspect of this treatment of consumer choice is of course novelty. Lancaster’s main motivation was to reconstruct consumer theory to enable it to cope with product innovation and differentiation. One of the weak points of his ‘characteristics model’, however, as indicated above, is the way the data of the consumption technology are derived. In the present approach the motivation arising from the pursuit of identity suggests a solution to this problem. Innovation from the supply side can be incorporated in the analysis of choice as the element which inserts dynamism into the definition of consumption alternatives for the satisfaction of particular needs. The creative contribution of consumers according to their personal and social backgrounds, meanwhile, defines the final form in which needs will be met – in Lancaster’s terminology, the consumption technology. Similarly, new commodities are not the only channel through which innovation affects consumption patterns, as stressed by Ironmonger. It is rather the case that new complementarities and the definition of use systems determine the success of new items of consumption and their diffusion paths. This, as much as technical change, can explain the concrete forms taken by the efforts of consumers to satisfy their needs.



CONSUMERS AND THE FORCES DRIVING MODERN CONSUMPTION Industrial transformation and consumption markets An important aspect of what makes plausible and appropriate the view of the consumer sketched above can be located in the process of industrial transformation and the new competitive pressures arising in global markets. From this point of view the redefinition of the economic role of consumers only reflects the reality of changing markets. Stanback et al. (1981) have argued that the relationships between industrial restructuring, the development of consumers’ markets in advanced industrial economies, and the role of consumers are essential aspects of the rise of the ‘service economy’. According to their analysis the new emphasis of the past two decades on quality and product differentiation reflects the increasing complexity of market environments from the point of view of the definition of needs and their spatial and social variability. Product tailoring and specialization follow from the activation of the self-image of differentiated groups and subcultures, within an increasingly complex society. The very pressure towards homogenization of consumption patterns experienced with mass production pushes individuals and groups to look for some distinctive character by means of differentiated products. In this setting consumer behaviour is dominated by the possibility of satisfying higher order needs, emerging with the existence of discretionary income. The analysis makes reference to Maslow’s classification of needs and argues that higher order needs are not satisfied ‘efficiently’, saving scarce resources, but rather by spending resources.8 ‘New skills and new interests – both time consuming – are characteristic of ‘utility’ in lifestyle/identity consumption’. (Stanback et al. 1981: 32). A consumer oriented economy must be a service economy because of the joint nature of modern consumption. Innovation is not purely a matter of goods, but also of delivery and characteristics, which in turn may depend on services. Newness becomes a much more complex and subtle matter. Complementarities take on a new meaning and so does the input of the consumer. ‘Where products are increasingly differentiated . . . or the individual’s special activities and lifestyle become more important, the decision making process becomes more complex and the consumer must be more involved’ (ibid. 1981: 41). Given these technological and organizational changes involved in the development of consumers’ markets, the emphasis naturally falls on participation, search and validation on the part of consumers. More than ever consumption is seen to be a process, rather than a single act, involving products as well as the time and the contributions of consumers themselves.



Changes in the competitive environment A new body of literature on management and organization also insists on the importance of serving the consumer and on a firm’s market orientation along lines which suggest a break with previous management theories. Quite obviously customer orientation and product strategies are not in themselves novel. However, new management approaches which focus on ‘total quality’ are the result of a semantic shift which needs to be clarified. ‘Total quality’ stresses product quality in a sense quite different from that of the theory of monopolistic competition. The new meaning is internal to the theory of organization and stands for customer satisfaction. ‘Lean production’ can similarly be understood as a move away from large organizations to more flexible, customer-oriented organizations. The effort to develop products in close contact with clients becomes in this perspective the key to a leaner organization. The customer and the market are central also to the notion of globalization. This is a rather complex notion. Though it can be taken to represent ‘a push towards an international dimension of the competitive process’ (Vicari 1989) the point is not the broader geographical dimension of markets, but rather the competitive requirements which lead such a process. Keinichi Ohmae (1990) has argued that the fundamental force driving competition in global markets is the weight of fixed costs and, in particular, the rising costs of developing winning ideas which implies a need to operate on a larger scale. At the same time, in this new market environment, product strategies must focus on the creation of new value for clients, not on the creation of barriers to competition. Consequently here too the new competitive paradigm emphasizes the emerging power of clients. However, a global market does not mean homogeneous products. On the contrary, with the exception of a few worldwide market segments, global markets require first of all an understanding of consumers, which are much more specifically defined than the reference to worldwide consumption standards might suggest. Successful firms will then operate globally while penetrating markets locally. It follows that the new competitive environment implies a closer monitoring of the market and a deeper involvement of consumers. Consumers’ contributions become more important because the very logic of the global economy makes them both more informed and more demanding. On the other hand, product strategies which do not include an essential input on the part of consumers may be too risky since consumers’ reactions are also becoming more difficult to forecast. It may be concluded that the new managerial theories stress the imperative of serving the market and tend to focus on redesigning a firm’s strategy and organization in that direction. Global marketing strategies consequently emphasize the interaction with consumers, whose



role is magnified by the market orientation of global firms. In turn that implies a much more intimate connection between product development and market, between technologies of production and technologies of consumption.

New products and social innovation In the new competitive environment the consumer is increasingly an interactive agent, responding to and modifying the stimuli coming from firms’ productive and marketing strategies. This aspect of consumers’ behaviour, which emerged already in the discussion of the consumer, has now been shown also to be part of marketing strategies. Consequently the development of new products must be seen in a new light. New products considered as investment opportunities can be validated only within the same interactive process. This clarifies further the sense in which we can think of the consumer as an active agent. Writing of new products Nina Shapiro (1986) has argued that in the introduction phase the consumer is crucial in establishing the product as an investment opportunity. In particular, consumers with high discretionary incomes are important, since they are able and motivated to buy new products. Their predisposition is a result of previous experience in consuming and a more developed experimental attitude due to the availability of resources. These consumers are for the same reasons more receptive to further learning and innovation. The creation of a new product is then the result of specific circumstances, including the technological and organizational know-how of the firm, or potential firm, and the capacity of such consumers to devise modes of use and social practices which determine the utility of the product. The introduction of a new product is therefore driven by an interplay between technology, defining the characteristic of the product, and innovation in consumption, determining its use. This interactive process leading to the establishment of a new product suggests that the forces accounting for the definition of consumption alternatives are considerably more complex than the pure reference to technological advances or changes in tastes driven by extraeconomic factors. It is for this reason that I have argued above that we should speak of ‘social innovation’. The notion of social innovation has been elaborated by Gershuny (1983) in close connection with a long-term view of changes in the division of labour and in the composition of final demand. Gershuny distinguishes between technical change originating in the workplace – ‘embodied in new production technologies, new machines, new ways of organizing work’ (1983: 1) – and that originating in the household. The former traditional notion of technical change assumes a clear line of causation running from production to consumption; the



latter instead focuses on the changes in the ‘mode of provision’ by which households satisfy needs. These modes include service functions such as ‘food, shelter, domestic services, entertainment, transport, medicine, education, and, more distantly, government, “law and order” and defense’ (ibid.: 2). This aspect of technical change, much less integrated into economic analysis, is what Gershuny calls ‘social innovation’, i.e. changes in the modes of provision which affect the ‘the household’s pattern of expenditure on final commodities’ (ibid.: 2). Gershuny’s amplified approach to technical change includes a simplified model of how households choose between ‘alternative technical means of provision’ (ibid.: 4). The model treats the question of item choice in terms of broad categories of needs, but it has other implications important for my thesis. Changes in the means of satisfaction of needs imply that the same need may be satisfied by different commodities at different historical junctures and therefore by different economic structures. This seems a promising way to analyse empirically the forces determining consumption patterns. Moreover it makes plain that the forms taken by technical progress depend on consumers’ contributions. The selection of needs to be satisfied and items of satisfaction is guided by the practice of consumption which becomes part of a specialized knowledge held by consumers, especially those enjoying discretionary income opportunities and larger access to information. Indeed, what Gershuny calls the ‘modal split’, i.e. ‘the statistic which indicates distribution of provisions for a particular function between the alternative modes’ (ibid.: 50) and the change from traditional to innovative models of provisioning (ibid.: 51), are the outcome of the use-system devised by consumers. They account for changes in the pattern of final demand and, in particular, for the proportion of consumer durables and service activities which serve to satisfy households’ service functions. Gershuny’s objective is to show that the notion of social innovation can be used to analyse changes in the division of labour. In the present context it helps to locate and specify the consumer’s contribution, going beyond an acknowledgment of the general influence of social dynamics on consumption patterns. For Gershuny leads us to attend to change in the ‘technique’ of need satisfaction, involving new products in conjunction with changes of lifestyle and modes of use of commodities. Modes of use in turn may be contingent upon the creation of a number of physical and social infrastructures.

CONCLUDING REMARKS I began by suggesting that the study of choice under assumptions like given preferences determined by extraeconomic circumstances is a particularly poor way to account for change



in patterns of consumption. Lancaster and Ironmonger, we saw, pioneered efforts to link change in the form of product innovation to consumption, but both continued to think of the want satisfying power of goods as something technical, ultimately reducible to a confrontation between given tastes and the physical characteristics of (new) products. The fundamental challenge therefore remained to find a role for product innovation while allowing also that consumers may experience changes of taste. Suggested resolutions, such as ‘adaptive’ tastes and ‘endogenous’ tastes, turn out on inspection to ignore any reference to consumer motivation, hence allow for virtually everything, and explain nothing. Earl’s notion of ‘lifestyle’ held out a promise that consumption might be connected to social and personal aspirations, but stressed adaptive and defensive behaviours rather than focusing on the possible content of the idea of change in consumption. My preferred alternative is the notion that individuals pursue self-identity and at the same time actively invent consumption practices in the context of an evolving system of commodities. This notion does not presuppose any particular psychological assumptions about behaviour. It is consistent with the lifestyle idea, but emphasizes creative, active interaction between consumers and the production system. And it resonates with a range of analyses of recent changes in international markets and more general considerations of what is required for any new product to be successful. Chief among these requirements is that consumers ‘validate’ the product. This identifies a determinative role for the consumer quite lacking in the traditional theory of consumer choice, a role that has recently been defined more precisely through two ideas: the notion that consumers build up expertise in consumption; and that the consumer increasingly selects the modes of provisioning which will enable his or her wants to be satisfied. The consumer thus begins to emerge as an expert consultant to the producer or would-be producer, and one whose signals reflecting fine distinctions of taste are essential to productive enterprises.

NOTES 1. Zamagni takes as reference point Houthakker’s 1961 survey. His purpose is to provide an overview and critical evaluation of the main theoretical contributions, not an exhaustive review of consumer theory. In particular he chooses to neglect the welfare approach and the theory of consumer choice under uncertainty. 2. A characteristic is ‘an objective, universal property of the good (or activity) . . . personal reactions are reactions to the characteristic, not reactions about what the characteristic is’ (Lancaster 1971: 114). 3. ‘Neither of these requirements is yet easily met, partly because of the conceptual problems in


CHANGE, CHOICE AND INNOVATION identifying relevant characteristics and partly because the appropriate data have not hitherto been available’ (Lancaster 1971: 113). 4. An intermediate step, ‘group analysis’, should define the boundaries within which substitution effects are likely to be strong. 5. Otherwise the gains with respect to the traditional approach would be more apparent than substantial. Zamagni argues that the novelty of the model is very much in doubt. A neoclassical economist could be in an even more embarrassing position if the list of characteristics were to change more rapidly than the list of goods, a circumstance which can easily arise from the experience of consumption. This difficulty may explain the small impact that the characteristics model has had on consumer theory. 6. ‘New commodities and quality changes in commodities can be brought out of the pound of consumer tastes so that tastes can provide a more constant frame of reference than before’ (Ironmonger 1972: 13). 7. Learning of their new preferences by consumers is central to Pasinetti’s structural dynamics (1981, 1993). The learning process, however, remains largely unexplored because of the reference to exogenous preferences (Gualerzi 1996). The view of innovation in consumption advanced here is a first elaboration of that issue. 8. The hierarchical sequence of human priorities, according to Maslow, is dominated by superior needs such as self-actualization, self-esteem, social affiliation. They come next to the need of security and biological survival.

BIBLIOGRAPHY Akerlof, G. and Dickens, W. (1982) ‘The economic consequences of cognitive dissonance’, The American Economic Review, 72. Bacharach, M.O. (1986) ‘The problem of agents’ beliefs in economic theory’, in M. Baranzini and R. Scazzieri (eds) Foundations of Economics, Oxford: Blackwell. Becker, G.S. (1965) ‘A theory of the allocation of time’, Economic Journal 75: 493–517. —— (1976) The Economic Approach to Human Behavior, Chicago: Chicago University Press. Duesenberry, J.S. (1949) Income, Saving and the Theory of Consumer Behavior, Cambridge, Mass.: Harvard University Press. Earl, P. E. (1986) Lifestyle Economics, New York: St. Martin’s Press. Gershuny, J. (1983) Social Innovation and the Division of Labour, New York: Oxford University Press. Gualerzi, D. (1996) ‘Natural system, endogenous structural change and the theory of demand: a comment on Pasinetti’, Structural Change and Economic Dynamics, 7. Houthakker, H.S. (1961) ‘The present state of consumption theory’, Econometrica, 29. Karasek, R. (1980) ‘New value economics’, working paper, Department of Industrial Engineering and Operations Research, New York, Columbia University. Ironmonger, D.S. (1972) New Commodities and Consumer Behaviour, Cambridge: Cambridge University


CHANGE, CHOICE AND INNOVATION Press. Lancaster, K.J. (1971) Consumer Demand: A New Approach, New York: Columbia University Press. Maslow, A.H. (1968) Toward a Psychology of Being, New York: Van Nostrand Reinhold. Ohmae, K. (1990) The Borderless Economy, New York: McKinsey. Parrinello, S. (1984) ‘Adaptive preferences and theory of demand’, Journal of Post Keynesian Economics, VI. Pasinetti, L.L. (1981) Structural Change and Economic Growth, Cambridge: Cambridge University Press. —— (1993) Structural Economic Dynamics, Cambridge: Cambridge University Press. Pollak, R.A. (1970) ‘Habit formation and dynamic demand functions’, Journal of Political Economy, 78. —— (1978) ‘Endogenous taste in demand and welfare analysis, American Economic Review, 68. Scitovsky, T. (1976) The Joyless Economy, New York: Oxford University Press. Sen, A. (1977) ‘Rational fools: a critique of the behavioral foundations of economic theory’, in A. Sen, (1986) Scelta, benessere, equita’, Bologna: Il Mulino. —— (1985) Commodities and Capabilities, Amsterdam: North-Holland. Shapiro, N. (1986) ‘Innovation, new industries and new firms’, Eastern Economic Journal XII. Stanback, T. M. Jr, Bearse, P.J., Noyelle, T.J., and Karasek, R. (1981) Services: The New Economy, Totowa, NJ: Allanheld Osmun. Vicari, P. (1989) Nuove dimensioni della concorrenza, Milano: Egea. Winston, G.C. (1982) The Timing of Economic Activities, Cambridge: Cambridge University Press. Zamagni, S. (1986) ‘La teoria del consumatore nell’ultimo quarto di secolo: risultati, problemi, linee di tendenza’, Economia Politica, III.


4 TASTE FOR NOVELTY AND NOVEL TASTES The role of human agency in consumption Marina Bianchi

INTRODUCTION How do consumers choose? Starting with this question may sound paradoxical when the whole economic edifice is built on individual consumer choice as paradigmatic. On the basis of this paradigm, and with the simple aid of given and well ordered preferences, the consumer is supposed perfectly able to select that combination of goods which equalizes at the margin their weighted benefits, and to alter it in response to a change in the weights, or relative prices, or a change in income. However, when we press the logical consistency of preferences assumption to yield a more detailed analysis of the consumer’s choice, all we learn is that the consumer simply prefers more to less, because of supposed non-satiation. In other words, we know that goods are goods. But how goods become goods, and are selected as goods, what role their differentiated properties play, these are questions which remain (still) unanswered. How do consumers choose new goods? What is the place of genuine novelty in consumer choice? This is an even more difficult question to answer on the basis of the traditional paradigm. Novel goods are by definition outside the range of experienced goods which belong to the selected set of choice. Since no learning skills are made explicit in the theory, there is no way consumers can apprehend their yet unknown characteristics. Moreover, why should they? Why should they break an equilibrium solution for the unknown gains of new goods and activities? That these questions are still unanswered should not surprise the economist, since several leading theorists have pointed out the as yet unexplored richness of consumer behaviour. Nor are they unfamiliar to other social disciplines: anthropologists, psychologists, economic historians, marketing specialists have all looked at the problem with original eyes. Still, on this



topic, and despite various attempts to bridge the gap, the different disciplinary results remain distinct, and an economic explanation is still wanting. Moreover, both economics and other social sciences seem too prone, when the object is the consumer’s behaviour, to accept explanations which involve more normative judgement than analytical strength. This chapter will address and try to answer some of these questions by introducing a different paradigm of economic choice. Three points will be argued.


The consumer is also a maker, an active producer of his or her own wellbeing.


The production of individual wellbeing is also an innovative, creative process.


Novelty, discovery and surprise can be pleasant and as such become determining elements of the consumer’s ‘utility function’.1

To illustrate the basic problem that we face, I shall start with an historical example – the story of the Indian painted calicoes which in the seventeenth century started to appear, along with many other new goods, on the European market. This will ease our way into a discussion of the ‘externalities’ and interdependencies that exist among goods and which allow for free and novel recombinations on the part of the consumer. A brief reconstruction of alternative economic analyses of consumer choice will follow in order to address point (1). This will provide the basis for an analysis of points (2) and (3), and for devising a new paradigm of consumer choice where novelty plays a role and is actively produced.

TASTE FOR NOVELTY When Indian printed and painted calicoes were first brought into England early in the seventeenth century, they were just ‘leftovers from the barter trade’ for spices, and known only to few (Irwin and Brett 1970: 3). By the end of the century, they had gained such a widespread success, invading home furnishing fabrics and then clothing, as seriously to threaten the domestic production of wool, and to generate a wave of protest and complaint from domestic textile manufacturers. What had happened? In England sumptuary laws which prohibited fashionable display in clothes and decorations were repealed in 1604.2 Subsequently (in 1678), the restrictions put on the importation of French fabrics, silks, satins, and brocades (Mukerji 1983) must have induced a search for exciting replacements. But the Indian cotton’s success was mainly due to its intrinsic innovative qualities. It was very light and sheer like silk,



but also figured with a variety of bright colours, which, with their intricate floral patterns and decorations, rivalled the best French brocades and the finest embroideries. Moreover, Indian calicoes were much cheaper than silk and easy to wash because, surprisingly, their colours were water resistant. This was a property which was unknown in contemporary European coloured fabrics. All previous experiments with painted and printed fabrics made in Europe using oilmixed pigments had yielded only ‘fugitive’ results (Irwin and Brett 1970: 1; Mukerji 1983: 195), and Europeans were far behind the art of using mordant and resist-dyeing which Indians had mastered centuries earlier. Finally, and importantly, Indian calicoes were a ‘novelty’, which combined the exoticism of Indian culture with the already existing fascination for Chinese art and artefacts (cane chairs, lacquer work, cabinets of curiosities, tea, porcelain).3 With these technically and aesthetically innovative characteristics, Indian calicoes captured the taste of both middle and upper classes, who soon discovered their multiple uses, as decoration for walls, floors, windows, beds, furniture, equipages, and bodies. In the years between 1680 and 1685 calico’s success was such as to represent a serious danger to landowners’ and artisans’ interests in the fabric trade. Several petitions by silk and wool manufacturers were aimed at halting the importation and use of calicoes in England.4 The resulting laws however did not stem the desire for calicoes. Now scarce, they became even more coveted than before, and constantly new ways of continuing their use were found: recycling old clothes, utilizing furnishing and wall coverings for dresses, creating patchwork and appliqué embroideries and, of course, smuggling. Only the home production of a competitive fabric with comparable characteristics of variety, gaiety and lightness could solve the problem. This was a propelling force for the invention of weaving and spinning machines (Mukerji 1983). Success came some forty years later, in 1760, with the Arkwright spinning machine, which could make strong and fine cotton warp as good as that of the Indian cottons. The centralization of work in a factory imposed by the machine’s waterframe, which could be used only with an external source of power, vastly increased production. The improvements in quality and the lower prices that followed quickly turned demand towards English rather than Indian calicoes, though first in this instance among the lower and later among the upper classes. Not until the end of the century, however, were new dyeing techniques and new and cheap ways of colouring discovered, replacing the time-consuming processes of bleaching and dyeing (see Thomas 1965). With the development of the roller printing machine in 1783, calico could finally be printed fast, and in large quantities, putting an end to the old and, until then unsurpassed, Indian supremacy.5 The versatility of cotton, however, and its infinite diversity of styles and colour patterns, won its permanent place as the most prominent textile in demand by consumers (Cipolla 1973; Lemire 1991a).6



The result of all this was that cotton, in itself a rather humble material, became a commodity which carried an appeal and embodied a degree of ingenuity that had seemed reserved not only for ‘superior’ materials, but also for other techniques and products such as painting, decoration and art. New aesthetic combinations entered everyday life and became available to a wider range of the population; this in turn multiplied the possibilities for this fabric to be used in changeable ways for home and personal adornment.7

CROSSING UTILITARIAN BOUNDARIES The first evidence that the story of painted and printed calico brings to light is that a ‘good’ cannot be adequately defined by some single function. In fact, goods appear to be a multilayered set of functions, properties, characteristics, whose system of interconnections extends far beyond the boundaries of the single good.8 It is exactly this decomposable complexity of goods which allows for innovation. A new good, by introducing some new characteristic, or simply by changing the internal order of the existing ones, alters this system of interconnections. This has two consequences. One is that a novel characteristic may carry novelty much farther and start a chain of change that involves all the other interacting goods. The second is that a new good is never completely new. The new functions are still part of a system of relations which continue to be the same and make the good still understandable and recognizable to its adopters. Together these implications mean that novelty is both more radical and less disruptive than is implied by the notion of a good as a self-contained entity. We have seen all these effects at work in the case of cotton calico. If calico represented a simple improvement on existing fabrics, which it could silently replace because cheaper and better, on the other hand it was immediately adopted in a variety of new ways that completely redefined the role of materials and their combinatory relations among themselves and with other goods. Washable, light and colourful when used as clothing material, calico provided for better personal care and more freedom of movement, as well as for change, variety and multiple aesthetic arrangements. When used as a home furnishing material, not only did it add new decorative dimensions to existing accessories (bedspreads, valances), but also created new functional properties, for example, by redefining home spaces through colour contrasts and similarities. The creation of these new properties and possibilities of uses is often obtained by crossing the ‘functional’ or utilitarian boundaries which define and distinguish a good from others. In



our story, painting descends from the walls to be applied to materials and goes back to the walls in the form of painted fabrics used as curtains and wall hangings.9 Art, nature, religion, voyages and history all clash and mix in such a new product. Because of the extensive chain of ‘network’ possibilities which exists among goods, new complementarities and rivalries among goods can be constantly experimented with and variants invented. And often small, marginal variations can yield great changes. For example, to produce a novel outfit or add originality and uniqueness, it suffices to recombine its elements slightly. As Lemire (1991a) notices, in eighteenth-century England the basic construction of men’s and women’s clothing was standard, but the number of components very large. Playing on, and varying, these components – shirts, handkerchiefs, stockings and, for women, aprons, caps and petticoats – all mainly made of cotton material, created the whole difference and added novelty even to old apparel. Such minutiae (ibid.: 166) were the substance of change and fashion.10 Finally, the story of calico evidentiates something that remains obscure in more recent but otherwise similar patterns of innovation in consumption: the determinative role of consumers in selecting the new set of goods, and discovering and improving on their combinatory uses. Not only did consumers privilege the fanciful Indian patterns, despite early merchants’ attempts to confine them to more common and recognizable designs, but their desire for calicoes resisted import and use prohibitions and the risk of ‘calico chasing’, of being caught in the street and stripped of one’s calico. As we have seen, cunning ways to circumvent the law and re-utilize old pieces flourished. The persisting unsatisfied demand provided the stimulus for the technology of the new domestic production to appear.

MORE EXAMPLES The calico example is not an isolated one in seventeenth- and eighteenth-century Europe. It was repeated in similar fashion in the case of Wedgwood pottery, with its new glazes and rich variety of patterns; in the case of the new exotic flowers, such as tulips, with their novel shapes and unexpected colours; in the case of kitchen and pleasure gardens filled with exotic plants and statues (Mukerji and Schudsor 1991). In all these instances we find the same crossfunctional multiplication of use possibilities, and the discovery of new combinatory properties among goods.11 Again, small changes produced big ones. The larger availability of home furniture such as beds, chairs, tables and chests completely changed the interior space of domestic structures, allowing for different functions and locations of rooms.12 Hot, sweet, energizing tea transformed the cold meal of an English labourer much more than did beer (see Mintz 1993: 266).



The array of new consumer goods was seemingly unending in eighteenth-century London and touched every aspect of daily life. From cane chairs to fountain pens, from trunks to toys,13 to laces, buckles, buttons, lamps, watches, glasses, china, wigs and hair arrangements, the commodities offered in the shops, now sapiently lit after sundown, were constantly new and surprising, mixing all the novelties that could be found in art and nature.14 How is one to explain this proliferation of new goods, their surprising variety and the rate at which they appeared on the market? Economic historians, cultural historians, anthropologists and demographers, in the last ten years, have started to reverse the traditional approach which favoured a production-led industrial revolution; more attention is now paid to an analysis of consumer demand patterns. This change of focus, its difficulties notwithstanding, has allowed historians to discover the enormous potential for economic understanding hidden in the analysis of consumption structures. The economic historian Jan de Vries (1993: 107) has coined the term ‘industrious revolution’ to capture the change in habits of work and leisure which enabled consumer tastes and demand to flourish.15 What starts to be recognized in this reversal of focus is the role of the consumer as an autonomous economic force, present from the beginning of industrialization. Still the question how to explain the unending chain of new goods which seemed to find consumers eager to buy and experiment, to be surprised and even deceived, remains difficult to answer within the received economic paradigm. The temptation to resort to some sort of exogenous explanation such as fashion and emulation, manipulation and advertising, all first symptoms of consumerism, is, with some exceptions, always present.16 The consumer, if rescued from oblivion, is still viewed as an other-directed agent. Do we find any answer in economic theory? Few economists have addressed the problem positively. But the ones who have done so have provided intriguing contributions, though they have yet to find much following in the profession.17

INNOVATIVE CHOICE IN ECONOMICS: THREE STEPS TOWARDS A RICHER ANALYSIS Lancaster’s model Kelvin Lancaster (1971, 1991), in formulating his new theory of consumer choice, introduced important modifications of traditional theory which allowed him to address two difficult logical problems. The first is that, since consumer rationality corresponds uniquely to the coherence between given means and preferred ends, and since economic theory deals only



with rationality so defined, nothing can be said about consumer behaviour when new goods appear. No guidance is provided as to how the consumer should reorganize his or her order of preferences in the new goods space, nor as to how the new ordering is related to the old. All the information embedded in the old preference map is lost with the advent of the new one. The second problem is that the traditional consumer, despite consumer sovereignty, never contributes actively to the production of his or her consumption goods, never seems to add to but only to subtract. Lancaster, by contrast, sees the consumer as an agent who, not differently from the firm, actively transforms inputs – goods bought in the market – into outputs. ‘Outputs’, for him, are bundles of characteristics, that is, all the distinctive, objective properties which distinguish one good from another. In the case of food, for example, calories, flavour, taste, nutritional value, and also the social dimension of eating, all represent distinctive properties. Goods having being decomposed into their characteristics, the consumer is assumed to have preferences over these characteristics rather than over the goods themselves.18 Within this framework, Lancaster can explain the consequences for consumer choice of the introduction of a new good. A new good can be thought of as a good which has the same characteristics as the old good(s) to which it is related, but in different proportions. Given the consumption technology and the characteristics associated with the new good, if its price is not so high as to be dominated by some (convex) combination of the old goods, the efficiency frontier will expand, favouring those combinations which include the new good. In Lancaster’s model then, the consumer is not only a productive agent who actively recombines goods in order to produce characteristics, but is also able to produce and gain from efficiency improvements over the existing set of characteristics when new goods appear. The model provides a first answer to the problem of the choice of new goods. These are adopted because they represent an efficiency gain, because they enable the consumer to produce, within the existing budget constraint, more of the desired characteristics. But this analysis stops too soon. First of all, the idea of considering new goods merely as an efficiency improvement over existing characteristics is too restrictive. Though the case of product differentiation may be treated in this way (here in fact the number of characteristics is less than the number of goods), most new goods are new just because they involve the appearance of previously unthought of characteristics. As we have seen in the case of Indian painted calico, the pictorial freedom which could be transferred on to the new light material, with its glazed finish and exotic resonance, was a completely new attribute of the good ‘fabric’. To cite another instance, the product ‘education’ which was offered by myriad new boarding schools in eighteenth-century England, introduced ‘characteristics’ in the form of teaching materials, such as Newtonian physics and accounting, which were new, as were the books that explained and diffused them among children and adults. Moreover, as in the cases discussed



earlier, these new goods activated a whole cascade of complementary changes. The new education system both reflected and stimulated new attitudes towards children and a taste for knowledge (public lectures, scientific demonstrations and entertainments came into vogue, as Plumb relates, 1982: 328). New goods, even if they seem to represent only an efficiency improvement over existing characteristics, are yet likely to generate a whole chain of new properties and unthought of sets of novel uses among both existing and new goods.19 The problem of consumer behaviour in the presence of new goods is not solved by Lancaster, then, but simply shifted on to the appearance of new characteristics and their new ordering. We end up facing the same difficulty as the one encountered in the traditional paradigm of choice: how to understand the consumer’s decision processes over characteristics that are new to him or her, and what is their active role in the process. These considerations have two consequences. First, if new goods involve the appearance of characteristics which did not exist before, these new attributes may well represent the dominant feature of the good and render it unique, non-substitutable. It may constitute an irreplaceable difference for the consumer to acquire them. The demand for new goods, then, is not simply derived from different combinations of existing characteristics. Correspondingly, because genuinely novel, the price of new goods may increase beyond that level which would be dominated by the existing efficiency frontier, an efficiency loss for Lancaster. The second, more important, consequence relates to the role of the consumer. In the model just described, the ability of the consumer is to transform goods into the desired combination of characteristics. Still, if the characteristics are new, not known in advance, if they are difficult to assess because multifunctional, and if the goods are new and not decomposable, then how to produce them must also be a matter of discovery. The potential of their gainful opportunities is not openly displayed but must be looked for, noticed, recognized. The consumer, besides having the technical abilities of a producer, must thus possess innovative entrepreneurial abilities in order to transform uncertain events into gains.20 If new goods are adopted because of the gainful opportunities one expects they will provide – they expand the feasibility set – still efficiency gains like these are the result of a process of discovery. Experimentation and learning, trial and error, are the ineluctable accompanying processes of choice.

Earl’s model If in Lancaster the problem of novelty in consumption, and the uncertainty and unpredictability related to it, is ingeniously sidestepped by reducing new goods to different combinations of



known characteristics, this is not the case in those models of bounded rationality which try to deal exactly with the behavioural feedback activated by the presence of uncertainty. In the face of uncertain events – which appear in the form of both a scarcity and an excess of information – deliberative agents respond with rules of choice that are only boundedly rational, but flexible enough to be revised and adjusted as experimentation and trials require (Heiner 1983; Simon 1988). Applied to the theory of the firm, these models have yielded very fruitful results for the understanding of the role of firm organization (Egidi 1992; Egidi and Marris 1992). The same method could be applied to the consumer, exploiting Lancaster’s (and Becker’s) insight of integrating production and consumption theories. A very good example in this direction is Peter Earl’s pioneering analysis of the consumer (1986). In an uncertain, turbulent world, consumers’ choices need to be strategic, both in the sense that they have to incorporate surprise and anticipate the unexpected and in the sense that they have to cope with the inevitable interdependencies that exist among choices. Confronted with such complexity, consumers devise sets of patterned responses within which, in a way not dissimilar from the firm, they might be able to yield choices that are flexible and adaptable to uncertain events. The choice of money safety margins, of diversification of commitment and the exploitation of synergies among activities, of internalization of risk (do-it-yourself), of modular products and mobile furniture, are all examples of uncertaintylimiting rules. Yet choices remain complex. To start analysing them, Earl uses Kelly’s psychological theory of personal constructs (Kelly 1955) in an original way. Within this theory ordinary people engage, as do scientists, in translating things and events from the environment into personal images and constructs as a way of gaining predictability and control. Choices among alternative options can be understood only within this larger personal framework which guides people to organize their ideas and construct mental replicas of anticipated events, and to devise repertoires in order to compare and contrast alternatives. Choices represent the tests of these individual constructions of reality, to the end of constantly refining and changing the power to predict and control. Earl’s effort to go beyond the image of orthodox consumer theory, venturing into the unsafe waters of psychology, is really notable and, though mainly unnoticed by economists, his insights into the multiple strategies used by consumers to cope with uncertainties and surprise are new. His idea of the consumer as a scientist is also worth pursuing. A limitation I wish nonetheless to note arises from my concerns here. The emphasis Earl places on the turbulence of the modern world leads him to represent the consumer’s strategic behaviour as comprising mainly defensive, protective moves. Thus, although in his model consumers evolve through time and therefore change and enlarge upon repertoires of patterned responses,



still they seem never actively to engage in making change, in modifying the set of available options. Strategies arise to cope with surprise when it appears, never to cause it to appear. From my perspective, Earl does not press the analogy with the firm hard enough to incorporate into consumer decisions the ability to innovate and produce new values. Without this, however, how can we explain the fact that novelty is not only avoided but also looked for?

Scitovsky’s model ‘What does an organism do when all its needs are satisfied, all its discomforts eliminated?’ One’s likely immediate answer, ‘nothing’, is now generally recognized by psychologists to be wrong, says economist Tibor Scitovsky (1992: 31). A situation of perfect rest and comfort, when protracted, is in fact, with its absence of stimuli, unpleasant and boring and the organism will try to break out of it. Scitovsky takes this insight as his own starting point for a theory of consumer behaviour which stresses the pursuit of novelty as one of the fundamental drives. ‘First love, the first taste of some special food . . . together with many other firsts’, he says, ‘are among our most cherished memories’ (1992: 58). Novelty represents a major source of human satisfaction. But why is novelty so pleasurable? Why do we continue to seek pleasure even beyond the satisfaction of our needs? Scitovsky’s answer relies strongly on the findings of a body of literature in experimental psychology and neurophysiology, and in particular on the work of Berlyne (1960, 1971) who is the first to have systematically explored this question.21 For Berlyne, novelty together with variety, complexity and surprise, are the different components of the stimulus or, as he calls it, the arousal potential of a specific situation or experience. They are responsible for changes in the level of arousal which are linked with the sensations of pleasure. However, novelty and the other stimulus properties are pleasant only within bounds. In particular, when stimulus is too low it is experienced as boring, when too high as bewildering. An inverted U-shaped curve, the so-called Wundt curve (1874), describes this relation: with an increase of novelty and surprisingness pleasure increases up to a point, then decreases, replaced by displeasure. By acting on the stimulus properties of a situation, pleasure can be generated in two ways: through a moderate increase in arousal when arousal is too low (arousal boost mechanism), or through a decrease in arousal when this is uncomfortably high (arousal reduction mechanism). Scitovsky documents and develops these hypotheses with a variety of examples which open up a completely new way of looking at consumers’ behaviour in their market and non-market relations. To include the dimension of novelty and stimulus seeking in the consumer’s set of



experiences changes, for example, the way search and exploratory behaviour is represented since these can be pleasant. It also sheds light on the problems of work–leisure preferences, on changes in preferences, and on the way welfare is measured. Two points in this analysis are of particular relevance here. One is that this model of what we might call ‘bounded novelty’ influences the way people select and process information and knowledge, and more generally the way they evaluate their experiences. Different mixtures of redundancy and subjective novelty, Scitovsky rightly emphasizes, by providing stronger or milder degrees of stimulation, determine the priorities and ranking of our information set. This explains why, for example, anything new or unexpected becomes the centre of our attention, rendering the familiar at once unimportant. This insight is what has led us, in commenting on Lancaster, to insist that novelty can become the focal point of consumer interest, something ‘unique’ that cannot be further decomposed into similar goods. The second point is that consumption, in order to be a source of pleasure, must be what Scitovsky calls ‘skilled consumption’ (1992: 225). A new activity, as well as a novel good, just because novel and often uncertain and complex, can be enjoyable only if recognizable, if some of its potentiality is understood, if tastes have developed and adapted in order to appreciate it. In this process learning and the acquisition of experience, as well as knowledge, culture and society, play important roles.22 For Scitovsky, however, the path to pleasure is not an easy one. He envisages the existence of a difficult-to-reconcile conflict between activities that aim at comfort and those which aim at pleasure. Comfort, for Scitovsky, is linked to the avoidance of pain and discomfort but discomfort and its relief (arousal reduction mechanism, in Berlyne’s terminology) is also a precondition of pleasure. Therefore, all those activities which tend to increase comfort have as well the result of diminishing pleasure. Since, for Scitovsky, our society abounds exactly in these discomfort relieving goods and activities, he sees little room for redressing the balance in favour of pleasure at the cost of some sacrifice of comfort (1992: 71ff.). This asymmetry that Scitovsky suggests exists between comfort and pleasure – to which corresponds a difference between harm-preventing, defensive products, and creative products which produce positive pleasure – is not entirely convincing (as the last section of Peter Earl’s chapter in this volume confirms). At least it is not convincing in terms of the Berlyne model on which Scitovsky mainly relies for his evidence. To reduce the stimulus potential – to look for comfort – can be as gratifying and pleasurable as to increase it, and requires the same amount of ingenuity and skills. What this model shows unequivocally, however, is that trying to remain close to the optimal level of arousal by repeating the strategies that brought one into its vicinity is simply self-defeating. The attainment of an equilibrium of satisfaction, or the repetition of satisfaction-



inducing activities, threaten the very pleasure that caused us to reach for them in the first place. Novelty seeking is a never-ending process.

CHOOSING NOVELTY What consequences can be drawn from this discussion? In the traditional consumer theory goods are compared and ranked, but they have been already preselected, according to a process of preference formation that is left unexamined. In the potentially infinite set of available goods, the consumer, when choosing, has already singled out the space of goods (or of characteristics, in the case of Lancaster), be it music or drugs, milk or drawings. But we have seen that the original selection of the choice set (the space of goods) is not without consequences for the value, uses and characteristics that can be produced with these goods, depending on their novelty potential and combinatory properties. What can be said of this process of selection and identification of the goods which become objects of choice? On the basis of the previous examples we can attempt to specify two elements in this process. One is recognizability. No matter how new are the objects of choice, they must carry some resemblance and analogy with the already known and experienced objects of choice. Indian calicoes were new, but fitted easily at first within the pattern of uses attached to other fabrics with their variety of colours and decorations. Path-dependent choices and binding rules stressed by the models of bounded rationality, Scitovsky’s notion of skilled consumption, and even Lancaster’s old preference mapping which helps identify the efficiency of new goods, all evidentiate this important point and suggest some of the procedures that restrict the set of the otherwise limitless available goods. Recognizability therefore expresses the given framework of rules. Its task is: first, to provide the selecting rules which delimit the space of goods (aesthetic, utilitarian or ethical, for example); second, to help in establishing the links and relations which exist among goods. In all our examples we have seen that goods do not travel alone, they are always part of an intricate network of relations with other goods. To choose a good is never to choose it on the basis of its characteristics alone, and therefore of its solitary marginal utility. To choose it instead implies also the recognition of the set of relations which exist among goods. This is why, before a good becomes a good and an object of choice, it has also to be framed as part of a set, as part of a network. The second element is much less stressed. It represents the potential for improvement, change and novelty that a specific set selection determines. From the previous discussion it appears that novelty has a double role in consumption. In the first place it is linked to



efficiency. A newly discovered good, characteristic or combination which has the power to improve goods’ functional properties represents an efficiency gain for the consumer. Because of the multiplicity of use possibilities and the complex of network interrelations which exist among goods, these efficiency gains, these ‘profit’ opportunities for the consumer, are never, nor can they be, fully known in advance. Search processes of updating information can surely reduce gaps and deficiencies, but do not eliminate them. Much depends on the consumer’s capabilities and alertness to discover the goods’ potentialities and connections through use. As Kirzner (1989) puts it, incomes are always ‘discovered’ incomes.23 Exploiting the novelty potential of a good contains moreover an additional incentive, besides that of gains in efficiency. The discovery process which generates novelty is enjoyable in itself. As the discussion of Scitovsky’s ideas has shown, novelty, change and variety – within limits, within recognizable boundaries – are pleasant. In the case of calicoes, there was a recognition that these were a suitable and superior form of clothing, but it was their flexible uses and mutable combinations that caused them to be selected as a new and permanent part of the choice set. They were generative of a ‘utility’ that could be varied through time by means of exploring an apparently never ending potential for innovation and change. Efficiency, on the one hand, and pleasure gains, on the other, are therefore dual incentives towards novelty seeking behaviour. This, together with recognizability, provides a framework for understanding how goods are selected and chosen. The differences with the traditional framework stand out very sharply if we reframe what has been said with the help of some simple tools of standard consumer choice. Thus, let S of Figure 4.1 be the set of the goods which fall within the consumer experience, and s ∈ be the chosen subset. Suppose, further, that the segment AB of Figure 4.2 represents the efficiency frontier of the chosen subset. This, in the usual way, describes the maximum outcome obtainable with given resources, that is, when all gainful opportunities have been exploited. But suppose now that the subset s contains the possibility of innovation. Innovation might be the result of a new good or characteristic which alters the existing combinations, and allows the consumer to discover better ones within the given resources (example: a CD-Rom version of a well-known gallery of paintings as compared with a traditional printed catalogue). Or, allow the consumer to be the initiator of profitable gains through a more productive combination of existing goods or characteristics (examples: a reordering of a filing system, trying a new recipe). As a result of such innovation imagine the efficiency set expands and reaches point C in the figure (we are assuming that innovation is ‘local’, as in Stiglitz 1987; Stiglitz and Atkinson 1969: learning is localized and does not require investment in a whole new consumption technology).



Figure 4.1 Selecting the choice set.

Figure 4.2 Local efficiency gains.



Figure 4.3 Local utility gains.

We have seen, however, that efficiency gains are not the only dimension of novelty seeking behaviour. Novelty is pleasurable and therefore can be thought of also as an argument in the consumer’s utility function. Let us assume that the utility function has the traditional form U=u(x), with U’>0 and U”1, where δ denotes the income elasticity of demand for time-intensive goods), this is probably incorrect. The increase in ‘leisure’ over the past one hundred years has to be interpreted as a consequence of this high income elasticity of demand for earnings intensive commodities (Becker 1965: 505ff, 517). This will have to be explored as a possible explanation of trends in time use common to OECD countries before the 1970s. The most challenging episode, however, is the present reversal of the secular trend towards shorter working hours (both in the USA and UK – cf. Schor 1991; Mulgan and Wilkinson 1995). While incomes have continued to grow – albeit at reduced rates – since the 1970s, annual working hours (of some groups) have increased. In the traditional framework, we would simply predict a somewhat slower trend towards even shorter hours. The Becker model, however, would argue that the present turnaround may be due to a similar set of factors as the rapid increase of working hours in the eighteenth century – slow productivity growth combined with a wave of new consumer durables, as well as an increasingly efficient retail sector. Cross-country comparisons will be particularly useful since neither the liberalization of the retail sector nor the productivity slowdown since 1973 affected all the developed nations to the same extent. What is clearly necessary is a new method for quantifying the productivity of consumption. Yet even without precise measurements of the productivity of consumption, relative rates of change compared with the productivity of production can sometimes be ascertained. As this chapter has sought to demonstrate, such an approach allows new insights into the interaction of consumer preferences, labour supply decisions, the consumption of consumer durables and the demand for leisure.

ACKNOWLEDGEMENTS This chapter is based on my doctoral research. I wish to thank my Oxford supervisors, John Landers, Avner Offer and Richard Smith, for their encouragement, support and useful criticisms. Albert Carreras, Liam Brunt, Paul A. David, Charles H. Feinstein, Jay Gershuny,



Timothy C. Leunig and John Styles helped at various stages. Seminar audiences at All Souls (Oxford), Corpus Christi (Cambridge), Istituto Universitario Europeo (Florence), Max-PlanckInstitut für Geschichtswissenschaft (Göttingen), the 1995 Cliometrics Conference (Kansas), Nuffield College (Oxford), the 1996 Social History Conference (Glasgow) and the 1996 Economic History Society Conference (Lancaster) asked penetrating questions. Financial support by Nuffield College ESRC, Sir John Hicks Fund, Clare College and DAAD is gratefully acknowledged. I am to blame for any remaining errors.

NOTES 1. Again, the difference is not statistically significant. Sample size varies between 34 and 48. 2. It should be remembered that, with a logit model, we test the difference of the coefficient from unity (since e to the power of one is zero), and not from zero. 3. Comparisons with Easter and Christmas are not possible for the second sample. For each period, I collected approximately 1,000 court cases; this required analysing the Old Bailey Sessions Papers from 1749–63 and from 1799– 1803. A larger volume of cases handled by the court sharply reduced the number of years represented in the sample. 4. A value of £20,000 for bankers is not included in the sample – the use of ‘capital’ is very different in this case. 5. The assumption here is, of course, that national trends are indicative of developments in London. In the absence of evidence to the contrary, it seems sensible to proceed on this assumption. 6. It should be remembered that such a long-term average is influenced by the exceptionally high prices in 1801. Some periods, such as the 1770s, showed a higher level of per capita spending than 1801. Yet even if we use 1770 as a benchmark, growth would have been a mere 13.5 per cent, or 0.18 per cent p.a. (Crafts 1985: 95). 7. Assume that, of 106 units of income in 1801, 95 are devoted to non-durables. Then there will be 11 units of income left for spending on goods. In 1700, there would be have been only 5 units available for this purpose. 8. As McKendrick has noted, Wedgwood often enhanced his brand name by charging more than marginal cost, thus exploiting a ‘Veblen effect’ (for some goods, demand increases with prices because the real commodity on offer is social exclusivity). In most cases, however, innovations in advertising and distribution widened the market. One could even go so far as to see both strategies as complementary – using Veblen pricing was necessary because the ability to buy pottery no longer distinguished one from those on the lower rungs of the social ladder. 9. Much improved cost-of-living indices (Lindert and Williamson 1985; Feinstein 1994) are only available from 1780 onwards and thus unsuited to our task. 10. The effect constitutes a lower bound since the Tucker series shows the smallest decline of all price


WORK AND CONSUMPTION series whenever comparable data are available. The Tucker index from the nearest year for which data are available was used as a regressor for the observations.

BIBLIOGRAPHY Primary sources Burnett, D. (1989) The Autobiography of the Working Class, New York: Harvester Press. Campbell, R. (1747) The London Tradesman, London: T. Gardiner. Fielding, H. (1751) Enquiry into the Causes of the Late Increase in Robbers, London: A. Millar. Marx, K. (1967 [1867]) Capital, vol. III, London: International Publishers. —— (1983 [1867]) Das Kapital, vol. I, London: Dietz. Smith, A. (1991 [1776]) An Enquiry into the Causes of the Wealth of Nations, Edinburgh: Campbell. The Whole Proceedings of the Sessions of the Peace, and Oyer and Terminer for the City of London and County of Middlesex, London.

Secondary sources Ashton, T.S. (1948) The Industrial Revolution 1760–1830, London: Oxford University Press. Becker, G.S. (1965) ‘A theory of the allocation of time’, Economic Journal 75: 493–517. —— (1976) The Economic Approach to Human Behaviour, Chicago: Chicago University Press. Berg, M. and Hudson, P. (1992) ‘Rehabilitating the industrial revolution’, Economic History Review 45: 24–50 Bienefeld, M.A. (1972) Working Hours in British Industry: An Economic History, London: Weidenfeld and Nicholson. Botham, F. and Hunt, E. (1987) ‘Wages in Britain during the industrial revolution’, Economic History Review 40: 380–99. Brewer, J. and Porter, R. (eds.) (1993) Consumption and the World of Goods, London: Routledge. Clark, G., Huberman, M. and Lindert, P. (1995) ‘A British food puzzle, 1770– 1850’, Economic History Review 48: 215–37 Crafts, N.F.R. (1984) ‘Patterns of Development in Nineteenth Century Europe’, Oxford Economic Papers 36: 438–58. Crafts, N.F.R. (1985) British Economic Growth during the Industrial Revolution, Oxford: Oxford University Press. Crafts, N.F.R. and Harley, C. (1992) ‘Output growth and the industrial revolution’, Economic History Review 45: 703–30. Crouzet, F. (ed.) (1972) Capital Formation in the Industrial Revolution, London: Methuen. Cuenca Esteban, J. (1994) ‘British textile prices, 1770–1831: are British growth rates worth revising once


WORK AND CONSUMPTION more?’, Economic History Review 47: 66–105. —— (1995) ‘Further evidence of falling prices of cotton cloth, 1768–1816’, Economic History Review 48: 145–50. Deane, P. and Cole, W. (1969) British Economic Growth 1688–1959, 2nd edn, Cambridge: Cambridge University Press. Demaris, A. (1992) Logit Modeling. Practical Applications, Newbury Park: Sage. De Vries, J. (1993) ‘Between purchasing power and the world of goods: understanding the household economy in early modern Europe’, in R. Porter and J. Brewer (eds). Consumption and the World of Goods, London: Routledge. —— (1994) ‘The industrial revolution and the industrious revolution’, Journal of Economic History 54: 249–70. Earle, P. (1989) The Making of the English Middle Class: Business, Society and Family Life in London, 1660– 1730, Berkeley: Methuen. Feinstein, C.H. (1994) ‘Changes in nominal wages, the cost of living and real wages in the United Kingdom over two centuries, 1770–1990’, unpublished manuscript, All Souls College, Oxford. Fine, B. and Leopold, E. (1993) The World of Consumption, London: Routledge. Fogel, R. (1993) ‘New sources and new techniques for the study of secular trends in nutritional status, mortality, and the process of aging’, Historical Methods 26: 5–43. Freudenberger, H. and Cummins, G. (1976) ‘Health, work and leisure before the industrial revolution’, Explorations in Economic History 13: 1–12 Gell, A. (1992) The Anthropology of Time, Oxford: Berg. Gilboy, E. (1967 [1932]) ‘Demand as a factor in the industrial revolution’, in A.H. Cole (ed.) Facts and Factors in Economic History, London (reprinted in R. Hartwell, ed., The Causes of the Industrial Revolution in England, London): Methuen. Hardy, M. (1993) Regression with Dummy Variables, Newbury Park: Sage. Harley, C. (1993) ‘Reassessing the industrial revolution: a macro view’, in J. Mokyr (ed.) The British Industrial Revolution: An Economic Perspective, Boulder: Westview Press. Harley, K. and Crafts, N. (1995) ‘Cotton textiles and industrial output growth during the industrial revolution’, Economic History Review 48: 134–44. Harris, M. (1984) ‘Introduction’, in The Old Bailey Proceedings. Parts one and two. A Listing and Guide to the Harvester Microfilm Collection, 1714–1834, Brighton: Harvester. Hauck, W. and Donner, A. (1977) ‘Wald’s test as applied to hypotheses in logit Analysis’, Journal of the American Statistical Association 72. Hecht, J. (1956) The Domestic Servant Class in Eighteenth Century England, London: Routledge Hunt, E.H. (1986) ‘Industrialization and regional inequality: wages in Britain, 1760–1914’, Journal of Economic History 46: 935–66. Jones, E. (1973) ‘The fashion manipulators: consumer tastes and British industries 1660–1800’, in L. Cain, P. Uselding (eds.) Business Enterprise and Economic Change, London: Kent State University Press. Juster, F. T (1985) ‘A note on recent changes in time-use’, in F. Juster and F. Stafford (eds.) Time, Goods,


WORK AND CONSUMPTION and Well-Being, Ann Arbor: Survey Research Centre, University of Michigan. Juster, F. T. and Stafford, F. (1985) ‘Introduction and overview’, in F. Juster and F. Stafford (eds.) Time, Goods, and Well-Being, Ann Arbor: Survey Research Centre, University of Michigan. King, P. (1996) ‘Pauper inventories and the material lives of the poor in the eighteenth and early nineteenth centuries’, in T. Hitchcock, P. King and P. Sharpe (eds.) Words of the Poor, London. Kusamitsu, T. (1991) ‘Novelty, give us novelty’: London agents and northern manufacturers’, in M. Berg (ed.), Markets and Manufacture in Early Modern Europe, London: Routledge. Lemire, B. (1991) Fashion’s Favourite: The Cotton Trade and the Consumer in Britain, 1660–1800, Oxford: Oxford University Press. Lindert, P. and Williamson, J. (1985) ‘English workers’ real wages: reply’, Journal of Economic History 45: 145–53. McKendrick, N. (1961) ‘Josiah Wedgwood and factory discipline’, Historical Journal 4: 30–55 —— (1974) ‘Home demand and economic growth: a new view of the role of women and children in the industrial revolution’, in N. McKendrick (ed.) Historical Perspectives: Essays in English Thought and Society, London: Europa Press. —— (1982) ‘Commercialization and the economy’, in N. McKendrick, J. Brewer and J. Plumb The Birth of Consumer Society. The Commercialization of Eighteenth-Century England, London: Europa Press. McKendrick, N., Brewer, J. and Plumb, J. (1982) The Birth of Consumer Society. The Commercialization of Eighteenth-Century England, London: Europa Press. Mathias, P. (1979) ‘Leisure and wages in theory and practice’, in P. Mathias (ed.), The Transformation of England, London: Columbia University Press. Mokyr, J. (1977) ‘Demand vs. supply in the industrial revolution’, Journal of Economic History 37: 981– 1008. Mulgan, G. and Wilkinson, H. (1995) ‘Well-being and time’, Demos Quarterly 5. Phelps Brown, E.H. and Hopkins, S.V. (1955) ‘Seven centuries of the prices of consumables, compared with builders’ wage-rates’, Economica 40. Reid, D. (1976) ‘The Decline of St. Monday, 1776–1876’, Past and Present 71: 76–101. Rule, J. (1981) The Experience of Labour in Eighteenth Century Industry, London: St. Martin’s. —— (1986) The Labouring Classes in Early Industrial England 1750–1850, London and New York: Longman. Schor, J. (1991) The Overworked American. The Unexpected Decline of Leisure, New York: Basic Books. Schwarz, L. D. (1985) ‘The standard of living in the long run: London, 1700–1860’, Economic History Review 38: 24–41. Shammas, C. (1983) ‘Food expenditures and economic well-being in early modern England’, Journal of Economic History 43: 89–100. Stone, L. (1994) ‘Leisure in eighteenth century England’, paper presented to the Datini Conference Il Tempo Libero, Prato. Styles, J. (1993) ‘Manufacturing, consumption and design in eighteenth-century England’, in J. Brewer and R. Porter Consumption and the World of Goods, London: Routledge. Thirsk, J. (1978) Economic Policies and Projects: The Development of a Consumer Society on Early Modern England, Oxford: Oxford University Press.


WORK AND CONSUMPTION Thomas, B. (1985) ‘Food supply in the United Kingdom during the industrial revolution’, in J. Mokyr The Economics of the Industrial Revolution, London: Rowan and Allanheld. Thompson, E.P. (1967) ‘Time, work-discipline, and industrial capitalism’, Past and Present 38: 56–97. Tucker, R.S. (1936) ‘Real wages of artisans in London’, Journal of the American Statistical Association 31: 21–35. Voth, H.-J. (1996) Time-Use in Eighteenth Century London: Some Evidence from the Old Bailey, unpublished D. Phil manuscript, Nuffield College, Oxford. Weatherill, L. (1988) Consumer Behaviour and Material Culture, 1660–1760, London: Routledge. Wrigley, E.A. and Schofield R. (1981) The Population History of England 1541– 1871, London: Edward Arnold. Wrigley, E.A. (1987) ‘Urban growth and agricultural change: England and the continent in the early modern period’, in E.A. Wrigley (ed.) People, Cities and Wealth, Oxford: Blackwell.


9 SILK PURSES OUT OF SOWS’ EARS Mass rarefaction of consumption and the emerging consumer–collector*

Guido Guerzoni and Gabriele Troilo

INTRODUCTION One of the phenomena which most permeates the economic environment of industrialized countries is, without doubt, that of complexity (D’Aveni 1994; Hamel and Prahalad 1994). Among the properties attributed to complexity, one is particularly relevant to our study: the simultaneous presence of contradictory events (Vicari 1994). The increasingly international dimension of cultural, economic and political phenomena coexists with highly localized events. Moreover, the increasing compression of time in productive phenomena (the reduction of the lifespan of products, the shortening of the time-to-market phase, etc.) is accompanied by a prolongation of the effects of firms’ strategic decisions (i.e. high investment in flexible production lines). We will focus on one of these contradictory phenomena, in order to analyse the simultaneous presence of two apparently contrasting consumption phenomena. The first of these, defined as ‘hypertech mass customization’ is based on the possibilities offered by an increasing number of innovative, sophisticated and progressive productive technologies. The second, ‘anti-tech mass customization’ concerns goods and services that are anachronistic, pre-modern, anti-technological – these are products whose production processes employ out-of-date, obsolete or disappearing technologies. Our main purposes are to discover whether it is possible to find a single reading key of the two phenomena mentioned above and to find out if this key makes it possible to propose an interpretative model of emerging consumption patterns.



CONTRADICTORY PHENOMENA IN THE ECONOMIC ENVIRONMENT: ‘ANOMALOUS CONSUMPTION’ In December 1994 Levi’s announced that in selected outlets clients could supply information regarding their favourite colour, size and other features and receive a personalized, custommade pair of jeans. The attention that this proposal attracted focused on the company rather than the event itself, because Levi’s then manufactured one of the mass goods par excellence, a globally recognized symbol of aggregation and homogenization. While the possibility of personalizing a product no longer attracts particular attention, a custom-made pair of jeans is an index of further changes. Several companies offer this kind of service in different sectors. From Japan, the National Panasonic Bicycle markets over ten million models of bicycles that are modified according to the preferences and individual measures of its clients. From Italy, Natuzzi, a leader in the sector of leather sofas, offers a million different combinations of its products. Hairdressers can give their clients the chance to see the final result of a new haircut on a computer screen before receiving a haircut. In many perfume shops it is possible to have a rapid skin test in order to choose the most suitable cosmetic. These are very common examples of a phenomenon which is going to modify the traditional purchase and consumption habits of western consumers. Several authors believe that advanced economies are experiencing a change in the prevailing technological paradigm. This change can be characterized as a move away from mass production towards flexible and automated production (Davis 1987; Cozzi et al. 1988). The former characterized industrial development from the time of the Industrial Revolution until the early 1970s. From the point of view of the company, mass production entails manufacturing standardized goods at moderate prices. The regulating principle of production is efficiency, obtained thanks to vast plants which make economies of scale possible. Given the inelasticity of these plants, the consumer is forced to accept a low price but standardized product. The role of consumers is therefore passive. If they want a product that meets all their requirements they must look for it outside the industrial circuit, in the handmade world – but they have to be prepared to pay larger sums. From the oil crises of the 1970s onwards, the emergence of the new information-based technologies (computers, telecommunications), resulted in considerable changes. The flexibility of plants relying on the new technologies allows consumers to overcome the ‘either/or’ logic of standard goods or high prices – and to shift to a ‘both/and’ pattern, in which they can obtain personalized products at moderate prices. The mass production era is turning into the mass customization era (Davis 1987; Pine 1992). The regulating principle in this case is not



efficiency but effectiveness, that is to say the production of goods and services which meet consumers’ requirements. The most interesting factor in this process is that the role of consumers changes radically. In the era of automated and flexible production they become active because they take part in the process of designing and creating the product, by supplying information regarding their exact requirements. The former mass production paradigm was characterized by information asymmetry in the relation between company and consumers. This favours the stronger subject, the company. This relationship is more balanced in the case of flexible production: both subjects have their own competencies. Although in the cases cited above the personalization of goods and services relies on increasingly refined, innovative, and ‘progressive’ technologies, there also exist other cases which do not share these characteristics – and their number is increasing. More and more frequently, the preferences of consumers are directed towards goods that can be defined as anachronistic, pre-modern, and anti-economic both in the performances they offer and in their purchase prices (these are often considerably higher than more recent products). The production processes of these products are based on out-of-date, obsolete and sometimes nearly extinguished technologies. However, the contrast is only apparent. These cases in fact demonstrate a further evolution of the mass customization paradigm of the technological matrix – perhaps the first collective reaction to its diffusion. We can therefore sketch a taxonomic grid of the different category of goods subjected to the above mentioned phenomena. We have listed these under the heading ‘consumption of anomalous goods and services’. In the following they are classified into specific types based mainly on consumers’ interpretations of them:

jurassic goods;

revival goods;

handmade goods;

nostalgia goods;

antique and discarded goods.

Jurassic goods ‘Jurassic goods’ are either products which have not gone out of production in the last thirty to forty years, maintaining their characteristics unaltered, or products which have been brought back into production without any modification. This category includes motorcycle models such as Harley Davidson, Guzzi, Ducati, Piaggio, Lambretta, Triumph and BSA, that only five years ago were nearing extinction or had disappeared altogether. Examples in the car sector



include the Mini Morris, the Fiat 500, the Citroen Dyane, the Volkswagen Sedan, several models manufactured in the USA, as well as the DDR, Trabant and the Lada produced in Romania which have acquired cult status in western markets. Further examples from other sectors include Kelvinator refrigerators, Roberts radios, General Electric fans, small electrical appliances by Quick Mill, several Rolex Models, Mont Blanc and Parker fountain pens, some Ray Ban and Persol glasses, Ronson lighters, Converse shoes by All Star and the Adidas ‘Gazelle’ line. In the field of fashion and cosmetics, examples can be found in the boom in secondhand clothes and international phenomena such as Barbour or Husky LTD coats, not to mention ‘classics’ such as the Hermes ‘Kelly’, Chanel Number 5, or Gucci moccasins. These goods have retained their technical and aesthetic features unaltered for the past forty years, never bowing to the demands of ‘progress’ and ‘innovation’, never attempting to keep pace with time. Despite going through phases of acute crisis, when they seemed to be doomed to disappear, to bow to the rigorously evolutionary destiny of goods, these products managed to remain the same, unaltered, stepping through the traps of time unharmed and taking unexpected and remarkable revenges.

Revival goods ‘Jurassic goods’ should be distinguished from another, very similar group, that of ‘revival goods’. ‘Revival goods’ are products which make their first appearance on the market now, but which are clearly inspired by products of the past, from which they copy colours, lines and shapes. These goods combine a technologically hyper-advanced core with aesthetic features referring back to those of the past. Examples of ‘revival goods’ include Mazda’s Spider Mx-5, Wurlitzer and Rok-ola jukeboxes, certain models of Bosch refrigerators and some electrical appliances designed by Alessi and produced by Philips. Further examples can be found in the furniture, design and fashion sectors, where revivals are very common.

Handmade goods The 1970s and the 1980s were notable for a return to handmade products in various sectors. Examples of such products include the reintroduction of tailor-made garments or handmade shoes, the success of handmade products in the field of furnishing or interior design, the reappearance of materials such as copper and wood for kitchens, and developments in ceramics for vases, plates and kitchenware. It should be emphasized that this was a very widespread phenomenon that involved not only the wealthiest consumer groups, but also attracted a vast and heterogeneous public.



Nostalgia goods We have now to take into consideration less obvious, but not less important, cases of anachronistic, pre-modern goods. These are characterized by an explicit declaration of mistrust in the fascination and the advantages of progress, technology and innovation. We refer to them as ‘nostalgia goods’.1 Examples of nostalgia goods include the booms in biological foods, chemical-free products, natural fibres, homeopathic medicines and other forms of ‘natural’ remedies, and the popularity of farm holidays. Nostalgia goods can also be limited mostly to packaging, for example, soaps, shampoos, teas, cocoa, chocolate, biscuits and jams. All are excellent examples of forms of consumption whose satisfaction is derived from, or focused on, a return (real or imaginary) to a pre-industrial, pre-modern past. The nostalgic contention is that food was genuine, healthy and tasty, and products were comfortable and made with care and affection, sixty or one hundred years ago. Nostalgia goods are perceived as having optimal levels of quality. Technology, progress and innovation put an end to this idyllic golden age, bringing pollution, pesticides, preservatives, terrible allergies, and so on. Therefore, it is not surprising that the post-modern era of goods starts with a curious return to the past, an exaltation of the pre-modern and most of all of the pre-industrial.

Antique and discarded goods Further confirmation of the phenomenon is given by the huge growth of the sector concerning antique and modern goods. Markets, fairs, outlets and specialized magazines crop up everywhere, focusing on the purchase and sale of objects belonging to the remote or near past. It is important to note the diffusion of this phenomenon on a mass scale, the inclusion of these kinds of goods among the preferences of a steadily growing number of individuals. For the first time they see in the ‘old’, the ‘antique’ and the ‘discarded’, a potential ‘treasure’ which places the past above the future, tradition above innovation, and the hand of man above the machine. ‘Anti-tech goods’ nowadays enjoy an undeniable but puzzling success. Why should consumers prefer goods that are technologically obsolete, aesthetically old-fashioned, uncomfortable, noisy, energy wasting and offer low levels of performance? What is it that persuades those who consume these kinds of goods to prefer them to more recent products? What is it that makes them reject the possibilities offered by new technologies?



ON THE INEFFICACY OF TRADITIONAL ECONOMIC ANALYSIS In an attempt to provide answers to the questions raised above, the traditional analysis of consumer behaviour offers no structure on which to lean.2 This is because it has only recently begun to face the question of ‘anomalous’ consumption patterns. In marketing studies, the prevailing paradigm is that of information processing (Bettman 1979). These models interpret consumer behaviour as a response to determined external and internal stimuli. As in the case of microeconomic models, it is assumed that the individual is a rational problem solver who is able to handle and manage a considerable flow of information. Moreover, it is assumed that the consumer’s aim is to meet well-defined needs thanks to the usefulness of the features of the good. As in the modern microeconomic theory of consumer behaviour (Lancaster 1966), these models assume that the individual is able to choose between and evaluate the elements forming goods and services. The functionality of the product to the need is analysed mainly by reference to economic criteria, and the aim is to maximize usefulness. Given these premises, attention focuses mainly on the handling and processing of data that the individual gathers in the presence of particular stimuli (internal data are concerned with the consumer’s needs, and external data with company communications). The analysis concentrates on the purchase behaviour that results from the consideration of these data. Of the three main factors in the consumer purchase process, cognition, affect and behaviour, most emphasis is given to cognition. The choice, the final result of the process, is considered to be principally affected by the cognitive process and only indirectly by affect. Preferences are developed as a result of the rational process of data elaboration. Variables concerning affect play a secondary and derivative role. At the same time, behaviour is reduced to the act of purchase, with very little importance attached to the consumption process. There is, therefore, an overlapping with microeconomic studies. At the end of the process, consumers, whether satisfied or not, achieve their meta-goal, i.e. the gathering of further information to face future problems or needs. The assumptions on which this paradigm is based, however, restrict its application to certain types of consumer behaviour. In particular, how is it possible to explain the purchase of those highly symbolic goods which are difficult to section in attributes, whose purchase is not wholly justified by reference to economic utility? Further, how can we explain that, in these kinds of products, consumer need is driven by aesthetic satisfaction, rather than traditional product functionalism? Is it therefore possible to establish some common features that can bring together two apparently contradictory phenomena such as mass customization determined by technological progress and the growth in anomalous consumption patterns at odds with the beneficial effects of progress and technological innovation? Is it possible to



unite them under a single interpretative key? In our opinion, the answer is positive. The common factor is that the relationship between producer and consumer becomes individualized and is expressed through unique and/or rare, virtually irreproducible products, which have a noticeable aesthetic and semiotic content. The personalized Levi’s jeans mentioned in the first case are unique and irreproducible because they are tailor-made, while the Volkswagen Sedan is unique and irreproducible because it is no longer in production and is therefore becoming rarer and rarer. Both cases represent ‘rarefaction processes’3 of firm–consumer relationships. By the term ‘rarefaction process’, we do not mean that the relationships decrease numerically, but rather that standardization is reduced, because there is a shift from a system in which the firm has relationships with anonymous and unknown buyers, to one in which there are relationships with identifiable and known consumers.4 This process is intended to create a unique and unrepeatable relationship between firms and consumers, based on the production and consumption of unique and unrepeatable goods. Looking again at previous examples, the Levi’s jeans buyer supplies his or her requirements that are recorded on computer to design and create his or her product; while in the case of the Volkswagen Sedan the buyer gains admission to the ‘club’, the community of Sedan owners (often institutionalized). In these cases, therefore, the firm–consumer relationship is different from all the others and is unique or rare. In this sense, we can conceive the spread of anomalous consumption as an early, logical reaction to mass customization of technological origin. If technology can produce unique and rare goods through ‘high-tech’, what can further distinguish a consumer from the former kind of owners than the possession of rare and unique goods that deny or refuse technology itself? Rarefaction processes are ascribable both to production and consumption. In the first case, increased competitiveness forces firms to take advantage of the potential offered by new technologies in order to distinguish themselves from competitors. In the second case, certain social–cultural phenomena occurring in industrialized countries, such as raised income and education levels and the variety and variability of social roles, increase the need for differentiation, spurring consumers towards innovative and highly individual consumption models (Christopher 1989). Figure 9.1 represents the variables that determine the rarefaction processes. Hyper-tech and anti-tech personalization can be ascribed to rarefaction processes that lead towards a world of unique and rare goods. But both these phenomena have properties that are identical to those of collectable goods. So, we believe it is also possible to analyse some general consumption processes through a consideration of goods as collectable goods and consumers as collectors.5



Figure 9.1 Rarefaction processes in the firm–consumer relationships.

PROPERTIES OF COLLECTOR ITEMS: HEURISTIC POTENTIALS Collectable goods are par excellence unique or rare. Their functions bear least relation to consumer need. It is interesting to note their principal characteristics, from an economic point of view.

Uniqueness and rarity/distinguishing features The collectable good is unique or rare. Sometimes it is the nature of the production process that decrees its uniqueness, as in the case of handmade goods which are imperfect but unrepeatable. In other cases, it is time that selects and rarefies a set of goods which were originally identical or very similar (Rheims 1956, 1990, 1992; Eccles 1968; Alsop 1982; Pomian 1987, 1993; Schnapper 1988).6 Finally, the producer’s distribution strategy can affect the destiny of a good, directing it towards collections or to the anonymity of a hard discount. A typical example is that of the introduction of CDs, an event that transformed a significant proportion of vinyl recordings into collectable goods. In our standardized society, which is obsessed with the need to make distinctions, the charm of the ‘unique piece’ is irresistible (Baudrillard 1968). Possession of a unique product reveals the uniqueness of the owner. A ‘single good’ underlines the singularity of its fortunate possessor. The collector protects and elevates his individualism thanks to the possession of something that is possessed by no one else or, at best, by only very few. Such a possession guarantees distinction. Moreover, the unique good is usually authentic. Copies and fakes



threaten it, but less effectively than occurs with other types of goods, such as luxury goods. Collectable goods reduce or limit the risks of participating in the quotation game between originals and fakes.

Reduced or lost utilization values and functions When an object becomes collectable, it is removed from the world, and from the ‘vulgarity’ of circulation (Pomian 1987). It loses its original functions and values of utilization. This loss or suspension of function in keys that do not lock, wines that are not drunk, china sets or silverware that will never touch lips or food, and stamps that will never frank a letter is evidence of their owner’s resources and of his power to keep them unproductive, outside consumption. In the collector’s purchases there is no evidence of other needs than pleasure. The pretext of ‘investment’ conceals a powerful intention to show the world the collector’s ability both to remove goods from the world, and to return them to it, according to whim. The good is therefore reduced to a mere sign, a piece of symbolic capital (Bourdieu 1971).

Seriality The single collectable good cannot be separated from the ideal and real series it belongs to: ‘Whether denied, forgotten, destroyed or virtual, the series always remains operative. As much in the humblest of everyday objects as in the loftiest of rarities, it is the indispensable nourishment of ownership and the passionate game of possession’ (Baudrillard 1994: 15). The meaning and value of a collectable good issues from its membership of a group, a class, or a set. These categories represent the first referent of the series which the collector always tries to complete. Hence, the missing piece will always be more important than the last acquisition. The perennial tension directed towards the creation of a series that is endless and impossible to complete is the most fascinating feature of the collector and it leads to the prototype of the ‘serial consumer’. As Muensterberger (1994: 11) observes: ‘Whatever the motivation, there is little question that collection is much more than the simple experience of pleasure. If that was the case, one butterfly, or one painting, would be enough. Instead, repetition is mandatory.’

Aesthetics Collectable goods are loaded with aesthetic properties. In fact, they are the main conveyors of aesthetic projects, first individual and later collective. The processes described above remove the product’s original utilization functions and values and replace them with aesthetic prop-



erties – shapes, colours, sounds and scents. These kinds of goods owe their survival in time to their aesthetic properties. A collector of fountain pens does not value them principally as writing instruments, and will usually not use them to write. Instead, he discusses the materials from which they have been made, the quality of their design, and their subtle or daring chromatic match. In praising the subtlety and the refinement of the design he draws attention to the pleasant sound produced by the nib stroking the paper, or the unique scent of a certain piece in his collection. The product, the object, the piece, leaves the world of goods to enter that of sensations, of messages, of evocations, in a complete aesthetic sphere. We are therefore entering a field which is still foreign to economic research. Neither price nor quantity, technical features nor technologies, can explain the success or failure of certain goods, the development of consumers’ preferences, their choice processes, and the evolution of their demand. Shapes, scents, colours, flavours and sounds have long been banned from the sacred enclosure of rational choices. And yet these underestimated variables which economists too often confine to the vague and treacherous analytic category of ‘taste’ (Douglas and Isherwood 1979: 24) affect the choices of an ever-growing number of consumers today, and will become crucial in the future.

Semiotics Collectable goods are outstanding as ‘signs’ (Pomian 1987, 1993). They convey meanings and symbols. These are signs of distinction and homologation, of tradition and innovation, of continuity and breaking off, of belonging and the extraneous. We cannot state that the phenomenon of collecting always and a priori represents a form of adaptation or submission to legitimate and prevailing models and cultural patterns (Bourdieu 1979). On the contrary, the excessive interest shown in the ‘high’ forms of collecting, in particular art collecting (Elster and Cardinal 1994), has drawn the attention of researchers away from other forms of collecting which are clearly antithetical to aristocratic or uppermiddle-class cultural referents. From an hermeneutic point of view, a ‘collection’ of tattoos or the ‘rings and bars’ of body piercing accomplish the same function as a collection of seventeenth-century faience. They identify the ownership group, its mechanisms of cooptation and legitimization, a communal linguistic, cultural and behavioural code, and a shared symbolic universe (Bourdieu 1971: 69). This evocative and communicative power explains the steady and inexorable growth in collecting as a mass phenomenon. A single good is no longer an unerring indication or reliable testimony of its owner’s identity from a socio-economic and cultural point of view. The combination of possibilities resulting from the relationships in the matrix of ‘goods owned/socio-cultural profiles’ is endless, as are the opportunities of using goods as shelters to



defend identities or weapons to attack social prejudices. Conversely, the series or collection could prove the loyalty of the owner to his or her reference group, the steadiness of his or her membership.

History and time A collectable good is a son of time, the amber that captures its moments, the totem and the hourglass of the contemporary and endless time of goods (Laurens and Pomian 1992; Olmi 1992). Ours is a society which has delegated its memory to machines, which has lost the sense of time and makes no distinction between history and the recent past. In such a society, in which history is perceived as a process without gradations or depth, the product of the day before yesterday can have an historical meaning and dignity equal to that of a fossil. Both belong to the past and both reify it, making it present. The fact that 1977 punk T-shirts or 1970 Barbies are already put up to auction and sold for considerable sums should stimulate reflection. ‘Survival in time’ has always been seen as an index of the quality and importance of a good. Time was the final and supreme arbiter of the value of a good, while centuries and decades were the parameters of its judgement. In a society such as the contemporary one, goods do not perish physically and are not ‘consumed’ – they are whimsically abandoned to their destiny as trash after a few hours or months.7 Thus, a time horizon often years seems as great as the glacial era. In such a society a 1978 television set or a 80/88 micro-processor seem unquestionably, unbearably or romantically ancient. Therefore, collecting 1992 telephone cards, early 1980s computers or 1989 records becomes for thousands of people a plunge into the past, a way to exorcize it, to beat it or to relive it (Baudrillard 1994). The goods, even more than feelings or emotions, are the notches that mark the passing of life, of time. If up to the last century objects reminded us of our past, of our roots, of a communal identity that took us painlessly back to nature and to the earth, today they can only talk about the future, anxiously projecting us towards something that is not yet here.

THE EMERGING CHARACTER OF THE CONSUMER-COLLECTOR However, if the world of goods is moving towards a ‘collectable’ dimension, the ‘old’ character of the traditional consumer is also shifting in the same direction. The final result is the emergence of the ‘consumer-collector’, whose characteristics are described in the following pages.



Affect-driven behaviour The relation of the consumer-collector to the world of goods and objects is mainly affective, sentimental, humorous, often fetishistic, never merely instrumental nor spurred by the search for economic utility. The consumer-collector is spurred by enjoyment, desire for gratification, self-complacency, never by need, at most by the need for pleasure. Most important are fantasies, imagination and daydreaming: in this sphere the borders of reality can be stretched at one’s leisure. The consumer-collector draws satisfaction from pure possession rather than use, rewarded not by the purchase act itself, but by the phases that precede it (the search for the piece, the discovery, haggling with the seller), and those that follow immediately afterwards (becoming acquainted with other consumer-collectors, meeting them, exchanges, pacts, rivalries). These elements of socialization and negotiation seem to represent the main behavioural motivations, rather than more technical and rational aspects.

Semiotics and aesthetics The consumer-collector is attracted more by the aesthetic and semiotic values of goods than by their functional properties. The meaningful stimuli are non-verbal and sign language is fully developed: the semantic contents of products and product communication are not as important as the syntactic ones, those concerning structure, style and the stimulation of the senses. The Swatch example is not an isolated one, but clearly illustrates the inexorable trajectory that an ever-growing number of goods is taking towards a merely communicative, immaterial dimension, which reduces the object to a pure sign, indication or signal.

Seeking individualism There are no objective parameters (utility, economic value) driving the individual, but only subjective borders, related to excitement, entertainment, sensory stimulation and joy. Through the objects owned, the consumer-collectors distinguish themselves from others, thus confirming their individuality, taste, flair and culture. The acquisition of a piece is thereby perceived neither as a vagary nor as consumerist excess, but as another membership card to be placed in the difficult formation process of an autonomous and recognizable identity. This does not prevent consumer-collectors from perceiving themselves as members of an élite of enthusiasts who share their interests and can recognize and value their flair and personal, individual talent.



Creativity: the ‘past as novelty’ seeker The consumer-collector demonstrates exploratory behaviour. The explorer is someone who dares to enter unknown places, who only has a vague idea of the destination and the path to be taken. They cannot make use of previously acquired information and therefore have to build up a set of data gradually, during the journey. Therefore we can talk about ‘novelty seeking’ (Hirschman 1980: 284), distinguishing between ‘inherent novelty seeking’ (e.g. the desire to search for the new piece of the collection, meeting other consumer-collectors to exchange information) and ‘actualized novelty seeking’ (e.g. the behaviour of acquiring a piece for the collection). The ‘new’ object to be consumed or collected is rarely a good that has never been produced in the past. More often it is an ‘old’ good, rediscovered and reinterpreted with new functions and meanings. Therefore we can talk about use innovativeness (Hirschman 1980: 289).8 The consumer-collector plays with the stock of existing goods, makes up new functions, new meanings and new combinations which subvert the previous order, hierarchy and relationships. In a revolutionary fashion they redefine the functions of whole categories of goods, changing not only their destination of usage, but even the category they belong to. They are free to create these kinds of features, to intervene subjectively, and to express their own personalities more explicitly.

Seriality of consumption The most important characteristic of the consumer-collector is probably the inclination to serial consumption. If the single good cannot on its own be the membership card of a social, economic or cultural group and cannot accurately define the life style adopted as its reference pattern, the series, the collection, can do this. Loyalty and attachment towards the goods that form a series are the irrefutable proof of the trust in which the group or reference models are held. The more the series is completed, the more the value of the ‘missing piece’ increases. In order to acquire it, the consumer-collector ‘does not mind the expense’, and is prepared to buy ‘at any price’.

Increasing or constant marginal inclination to consumption and demand inelasticity In cases such as these, the marginal inclination to consume does not decrease with an increase in the number of repetitions/amount owned.9 Instead, it is positively connected to the size of the series/amount possessed and is therefore increasing or steady anyway. The desires of the consumer-collector have no end. From this perspective, the consumer-collector might be seen



as the prototype of the ideal consumer, from a company point of view. However, consumercollectors are unpredictable, demanding and fickle, because of attention to changes and the most subtle differences. They can thus become a nightmare or an irritant to a company because they demand constant changes, differentiation, innovations and attentions. They impose never-ending changes and are uncontrollable subjects of environmental transformation. The consumer-collector is a strong example of the active consumer. Like any great collectors, they do not only make purchases, but become patrons. Rediscovering the pleasures of patronage, they cooperate in the creative process, defining the details and specifications of the product, and imposing their own taste on the producer. The factors discussed above explain the consumer-collector’s relative inelasticity of demand with respect to price and quantity. The unique/rare good is ‘priceless’, and has to be bought ‘at any price’. It is impossible to resist the temptation to purchase. The market is therefore reduced to a virtually endless series of auctions for unique pieces. Variety stretching out towards infinity becomes the main criterion determining preferences.

DIFFERENT TYPES OF RAREFACTION PROCESSES Although the rarefaction processes in the relationship between the company and consumer and the emergence of the consumer-collector can be considered as the interpretative keys of these phenomena, it is nonetheless clear that the two phenomena also have different characteristics. In our opinion, what distinguishes them are the different types of rarefaction processes involved. If rarefaction derives from developments in the world of production and consumption, certain processes are more clearly ‘directed’ by enterprises and others by the consumer, while others stem from circular dialectics. We will therefore define the first type as firm driven, the second as consumer driven and the third as market based. Figure 9.2 illustrates these three types along a continuum that stretches from the enterprise to the consumer.

Figure 9.2 Types of rarefaction processes.



Firm-driven processes In the case of several products, the possibility that a relationship might become unique depends both on a specific strategic plan and on the technological know-how of the enterprise. This allows a simplification in the specification of consumer needs and their translation into coherent product attributes (Cozzi et al. 1988). Unique and rare goods are the natural result of contemporary production technologies and represent the response of enterprises to competitors and consumer pressures. However, it should be emphasized that the choice to favour the ‘collectable’ dimension is not invariably accompanied by positive results (serial consumption, loyalty, indirect advertisement). It also exposes the enterprise to considerable risks due to the perennial renewal of the products, to a general atmosphere of constant innovation and to the volatility and short duration of position profits. When technological elements cease to play a fundamental role in the formation process of consumer choice, and are replaced by semiotic and aesthetic features, then creativity, imagination, willingness to risk, hazard and bet become crucial elements. These factors, apart from very limited cases (Ben and Jerry ice-cream) are still penalized both by the selective mechanisms of enterprises and by educational strategies adopted by high schools and universities. Apart from the causes and hypothetical consequences which are still largely unpredictable, we believe it is necessary to focus attention on the strategies that enterprises can adopt to handle the rarefaction process. We have identified some of them.

Positioning strategies on collectable markets Enterprises can decide to put their products directly into collectable markets.10 This is a very common procedure in the field of cultural industries. In the record and publishing industry, for instance, ‘productions’ addressed to collectors often coexist with ‘normal’ ones: special editions, republications and ‘bootlegs’ are typical examples. Even in the film industry the whole sector of gadgets linked to merchandising is addressed mainly to collectors, as are specific movies in the home video sector.

Product family strategies Companies can apply rarefaction techniques either directly by producing unique goods or by aiming at rarity, as in the case of a limited series. A minor but interesting form of rarefaction is the specialization of the market achieved by extending and deepening choice. Such ‘series productions’ are introduced into the market as ‘winter-autumn collections’: shoes, clothes, perfumes and cosmetics are only a few minor examples. The emphasis put on product lines,



through the filter of the brand, of the name, is a polite though tacit form of invitation to collecting, to serial consumption. If someone already uses an aftershave or a deodorant by Vichy or Clarins, why then not buy the other products in the series and ultimately the whole collection? If a consumer already owns the whole Nike Air collection, why not buy the latest models in the 1998 collection?

Price and distribution strategies Price and distribution strategies usually involve fixing a high price and exclusive distribution networks to restrict ex ante the segment of potential purchasers. This strategy creates an élite of consumer-collectors selected according to income – the rarity of the good becomes a warranty of distinction.

Communication strategies Companies always try to profit from the latent collection potential of certain forms of consumption by resorting to promotional mechanisms which evoke ‘game-like’ or, more specifically, collection elements. Such strategies include tokens or picture cards (Castaldo and Mauri 1994), competitions, lotteries, drawing of special prices. These are important components of extra gain, designed to shift the act of consumption towards a totally different sphere, the sphere of entertainment, bargaining, gambling and loyalty with the consumer-collector. They are an attempt to take advantage of the inclination to serial consumption. Therefore, it is not by chance that the dominant promotional strategies are based on tokens that allow the consumer to obtain pieces of a reference ‘collection’ (kitchenware, garments, and so on).

Consumer-driven processes In other cases, rarefaction processes are imposed by the consumer. The causes of this phenomenon are as follows.

The need for hyper-differentiation Hyper-tech mass customization can perhaps appease the need for distinction of large numbers of consumers, but it can also paradoxically drive other large groups to look for even more subtle, refined, snobbish and ingenious forms of distinction.



The affirmation of the consumer-collector is probably the symptom and the most blatant proof of this phenomenon. Their choices and their behaviour are fascinating, surrounded by an invulnerable aura of uniqueness, of exclusiveness. They imply a search, a selection, and an individual story.

Collecting behaviour as a mass phenomenon One of the main reasons for the emergence of the consumer-collector is the transformation of collecting into a mass phenomenon. The first symptoms began to show up at the end of the last century. In the ancien régime society, dominated by poverty and the spectre of famine, collecting remained the exclusive prerogative of the clergy, the aristocracy and financial elites. It had a very precise social and political meaning. At the end of the nineteenth century, while the small and middle bourgeoisie were consolidating and average incomes were increasing, a growing number of individuals permanently escaped the grasp of material needs, therefore increasing the amount of surplus resources usable for purposes previously precluded. Mass consumption was born, together with mass society and its distinction phobia, its maniacal attention to symbols which indicate social hierarchies, barriers, inter-class connections and neutral territories. Is any behaviour nearer to the high-bourgeois and aristocratic ethos, providing the fallacious or real perception of social distinction, of uniqueness in the greyness of the mass, than collecting? Thanks to the remarkable increase in average incomes, collecting has become, in the last few years, a mass phenomenon. It does not only involve the holy of holies of traditional artistic production (elevated, not accidentally, to the role of economic fetishes), but it affects every part of the huge world of goods. Simply reading the catalogues of minor or specialized auction houses, it is easy to perceive the uncontrollable growth in new collectors’ items (Stewart 1984; Opie et al. 1994). This increase in the number of people prepared to devote part of their resources to collectable items extends the concept of such items until it becomes omnicomprehensive, contaminating even the sphere of ordinary, everyday consumption.

Globalization of markets, fragmentation of life styles and reduction of product life cycles Another reason for the emergence of the consumer-collector is the globalization of markets and the presence of more and more articulated and unsteady consumption models. These phenomena allow the consumer to play on international mismatching. The ‘last novelty’ on the Bulgarian market can be a potential good in Azerbaijan, out of date in Paris, venerated like a fetish in Japan, and in full revival in the USA. This possibility extends the potential length of economic life of a good or category of goods. Therefore, the normal life cycle of the



product and its subsequent collectionistic life cycle cannot be distinguished. The permanence in the intermediate position of refuse, of trash, becomes shorter and shorter and sometimes even non-existent. This phenomenon subverts traditional views considering the ‘new’, in which the ‘last’ product is seen as the best and most desirable for consumers. What happens if the myth of ‘progress’ breaks down before an implicit but significant declaration of distrust in contemporary productions, if the ‘old’, in the form of collectable goods, bursts successfully into the limelight of mass consumption? We believe that these questions are worth the attention of economists. This speeding up is surely a novel thing. It was only recently that at least twenty or thirty years had to pass before an object could be ‘rediscovered and relaunched’ as a collectable good. Nowadays, only a few months, or even days, are sufficient. Goods that are still in production in certain countries have already become a cult and highly collectable in other geographic areas, and the same phenomenon involves different groups of consumers. The possibility of playing on different tables is endless, thanks to very different consumption and production models and to the ubiquitous collectors’ willingness to guarantee a positive final outcome to the process.

Institutionalization of collectionistic consumption Not only are more and more categories of goods being elevated into the pantheon of collectables more quickly than ever before, but most important of all they are rapidly institutionalized as such. Up to a few years ago it was possible to distinguish the different phases of this process very clearly. Originally, goods that had not yet acquired collectable status were available only in junk shops, dumps, flea markets, car boot or yard sales. Gradually, the most innovative antique dealers started participating in mid-level sales, the first collectors’ fairs were organized, and the first annotated catalogues published. The final consecration was achieved with the publication of specialized magazines, theme auctions, newsletters and permanent exhibitions. In the past this process required years to be completed and therefore prices remained low for a long time, allowing remarkable collections to be built up at moderate cost. This does not happen any more. The processes of institutionalization and canonization are amazingly rapid. In a few months, hyper-specialized magazines crop up, fanzines and newsletters mushroom, ‘collectors’ stock exchanges’ proliferate, together with exchange weeks, international conventions and sales in the most prestigious auction houses. Monographs and catalogues are published more and more often, the number of experts and opinions grows and, most important of all, specialized museums are opened. These phenomena speed up the process of ‘historicization’ of the objects and guarantee to collectors the exciting sensation of ‘taking part in history’, of possessing a sempiternal niche in the immortal pantheon of collectors, whether specialized in billboards of hillbilly groups, 3D postcards or 1970s porn picture-stories.



Moreover, these factors make the community, the collectors’ group, immediately distinguishable and they are more rapidly provided with a common and fundamental identity. The relationship with other collectors involves friendships or idiosyncrasies, exchanges of pieces and information, trips and tours together. The collectors’ community becomes an important social point of reference. These phenomena do not, however, affect the dimension of privacy. The collection is also the last, inviolable personal and private space of people which remains constantly on call. If compared to the imaginary communities described by Uusitalo (1998), collectors’ communities also have the undeniable advantage of being real. The collection can reflect imaginary or dreamlike social models, but nothing is more concrete than the group of collectors with whom one shares the same passion. In the above cases, the consumers utilise goods in a creative way on the basis of codes and symbols of their own sociocultural reference group. Even standard goods acquire features of uniqueness, because they can express the uniqueness of the individual who uses and consumes them. In this case companies have a very limited chance to intervene. They can either adapt their production levels to meet market demand, recuperating the productions once decentralized in underdeveloped countries, where these goods were sold as ‘western latest fashion’ or restart production lines that had long lain idle. In either case, the consumer-collectors subvert the strategic choices of enterprises, forcing them to follow, flatter and court them. Buying whole series of objects rather than isolated pieces, acting on the matching combinations offered by the series and playing their inherent semiotic games, the consumercollectors can invent new forms of consumption. They can save goods from the extinction to which a strictly evolutionist destiny would condemn them, once they have reached a certain age. They can restore goods and enterprises from anonymity and can specify and impose their requirements on companies in the same way as a consumer-client. Thus, they force producers to alter their production methods constantly, by renewing supply or by changing other strategies. From this point of view, consumer-collectors are exceptionally active. Their range of choice is endlessly extended and the ability to impose themselves on the market increases constantly. They play a fundamental role as decisional rather than economic subjects.

Market-driven processes Finally, we would like to analyse those processes involving situations which fall between the two described before. The rarefaction process has its origins in the playing of a deferment game between enterprise and consumer, a game in which each player tries to enter the symbolic field of the other, creating in turn new codes and languages. Swatch watches, for example, are manufactured in limited collections and their distribution therefore ‘plays’ on the simultaneous presence of different groups of consumers: collec-



tors and common buyers. When the occasional buyer can afford to buy a piece with potentially a very high economic value, the watch virtually becomes a lottery ticket and consumption enters the ‘game-like’ sphere, attracting consumers stimulated by the gambling, extragain component. The same holds true in the car and motorcycle sectors, where many producers manufacture models in limited series. A notable example is Ferrari, which not only restricts the production of certain models to a predefined number of pieces, but assigns its product to a small élite. It would be fascinating to know the criteria adopted to conceive the list of customers. This is an intermediate area, whose borders are still uncertain and which surely needs further studies, especially empirical surveys. We believe that this sector will soon become not only more and more crowded, but crucially important in a future dominated by unique and rare goods, where ‘transactions involving signs will form the continuously expanding core of economic activity, while the production and consumption of material goods will stagnate’ (Hutter 1998).

ACKNOWLEDGEMENTS We are grateful to Marina Bianchi who gave us helpful comments as well as the opportunity to write our paper. We must thank J.T.F. Butler and Gioia Guerzoni for their assistance during the translation phase, Neil De Marchi, Salvatore Vicari and two anonymous referees for their crucial comments; and our dear friends Carlo Gregori, Mario Bertoni and Peppino Soda for their criticisms and moral support. This does not discharge us from the responsibility for all remaining errors.

NOTES 1. On ‘nostalgia goods’ see the articles and related bibliographies of Havlena and Holak (1995) and Holbrook and Schindler (1995). 2. From a theoretical point of view criticisms of the traditional economic analysis of consumer behaviour are expressed in this book by Bianchi, Ranchetti and Gualerzi. The only exceptions in the previous panorama are represented by Thompson’s marvellous book (1979), in particular Chapters 2, 5 and 6 and the article of Belk et al. (1991). 3. In Anglo-Saxon languages, the verb to rarify means ‘to make or become less dense or solid’. In our interpretation, we restore the term rarefaction to its original etymological meaning, derived


SILK PURSES OUT OF SOWS’ EARS from the Latin rarefacere, rarum-facere – ‘to make rare’. In this sense, we want to stress the processes or actions which transform a normal, common good into a rare good. This phenomenon can be ascribed both to factors independent of human will (the disruptive action of time, which physically destroys goods) and to deliberate strategies and actions (i.e. in philately and numismatics certain collectors deliberately destroy certain pieces in order to increase the value of the remaining ones, as well as firms which produce ‘limited series’ to create artificial conditions of scarcity). 4. This trend is empirically demonstrated by the fact that firms in industrialized countries are moving from traditional communication instruments (mass advertising) to direct marketing, which allows individual and interactive contact with the consumer. Further, marketing literature nowadays assumes the shift from transactional marketing (based on spot exchanges) to relational marketing (based on the construction of long-term relationships). In this regard, original contributions have been introduced by European scholars, such as Gummesson (1987); Fiocca (1988); Gronroos (1989, 1991); Grandinetti (1993); Castaldo (1994). 5. Our hypothesis can be seen as an extension of a paradigm already mentioned in consumer behaviour studies, even if it is a somewhat marginal one – the paradigm of the experiential consumer and of hedonistic consumption (Hirschman and Holbrook 1982; Holbrook and Hirschman 1982; Unger and Kernan 1983; Hirschman 1984; Holbrook et al. 1984; Havlena and Holbrook 1986; Holbrook and Grayson 1986; Arnould and Price 1993; Lacher and Mizerski 1994). Holbrook and Hirschman’s article (1982), which can be considered as the ‘manifesto’ of the new paradigm, demonstrates that several purchase acts can be better understood if this hypothesis is considered. 6. For further information see also the articles printed in the Journal of the History of Collections. 7. On these aspects, see the observations of Thompson (1979), in particular Chapters 2 and 3. 8. Another behaviour which is traditionally recognized in the marketing literature as an expression of consumer creativity is adoptive innovativenes (Rogers and Shoemaker 1971; Hirschman 1980). This occurs when an individual buys a new product to satisfy his/her functional or hedonistic needs. 9. On these aspects see Haskell and Penny (1981), Laurens and Pomian (1992), Saarinen (1958), Taylor (1954). 10. On collectable markets see Stoller (1984), Colonna and Henry (1989), Guerzoni (1995, 1996).

BIBLIOGRAPHY Alsop, J. (1982) The Rare Art Traditions: The History of Art Collecting and its Linked Phenomena Wherever These Appeared, New York: Harper & Row. Appadurai, A. (ed.) (1986) The Social Life of Things, Commodities in Cultural Perspective, Cambridge: Cambridge University Press. Arnould, E.J. and Price, L.L. (1993) ‘River magic: extraordinary experience and the extended service encounter’, Journal of Consumer Research 20, June: 24–44.


SILK PURSES OUT OF SOWS’ EARS Baudrillard, J. (1968) Le Système des objets, Paris: Gallimard. —— (1994) ‘The system of collecting’, in J. Elsner and R. Cardinal (eds) The Cultures of Collecting, London: Reaction Books: 7–24. Belk, R.W., Wallendorf, M., Sherry, J.F. and Holbrook, M.B. (1991) ‘Collecting in a consumer culture’, in R.W. Belk (ed.) Highways and Buyways: Naturalistic Research from the Consumer Behavior Odyssey, Provo, UT: Association for Consumer Research: 178–215. Bettman, J.R. (1979) An Information Processing Theory of Consumer Choice, Reading, MA: Addison-Wesley. Blumer, H.C. (1969) ‘Fashion: from class differentiation to collective selection’, The Sociological Quarterly 10 (3): 275–91. Bourdieu, P. (1971) ‘Le Marché des biens symboliques’, L’Année sociologique 22: 49–126. —— (1979) La Distinction, Paris: Les Éditions de minuit (Italian translation 1983, La distinzione. Critica sociale del gusto, Bologna: Il Mulino). Castaldo, S. (1994) Le relazioni distributive, Milan: EGEA. Castaldo, S. and Mauri, C. (1994) Supermarket Customer Loyalty Promotions. An Empirical Study, Milan: EGEA. Christopher, M. (1989) ‘2001 – the existential consumer’, European Journal of Marketing 23 (8): 80–4. Colonna, C.M. and Henry, L.H. (1989) ‘The derived demand for collectibles’, in D.V. Shaw, W.S. Hendon, and V. Lee-Owen, (eds) Cultural Economics 88: An American Perspective, Akron: Association for Cultural Economics: 137–42. Cozzi, G., Di Bernardo, L. and Rullani, E. (1988) ‘Marketing e tecnologie dell’informazione: dall’economia di massa all’economia della varietà’, Scritti in onore di Luigi Guatri, Milan: Bocconi Comunicazione: 361–417. Dance, S.P. (1966) Shell Collecting, London: Faber and Faber. D’Aveni, R. (1994) Hyper-competition. Managing the Dynamics of Strategic Maneuvering, New York: Free Press. Davis, S. (1987) Future Perfect, Reading, MA: Addison-Wesley. Douglas, M. and Isherwood, B. (1979) The World of Goods: Towards an Anthropology of Consumption, London: Allen Lane (Italian translation 1984, Il mondo delle cose, Bologna: Il Mulino). Eccles, Lord (1968) On Collecting, London: Longman. Elsner, J. and Cardinal, R. (1994), ‘Introduction’, in J. Elsner and R. Cardinal (eds) (1994) The Cultures of Collecting, London: Reaction Books: 1–5. Fiocca, R. (1988) ‘Dalla gestione del marketing mix al marketing relazionale nei settori ad alta tecnologia’, Sinergie 24: 131–9. Frey, B.S. and Pommerehne, W.W. (1989) Muses and Markets: Explorations in the Economics of the Arts, Oxford: Blackwell. Grandinetti, R. (1993) Reti di marketing, Milan: ETAS. Gronroos, C. (1989) ‘Defining marketing: a market oriented approach’, European Journal of Marketing 23 (1): 52–60. —— (1991) ‘The marketing strategy continuum: towards a marketing concept for the 1990s’, Management Decision 29 (1): 7–13. Guerzoni, G. (1995) ‘Reflections on historical series of art prices. Reitlinger’s data revisited’, Journal of


SILK PURSES OUT OF SOWS’ EARS Cultural Economics 19 (3): 251–60. —— (1996) ‘The British market of paintings. 1789–1914’, in M. North (ed.) Economic History and the Arts, Koln: Böhlau Verlag: 97–132. Gummesson, E. (1987) ‘The new marketing – developing long-term interactive relationships’, Long Range Planning 20 (4): 10–20. Hamel, G. and Prahalad, C. K. (1994) Competing for the Future, Boston: Harvard Business School Press. Haskell, F. (1976) Rediscoveries in Art, London: Phaidon Press (Italian translation 1982, Riscoperte nell’arte, Milan: Edizioni di Comunità). —— (1989) Le Metamorfosi del Gusto, Turin: Bollati Boringhieri. Haskell, F. and Penny, N. (1981) Taste and the Antique. The Lure of Classical Sculpture 1500–1900, New Haven: Yale University Press (Italian translation 1984, L’antico nella storia del gusto, Turin: Einaudi). Havlena, W.J. and Holak, S.L. (1995) ‘The good old days: Observations on nostalgia and its role in consumer behaviour’, Advances In Consumer Research 18: 323–9. Havlena, W.J. and Holbrook, M.B. (1986) ‘The varieties of consumption experience: comparing two typologies of emotion in consumer behavior’, Journal of Consumer Research 13, December: 394–404. Hirschman, E.C. (1980) ‘Innovativeness, novelty seeking, and consumer creativity’, Journal of Consumer Research 7, December: 283–95. —— (1984) ‘Experience seeking: a subjectivistic perspective of consumption’, Journal of Business Research 12: 115–36. Hirschman, E.C. and Holbrook, M.B. (1982) ‘Hedonic consumption: emergent concept, methods and propositions’, Journal of Marketing 46, Summer: 92–101. Holbrook, M.B., Chestnut, R.W., Oliva, T.A. and Greenleaf, E.A. (1984) ‘Play as a consumption experience: the roles of emotions, performance, and personality in the enjoyment of games’, Journal of Consumer Research 11, September: 728–39. Holbrook, M.B. and Grayson, M.W. (1986) ‘The semiology of cinematic consumption: symbolic consumer behavior in Out of Africa’, Journal of Consumer Research 13, December: 374–81. Holbrook, M.B. and Hirschman, E.C. (1982) ‘The experiential aspect of consumption: consumer fantasies, feelings and fun’, Journal of Consumer Research 9, September: 132–40. Holbrook, M.B. and Schindler, R.M. (1995) ‘Echoes of the dear departed past: some work in progress on nostalgia’, Advances In Consumer Research 18: 330–33. Hutter, M. (1998) ‘On the consumption of signs’, in M. Bianchi (ed.) The Active Consumer, London: Routledge. Lacher, K.T. and Mizerski, R. (1994) ‘An exploratory study of the responses and relationships involved in the evaluation of, and in the intention to purchase new rock music’, Journal of Consumer Research 21, September: 366–80. Lancaster, K.J. (1966) ‘A new approach to consumer theory’, Journal of Political Economy, April: 132–57. Laurens, A.F. and Pomian, K. (eds) (1992) L’Anticomanie, Paris: Editions de l’Ecole des Hautes Etudes en Sciences Sociales. MacCain, R. (1986) ‘Game theory and cultivation of taste’, Journal of Cultural Economics 10, June: 1–16. Midgley, D.F. and Dowling, G.R. (1978) ‘Innovativeness: the concept and its measurement’, Journal of Consumer Research 4: 229–42.


SILK PURSES OUT OF SOWS’ EARS Mueller-Heumann, G. (1992) ‘Market and technology shifts in the 1990s: market fragmentation and mass customization’, Journal of Marketing Management 8: 303–14. Muensterberger, W. (1994) Collecting. An Unruly Passion, Princeton: Princeton University Press. Olmi, G. (1992) L’inventario del mondo. Catalogazione della natura e luoghi del sapere nella prima età moderna, Bologna: Il Mulino. Opie, R. and Eisner, J. and Cardinal, R. (1994) ‘Unless you do crazy things’, in J. Elsner and R. Cardinal (eds) (1994) The Cultures of Collecting, Reaction Books: London: 1–5. Park, C.W., Jaworski, B.J. and Maclnnis, D.J. (1986) ‘Strategic brand concept-image management’, Journal of Marketing 50, October: 135–45. Pine, B.J. (1992) Mass-Customization, Boston: Harvard University Press. Pomian, K. (1987) Collectionneurs, amateurs et curieux: Paris-Venise, XVI-XVIII siècle, Paris: Gallimard, Paris (Italian translation 1989, Collezionisti, amatori e curiosi, Milan: Il Saggiatore). —— (1993) ‘Collections et musées’, Annales 48e année, 6, Novembre-Décembre: 1381–1401. Porter, M. (1985) Competitive Advantage, New York: Free Press. Rheims, M. (1956) La vie étrange des objets, Paris: Gallimard. —— (1990) Les Fortunes d’Apollon, Paris: Seuil. —— (1992) Apollon à Wall Street, Paris: Seuil. Rogers, E.M. and Shoemaker, F.F. (1971) Communication of Innovations, New York: The Free Press. Saarinen, A.B. (1958) The Proud Possessors, New York: Random House (Italian translation 1977, I grandi collezionisti americani, Turin: Einaudi). Schnapper, A. (1988) Le Géant, la licorne et la tulipe. Collections at collectionneurs dans la France du XVIIe siècle, I. Histoire at histoire naturelle, Paris: Flammarion. Singer, L. (1988) ‘Phenomenology and economics of art markets: an art historical perspective’, Journal of Cultural Economics 12, June: 27–40. Stewart, S. (1984) On Longing: Narratives of the Miniature, the Gigantic, the Souvenir, the Collection, Baltimore: Johns Hopkins University Press. Stoller, M.A. (1984) ‘The economics of collectible goods’, Journal of Cultural Economics 8, June: 91–104. Taylor, F.H. (1954) Artisti, principi e mercanti. Storia del collezionismo da Ramsete a Napoleone, Turin: Einaudi. Thompson, M. (1979) Rubbish Theory. The Creation and Destruction of Value, Oxford: Oxford University Press. Unger, L.S. and Kernan, J.B. (1983) ‘On the meaning of leisure: an investigation of some determinants of the subjective experience’, Journal of Consumer Research 9, March: 381–92. Uusitalo L., (1998) Consumption in Postmodernity: Social Structuration and the Construction of the Self, in M. Bianchi (ed.) The Active Consumer, London: Routledge. Vicari, S. (1989) Nuove dimensioni della concorrenza. Strategie nei marcati senza confini, Milan: EGEA. —— (1994) ‘Acquisition as experimentation’, in G. von Krogh, A. Sinatra, and H. Singh (eds) The Management of Corporate Acquisitions, London: Macmillan: 337–58. Zeithaml, V.A. (1985) ‘The new demographics and market fragmentation’, Journal of Marketing 49, Summer: 64–75.



SCITOVSKY ON CONSUMPTION Scitovsky’s theory of consumption departs radically from the standard utility-based approach that has dominated the thinking of economists for over a century. A problem Scitovsky identifies with the standard theory is the insufficient attention given to the importance of novelty, interdependence and habits in consumer behavior (Scitovsky 1976: 150). We address these neglected issues in a formal model, an approach which Scitovsky considered but decided was out of place in his own exposition (ibid.: 149). Central to his approach is the idea that human beings, like other animals, seek optimal levels of arousal: a quantity that ‘depends on the stimulation the central nervous system receives from outside, through the senses . . . internal organs . . . and from within the brain itself (Scitovsky 1976: 18). Too high a level of arousal gives rise to strain and discomfort, and induces actions on the part of the subject to try and reduce arousal levels. Too low a level of arousal leads to boredom and induces actions directed at raising arousal levels.1 To take an extreme example, experimental subjects who are otherwise well provided for are found to experience, after a few hours of sensory deprivation, levels of boredom so painful as to induce ‘headaches, nausea, confusion, fatigue, hallucinations, and a temporary impairment of various mental faculties’ (ibid.: 22). Consumption activity is one avenue for the attainment of changes in arousal levels.2 In looking at consumption, Scitovsky employs Hawtrey’s important distinction between defensive and creative products (Hawtrey 1925: 189–92). Defensive products serve to remove discomforts such as pain, hunger and the like, whereas creative products occasion pleasure.



The attainment of comfort and the sensation of pleasure are not the same thing (Scitovsky 1976: 59–60), nor are they valued in the same way. In Hawtrey’s words:

It is possible for a rich man to incur heavy expenditure . . . securing the minimum of discomfort. . . . But the whole yields no positive good; it merely brings him to the zero point, at which he is suffering from no avoidable harm. He has weeded his garden and still has to choose what he will plant in it before he can be said to have made anything at all of his life. (Hawtrey 1925: 191; cited by Scitovsky 1976: 11)

If these distinctions hold true, the study of activities promoting pleasure will likely differ from that of consumption aimed at increasing comfort. We focus on the analysis of creative products and the pleasure they give, while sharing with both Hawtrey and Scitovsky the understanding that these distinctions are rarely clear-cut (Scitovsky 1976: 108–12). We are therefore interested in the analysis of consumption in aid of activities such as hobbies, sports and recreation, entertainment, cultural and intellectual pursuits. In addition to these clearly creative activities, we include a wide variety of mundane objects, such as clothing and even consumer durable goods. These items are chosen by consumers not only with an eye to pure functionality, but also to their design, fashionability and aesthetic appeal. Consequently, their consumption can also alleviate boredom and promote pleasure. In this manner, the model we propose applies to a wide variety of consumption activities. Such activities have a number of interesting characteristics. In order for a creative activity to yield the pleasure of novelty it must surpass a certain level of complexity. With complexity, the consumer finds stimulus in the discovery of the new aspects of the activity. On the other hand, maximum enjoyment of a complex activity requires some familiarity with it; familiarity can come with consumption, but the activity should also not be so highly complex as to be bewildering. Thus ‘a flow of information that is either too easy and too familiar or too fast and too unfamiliar should be less pleasant than something in between, and the most pleasant seems to be what fully engages the brain’s information-processing capacity’ (Scitovsky 1976: 35). Learning is then typical of such ‘skilled consumption’ (ibid.: 58) which involves attention and care for maximum enjoyment. Since the enjoyment of a complex activity is itself an accomplishment, the process of satisfying a desire for novelty can itself be pleasurable. Pleasure is derived both from the result of doing a crossword puzzle – finishing it – as well as from the process itself (ibid.: 76–8). The pursuit of novelty in consumption can then be seen as a widely used method of fighting boredom in modern society, and in the attainment of an optimal level of arousal.



The alleviation of boredom through the pursuit of novel consumption activities is a process that can never lead to satiation, since ‘novelty, like food, is used up in the act of its enjoyment’ (Scitovsky 1976: 58). The choice of a novel consumption activity gives rise to high levels of arousal at first, but the degree of arousal declines with every repetition of the activity. This decline is attenuated but not halted for highly absorbing and fascinating activities: Scitovsky mentions serious music as having this quality of complexity and high information content (ibid.: 54). The act of repetition has the powerful secondary effect of habit formation. Scitovsky gives several reasons why:

One of the most firmly established findings of psychology is that any act immediately followed by a reward . . . is reinforced; that is, the likelihood and frequency of its future performance is increased. Moreover, the reinforcement becomes the stronger, the more often the sequence . . . is repeated. The slower, therefore, the realization of the loss of pleasure, the stronger will become the habit, and the smaller the likelihood that the decision, once made, will be reconsidered and reversed. (Scitovsky 1976: 73)

Scitovsky further explores the ‘comfort of sticking to our habits’ with the aid of the psychologists’ opponent-process theory of motivation (Solomon and Corbit 1974; Solomon 1980), on which other economists have also occasionally drawn.3 According to the opponent-process theory, many feelings of pain and pleasure are followed by a contrary aftereffect. Thus pleasure can be followed by a feeling of emptiness. As a pleasurable activity is repeated, the pleasure derived declines in magnitude but the aftereffect strengthens. If the aftereffect becomes so strong that the subject re-engages in the activity for the (temporary) relief offered by the pleasurable effect, the subject is habituated or addicted to the activity. Hence the pursuit of novelty may be countered by the formation of habits, and the discomfort resulting from the interruption of habitual patterns of behavior. In the chapter, we speak of habits rather than addictions. Habits seem to refer to ‘normal’ activities, addictions to ‘pathological’ ones. We understand that addictions may either have physical or psychological bases, and so a distinction made on those grounds between habits and addictions is not clear-cut. Yet we are not concerned with defending a precise distinction between habits and addictions. Indeed, Scitovsky (1976: 129–30) stresses the commonality between habits and addictions. This also is a theme of the opponent-process literature itself, in which it is argued that its ‘novel feature’ is ‘that it sees the behavioral phenomenon of



addiction as an empirical model for all acquired motivation’ (Solomon and Corbit 1974: 144). In one sense, this is a great strength of the theory since it allows an overarching conception to explain a wide variety of observations. Nevertheless, we can acknowledge a subtle, yet hard to define, difference between habits and addictions. Related to this issue is the idea that not all consumption activities are likely to devolve into habits or addictions. This can either be due to differences among individuals or among activities. While Solomon (1980: 708) realizes that the same activity can leave different individuals differentially susceptible to becoming addicted, he admits that this observation is not adequately explained by the opponent-process theory. Moreover, in addition to differences among individuals, there are likely to be differences among activities in their propensities for invoking habituation, as was also realized by Solomon and Corbit (1974: 142–3). These differences could well be explored. For the purposes of this chapter, however, since we do not appeal to any theory which captures these differences, introducing individual- or activity-specific differences would not gain us any true generality, and we opt for the symmetric case. Alongside the pursuit of novelty and the adherence to habit, there is a third source of motivation which Scitovsky identifies as being important: ‘the desire to imitate and conform to the behavior of the group’ (Scitovsky 1976: 115). Although Scitovsky attributes the persistence of imitative behavior to the ‘comfort of belonging,’ there are other indications in his work why imitation might be a natural complement to the pursuit of novelty when habit formation is prevalent. Recall that the repetition of an activity causes a reduction of induced arousal levels while at the same time tending towards the creation of a habit. When arousal levels have fallen sufficiently low, assuming that the habit has not reached the level of an addiction, individuals will tend to seek out alternative activities with a view to attaining higher arousal levels. There will generally tend to be a large menu of such activities available, about which any given consumer will have very limited information. Observing the choices of others may therefore serve an informational purpose, bringing new activities to the attention of a novelty-seeking individual and signaling to that individual the fact that these activities are pleasurable to those who persist in performing them. When individuals change their consumption patterns therefore, it tends to be through the adoption of those activities which others have been observed most frequently to perform. However, a consumer’s own consumption history may also provide information about which activities are chosen next. The complexity and learning aspects of activity choice suggest that activities allied to those currently chosen may have a higher probability of adoption. For instance, an interest in racquetball may lead to squash and then to tennis. To model this aspect adequately would require the specification of the relations among activities, indicating which are likely to lead to others. With such a network, the evolution of consumption among related activities could be studied.



That consumers do not have full information about the entire range of activities available, as is assumed in the standard theory, was quite obvious to Scitovsky, who stressed the importance of limited information about choices, and the role of observations of others’ consumption in providing fuller information:

The economist’s traditional picture of the economy resembles nothing so much as a Chinese restaurant with its long menu. Customers choose from what is on the menu and are assumed always to have chosen what most pleases them. That assumption is unrealistic, not only of the economy, but of Chinese restaurants. Most of us are unfamiliar with nine-tenths of the entrees listed; we seem invariably to order either the wrong dishes or the same old ones. Only on occasions when an expert does the ordering do we realize how badly we do on our own and what good things we miss. (Scitovsky 1976: 149)

Indeed, Scitovsky thought that we learn a great deal from others (Scitovsky 1976: 137), and that interdependence expresses itself in many ways in consumption choices (ibid.: 131–2). Moreover, since the activities consumers currently pursue are typically complex and involve care, attention and repetition, it makes sense that they block out other possibilities as being extraneous for some time (ibid.: 44–6) since information-processing capacity is limited (ibid.: 42–4). With these preliminaries in mind, we can turn attention to formulating a model capturing some of the key elements of Scitovsky’s outlook. In order to do so, we shall neglect a number of important determinants of consumption that have been stressed by traditional theory, such as the role of relative prices. Ways in which the two approaches could be combined to yield a more complete theory of consumption are discussed in the concluding section.

A DYNAMIC MODEL We propose to combine the three motives discussed in the previous section into a single model of activity choice. Consider the case of m individuals who can choose among n distinct activities.4 In any given period t, each individual can engage in at most one activity. At the start of any given period, each individual will have a history of choices. Let s ij represent the share of activity j in the consumption history of individual i. Hence if individual i has never chosen



activity j, we would have s ij = 0. Similarly, if individual i has chosen nothing other than activity j throughout her history, then we would have s ij = 1. We refer to s ij as individual i ’s consumption stock of activity j. For each individual i, the sum of stocks equals unity: n

∑ s ij = 1 for all individuals i. j=1

The idea that one’s consumption history has an impact on current preferences is commonly made in the economic literature on ‘rational addiction’ (Becker and Murphy 1988), where it is also represented by a consumption stock. Furthermore, it is usually supposed that more recent consumption choices are weighted more heavily than those made on the distant past. This latter effect can be represented by assuming that the consumption stocks corresponding to the various activities depreciate at some rate δ over time, where 0 ≤ δ 0.

h ij = g(s ij),


In any given period, the choice of activity is made in two steps. First, the individual decides on whether or not to adopt an activity different from that adopted in the previous period. If she decides in favor of repetition, then the choice is made. If she decides against repetition, and chooses instead to ‘try something new,’ there remains the decision of which alternative to adopt. Suppose that activity j was adopted in the previous period. Then the probability of repetition is specified as follows: π ( a i (t ), h i (t));

π 1 > 0,

π 2 > 0.


In other words, the probability with which the most recently chosen activity is repeated in the current period is highest when the arousal level that it induces is high, and the degree to which the individual is habituated to the consumption of that activity is high. These two effects pull in opposite directions. The higher the stock of consumption associated with a given activity, the lower will be the arousal level associated with it (the novelty effect), but the higher will be the degree to which that activity has become a habit. It is therefore conceivable that the probability of repetition will be highest for non-extreme values of the consumption stock. For simplicity, we may use (10.1–10.2) to write the probability that agent i will repeat activity j as π(s ji ). We assume throughout that the novelty-seeking impulse is strong enough to prevent complete addiction to one activity. Formally this means that π(1) < 1, so that an individual who has experienced only one activity will not continue to repeat that with probability one. What if the individual decides to seek an alternative activity? In this case a choice has to be made with regard to which of the available activities is to be selected. We assume that this choice will be made on the basis of observing the recent choices of others. Define the aggregate, economy-wide, consumption stock of activity k as follows: m Sk = ∑ s ik , i =1

k = 1, 2, . . . n.

The aggregate consumption stocks contain information about the activity choices made by the population as a whole. An activity with a high level Sk is one that has been chosen by numerous people on numerous occasions in the recent past. It is likely, therefore, that an individual looking to try something novel will be attracted to such activities. Formally, the probability that a novelty seeking individual i will switch to activity k can be represented as: γ(Sk );

γ′ > 0,

γ(1) < 1.




In order that the values γ(Sk) are interpreted as probabilities they must satisfy: n

k =1

γ(Sk ) = 1.


Since γ(Sk) will generally be positive even when k = j, the above specification does not rule out the possibility that an individual who seeks a new activity will nevertheless come to adopt the same activity chosen in the last period.6 This will be most likely to occur when Sj is close to 1, so that imitative behavior is not likely to lead to novel choices. Nevertheless, as long as γ(1) < 1, complete homogeneity of behavior cannot persist, since there will be a strictly positive probability, however small, of adopting something new. Putting (10.3) and (10.4) together, we obtain:

Pr[c i(t + 1) = k | c i(t) = j] =


π(s ij ) + (1 – π(s ij )) γ(Sj ) if k = j, (1 – π(s ij )) γ(Sk )

otherwise. (10.6)

It is a straightforward matter to verify that if (10.5) holds, then the sum of the above probabilities over all values of k will equal 1 as required. Given the values of ci and si for all agents i at period t, equation (10.6) fully defines the probability distribution governing the random variables ci (and hence si) for all agents i at period t + 1. The dynamics of consumption over time are therefore characterized by a Markov process in the 2m variables c1, . . . cm, s1, . . . sm, for which equation (10.6) is the one-step transition function. Given any initial values of these variables, it is possible to simulate an indefinitely long time path of consumption for each agent using this transition function. No two realizations of the process will be identical, since the dynamics are not deterministic. Nevertheless, the long-run properties of the time path generated may be characterized. The state space of the Markov process described above is the set of all admissible values of the vector (c1, . . . cm, s1, . . . sm). All values of ci must lie in the set J = {1, 2, . . . n} since there are n distinct activities. All values of si must lie in the interval I = [0,1]. Hence the state space is the set Ω = Jm × Im. It is a straightforward matter to verify that Ω is also the unique ergodic set of the process: any trajectory originating in Ω remains there for all t, and no proper subset of Ω has this property.7 Since the process has a unique ergodic set, it will have a corresponding invariant distribution. Intuitively, this distribution tells us, for each subset of Ω, the proportion of time periods during which the process will be found in that set in an infinitely long realization of that process.



SIMULATION RESULTS While the properties of the invariant distribution cannot be derived analytically with arbitrary functional forms, it is possible to simulate the model over long periods of time to examine the resulting time series. Estimates of the invariant distribution are obtained by computing the frequency distributions corresponding to the simulated values. Over a sufficiently large number of time periods, convergence can be attained to an arbitrary degree of accuracy. We present simulations based on the following function specifications:

f (s) = e–s/(1-s) g (s) = 10s2 π(a, h) = ah =10e–s/(1–s)S2 γ (S) = (1 – ε )S + nε As long as ε > 0, no commodity can be driven to ‘extinction’ since there is always a nonzero probability, given by ε/n, that a commodity with a zero aggregate consumption history will be chosen by a consumer who decides to switch. The following parameter values were chosen for the simulation: n = 3;

m = 5;

δ = 0.85;

ε = 0.14;

Hence there are three goods and five consumers. A time series showing the aggregate consumption histories Sj , j = 1, 2, 3 is depicted in Figure 10.1. The stochastic nature of activity choice gives rise to a rather erratic pattern of consumption for each of the three activities. However, certain patterns are discernible in the time series generated. When the aggregate consumption history of a particular activity drops close to zero, indicating that it has been chosen very infrequently in the recent past, it tends to remain very low for a number of periods before recovering. This effect arises from the fact that such activities are not very visible in the population and are therefore less likely to be adopted by those who are seeking to alter their current activity choice. Hence while one activity remains neglected, the other two maintain market shares of close to 50 per cent. Over time, accumulated random shocks cause a change in the pattern as a previously popular good falls into neglect with a previously neglected good becoming popular. Hence the consumption histories of each good are either close to zero or close to 50 per cent most of the time.



Figure 10.1 Time paths of consumption stocks sj , j = 1, 2, 3.

Figure 10.2 depicts the estimated invariant distribution of the process, computed on the basis of 500,000 iterations. Symmetry requires that the distributions for each of the three variables should be identical, so only the frequency distribution of good 1 has been shown. The pattern of neglect and popularity which can be discerned in Figure 10.1 shows up much more sharply in Figure 10.2 in the form of a bimodal invariant distribution. With a larger number of available activities, the invariant distribution may be expected to have multiple peaks.8 Activities are likely to be in widespread use or to be virtually abandoned, with moderate levels of adoption less likely to be observed. Thus we are able to have knowledge of consumption in a statistical sense, in the long run. The bimodality is consistent with the observation that many products undergo periods of disfavor and periods of fashion with relatively little by way of intermediate levels of consumption.

DISCUSSION Our model is very simple in many ways, and could be generalized. However, it does admit at least three distinct motives for choice – habits, novelty and interdependence – while establishing regularity at the macro level, in the form of a limiting invariant distribution. An interest-



Figure 10.2 Estimated invariant distribution, consumption stock of good 1.

ing avenue for future research would be to combine the Scitovskian effects considered here with more traditional determinants of consumer choice. For instance, introducing a production function for each good, with some price elasticity in consumption choices, could give rise to a richer model of consumption. The effects of multiple peaks in the invariant distribution of consumption histories could be reinforced if there were scale economies in the production of some goods. In this case a neglected good would also become more expensive and hence face greater obstacles in recovering popularity. We have modeled the likelihood of adoption of an activity as being greater the more common it is in the general population of consumers. The motivation for this is twofold: it not only represents the desire to imitate and conform, but also the overcoming of information limitations by observing others’ choices. However, another effect is possible. Consumers may want actively to differentiate themselves from commonly employed activities, and will therefore shun ‘mass’ consumption patterns. To the extent this is the case, the effect of an increased level of consumption in the population will diminish or reverse the effect we have postulated. The passage of time in the present framework has important effects which do not have to do with the time-constrained aspect of consumption emphasized by traditional theory. With time, the product becomes familiar, leading both to a tendency towards habituation and the



atrophy of the novelty effect. Consumption is then renewed. Present consumption depends also on consumption histories, or past choices. Thus consumption here has a fully dynamic aspect. The picture which emerges of satisfaction with consumption is also different. Satisfaction has to do with consumption, but the relation between the two is complex. There is an initial burst of arousal with a new product; this decays with time and is re-achieved when the next product is consumed. These fluctuations occur even though – as in our model – the money value of consumption is constant. It is not straightforward, then, to link the quantity of consumption or its value with the well-being derived from consumption. There is an important asymmetry between the actor and the observer of the consumption process. For the actor, the choice of a consumption item may be full of meaning, the result of careful deliberation, with a clear purpose in mind, and showing all the characteristics of individual volition. We interpret our model to include such a view of the agent. Yet the very fact that the consumer is active means that certitude in response to stimuli is lost, and from the point of view of the observer the actions take on a random character. On the other hand, it may seem that in traditional consumer theory the individual agent is a stimulus-response machine. In the words of Pareto (1971 [1906]: 120), the ‘individual can disappear, provided he leaves us with this photograph of his tastes.’ For a theory which is often called individualistic, there is very little of individual that is needed. The consumption theory based on this view sees the agent as passively responding to price and income stimuli. It has been argued that this point of view goes back to at least Locke and the other British empiricist writers (Rizvi 1992). Locke, for instance, saw the newborn child’s mind as a tabula rasa upon which sensations from the outside world were imprinted. These perceptions were somehow ordered into coherent patterns by a process called the association of ideas. Karl Popper has even termed this the ‘bucket theory’ of knowledge for its insistence on seeing the agent as a passive receptacle for sensations (Popper 1972). Popper contrasts this view with the ‘searchlight theory’ of knowledge which sees the agent as an active learner who explores the environment, and revises previous beliefs in the light of new data. Popper views this presupposition of an active agent to be the outcome of an evolutionary process – it is precisely an active agent who would better be able to survive. The point of view taken in this chapter, in which consumption is seen as a way actively to promote pleasure in the form of arousal levels – Scitovsky’s approach – is consistent with Popper’s searchlight metaphor.

ACKNOWLEDGEMENTS We thank Marina Bianchi and an anonymous referee for very insightful comments on an earlier draft. We also thank seminar participants at the University of Vermont.



NOTES 1. A similar mechanism has been acknowledged by a variety of authors. For instance, Durkheim (1933 [1893]: 235) considers it a ‘truth generally acknowledged today’ (making reference to Spencer, Wundt, Weber and Fechner) that there is ‘pain when the functional activity is insufficient, but excessive activity produces the same effects . . . Pleasure then is situated between those two extremes.’ Also, Gombrich (1979: 9): ‘we must ultimately be able to account for the most basic fact of aesthetic experience, the fact that delight lies somewhere between boredom and confusion.’ 2. Some authors have reported experimental findings which, on the whole, do not find that hedonic satisfaction is a single-peaked function of arousal (Martindale et al. 1990); see also Zajonc (1980). We might add that there are other approaches to aesthetic satisfaction based on psychology, for example see Crozier and Greenhalgh (1992). 3. See in particular, the model of consumption developed by Taylor (1988). 4. We refer interchangeably to activities and goods below; the model may be interpreted to apply to either. 5. Generalization to agent- or activity-specific arousal levels is trivial, and is omitted here only to prevent the proliferation of symbols. 6. This facilitates exposition and programming for simulation purposes without involving any loss of generality. 7. This is a consequence of the assumptions that γ(1) < 1 and π(1) < 1. Together, these assumptions imply that there is a nonzero probability that any subset of Ω having positive measure can be reached in a finite number of steps from any point in Ω. See Sethi and Franke (1995) for further details of a proof along these lines. 8. This has been verified for the present model in the case of four activities. With four activities, one could be neglected while the other three are of roughly equal popularity, or two could be neglected while the remaining two are of roughly equal popularity. This would give rise to peaks in the invariant distribution at approximately 0, 33 per cent and 50 per cent.

BIBLIOGRAPHY Becker, G.S. and Murphy, K.M. (1988) ‘A theory of rational addiction,’ Journal of Political Economy 96: 675–700. Crozier, W.R. and Greenhalgh, P. (1992) ‘The empathy principle: towards a model for the psychology of art,’ Journal for the Theory of Social Behavior 22: 63–79. Durkheim, E. (1933 {1893}) The Division of Labor in Society, trans. George Simpson, New York: Free Press. Gombrich, E.H. (1979) The Sense of Order: A Study in the Psychology of Decorative Art, Ithaca: Cornell University Press.


NOVELTY, IMITATION AND HABIT FORMATION Martindale, C., Moore, K. and Borkum, J. (1990) ‘Aesthetic preference: anomalous findings for Berlyne’s theory,’ American Journal of Psychology 103: 53–80. Pareto, V. (1971) Manual of Political Economy, New York: A.M. Kelley. Popper, K.R. (1972) ‘The bucket and the searchlight, two theories of knowledge,’ in K.R. Popper Objective Knowledge: An Evolutionary Approach, Oxford: Clarendon Press. Rizvi, S.A.T. (1992) ‘Ricardo’s resistance to utility theory and the association of ideas debate,’ paper presented at the Duke University Economic Thought Workshop, 1 October. Scitovsky, T. (1976) The Joyless Economy: An Inquiry into Human Satisfaction and Consumer Dissatisfaction, Oxford: Oxford University Press. Sethi, R. and Franke, R. (1995) ‘Behavioural heterogeneity under evolutionary pressure: macroeconomic implications of costly optimisation,’ Economic Journal 105: 583–600. Solomon, R.L. (1980) ‘The opponent-process theory of acquired motivation,’ American Psychologist 35: 691–712. Solomon, R.L. and Corbit, J.D. (1974) ‘An opponent-process theory of motivation,’ Psychological Review 81:119–45. Taylor, L. D. (1988) ‘A model of consumption and demand based on psychological opponent processes,’ in Paul J. Albanese (ed.) Psychological Foundations of Economic Behavior, New York: Greenwood Press. Zajonc, R.B. (1980) ‘Feeling and thinking: preferences need no inferences,’ American Psychologist 35: 151–75.



11 CONSUMPTION IN POSTMODERNITY Social structuration and the construction of the self Liisa Uusitalo

The purpose of this chapter is to examine how the understanding and dominant narrative of consumption have changed along with diverse conceptions of society. In particular, we should note how the ‘story of consumption’ – like any scientific explanation of activity, for that matter – follows or is strongly influenced by contemporary conceptions of society and social order. The idea of this chapter is to see consumption as a form of socially structuring, as well as identity constructing, activity. Special attention will be paid to the way consumption should best be conceptualized as society enters the late phase of modernity, usually described as the postmodern era. The term modernization refers here to the societal processes which lead towards a differentiated society. Its typical features include the separation of different life spheres (e.g. science, law, cultural life); division of labour; effective, often hierarchical organization structures; and social class division based primarily on position in the production process. By postmodernity is meant not a mixed cultural style but rather the emerging societal conditions that take the form of the de-differentiation and mingling of various life spheres and class distinctions, as well as the disruption of universal norms and ideologies. Before turning to the special aspects of consumption in the postmodern era, let us note some aspects of the present status of consumer research.

ALTERNATIVE DISCOURSES DEVELOPING IN PARALLEL It is normal to classify consumer models according to the differentiation of disciplines into economic, psychological and sociological models, each of them having their own proponents



and patrons. They have hardly established any common core, common issues or standpoints. Economists continue to examine consumption from a neoclassical position as utility maximizing behaviour or time allocation under budget constraints and given preferences. However, among macroeconomists some interest has also been shown in the psychological theory of expectations, and in the question of preference formation through habit formation. Psychology-oriented consumer studies, strongly dominated by cognitive psychology, continue to inquire into the structure of human perception, memory and attitude formation. For the last twenty years, cognitive psychological models have been the most dominant paradigm in consumer studies. Their strong position has been supported since the 1970s by the growth of marketing oriented consumer studies, and their institutionalization by worldwide marketing and consumer research associations. They heavily emphasize aspects of the buying process, especially product and brand choices. Moreover, in economic psychology, a new common interest with economists has emerged in the area of decision-making theory, especially in applying the game-theoretical approach to situations in which consumers have to choose between cooperation and individual utility-seeking behaviour. Sociology oriented consumption studies have developed more or less separately from the above mentioned economic and psychology based approaches and, mostly, also separately from each other. Traditionally, they have conceptualized consumption as determined to a great extent by cultural and social structures. However, in present-day approaches there is a clear shift toward emphasizing the active role of consumption in constructing the actors’ social identity and in constructing social structures. Consumption is also taken to be an indicator of ways of life and taste differences and as a means of social distinction. In all these studies, symbolic or communicative aspects of consumption are important, and consumption activities are often treated in a holistic way. Patterns of behaviour are seen as more interesting than individual product choices. Actually, consumption studies in sociological and cultural research illuminate the development of a change of approach in a whole discipline which has followed the linguistic turn in philosophical thinking. For sociological models of consumption, this means a radical turn away from structurally deterministic models in which consumption was explained in terms of social class, reference group, culture, or some other social background factors. Now the interest focuses on the ways in which people constitute their lives in everyday practice and – by their activity – deconstruct old social structures and reconstruct the new. In this way, consumption has received an important role in social structuration. The above classification of consumption studies clearly shows how societal differentiation, the most typical feature of modernization, is correspondingly reflected in a marked differentiation of various scientific paradigms in consumer research. The three aforementioned main paradigms, each of them further divided into a number of mid-level models of consumption,



have developed in parallel but fairly isolated from each other (for a review of the various paradigms that mainly covers the research conducted in the UK and USA, see Miller 1995). This kind of simultaneous development of several paradigms is certainly not in conflict, for example, with Kuhn’s famous concept of paradigm development (Kuhn 1970). Kuhn does not insist on a single paradigm model of science. He never maintains that one paradigm should cover a whole discipline. Rather, he sees paradigms as governing various specialities of communities of practitioners. However, it is possible that a misinterpretation of Kuhn’s theory has contributed to the separation of paradigms, with every upholder insisting on being ‘the’ paradigm of consumer research and usually rejecting any interdisciplinary discourse. Emphasizing problem centredness and analogical reasoning, some attempts have been made to overcome the barriers between various paradigms (e.g. Laudan 1977; Uusitalo and Uusitalo 1981, 1985). However, truly interdisciplinary discourses in consumer science are still very rare, and they usually fall outside the scope of the established journals of the differentiated paradigms. Also, the quasi-naturalistic models in the economic and psychological consumer paradigms have obstructed the treatment of consumer behaviour using a more holistic, qualitative approach and stressing connections with societal and cultural development.

NEGLECT OF COLLECTIVE GOODS AND DECISION-MAKING While this debate has been going on, many phenomena at the social level have still remained anomalies in the frameworks of consumer research. Many problems demand new interpretations using a more holistic or interdisciplinary approach to consumption and a rethinking of the concept of consumer rationality. For example, awareness of increasing ecological problems has been an important reason to challenge and criticize all existing paradigms for their lack of interest in and lack of tools for addressing the collective consequences of individual consumer choices and activity (Uusitalo 1986). It is not the task of this chapter to develop a theory of ecological aspects of consumer behaviour. However, it can be pointed out that broadening consumer theory in this direction will inevitably also mean revisions in many parts of the present theory. For example, the attitude-behaviour models of consumer behaviour need revision. These models have been developed to explain choices of private goods, and not choices of collective goods that often involve a choice between cooperating and free riding. Likewise, new methodological approaches are needed. For example, game theoretical approaches are far more relevant for understanding choices of collective goods than choices of private goods.



More generally, the role of consumers as users or coproducers of common goods, such as environmental quality or cultural atmosphere, has long been a neglected topic of research. It is possible that when libertarian and individualistic attitudes gain in prominence, consumers’ willingness decreases to contribute to ‘cooperative games’ in favour of collective goals. Instead, consumers may increasingly strive only for immediate individual utility and pleasure. This may also produce crucial practical questions for the future development of the social welfare states: how to justify environmental and social programmes or public spending on culture if consumers’ willingness to contribute through taxes or by their own cooperative action is decreasing; how to create new behavioural norms in favour of environmental quality if consumers wish to pursue only their private interest. The discourse on consumers’ willingness to act as coproducers of collective goods is closely connected with one of the fundamental philosophical issues of today’s social theory. The issue is whether it is at all possible to define and strive for common goals, principles and norms by way of rational public discourse. The first basic question here is whether it is possible to reach consensus about common goals or universal values in a highly differentiated society, and by what processes just agreement can be reached. Second, it has been questioned whether such universalism is at all necessary. For example, Habermas (1981) and Lyotard (1984) present two mutually opposing views. Habermas represents a view in favour of discursive ethics and procedural justice. We can and should strive for a consensus on some universal goals and principles in an open discourse in which each participant is assumed to follow the rules of rational argumentation. This means that, for example, universal norms have to be justified by evaluating their rightness for all, not only based on how they favour the person or lobbying group in question. Lyotard rejects the whole idea of striving for a consensus. He represents the view that (postmodern) society is constituted by a number of incommensurable language games. Their function is not to build up consensus but rather to break it down, create ‘dissensus’, and in that way bring up new ideas. Pragmatists such as Richard Rorty (1989) reject the idea of universal goals and norms, not because it will be difficult to reach agreement about them but because – as pragmatists claim – open discourse is sufficient for democratic development. There is thus no need for universally shared common values. All contemporary Kantians represent an opposing view: some common moral principles are certainly needed. However, it is often emphasized that securing procedural justice (via an open rational discourse, as Habermas proposes) is necessary but alone not a sufficient element for reaching collective goals in society. More communitarian theorists add that selfhood, virtues and virtuous behaviour (or social norms) are often learned from tradition or local practices and institutions such as professional practices, schools and families (e.g., MacIntyre 1987; Taylor 1989).



In the economic and social debate, these philosophical problems are often captured by the ongoing argument between the various proponents of extreme market liberalism and those representing a more communitarian view. They first emphasize the freedom of individual choice and the importance of the market mechanism. This approach faces problems in treating collective goods, for example, treating negative environmental external effects for which a market cannot be created. In contrast, the communitarian approach emphasizes the importance of collectively shared values and norms as prerequisites for collective goods. This approach involves the problem of whose values should be accepted if – as is often the case – no attention is paid to the procedures of public argumentation and discourse (see for example, Walzer 1990; Bellah et al. 1992; Etzioni 1993). It is apparent that the philosophical literature contains very divergent views of decisionmaking concerning collective goods in society. These ideas will also have an effect on the theory of consumer choice, especially concerning choices associated with collective goods. They suggest a rethinking and broadening of the concept of consumer rationality as well as studying its conditions (Uusitalo 1989, 1997). Such new research topics in consumer theory are closely connected to more general studies of rational decision-making and societal rationality (e.g. Kahneman and Tversky 1979; Sen 1973, 1982; Elster 1979, 1983; Habermas 1981; Margolis 1982; Tversky and Kahneman 1986). Consumer choices are thus relevant beyond the narrow individual sphere.

THE STRUCTURING CAPACITY OF CONSUMPTION Another type of critique of consumer models comes from both the structuration theorists and interpretationist approaches. By these terms I intend various forms of hermeneutic approach which try to understand consumers and treat them as depending on, but also actively modifying, external conditions and structures. Consumption in post-industrial societies – perceived as consisting of a broad scale of activities including cultural and leisure activities – will have an increasing role in cultural identification and the structuring of society. These aspects have been neglected by traditional consumer choice or decision-making models. The role of consumption in social identification and structuration will become more important, because previous societal class differentiation is becoming more and more diffuse and work is gradually losing its (class) structuring and integrative capacity. On this view, society as a whole will be increasingly structured around consumption instead of production. Society’s economic prospects will be increasingly dependent on con-



sumer demand. This has not been fully realized by contemporary economic policy-makers. For example, after recessions in the early 1990s, some countries have experienced a lack of growth in consumer demand, despite consumers’ rational expectations that the economy at large would recover. This reluctance to consume was due to consumers’ mistrust of any improvement in their own personal employment prospects and has seriously postponed recovery (Uusitalo and Lindholm 1994). As Baudrillard (1988) has suggested, as workers citizens can be replaced, but as consumers they are irreplaceable. As consumers they exert what he calls ‘consumption power’: this means that the economy is dependent on consumers’ decisions to spend or to save at just the times needed to keep economic growth in balance. Consumption is also a structuring force in that it intermingles with production, especially in the service sector. The division between production and consumption will be increasingly blurred, which emphasizes the structuring role of consumers in the production process. Many service production units are mingling, or even dissolving into, consumption activities (for example art services such as exhibitions and performances, or services for the strollers in a shopping centre). Production and consumption are taking place simultaneously (Shields 1992). Consumers who stroll in shopping malls, cafeterias or music festivals form an integral part of the consumption experience of other consumers. The mingling of consumption and production activities is perhaps the most typical form of ‘postmodern’ de-differentiation taking place in consumer services. Another new form of de-differentiation is the difficulty in distinguishing products from their marketing. Product images created with the help of marketing and advertising become an integral part of all designed and branded products. Earlier, we used to think of social structures, for example, social class, as determinants of consumption. However, the new idea of consumption as a structuring and integrating force in society means that consumption and activity patterns will create new social structures and not vice versa. Consumption in the form of cultural taste differences (Bourdieu 1984), illusory and hedonistic consumption ideals or schemes (Campbell 1987; Schulze 1992) or way of life patterns (Uusitalo 1979) will create new social bonds and feelings of belonging to or of distinguishing oneself from a group. The resulting structures, the new ‘communities’, are often very temporary in character. Moreover, they may be based on merely imaginary aspects conveyed by the mass media. Similar tastes or consumption patterns do not necessarily involve actual interaction between the consumers. It has been claimed that these shared consumption interests and tastes – especially in the area of leisure activity and media – and the imagined affective bonds thus created between people are the only integrating forces left in post-modern societies (Maffesoli 1993).



As such, there is nothing new in claiming that consumption has a structuring and integrating role in society. However, consumption as a socially integrating force worked quite differently in traditional societies, where the ‘correct’ shared ways of life used to be stable and unquestionable, or in societies of economic growth, in which consumption often acted as a sign of class identification or as a vehicle of social mobility. The social bonds were strong and depended mainly on one’s position in the production sphere, and not on consumption. In present post-traditional culture, the social bonds based on consumption are vague or ambiguous. Even if people share tastes and consumption patterns, they also often shift between them or have several interests in parallel. The spatial pattern or dimension of consumption is also important as a structuring element. Consumers often share physical spaces of consumption such as concert or exhibition halls, tourist resorts, sports activity sites, shopping malls and cafeterias, without really having any actual social interaction. This type of coexistence is called sociality by British social theorists and often replaces former types of socializing (Shields 1992). Whenever considering the integrating function of consumption in present-day society, we should thus remember that the ties created are loose and not necessarily strong enough to create any kind of enduring cooperation (for example, for the purpose of environmental protection). However, sometimes consumers’ sociality and gathering to express emotions can have collective consequences. Thus gathering that starts without having any definite purpose can lead to some common cause. This is typical of many current social movements. They may start with demonstrations having mainly an expressive value for the participants, but come to involve a greater number of participants and find more clearly designated common goals. This was probably the course of development in the worldwide demonstrations against the French nuclear tests on Mururoa Island in 1995. An example in the political field is the unification of West and East Germany, which was also an escalatory process. At the beginning there seemed to be no overarching rational political discourse on the unification issue. Everything started from spontaneous demonstrations, that is, from affective sociality. Hirschman (1995) has described this escalatory process from the point of view of the two main channels of influence that consumers can employ (exit and voice). The process in East Germany started from affective ‘we want out’ demonstrations (emphasizing the exit option) and developed into a mixture of opinion that now also included ‘we want to stay in’ demonstrations, expressing the wish for profound changes in society (thus emphasizing the voice option). The joint development finally led to a demand for unification. Nevertheless, there can also be dangerous consequences if this type of loose ‘affective sociality’, based on shared emotions, increases in importance and rational argumentation (rational public discourse and choice of common goals) is totally dismissed. For instance,



affectivity can take on forms of destructive collective behaviour (gang criminality, vandalism, environmental destruction, not to mention fanatical nationalistic or racist political movements). Excessive free riding can also increase because short-term affective sociality cannot usually create a sufficient level of commitment or cooperation. Among social theorists then, there are high expectations that consumption can and will act as a structuring and integrating force in postmodern society. This reflects the fact that other structures (class structure) or institutions (shared social norms and traditions) are gradually dissolving or diminishing in importance. However, as I have tried to show, resulting structures and social bonds that are based on imaginary taste similarities or shared affective behaviour are not strong or lasting. Therefore, consumption may merely act as a weak integrating force in society.

THE SOCIO-SEMIOTIC MEANING OF CONSUMPTION Consumer as self-constructor Another feature that has been emphasized in contemporary literature is consumption’s role in constructing the self, that is, its role in individual identity building. Each individual holds a plurality of social worlds or roles which are often connected to some leisure or cultural/taste groupings. Consumers can become attached and belong to several of them simultaneously with a varying degree of intensity (Strauss 1978; Becker 1982). They can easily join new social worlds and leave others. If this pluralism and the ensuing constant exchange of consumer roles is to be taken seriously, the theories of distinct consumer types must also be questioned. It becomes extremely difficult to identify any theoretical ideal types of consumption or consumers, whether classified as ‘styles’, ‘schemes’, ‘taste groups’ or ‘segments’, unless on a very broad, aggregate basis. Perhaps this difficulty in classifying postmodern consumers at the macro level according to their taste or behaviour is why recent consumer studies, including those within the social scientific approach, have become so interested in the mental aspects of consumption: the social, semiotic or even psychoanalytical meanings of consumption. This analysis of meanings, instead of looking at structural patterns, is often an individual one, but the individual is seen in a social context. We can give some examples of such characterizations found in the literature. For instance, Giddens (1991) has suggested that change in the meaning of some forms of consumption



behaviour – he refers especially to addictive consumption – is a consequence of the emerging post-traditional society. When external social norms and traditions are eroding, individuals are left to handle all the information around them and to refect on the ethical side of their choices. In post-traditional society, when reflectivity thus becomes an individual project, individuals try to create their own ‘inner traditions’ to guide their behaviour. This can lead to various addictions (drugs, alcohol, eating, work, sport). Hence, Giddens also sees some negative consequences of consumers being released from normative pressures and traditions. In contrast, other writers who usually favour extensive market liberalism, consider this primarily as a positive development towards increased autonomy and self-determinacy in consumers. Noro (1995) has collected the various socio-semiotic meanings that consumption has for the individual under the headings of four main functional figures:

• •

the self-demonstrator (status or distinctive taste displayer); the self-completing figure (a notion based on psychoanalytic theories of oral consump-

tion); the self-signifying player (sovereignly playing with and combining different signs and

roles); the self-producer (whose own life and work are constructed as ‘a product of art’).

However, all these types contain one and same idea of self-construction looked at from slightly different angles. The oldest way of looking at the social meaning of consumption for the individual is perhaps the demonstrative function (of the self). For Veblen (1899), consumption was a demonstration of (the nineteenth century aristocracy’s) status through idleness: spending time in other activities than work. For Bourdieu (1984), it takes the form of a more hidden social distinction of the upper and the ‘new middle classes’ from less-educated mass consumers with the help of acquired cultural competencies and more developed, differentiated tastes. Bourdieu also describes the dynamic of this distinction process: a higher social status in a cultural field can be achieved by bringing new tastes from the margin to the centre. Earlier, Hirsh (1977) presented similar ideas about the importance of striving for relative positions in society. He used the concept of positional goods: the broader the distribution of certain goods or tastes, the less satisfaction they give to the individual consumer and the less they possess distinctive capacity. Consequently, the enjoyment of goods depends on one’s position relative to others. Some goods are enjoyable only insofar as they remain rare (compare Guerzoni and Troilo, Chapter 9). Examples of such goods are collectors’ items or exclusives such as tours with unknown destinations. As soon as the majority of consumers start acquiring them, physical congestion will take place, as in the case of car driving, popular tourist resorts, etc. Or alternatively, there will be social congestion. Goods/ tastes will lose their



capacity to distinguish the person from those with a less differentiated and refined taste as in the case of cultural products. At least some aspects of cultural change, for example, changes in cultural tastes or the pursuit of better education to secure upward mobility, are thus due to the search for new items or signs of social distinction. One feature of this dynamic is that especially the new middle classes – the new cultural intermediaries – who often work in media, fashion or art occupations, seek to bring totally new aspects from the cultural margin to the centre. Thereby, they try to improve both their own position and the position of the new cultural genre they promote (Featherstone 1991). For example, many things originating in ‘lower’ or popular culture have been upgraded into high culture through such a process (photography, detective novels, comic books, rock and folk music). It has also been suggested that women may take special advantage of this type of dynamic for attaining upward social mobility. Cultural competencies acquired particularly by additional education and cultural activities, but also by taking care of one’s health and body, or a creative interest in the aestheticization of everyday life (clothing, fashion, home decoration, art crafts, cooking skills, children’s culture) can all also have social consequences for women as a gender. These competencies may help women to improve their position in that they succeed in making those areas of life more valued than before. The basic human striving for social distinction postulated in Bourdieu’s theory has also received much criticism. I suggest, therefore, that the theory of social distinction should be interpreted as closer to an idea of the consumer as self-producer than as self-demonstrator. For example, several studies of cultural consumption show that consumers often prefer solitary experiences because they want to learn to understand art. They attend art exhibitions more with a view to self-production and acquiring new competencies than to demonstrate their distinctive taste or position, as compared with consumers who do not have art interests (Uusitalo and Ahola 1994). The whole idea of openly demonstrating one’s distinct position does not fare very well, for example, in the Nordic democracies, where education systems have long tended towards equal opportunity and where demonstration of distinctiveness often violates the common norm of behaviour. Bourdieu’s theory of social distinction has thus been much criticized there. However, if we interpret the dynamic behind social distinction to refer also to consumers’ conscious self-production or self-completion instead of self- or class demonstration, the theory can be accepted more easily. Consumption patterns may thus be regarded as indicative of both self-identity and of how consumers wish others to regard them. They can also be used to improve the actor’s position in the social hierarchy or to break old hierarchies and bring in novel criteria for new ones.



MODERN HEDONISM: IMAGINARY CONSUMPTION One of the most interesting contributions to the new approaches to consumption comes from Campbell (1987), whose historical analysis reveals the neoromantic character of presentday consumer hedonism. Campbell claims that the middle classes have always been the carrier of the romantic consumerist tradition. The taste and aesthetic experience of the bourgeois middle class were based on the same sentimental ethics of inner experience as their religious feelings of sympathy and pity. While, in the last few centuries, the upper class regarded taste as the application of established strict standards, the middle class considered taste to be a sign of the mental capacity to enjoy beauty and react to it emotionally (Campbell 1987: 205). Imagination and feelings have thus always accompanied consumption. However, a typical feature of the late modern consumer society is that consumption increasingly becomes a pursuit of pleasure. Seeking pleasure means that consumers no longer seek ‘utilities’ in the products nor evaluate product attributes but rather try to expose themselves to various types of stimuli to create emotions. These pleasurable emotions can also be created on a purely imaginary basis, for example, when, with the help of advertising or other media, the consumer fantasizes or daydreams about the product or activity. The consumer can ‘consume’ the product on an illusory level perhaps without any buying intentions, by looking at home decoration magazines, product catalogues or travel brochures. What makes illusory consumption so characteristic of our time is the increasing capacity of electronic media to provide visual stimuli for this purpose. Nowadays, it is possible that surfing on the Internet comes closer to illusory consumption than information seeking in the traditional sense. Campbell also makes a distinction between traditional and modern hedonism. Traditional hedonism can be considered as the pursuit of sensual pleasures obtained from the consumption activity itself: eating good food, listening to music, and so on. Modern hedonism, by contrast, takes place on the illusory level and is very much based on aesthetic visual images and anticipation of consumption. Sometimes, illusory consumption can be more pleasurable than real consumption. Consumers’ disappointment with the satisfaction of acquiring the real product as compared with the great expectations on the illusory level is one explanation of consumer dynamics. It could be the reason why consumers constantly expose themselves to new items in the hope of experiencing emotional pleasure. To this may be added that modern illusory hedonism closely depends on the mediatization and symbol producing capacity of society. A society or some of its subsystems, for example, a distribution system of goods and services, will also have to satisfy the needs of symbol production. If the system is very poor in the stimuli needed for this type of modern hedonism, it will cause dissatisfaction even if the basic functional needs are satisfied.



It has often been claimed that such dissatisfaction was one reason for the upheaval of the socialist systems in eastern Europe. But also in western capitalist countries, a ‘reculturalization’ or ‘aestheticization’ process of production and distribution systems will be needed. Both private and public providers of products and services will have to pay much more attention to symbolic and aesthetic elements, and not only to cost efficiency. Product and package design, advertising aesthetics and design management of corporations (company emblems, architecture, environments) are becoming more important than ever before. Above all, one can expect to see the increasing aestheticization of distribution systems that daily provide sites for modern imaginary hedonism. Important in the shopping experience are not only the interior architecture of shops and the appearance of the personnel, but also the aesthetics and care of the whole environment, including the shop’s surroundings. In brief, the theory of modern consumer hedonism directs much attention to the aesthetics and differentiation of the systems of provision and mediation, both as a prerequisite and as a consequence of imaginary consumption. The processes of social distinction in consumption are also connected to this development, because places of consumption and media products are now also included in consumers’ tools of social distinction or self-construction. Subcultures and distinctive taste groups are gathered around certain types of cultural genres including, say, restaurants or other distinctive symbolic places. The ‘genius loci’, the overall images of various places, become important signs in the consumer differentiation and distinction processes.

THE STRUCTURE VS. AGENCY DILEMMA The above examples show that many processes which serve self-constructing purposes at the individual level can have a socially integrating or structuring function at the aggregate level. This fact has been ignored in many theories of postmodern sign culture, such as Baudrillard’s theory of simulacra, or in theories that primarily concentrate on describing the new illusory consumer hedonism and explicate consumption only as ‘text’ or ‘language’. These studies are interested only in consumption’s communicative and demonstrative functions from an individual consumer’s point of view, but not in its socially structuring function. The view proposed here of the double role of consumption in self-construction and social structuration can also shed some light on an old dilemma in the social scientific explanation of behaviour, the structure vs. agency dilemma. The entire history of consumer theory in the social sciences can be seen as a pendulum movement between structure and agency explanations. The question is whether consumer behaviour can best be explained by structural conditions of behaviour, or by consumers’



independent, individual decision-making. Neoclassical economics and behavioural psychological consumer research both posit a sovereign consumer who makes individually rational decisions independently from other consumers. In contrast to this, the Frankfurt critical theory and Galbraithian political economy see consumers as determined by the market and the mass production of goods and services, with a rather limited range of individual ‘free’ choice. Some postmodern theories, again, assign to consumers a total freedom from market forces or social structures. Consumers are conceived as playing with market signs. They construct their own meaning for every single product and activity. (For a critique of this somewhat romanticized view of consumers in the reception of cultural and media products, see Morley 1995). The delight in finding the ‘active consumer’ is great, and taken to extremes in the analyses of consumers as sovereign ‘shoppers’ or ‘tourists’ in the market, or ‘counterculture meaning-creators’ when they watch television programmes or advertising. This optimistic idea of the all-empowered active consumer has been backed by the belief that new information technologies and new urban planning provide proper means and places for the expression of this new kind of consumer liberty. Only lately has attention been paid to the many problems inherent in this view, such as the uneven distribution of access to information technologies, as well as the segregation of urban areas (e.g. Sassen 1995). Consequently, consumers are admittedly active and free to choose, but this should be seen as only part of the truth. The assumption of a total freedom (or total indifference) of consumers is as idealistic as the theory of free competition and laissez-faire economy. Before a consumer can start to play with the signs, the signs will be produced by often highly structured and concentrated industries, media and retail chains. Fine and Leopold (1993), therefore, have suggested a vertical analysis of consumer behaviour that pays attention to the ‘system of provision’ behind consumer activity. Such an analysis should, however, be product specific, since the system of provision varies greatly from one type of product to another and no universal model is valid for all types of consumption. Finally, we can claim that consumers are and always will be interdependent, even in a posttraditional society with loose social bonds and vague commitments. First, consumers do still note what their fellow consumers are doing. They consciously and unconsciously adopt ideas from others and from the media when constructing their individuality and consumption patterns. Moreover, the scarcer public goods become such as a clean environment, the more profound will be the interdependence with other consumers and the political system in the satisfaction of corresponding needs. Also, the increasing aestheticization of life will accentuate rather than eliminate social interdependence because, in producing aesthetic environments, cooperation is needed from everyone.



However, it must be noted that this interdependence is more or less hidden from consumers and no longer takes the form of strong normative pressures as it did in hierarchical class societies. Especially if affective communities take the place of real communities, there will be no sanctioning mechanism to back up normative pressure. At the individual level, this has been seen predominantly as a relieving and positive development. On the other hand, the erosion of social norms increases the individuals’ needs and responsibility to create their own norms of behaviour. Not all people are capable of this kind of constant self-reflection, as we saw earlier in the example of addictive behaviour. From the point of view of producing collective goods or creating cooperative behaviour, social norms are still of great importance and often necessary in supplementing individual (economic) incentives. Therefore, a balanced theory of the active consumer should draw on Giddensian structuration theory (Giddens 1984): structures and social norms do influence the range of choices that is available and considered good, but consumers still have a considerable autonomy in their activity and choice based on personal preferences and tastes. Frequently, structures may be viewed as ‘enabling’ rather than as strict determinants of behaviour. Moreover, as stressed by the theory of collective goods or the social distinction theory, consumers can actively influence existing structures and institutions by their choices and by using their voice. The structuration approach thus tries to avoid one-sided social determinism as well as empty descriptions of consumers as individual flâneurs or sign players. For choice theorists, in turn, structuration theory stresses the interplay of structures with autonomous intentional choice. Even when perceiving consumers as autonomous constructors of their own way of life, it is also justified to look for invariances or patterns of behaviour and to try to understand their cultural background. As several empirical studies show, we can find shared consumption patterns and tastes that are socially determined (Uusiralo 1979; Bourdieu 1984) or groups of consumers who, due to their background, have similar expectations and reactions, for example, to a recession (Uusitalo and Lindholm 1994). These studies apply a structuration approach. Several anthropological and ethnographic studies also seem implicitly to rely on such an approach. For instance, it has been found that when elements of global cultures are imported to traditional communities, they are given totally new local meanings by the inhabitants (Miller 1992). It can be concluded that consumer behaviour should be seen as something between the two poles. It has elements both of autonomous self-construction of life and of social dependence. In empirical analyses, however, one has to start more or less from one pole only. For example, one can be interested in the relative importance of social background factors in explaining specific consumption behaviour without believing that these factors will explain all the variance in behaviour. Or one can be interested in consumers’ individual preferences or distinctive strategies without believing that they have emerged in a totally autonomous way.



SHIFTS IN THE NARRATIVE OF CONSUMPTION AND THE IDEAL CONSUMER As suggested above, postmodern consumption is actually no more self-determined and autonomous than was modern consumption. Neither can we claim that postmodern consumption is a great deal more communicative and social in character than previous forms, because consumption in traditional, agrarian and industrial growth societies already had many social, communicative and socially constitutive functions. Consumption has always been an integrative force, in an even stronger sense than at the present time when normative pressures have weakened and consequently the affective bonds and commitments created in the field of consumption are rather weak. Ultimately, it seems that it is the ‘story of consumption’, rather than actual consumption patterns or the actual functions of consumption, that has changed along with the concepts of society. This narrative always deals with an abstract, ideal type of consumer, in a Weberian sense, reflecting different concepts of the human being. We can summarize the story of consumption as having undergone the following main shifts. First, we had the story of a rationally choosing, utility maximizing individual consumer, which is obviously a typically ‘modern enlightenment narrative’. Gradually, critical theory changed the conception into a structuralist view of a consumer (over)determined by social structures. Later on, in the poststructuralist view and Foucaultian sense, structures took a much more hidden and non-recognizable position. The structures, the logic of culture, or institutions such as social norms are considered as internal; they think within the consumer rather than being shown as measurable external influences. After the linguistic turn in the social sciences, when communicative and aesthetic elements began to be emphasized, consumers became style and taste managers and constructors of their own life project. Emancipatory elements are emphasized in the – basically still structural – theories of social distinction and life styles. Through their activity, consumers can influence structures as well as their own position. Finally, when approaching the postmodern world of hyper-reality and simulacra (if Baudrillard’s concepts are to be used), consumption is presented as some kind of mentality. Consumption takes place in the fantasies of the fragmented consumer, as in the case of imaginary consumption. In postmodern society, social structures are replaced by weak emotional sociality and structural influence takes the form of (visual) seduction. Postmodern consumers are mostly conceptualized as autonomous sign players or self-constructors who are indifferent, consciously distancing and detaching themselves from any serious collective commitment. They hold a cool or ironical attitude towards the rest of society. According to



the detachment hypothesis, consumers often act as if they were tourists in their own society (Bauman 1993). The ideal postmodern consumers are also ascribed the capabilities of self-reflection and meta-analysis of their own preferences, both being necessary in a society where social norms and traditions are fading away. Consumers are also conceived of as being able to reinterpret and accommodate signs to their own culture. Actually, the postmodern consumer thus conceived is paradoxically also somehow extremely modern: autonomous, self-reflective and in that sense very rational. There is no reference here to utilitarian, economic rationality in the traditional sense but to the ability to reflect on one’s own preferences which is typical of modern, enlightened persons. We are precisely where the story started in the first place.

CONSUMPTION UNDER POSTMODERN CONDITIONS If we can in any way claim that a shift towards postmodern consumption has taken place, we should perhaps not so much try to describe the postmodern consumer as a distinct ideal type, but rather try to study in more detail to what degree people’s activities in consumption reflect the shift of society from modern to postmodern or late modern. More precise hypotheses are needed on how consumption is affected by the postmodern condition. These can be based on theories of postmodernity. For example, it could be assumed that consumption patterns and life styles should reflect the shift from a modern differentiated society towards not only a pluralistic, but especially a de-differentiated society, in which old borders and categories are transcended and hierarchies are replaced by loose network relations (e.g. Lash 1990). Postmodern consumption activities may also be assumed to be indicative of changes in the type of reflectivity, from public discourse and universal rationalization towards individual self-reflectivity and the privatization of morality. The consumer’s decreasing participation and commitment to collective causes should in some way reflect this focused morality of the fragmented self in which the subject does not act as a ‘total person’ but is partially interested in several issues, the focus shifting from one issue to another (Bauman 1993: 198). A typical feature of postmodern culture is the emphasis on authentic experiences and the authenticity of oneself (Taylor 1991). We should, therefore, also explore what forms this emerging ethics of authenticity takes in the field of consumption. How does consumption reflect the universal need for the recognition of difference on the social level, and the need for



identity forming relationships in the intimate sphere? Both needs will lay heavy emphasis on the aesthetic and the affective as socially incorporating elements. Above all, consumption activities should more than before be examined and interpreted against the background of the development towards mediated reality and mediated participation, one of the most typical features of the postmodern condition. Therefore, some comments are needed to emphasize and describe this mediation and the resulting new type of publicity.

THE CONSUMER IN MEDIATED REALITY As stated earlier, consumed goods increasingly consist of simulations. Instead of representing any real thing, they build up a world of simulacra that becomes a new reality which has lost its reference to the original. Disneyland’s mini-Norway becomes the real tourist attraction, shopping centre ‘Italian piazzas’ are taken as the ‘real’ piazzas, and New York, Dallas or American life as represented in television serials and advertisements become the real USA for consumers in other parts of the world. This is Baudrillard’s world of hyper-reality, in which signs refer to other signs. Citizens’ recognition of both private and public events is also mediated by television and the press, and thus private domestic settings have become the main public site nowadays. Although the media offer some possibilities for interaction, these are very limited and thus merely a kind of quasi-interaction (Thompson 1990: 238–54). Extended mediatization takes place when recipients or the media pick up the mediated messages for further discursive elaboration. Thus, a circle of secondary recipients is formed who have had neither direct contact with the original event nor with the event as it was mediated in the first place. This is Thompson’s way of describing the world of hyper-reality. He also takes up the new demonstrative publicity as a new form of political participation. Demonstrative publicity was discussed by Habermas (1962) and refers to artificial publicity, a practice of strongly personified lobbying and efforts to reach the attention of the public by ‘knocking out’ possible competitors in the media. This personified self-demonstration has often come to replace genuine argumentative discourse. Habermas considered it mainly as a disruption of bourgeois publicity, taking place as a consequence of the entwinement of state and society in the phase of late, organized capitalism. The conditions of ideal argumentative publicity are vanishing and being replaced by the new demonstrative publicity. However, Thompson represents a slightly more optimistic view of the role of demonstrative publicity than Habermas, while claiming that it can to a certain degree act as a surrogate



for individual participation, that is, demonstrative publicity can also be conceived to include those persons who act on behalf of the ‘distant others’ (silent audiences) for some common cause. Social movements such as Greenpeace can be cited as instances. They have adapted themselves to the new means of political action and demonstrative publicity, in which the management of visibility and affection is most important. Still, the affective attachment and solidarity of television viewers is not concentrated on the persons but rather on the (concrete) collective task which they represent. As to consumption patterns, demonstrative publicity tends to personify and iconicize certain consumption and life styles as well as cultural genres. Not only in commercials but also in other television programmes (for example, in TV series) are personal relations constituted through products or framed in various stereotypical product surroundings. These simulated life styles and life settings can act as ‘totems’ or signifiers to consumers. The visual sign or picture represents the collective understanding of oneself and society and acts as a call to join the affective, imaginary community (Maffesoli 1993; Leiss et al. 1986).

HAVE WE EVER BEEN POSTMODERN? Latour (1993) has strongly criticized the tendency to interpret social phenomena in reductionist ways from one single point of view and not to see them as networks that include elements of natural–scientific, semiotic and collective characteristics. The same error can occur when trying to understand consumption phenomena: consumption treated as purely rational behaviour or – as in postmodern theory – as purely symbol producing, mental activity. We have listed possible tendencies towards postmodern consumption. Still, it is extremely difficult to devise empirical tests for any of the aforementioned propositions about the features or meanings of consumption under the postmodern condition, because both external behaviour and the mental processes of consumers should be included in the analysis. So far, conclusions about postmodern tendencies in consumption have been based on the study of signs (advertisements, TV programmes), not on consumers’ interpretation or how their behaviour has in fact been changed. Were we to hold a more sceptical attitude towards the whole theory of postmodernity, there would be even more profound reservations. Perhaps imaginary consumption, affective communities and consumers’ new self-reflectivity and self-construction are just artefacts for which only partial testable support can be found in everyday practice. Perhaps the majority of consumers are not that interested in, or capable of, playing with signs but continue to make choices mostly on the basis of economic constraints. Perhaps people still crave for real



communities and engagement instead of the imaginary, and prefer argumentative to demonstrative publicity. The theories of the symbolically rich consumption and ‘baroque culture’ of postmodernity were formulated in the top upswing of the post-World War II economic growth cycle. Ideas of postmodernity were based on great hopes for continuous economic growth along with the emergence of a service and information society. If this development has partly come to a halt, will postmodernity also receive a more modest interpretation? Instead of a totally new epoch or time period, it could then represent a reflective attitude: a disillusioned way of looking at the limits of modernity and enlightenment ideology, but without abandoning its core.

REFERENCES Baudrillard, J. (1988) ‘Consumer society’, in M. Poster (ed.) Jean Baudrillard: Selected Writings, Oxford: Polity Press, 29–56. —— (1994) Simulacra and Simulations, Ann Arbor: University of Michigan Press. Bauman, Z. (1993) Postmodern Ethics, Oxford: Blackwell. Becker, H. (1982) Art Worlds, Berkeley: University of California Press. Bellah, R. et al. (1992) The Good Society, New York: Vintage Books. Bourdieu, P. (1984) Distinction. A Social Critique of the Judgement of Taste, London: Routledge and Kegan Paul. Campbell, C. (1987) The Romantic Ethic and the Spirit of Modern Consumerism, Oxford: Blackwell. Elster, J. (1979) Ulysses and the Sirens. Studies in Rationality and Irrationality, Cambridge: Cambridge University Press. —— (1983) Sour Grapes: Studies in the Subversion of Rationality, Cambridge: Cambridge University Press. Etzioni, A. (1993) The Spirit of Community, New York: Touchstone. Featherstone, M. (1991) Consumer Culture and Postmodernism, London: Sage. Fine, B. (1995) ‘From political economy to consumption’, in D. Miller (ed.) Acknowledging Consumption. A Review of New Studies, London: Routledge: 127–63 Fine, B. and Leopold, E. (1993) The World of Consumption, London: Routledge. Giddens, A. (1984) The Constitution of Society, Cambridge: Polity Press. —— (1991) Modernity and Self-identity. Self and Society in the Late Modern Age, Cambridge: Polity Press. Habermas, J. (1962) Strukturwandel der Öffentlichkeit, Newied und Berlin. —— (1981) Theorie des kommunikativen Handelns, I-II, Frankfurt am Main: Suhrkamp. Hirsh, F. (1977) Social Limits to Growth, London: Routledge and Kegan Paul. Hirschman, A. (1995) A Propensity to Self-subversion, Cambridge, MA: Harvard University Press. Kahneman, D. and Tversky, A. (1979) ‘Prospect theory: an analysis of decision under risk’, Econometrica 47: 263–91.


CONSUMPTION IN POSTMODERNITY Kuhn, T. (1970) The Structure of Scientific Revolutions, Chicago: University of Chicago Press. Lash, S. (1990) Sociology of Postmodernism, London: Routledge. Latour, B. (1993) We Have Never Been Modern, London: Harvester Wheatsheaf. Laudan, L. (1977) Progress and its Problems: Towards a Theory of Scientific Growth, London: Routledge and Kegan Paul. Leiss, W., Kline, S. and Jhally, S. (1986) Social Communications in Advertising, New York: Macmillan. Lyotard, J-F. (1984) The Postmodern Condition, Minneapolis: University of Minnesota Press. MacIntyre, A. (1987) After Virtue. A Study in Moral Theory, 2nd edn, London: Duckworth. Maffesoli, M. (1993) La contemplation du monde. Figures du style communautaire, Editions Grasset & Fasquelle. Margolis, H. (1982) Selfishness, Altruism and Rationality: A Theory of Social Choice, Chicago: University of Chicago Press. Miller, D. (1992) ‘The young and the restless in Trinidad’, in R. Silverstone and E. Hirsch (eds) Consuming Technologies, London: Routledge. Miller, D. (ed.) (1995) Acknowledging Consumption. A Review of New Studies, London: Routledge. Morley, D. (1995) ‘Theories of consumption in media studies’, in D. Miller (ed.) Acknowledging Consumption. A Review of New Studies, London: Routledge: 296–328. Noro, A. (1995) ‘Uudemman kulutussosiologian mallit ja figuurit’ {Models and figures of contemporary consumer sociology} Sosiologia 32(1): 1–11. Rorty, R. (1989) Contingency, Irony and Solidarity, Cambridge: Cambridge University Press. Sassen, S. (1995) ‘Power and marginality in cyberspace’, paper presented at the conference Culture and Identity: City, Nation, World, Berlin, 10–14 August. Schulze, G. (1992) Die Erlebnisgesellschaft. Kultursoziologie der Gegenwart, Frankfurt am Main: Campus. Sen, A. (1973) ‘Behavior and the concept of preference’, Economica 40: 241–59. —— (1982) Choice, Welfare and Measurement, Oxford: Basil Blackwell. Shields, R. (ed.) (1992) Lifestyle Shopping: The Subject of Consumption, London: Routledge. Strauss, A. (1978) ‘A social world perspective’, Studies in Symbolic Interaction 1, 119–28. —— (1989) Sources of the Self. The Making of the Modern Identity, Cambridge: Cambridge University Press. Taylor, C. (1991) The Ethics of Authenticity, Cambridge, MA: Harvard University Press. Thompson, J. (1990) Ideology and Modern Culture, Cambridge: Polity Press. Tversky, A. and Kahneman, D. (1986) ‘Rational choice and the framing of decisions’, in R. Hogarth and M. Reder (eds) Rational Choice. The Contrast between Economics and Psychology, Chicago/London: University of Chicago Press: 67–94. Uusitalo, L. (1979) ‘Consumption style and way of life: an empirical identification of and explanation of consumption style dimensions’, Acta Oeconomica Helsingiensis A: 27. —— (1986) Environmental Impacts of Consumption Patterns, Aldershot: Gower. —— (1989) ‘Economic man or social man – exploring free riding in the production of collective goods’, in K. Grunert and F. Oelander (eds) Understanding Economic Behavior, Dordrecht/Boston/ London: Kluwer Academic Publisher, 267–283 —— (1990a) ‘Consumer preferences for environmental quality and other social goals’, Journal of


CONSUMPTION IN POSTMODERNITY Consumer Policy 13, 231–51. —— (1990b) ‘Are environmental attitudes and behavior inconsistent? Findings from a Finnish study’, Scandinavian Political Studies 13(2): 211–26. —— (1997) ‘Consumer choices and the environment’, Finnish Journal of Business Economics 1/1997: 15– 31. Uusitalo, L. and Ahola, E.-K. (1994) ‘Images of art museums. Consumer perception of the museums of the Finnish National Gallery’, Helsinki School of Economics W-94. Uusitalo, L. and Lindholm, M. (1994) ‘Kulutus ja lama’ [Consumption and recession], Helsinki School of Economics D-209. Uusitalo, L. and Uusitalo, J. (1981) ‘Scientific progress and research traditions in consumer research’, in K. Monroe (ed.) Advances in Consumer Research VIII, 559–63. Uusitalo, L. and Uusitalo, J. (1985) ‘Which sense of paradigms makes sense in marketing?’, in N. Dholakia and J. Arndt (eds) Changing the Course of Marketing: Alternative Paradigms for Widening Marketing Theory, Greenwich/London: JAI Press: 69–85. Walzer, M. (1990) ‘The communitarian critique of liberalism’, Political Theory: 1, 6–23 Veblen, T. (1899) The Theory of the Leisure Class, London: Macmillan.



INTRODUCTION Consumers in contemporary industrialized societies spend large portions of their purchasing power and their available time on the consumption of products that have few of the characteristics of old-fashioned material goods. People buy pictures, sounds and stories in the shape of films, discs and books. They process electronic signals in computers, and they use an increasing number of other electronic media for purposes of entertainment, information or control. All these products do not accomplish material transformations, as is the case with tools or food, for example. They relate to the processing of signals or, to use the most general term, to signs. The consumption of such immaterial products or products that generate the desired signal performance, like sound systems and monitors, has grown at remarkably high rates over the past decades. As the phenomenon increases in relevance, a better understanding of the nature of the process will be called for. One might try to use the conventional optimization paradigm to deal with the phenomenon.1 But there are problems. To name the two most obvious ones, consumption of sign or signal use occurs simultaneously, and the marginal production cost of information or signals is close to zero since no additional material resources are used up. Such conditions may possibly be modelled, but there remains a deeper epistemological problem: in general equilibrium theory anything related to information can be considered only as an environmental condition that has to exist in order to insure the allocation of goods. In the case of movies, video clips and software, however, something that clearly consists of information is consumed as a good. Thus, the dividing line between constraint and maximized variable dissolves, with fundamental consequences for the explanatory power of the paradigm. I will begin the discussion by briefly examining some of the current approaches in economic theory that deal with sign-related phenomena. I will then introduce some new ideas



about the nature of signs that have emerged in semiotics as well as in social systems theory. Equipped with the new distinctions, I will reexamine those actions which we define as the ‘consumption of signs’. I would like to note explicitly that signs are treated as a separate class of goods in this chapter. Of course, one might discuss signs as a dimension that is attached to any product, be it a bag of potato chips or a Broadway play. But the point is that there are products that, although they may have some material carrier, consist primarily and decisively of signs – like a music CD, a TV programme or an online service. Therefore, the usual order of priority is reversed: the goods to be discussed are essentially made of signs, and their material dimension is neglected.

SIGNS IN SOCIAL THEORY The treatment of signs in conventional economic approaches In traditional theory, signs are goods and assets just like all the other means of utility maximization. There is no other category available. However, some explanations and rationales for the consumption of signs are found in parts of the literature. One of them is the field that has been labelled ‘transaction cost economics’. The introduction of the term ‘transaction cost’ has been a breakthrough in the development of modern economic theory. The attachment of the cost label to an interaction links a host of activities that do not involve material production – like searching, negotiating and controlling – to the maximization calculus.2 The continuing discussion has led to an increased awareness of the complexity of transactions. The paradigmatic case is not an effortless, virtually instantaneous ‘nirvana exchange’ anymore, but a lengthy, contractually defined process, highly contingent on the existing context of trust and credibility, and thus on the ways in which trust and credibility are communicated or ‘signalled’. Yet signs and signals are only rarely attended to in transaction cost models. There is a strong emphasis on search and control costs, but the underlying assumption is that whatever is sought or controlled is something that existed prior to the information. Information is usually defined in the Shannon–Weaver version, as a given set of messages. A branch of the literature that focuses explicitly on signalling follows the work of Spence (1973). The notion of an activity undertaken not for its direct output but for its signifying quality is explored. Signalling activities refer to reputation and competence, or to credible commitments in future action, usually placed in the context of labour markets. The treatment remains limited, however, to cases with stable strategies and to models with very few signs (Riley 1987).



The distinction between sign and object plays an obvious part in monetary theory. It is no coincidence, therefore, that monetary theory has developed at a distance from the theory of ‘pure’ exchange. Only if the relationship between material object and monetary sign is unambiguous may the sign be substituted for the object with no distortion to the results of such an analysis. But what if expectations can be influenced by signs, as Keynes assumed, or if signs are generated autonomously in the banking system, as Schumpeter surmised? Money generation is difficult to model consistently in a general equilibrium framework.3 In all three discussions, the peculiar structure of the problem comes to light: there seems to be a basic ambiguity involved. In transaction cost analysis, it is ambiguous whether the aim of search is a transaction opportunity that exists already, or something that has yet to be invented and given a name. In signalling theory, it is ambiguous which activities are to be considered as done for their own sake, and which are to be considered as signifying something apart from themselves. In monetary theory, the ambiguity between signified value and signifying value is at the very core of the faculty of objects like copper rings or bank notes to indicate economic value.4 To shed light on the issues involved I propose a detour. A more detached look at the nature of signs should help us in understanding the difficulties encountered in modelling their economic impact.

The treatment of signs in semiotics and social systems theory Beginning with the work of Charles Peirce and Ferdinand de Saussure, there has been a growing awareness that ‘we move in signs, the way a body moves in space and time, we think and feel in signs and are ourselves, including our analysis of signs, produced by signs’ (Baer 1981: 305). The study of signs has become a legitimate scientific field, with extensive empirical contributions, as in the work of Bourdieu (1979), and sophisticated theoretical treatment, as in the work of Eco (1968), Sebeok (1979) and MacCannell and MacCannell (1982), to name only a few. Semiotics has a common base in treating signs as a set of relationships rather than as objects in themselves. The set of relationships is structured in two principal modes. The one following Peirce distinguishes between the sign itself, its relationship with the object, and its relationship with the ‘interpretant’: ‘that which a sign generates in an interpreter ‘ (Oeler 1981: 24).5 The one following de Saussure distinguishes between the signifier and the signified, and then interprets the sign as their dialectic unity. The signifier receives significance through its relationship with something to be signified (like the knight in a chess game), but that which



is signified, in turn, receives significance through another level of signifiers (like the rules of the chess game), and so on. In this potential infinite regress, the perspective of the subject assumes the part of the Peircean interpretant (Prieto 1975, cited in Krampen 1981: 120). Although the work of Pierce and de Saussure is so different in volume, detail and basic structure, there are still fundamental similarities. Both of them provide a structure that makes it possible to extend the scope of the phenomenon far beyond language, to include all types of communication, from genetic codes to social habitus. Both emphasize the crucial role of interpretation. But whereas the Peircean interpretant has the connotation of an organic, natural procedure, de Saussure’s subject is more evidently a construction in itself. Contemporary authors who emphasize the thorough constructedness of any social reality tend to follow de Saussure. The interpreter thus becomes a ‘constitutive component of meaning’ (MacCannell and MacCannell 1982: 10). Recently, the semiotic approach to sign relationships has been transposed into the terms of a fundamental ‘logic of forms’. That logic begins with the instruction to sever space and to indicate one of the sides as the ‘inside of the form’ – any form. This approach makes it possible to start observations with various distinctions. The traditional distinction between subject and object is only one possibility.6 Another possibility is the distinction between signs and the rest of the world. The approach shares the preoccupation of semiotic theory with the relevance of relations. The logic of forms, however, uses the notion of distinction rather than relation. The sign, then, is interpreted as the form of the distinction between the signifier and the signified. The act of using the sign constructs or invents a form whose inside (the marked state, the signifier) is distinguished from its outside (the unmarked state, the signified) (Luhmann 1993).7 The sign in Figure 12.1, for instance, distinguishes itself from the surrounding space. By distinguishing itself, it gives that space an interpretation of direction called ‘arrow’.8

Figure 12.1 Arrow.



Such an autonomous, independent status of the world of signs implies self-referentiality because reference to signs is made through other signs.9 The dimension of self-referentiality explains the interest of social systems theorists like Luhmann. In his theory, the basic operations are ‘communication events’. Just like signs, such communication events are not to be interpreted as separate entities. They are seen as a continuous series of distinctions, used by ‘observers’ to give meaning to their world. Every series or chain of signs is self-referential, since only signs can signify the unifying feature of a specific series. Only signs-in-use can continue signs-in-use. In this sense, all self-referential processes are operationally closed. Saussure speaks of codes, Bourdieu of cultural and intellectual force fields. They are closed in the same way that social systems or plays are closed. Plays, from children’s plays to court ritual, are an intuitive way to grasp the experience of ‘meaningful’ continuities of gestures, words or tokens, flows of musical variations or the course of a tennis ball in a grand slam match.10 Signs, then, are the stuff which the networks of plays that constitute our society are made of. On the basis of the body of literature introduced above, one must come to the conclusion that the relevance of signs for economic development is even larger than estimated. Signs are not simply a set of signifiers for a given world of signified objects. They are a way of drawing distinctions between signifiers and objects. The use of signs is not just another human activity. The use of signs is the kind of event in which the social world is continuously created and recreated. The social world includes the economic world. In which ways will the new understanding of signs influence our understanding of the consumption of signs, and our ability to model adequately its effect on economic change? The following sections are an attempt to outline a number of relevant aspects.

FOUR OBSERVATIONS ON THE CONSUMPTION OF SIGNS On the distinction of consumption and of signs To begin at a rather fundamental level, we observe that the notions of ‘consumption’ and of ‘signs’ are artificial since they depend on certain distinguishing signs to be recognizable in the first place. For these distinguishing signs, the new interpretation of signs holds as well: they are themselves changeable and, as they change, they change the notions to be signified. I begin with the term ‘signs’. It was argued above that signs are distinct from the rest of the world because of their self-referentiality. Their existence is not due to an outside item to be represented. It is due to a continuous reproduction of new signs. Consumers communicate



through their signs with others, thus maintaining the borderline between the world of communication and its environment. They may depend on professional providers of events and channels, but they are still the actual participants in the process. Two conclusions can be drawn. First, it is not entirely clear whether consumers use signs for their own objectives, or whether consumers themselves are used to assure the continuity of sign reproduction plays. Although this alternative assertion may seem far-fetched, there are good reasons for accepting it. The second conclusion refers to the structure of the reproduction process. If the reproduction of signs is so different from the manufacture of objects, then there should be characteristic features of that process. These features, in turn, will determine what is communicable: what possibly can be expressed through signs. Some of these features will be discussed in the third part of this chapter. I now turn to the notion of ‘consumption’. The mere use of signs need not be a concern for economic analysis. We need to isolate those uses that qualify as consumption. The criterion applied is straightforward: only those uses which are paid for qualify as consumption of signs. Payment marks the economic distinction. To use a concrete example, the playing of a basketball game may be a completely personal, self-serving event. But it turns into a production process when there are persons who pay for the provision of the excitement in a game with winners and losers. The act of payment turns the users into consumers. The use of a specific distinction has generated both sides of the distinction – production on the outside, consumption on the inside of the distinction. Again, two conclusions can be drawn: first, we note that consumers are involved in at least two types of sign codes or sign plays. When someone buys a Warhol print, he or she demonstrates the use of the signs that constitute money. But at the same time, this person demonstrates the use of signs in at least one of the games that make up his or her lifestyle world. The second conclusion refers to the form of payment. Given the public good properties of signs, it would be surprising if payment forms had evolved just like the private payment forms for material products. Some peculiarities of those payment forms which are actually in use will be discussed in the fourth part of this chapter. Given these two distinctions, we will now examine the potential for economic change in three major areas of sign use.

Signs used for contexts of transactions It is well known that the acquisition of rights to any good or service involves activities of search, negotiation and control. It is not so well understood that most of these activities consist in the use of signs.



Let us consider the three acquisition phases one by one. Search is a process that rarely consists in actual physical activity, as it does in the case of a shopper driving from store to store. Usually it is observed as a search for the right signs, or even a version of search where the aim consists in avoiding information: the consumer is inundated with product claims and must select the most credible sources. The signs generated refer to qualities and often consist of verbal descriptions or standardized indicators. For most modern consumer goods, the signals are not quite as straightforward. Colour values, testimonials or product shapes are used in trying to convey complex qualities. Signs like the Bacardi commercial are used to communicate the properties of other signs, in this case the sense of belonging to the right group when sipping a Bacardi drink. Search, then, is an activity that is not only undertaken by prospective buyers, but strongly involves those whose success depends on the outcome of buyers’ search. The negotiation activity in transacting standard consumer goods is negligible. Prices and product qualities are reasonably fixed. Negotiation is more extended in the case of investment goods, whether family homes, aeroplanes, steel mills or opera houses. The negotiation of such goods involves meetings, and meetings involve talk, text and numbers. The meetings are full of moves and countermoves. Recently, attention has been drawn to the strategic dimension of negotiations, to the relevance of notions such as commitment, threat and persuasion. Such ‘subtexts’ of negotation demand a subtle and imaginative use of signs, on both sides of the table. Control rarely takes the physical form of fences and walls anymore. The user wants to be informed about violations of his or her rights, or about possible unpleasant surprises that might simply affect the general environment of consumption. In order to get the information, a range of channels and sources is employed, from general radio and TV stations to specialized fan magazines. In the commercial field, whole departments filled with investigators and legal advisers provide such services on an inhouse basis. This short discussion was intended to demonstrate the extent of sign consumption taking place continuously around the seemingly simple process of economic exchange. Signs constitute the context within which transactions can be performed. The signs signify innumerable aspects of the real transaction. But that is not all. The new interpretation of signs outlined in the first section suggests that signs actively draw distinctions and thus define what functions as signifier, and what as signified object. Do observations corroborate that hypothesis? There is, even at this early stage in the research, considerable positive evidence. The redefinition of goods is rather direct in the efforts mentioned of trying to influence consumer search. The ‘path dependency’ of negotiation outcomes on negotiation procedures has been noticed. Control works in a roundabout way: when transactions results are not enforced, the terms of the next round of transactions are changed.11



We can conclude that the consumption of signs not only determines the context and thus the viability of transactions, but is also a central source of change for their content.

Signs used for direct consumption Every human being is faced with the basic task of keeping his or her brain occupied. Neurological studies have shown that the brain is in constant processing activity. It needs something to turn its attention to just in order to maintain the continuity of its operation (Hesse 1990). This is where the direct consumption of signs comes into play. For material products, like food, signs are only an attribute, more or less expendable. However, for communication products, like entertainment and information, signs are the primary content. The brain is satisfied when it gets a constant flow of manmade signs, whether texts, images or sounds. The simplest way of occupying the brain seems to be the viewing of stories12 via television. The stories are designed to bring a constant flow of new, slightly contingent variations of a format that is familiar to the viewer. The consumption of entertainment signs, then, has the effect of soaking up excess leisure time – time that is a negative asset because it is costly to dispose of. It is doubtful whether the current unemployment levels in the first, second and third world could be absorbed with such little social turbulence if entertainment consumption, particularly TV and recorded music signals, did not exist. In Brazil, for instance, more than 30 per cent of homes are without refrigeration, but less than 10 per cent without TV, even in the favelas. As one favelero is quoted as saying: ‘If I open the refrigerator it is empty, if I turn on the TV it is always full.’ We return to the new interpretation of signs. It matters which signs are used to fill up leisure time. The ‘soaking up effect’ of TV vis-à-vis alternative physical pastimes, like sports, crafts or homemade music, is therefore relevant. Personal, internal experience is substituted by fictional, external experience. Accordingly, the pattern of social belonging shifts to groups and fan communities created by the medium that connects the participants. Another aspect is the observation that TV content follows its own rules of survival: consumers can switch between viewing channels with great ease. As a consequence, stories are constructed with a maximum of action and suspense to keep their grip on consumers’ attention. That, in turn, determines the type of stories fit for entertainment and information in a competitive environment. A last observation relates to the use of material resources in consumption. Signs are virtually immaterial. Even though their media of transmission demand material inputs, the ratio is constantly shifting: increasing sign volumes can be processed with decreasing hardware volume. There are obvious, yet unexplored consequences of this shift towards the consump-



tion of immaterial goods. These consequences concern natural ecology and social ecology – the structure of occupations and their purchasing power.13

Using signs to identify oneself In societies with a high degree of division of labour, the status of the persons acting is not obvious. Differences are not established by birth and are not maintained for a lifetime. In any given event, it is not clear whether a natural or a fictitious person should be recognized as an actor. Therefore, signs are needed. Signs make people observable: titles like PhD set individuals apart; new companies, like Exxon, spend large sums to establish their name. Signs help to distinguish one similar person from another: for instance, differences can be indicated through privileges that serve as signs for access to sites and resources.14 Often it is difficult to distinguish between the opportunistic use of a sign and an indication of actual reputation of a person. At any given social event, there is a mixture of those who belong and those who want to signal that they belong. The use of signs for purposes of identification is still quite in line with a static, representational interpretation of signs. But sign use is changeable. Signs are appropriated by others and new ways of differentiating one’s identity have to be found.15 Fashion, in all its emanations, is the medium in which many of the identity creating sign uses take place. Haute couture is just a narrow example, but it demonstrates the principle: a continuous flow of variations, spread over a limited range of styles, from Gaultier to Chanel, is pumped through complex communication channels – shows, magazines, retailers. People can demonstrate their personal style through their choices. They have to renew their choice every season. If they keep their old garments, this also makes a statement.16,17 The new interpretation of signs goes beyond the aspect of change. It predicts that the signs used to indicate identities also create identities. However, it is difficult to prove that claim. A person’s self-consciousness operates in a way which presumes (with rare exceptions)18 the unbroken continuity of one’s identity. There is always a presumption that it is the environment, not oneself, which has changed. Yet, over the timespan of centuries one can find indications that people experience themselves in changing ways. The sign uses connected with such changes are usually taken from the arts. Literary figures seem to be an effective way of self-interpretation. Defoe’s Robinson Crusoe, for instance, was an avant-garde figure in his calculating rationality when the book was published in 1720. In the early nineteenth century, the (radically shortened) story had become part of education’s standard sign repertoire. Today, Crusoe seems simpleminded and quaint in comparison with Proust’s Marcel or Joyce’s Leopold Bloom.19 The explanation lies in the self-referential nature of sign consumption: as readers



absorb these stories and use them in turn during later conversations, they experience their own change in self-interpretation. The effect on self-change is easier to recognize in the case of collective identities. Tribes, teams, families, households, corporations and nations all have their own ways of drawing individuals into their particular collective identity. All these organizations need identity in order to coordinate action within the enterprise. One can try to achieve the same performance by setting rules and sanctions. But it has been found that such measures are not effective in complex and quickly changing corporations. Large companies are beginning to draw the consequences: they direct their ‘mission campaigns’ towards their own employees in order to increase their motivation and sense of belonging. To be successful, there has to be close correspondence between the signifying means used to implement ‘corporate identity’ and the type of organization intended to be signified.20 The intended and inadvertent change of individual and corporate identities is, in my opinion, the most far-reaching and relevant effect of the increasing consumption of signs.

SOME FEATURES OF THE REPRODUCTION OF SIGNS Material artefacts can be manufactured in one place and then be moved to their place of use. In the case of signs, the users are actively involved in selecting certain forms out of an environment of possibilities. They use distinct words of a language, selected items of a fashion trend, certain shows of a (serial) programme. The carriers or media of signs, being hardware, can be produced like other goods, but the signs in use are events that have to be performed by the users themselves. Some of the characteristics of that performance will be discussed below. The discussion does not consider the economic distinction. In other words, it remains open whether payment is involved, making the process of provision a process of production.

The medium of play I will define a medium as any set of loose elements out of which forms can be made by coupling some of the elements (Luhmann 1993). Language, for instance, is a medium out of which words and sentences are formed, the notes of the harmonic scales are a medium, and so forth. Through individual or industrial activity, loose signs are generated out of which persons form their own, private uses of signs. Using the definition just introduced, sign



provision generates the medium for the use of signs. That process may take place on a large industrial scale, as in news networks, or it may have more personal characteristics, as in the case of art academies or fashion centres. In dress fashion, for instance, individuals are able to write their personal style into the fashion medium provided by fashion houses and journals. The medium, then, is not a channel made out of a material that is unable to interact with its content, as steel is impenetrable to water. It is the next layer of meaning, used to distinguish a specific communication situation from its outside, its environment, its ‘unmarked state’. The unmarked state may be a cocktail party, a newscast, or a spreadsheet program – flows of signs into which the messages inscribe themselves. The unity or combination of the two, medium and incribed forms, constitutes a play, a self-organizing communication process that contains its own meaning (Hutter 1996). It is important to emphasize that the providers cannot control the plays since each necessitates the participation of the users. All they can do is to influence, maybe even manipulate, the plays while they continuously float loose signs through the networks. The major kinds of plays have been mentioned several times in passing. Systematically, one might distinguish between style plays, story plays and language plays. In style plays, the challenge is to come up with a steadily paced flow of variations; dress fashion is an example. In story plays, an observed or invented sequence of events is transmitted in a continuous or periodical pattern; examples are newscasts and TV sitcoms. In language plays, artificial languages are electronically coded and the use of such ‘operating systems’ is constantly serviced; examples are the ‘DOS world’, or the ‘Macintosh world’.21 Plays, then, are generated, reproduced and continued. They change in close interaction with their users. It is a matter of perspective whether one considers the users as employing the signs for their objectives, or whether one considers the plays as the driving force. The plays maintain themselves and employ the participation of users to accomplish the reproduction of their elements.22 These aspects are specific to the world of signs. They explain a good part of the particularities found in sign provision.

The emergence of new plays Plays are either found or invented. Even when they are found, they still have to be filled with meaning or ‘significance’ in order to make them attractive to users. One example is reports about the lives of people which show similarities to the lives of those who read them, or reports about the lives of prominent and influential people. Another example is sport events. Sports, especially team sports, are games of chance and skill.23 Originally, they were self-provided. Today, they have become the core events of major industries, like baseball, football, soccer,



tennis and car racing. Although the staging of these plays takes place with professional, even industrial, efficiency, the teams and athletes have to retain a basic element of self-provided play. Therefore, in all these games, it is strictly forbidden to manipulate the events in order to ‘produce’ predictable results. Once that is done, the game becomes a racket and loses its social value. The relevance of internal rules holds for all the plays invented within the ongoing communication of a society. Sign construction is inherently ambivalent or even polyvalent and leads continuously to new interpretations and variations. This property is most clearly evident in works of art, whether compositions, images or texts. But the criterion of polyvalence applies not only to high art. Rock music is an example of a new form of play that emerged even against the intentions of the music industry. The condition is simply a self-generated rule that is followed in executing a ‘performance’ of the invented play. Even in cases of outright commercialization, as in popular music or TV soap operas, there are remnants of rules, despite the fact that the design and execution of such forms of entertainment is severely predetermined. The success of movies, rock bands and novels cannot be predicted because it remains unclear which combinations the users will discover and which they will find attractive to participate in. One may ask about favourable conditions for the emergence of new plays. In the case of art, there is some evidence that styles of painting and literature emerge in circles of a few hundred persons (Schücking 1962). Similar though larger circles emerge in cultural industries like the movies and music, or in scientific research. The formation of such small, closely knit networks happens by chance, but the probability of their ignition is certainly increased by adequate technical and social conditions. Paradigmatic for the self-growth of communication networks under supportive technical conditions is the present growth of Internet. The large mainframe computers did not link up, as the science fiction literature had predicted; rather, a multitude of small, decentralized personal computers are linked together and their users support the linkage with their ingenuity.

The recognition of signs Signs have to be perceived or registered by means of conscious cognition. Re-cognition is another term for conscious cognition: that which is registered by physiological means is the event of repetition – a duplication of an earlier form. Recognition can be trained to a competence for reading the meaning of specific combinations of sign forms. The ability to read sequences of signs may relate to the noises of jungle animals, traffic signals, or to the appropriate snowboard wear. In every case, sign provision depends on the literacy and interpretive skills of the users.



Potential producers of signs have an interest in increasing recognition skills. Such efforts may use two strategies. The first aims at the physical availability of signs. Signs must be distributed and reproduced at a level of distinctiveness that makes them accessible for human cognition. Until less than a century ago, the only reproduction mechanism was the printed text and simple images. Today, sound and image can be reproduced electronically at high levels of ‘fidelity’. As a consequence, the skill requirements for recognizing signs have decreased. The second strategy is concerned with the users’ ability to read a specific set of signs. The public provision of training services for reading and writing in schools over the last hundred years has led, in fact, to a huge accumulation of social capital and to a fundamental change in the accessibility of people’s lives through signs. But the contemporary rate of change in the vocabularies of modern business and entertainment scenes is much quicker than the rate of change in state education institutions. Therefore, private providers initiate their own learning activities. Major movies or major software programs, for instance, are introduced with publicity campaigns; albums are introduced through concert tours, books through reviews and talkshows. Potential users have to be provided with some basic knowledge in order to be capable of reading or interpreting the new story. That knowledge, however, may have a high obsolescence rate, which makes it difficult for the user to decide in which trend to invest learning time. Once users have mastered some of the rules of a certain communication play, they tend to consume more of its signs. There are two explanations. The reading skill might make works available of whose existence the user had been unaware. That would cause a shift in consumption pattern since material items consumed previously are substituted by more sign consumption. But it might also be that mastering the rules of a play causes pleasure in itself. Practitioners speak of a feeling of testing and identifying one’s own boundaries. Even in less ecstatic cases, we observe ‘acquired taste’. The ability to play by certain rules has led to a source of pleasure that was previously unknown.

The star pattern Signs reproduce themselves in communication. A story is duplicated as it is retold, music tunes can be repeated anywhere and any time, the rate of growth is potentially exponential. However, not all the simultaneous offers of signs can experience such growth. Only a few movies are the blockbusters of the season, only a few operas fill the standard repertoire of most opera houses around the world. In short, a few ‘star events’ – Grand Prix car races, famous paintings, bestsellers, murder trials of the century – dominate each consumption play, surrounded by a periphery of regional, specialized and little-known events.



The star pattern is typical for the whole range of the environment of sign provision. Its advantages are convincing. If a specific combination of signs is repeated again and again in a field, like a pop song on radio ‘powerplay’, one can be sure that others recognize the signs. As so many others use the same set of signs, the specific combination becomes standard – for three weeks in the pop charts or for decades in a scientific citation index. One is understood if one pays attention to the star event, and there are others trying to communicate via the event. There are star events that petrify because their environment is sufficiently static – the opera repertoire is an example. In other cases, the star sign is the standardization of the image which which a historical event is depicted. War photographs are an example: one shot becomes symbolic, like the soldiers at Iwo Jima as symbols for the American Pacific victory, or the flag setting at the Reichstag for the Soviet victory in World War II.24 Most of the signs used by an individual or corporate person in the plays of daily life are still self-supplied. They are processed without an economic value connotation. Yet, in a growing number of plays the difficulties of payment are overcome by industrial providers. Once the industrial providers have established themselves, the positive feedback works in their favour: they are able to provide signs at a fraction of the opportunity cost of self-provision.

MAJOR PAYMENT FORMS OF SIGNS Payment as a particular form of sign use As noted above, the notion of consumption only makes sense if we already think in an economic context. Consumption is a construction. It is constructed by drawing a distinction whose two sides are then called consumption and production. The distinguishing event is the event of transaction. In the course of a transaction, a good changes assigned property rights and is paid for. A two-sided understanding about economic value takes place. Payment generates a peculiar kind of interdependence between income and expenditure, between activities intended to procure saleable goods and other activities intended to use them, either immediately or over time. This is why, in the discussion above, care was taken to distinguish between the use of signs – an activity like walking on the sand – and the consumption of signs, an activity that involves production and sale. There can be no consumption unless an acceptable form of payment has been found and that form, being a distinguishing sign, will have its effect on the identification of the consumption and the production side. Forms of payment are easily found in the case of objects or contractually defined services. In the case of potatoes sold or labour services performed something can be signified or



‘pointed to’ with sufficient certainty. The transfer of access rights to the particular good is selfevident and apparent. In the case of signs, that simplicity is lacking. It is difficult to specify the ‘object’ of transaction and to keep the signs from multiplying once they are in the world. Such conditions hamper the determination of payment. As a consequence, adequate markets are slow to develop, and their emergence is accompanied by controversial discussion. In general, one can say that more complex, sign-specific payment forms have indeed evolved. The following section will discuss some currently used forms of sign payment.

Payment by immediate users Despite the general thrust of the argument above, there are many types of sign consumption where conventional pricing can be practised. Tickets for the cinema or the opera and prices paid for computer programs or oil paintings are examples. But even here we observe difficulties: not all the seats in the opera are equal, and different performances are certainly not equal. Price differentiation can have countless variations and it might even become a sign in itself, as, for instance, when the prices for seat categories signal the social status of the temporary owners. Computer program copies do have prices, but we know the difficulties of excluding (almost) costless transfer. The works of a painter reach specific prices at auctions, but the prices depend on the public discussion and certification surrounding the object. There are ways to deal with these difficulties without abandoning the principle of user payment. Price differentiation has already been mentioned. The term is not quite fitting since we are not discussing the sale of homogeneous products. Different seats and positions in a time sequence do constitute different goods. The questions are therefore: how is it possible to find differences in prices that match the differing economic evaluations of these products? How is it possible to take into account the self-signifying effect of the signals involved in the pricing mechanism such as the effect of announcing ‘sold-out’ performances? In other cases, payments do not refer to single units. The users might pay for a series of events. They might pay a channel fee for a certain time period or become members of an association or organization that promotes access to certain signs, like first nights at the opera. In electronic networks, charges are often related to consumption time or even to the number of bytes transferred. Such practices, however, lead to a difficulty: in the material world, payment is a single communication event, an intrusion; afterwards, buyer and seller are disconnected, nameless, anonymous. In the world of signs, payment is one of many communication events. It is linked not only to a separate exchange act, but also to channel or network use. The connection between buyer and seller, therefore, remains and anonymity disappears. Anonymity has to be consciously reintroduced to assure the independence of the user.25



Payment through advertising One can attach the sign of a product to an event that carries positive connotations for its participants. The logo of the sponsor appears on the jerseys of soccer players, on racing cars, on festival programme brochures. In the case of radio and TV spots, the advertising message even uses the same medium as the event: it takes the form of a voice or film recording. In consequence, the design of advertising spots and entertainment forms has converged. On video clip stations, the two have become virtually indistinguishable. In the case of commercial spots, the environment of the surrounding programme is of limited importance. However, game shows or news casts determine the size and composition of the overall audience. Within these constraints, the actual impact of the spot determines its performance. The spot can ‘load’ the product with messages and significations that are a world apart from the material composition. The products with attached meanings and interpretations are possibly material goods, like cars, drinks, detergents, perfumes, or they are companies that want to create an observable image for their invisible performance, like banks, insurance companies, providers of health services. The cost of internationally employed commercials has by now reached the level of budgets for average European movie films. It remains unclear, however, to what extent such advertising efforts actually influence the sales of the company. Among those who actually invest in mass media marketing, there are standards for minimal exposure of a new or continuing product or brand, but beyond that trends and opinions vary rapidly. Companies sometimes succeed in attributing cult value to their self-created signs, their trademarks and brand names. Brands like Gucci or Porsche stand for an identity that is defined by the use of such products. In these cases, imitators do not bother with the material source of the sign. They sell gadgets, not sports cars or evening gowns. Of course, they dilute the credibility of the brand signs. Fashion houses, for instance, fight constant legal battles against T-shirt producers who imitate their trademark products. In order effectively to use mass media with a wide distribution, the goods to be advertised must also be mass products. Only part of the industry can make use of the wide distribution, so that part has considerable bargaining power. Its companies are the only ones that offer the media networks payment to insert their own messages. Given the fact that there are no other demanders, the media networks sell the rights to insertion relatively cheaply. They are pressured to tailor the programme content to the interests of the sponsor, thus increasing the performance value for the sponsor.26 Once the market price level for specific types of insertion stabilizes in a regional or national context, single networks find it attractive to lower production cost per programme in order to increase profit. Such a change in context can be detrimental for the advertising ‘principal’. State-supported radio, TV or telecommunication networks can function as a constraint on lowering cost and quality if they use their advertising-free space to their com-



parative advantage. It is quite possible that the real competition to the advertisers’ monopsony will come from direct payment for channel use. Commercial advertising and sponsoring will then survive only in networks with clearly defined target audiences, like music stations in metropolitan areas.

Payment through public purses Signs are generated in plays. Plays with sizes beyond personal contact need networks. Whenever the size of the network coincides more or less with the territory of political power, provision through government on a local, regional or national level has been considered an efficient solution for the past century. The argument is still considered valid for the law, for general education and for cultural institutions, but it has become doubtful for utilities, telecommunication and broadcasting networks. In such cases, an entire political entity demands a certain service. The entity pays for it and is formally the consumer of the signs. In order to be in a position to pay, money taxes are collected from households and firms. The actual payment decision is made in an assembly session run according to recognized political rules. Such a situation has severe consequences for the shape and quality of the signs provided through the ‘publicly’ financed transactions. The first and foremost use of signs by political entities is for representation of identity.27 Political action is largely invisible and must be at least partially opaque to public observation. It is then the more necessary to find images for the stability, continuity, power and legitimacy of a political regime. Medieval city states deployed a wide spectrum of parades, fireworks, music and poetry (Strong 1986). Modern territorial states have a changed yet even wider range of instruments at their disposal. They have a strong interest in controlling newly emerging news and entertainment media. The cultural ‘mission’ of such media organizations is something that is able to survive in the shelter of a non-competitive organization and from it may emerge the sources of new signs. But the cultural mission is usually sacrificed when it hinders the effectiveness of political control. Still, the news and entertainment networks have slipped from the control of public authorities, even in many European countries; such is the effect of immediate competition with international, commercial advertising-financed providers. Due to the global distribution of signals via satellite, territorial boundaries can no longer be enforced. The direct competition demonstrates classical drawbacks of public institutions. Economic valuations can be carried out, but accounting rules and incentives limit the impact of economic arguments on decisions. The result is a loss in efficiency. Second, large broadcasting networks offer opportunities for private gains, either to individuals or to political power-holders. These gains have been



exploited, often with blatant openness. It is not surprising that the European public networks lose territory against the commercial networks. Their positions are filled according to party interests. The competence that remains in the institutions is not sufficient to compete with the know-how and flexibility in the private companies. The result is state-enforced resistance, anxious overadjustment or a drifting off into social irrelevance.28 I will only briefly touch on an entirely different kind of public support, namely support through legal rules. Payment becomes more probable when the rights to specific signs and the conditions of their use are recognizably assigned, and when enforcement can be made credible. The ongoing discussion about appropriate modifications, particularly of intellectual property law, reflects the importance assigned to property rules (Besen and Raskind 1991; Hutter 1995).

CONCLUSIONS My starting point was the observation that signs might be a worthwhile topic for economic analysis. Semioticians tell us that signs are most adequately interpreted as relationships between a signifier and its signified object, which would be relevant enough given that most people, including economists, see signs as ‘attached’ to objects in some sense. In concert with logicians and epistemologists, they also tell us that signs, by drawing distinctions between signifier and signified, change the shape and content of signifiers as well as signified objects. If that observation is correct, it becomes crucial to explore how the use of signs shapes economic change. Signs appear in the economy in the form of consumption. They operate as events involving several persons and are part of self-organized communication plays. I have examined different distinctions of consumption and the differences between sign and consumption for their potential to influence economic change. All the distinctions have the potential to change that which they signify: transactions, communication content, identities, play forms and payment forms. In changing these institutional parametric forms, they change nothing less than the pattern of reproduction, the very genetic code of the economy.

ACKNOWLDEGEMENTS I thank Marina Bianchi and Michael de Vries for comments and criticism.



NOTES 1. For a different set of drawbacks, not directly related to signs, see Bianchi and De Marchi (1997). 2. A historical version of this approach has been put forward by North (1981). For the new theory of the firm, transaction forms, particularly contracts, and transaction costs are central notions (Williamson 1985). 3. The issue is treated extensively in Hutter (1994a). 4. There are, however, attempts to come to grips with the issue outside of conventional theory. One example would be the work of Veblen. 5. Again following Peirce, the three constituents are distinguished in three categories each. Relationships with objects, for instance, appear as symbols (crown), index (arrow) or icon (crossed out cigarette). Going through all the permutations of relationships, a rich variety of different cases can be analysed. 6. See Hutter (1994b: 113). 7. The notions of form, marked state and unmarked state originate with Spencer Brown (1969). 8. The shape of this particular distinction is not a somehow available symbol. It is a rather modern way of signifying direction, made popular by the Bauhaus designers who, in turn, appropriated it from the work of Paul Klee. 9. The literature since Hofstadter (1979) and Bateson (1979) tends to accept the logical paradox involved in signifying the signifying distinction. The self-referential nature of the relationship between signs is considered not as a short-coming but as an existential feature. For a concise formal account of the informational content of paradox see Krippendorf (1984). 10. For a detailed introduction of the notion of play into social theory, see Hutter (1996). 11. To go one step further, one might also examine the distinction-making power of the channels and media used for information. As computer networks expand globally at exponential speed, the duplication and worldwide distribution of signs appears without limit. However, the scope of the signs used is still severely limited. To date, written letters, numbers and simple pictograms dominate. As a consequence, we observe a shift in the relation of telephone (voice) communication to fax and e-mail (text) communication, combined, of course, with a rapid growth in the total volume of media consumption. Similarly, the use of electronic mass media changes the nature of the signs in use. Whereas traditional print media favoured language signs, the contemporary media favour images and sounds. They in turn have to be woven into their own stories or plays often using the code of art (Hutter 1992). 12. TV stories are a combination of images, words and sounds. 13. We are, however, observing a process that has a long historical tradition. For instance, McCracken (1990: 19–20) reports on a large-scale introduction of life-style goods in the eighteenth century: ‘With the growth of fashion grew an entirely new habit of mind and pattern of behavior. Increasingly, aesthetic and stylistic considerations took precedence over utilitarian ones. . . . Fashion had the effect of requiring that objects be replaced over and over again. . . . The consumer had to devote more time to consumer learning. . . . Consumers now occupied a world filled with


ON THE CONSUMPTION OF SIGNS goods that carried messages.’ 14. An example is the different ‘passes’ given out to those who take part in a Grand Prix racing event. 15. The observation was first made by Simmel (1904: 543): ‘Fashions differ for different classes – the fashions of the upper stratum of society are never identical with those of the lower; in fact, they are abandoned by the former as soon as the latter prepares to appropriate them.’ 16. Similar mechanisms characterize other fields, like the car market or art markets. 17. The constant change leads to a demand for learning aids. Miller (1981: 183) reports that, in the late nineteenth century, the catalogues of department stores like the Bon Marché taught their clients ‘how they should dress, how they should furnish their home, and how they should spend their leisure time’. 18. One rather pervasive exception would be the use of initiation rituals. They constitute an active strategy to create a discontinuity in identity. 19. The reception of Robinson Crusoe is dealt with at length in Hutter (1993). 20. The standard negative example is the campaign that stops with the standardization of stationery and colour code. However, there are many cases reported where relatively simple discussion techniques have led to remarkable changes in the self-definition and subsequent behavior of the employees – or, rather, co-workers, since the choice of term is in itself a signifying distinction. 21. One could apply the medium–form distinction once more: styles, stories and computer programs contain concrete forms of signs. These forms, in turn, have been assembled out of the loose elements of some technical medium, like silk, comic pictures or chips. I will limit the discussion to the provision of the forms, keeping in mind the need for the existence of material resources in order to express these forms. 22. The relation between individual choices and action on one side and collective play on the other may be likened to the relation between organisms and genes in the reproduction of life forms. Both are part of a circular relationship, both are necessary for the continuity of social or biological systems, respectively (see Hutter 1996). 23. The distinction between games and play is really one of emphasis. Play emphasizes the criterion of closure and, thus, of self-contained meaning. Game emphasizes the limited extension of a particular event, and the fact that certain results may have instrumental value outside the game. 24. In the economics literature there is some treatment of the star phenomenon. Adler (1985), for instance, starts from the assumption of small differences in talent among performers in order to explain large earning differentials (see also Rosen 1985). But it seems that the initial difference is contingent. Many differences will do to gain an advantage in attracting attention to a new sign variation. The quality of the difference consists in its ability to generate a stream of changing interpretations. 25. In the case of ‘cyberbucks’, a money form used on the Internet, there is, in fact, a mechanism to disconnect the information about the origin of a payment (see Chaum 1992). 26. An actual example is the attempt to change the rules of soccer in order to have more ‘time outs’ for commercial insertions. 27. Political representation is, of course, just one of the cases of general semiotic representation.


ON THE CONSUMPTION OF SIGNS 28. A noteworthy effect takes place when public authorities act as providers of the background of signs. Space is needed to place signs in the immediate living environment of consumers: space on walls, on sidewalks, in the air, on trams and other vehicles. This space, as long as it is not claimed by private rights, is public space. The disposal rights to public space are held by state institutions, for instance, a town commissioner or a civil servant. As cash is offered for the use of such public space, the officeholders see a new source of income. Space is leased cheaply and generously. The private negative effects, in the shape of unwanted signals, do not affect the bargaining position. Thus, particularly, city environments are filled up with signals beyond the level preferred by the individuals living in such environments.

BIBLIOGRAPHY Adler, Moshe (1985) ‘Stardom and talent’, American Economic Review 75 (1): 208–12. Baer, Eugen (1981) ‘Die Zeichenlehre von Thomas A. Sebeok’, in M. Krampen et al. (eds) Die Welt als Zeichen, Berlin: Severin und Siedler: 281–322. Bateson, Gregory (1979) Mind and Nature. A necessary Unity, New York: Bantam. Besen, S. and Raskind, L. J. (1991) ‘An introduction to the law and economics of intellectual property’, Journal of Economic Perspectives 5: 3–28. Bianchi, Marina and De Marchi, Neil (1997) ‘The taste-less theory of consumer choice meets novelty’, in Andrea Salanti and Ernesto Screpanti (eds) Pluralism in Economics, Cheltenham: Edward Elgar: 177–90. Bourdieu, Pierre (1979) La Distinction, Paris: Editions du Seuil. Chaum, David (1992) ‘Achieving electronic privacy’, Scientific American 267 (8): 96–101. Eco, Umberto (1976) A Theory of Semiotics I. ed. 1968, Bloomington: Indiana University Press. Hesse, Günter (1990) ‘Evolutorische Ökonomik oder Kreativitiät in der Theorie’, in U. Witt (ed.) Studien zur Evolutorischen Ökonomik I, Berlin: Duncker & Humblot: 49–73. Hofstadter, Douglas (1979) Goedel, Escher, Bach. An Eternal Golden Braid, New York: Basic Books. Hutter, Michael (1992) ‘Art productivity in the information age’, in R. Towse and A. Khakee (eds) Cultural Economics, Heidelberg: Springer: 115–24. —— (1993) ‘Literatur als Quelle wirtschaftlichen Wachstums’, Internationales Archiv für Sozialgeschichte der Literatur 16: 1–50. —— (1994a) ‘Wie setzen sich Theorien durch? Zur Entwicklung deutschsprachiger Geldtheorien in der Zwischenkriegszeit’, in Ch. Buchheim, M. Hutter and H. James (eds) Zerrissene Zwischenkriegszeit, Baden-Baden: Nomos: 123–55. —— (1994b) ‘Communication in economic evolution: the case of money’, in R. England (ed.) Evolutionary Concepts in Contemporary Economics, Ann Arbor: University of Michigan Press. —— (1995) ‘On the construction of property rights in aesthetic ideas’, Journal of Cultural Economics 19:


ON THE CONSUMPTION OF SIGNS 177–85. —— (1996) ‘The value of play’, in A. Klamer (ed.) The Value of Culture, Amsterdam: University of Amsterdam Press: 122–37. Krampen, Martin (1981) ‘Ferdinand de Saussure und die Entwicklung der Semiologie’, in M. Krampen et al. (eds) Die Welt als Zeichen, Berlin: Severin und Siedler: 99–142. Krippendorf, Klaus (1984) ‘The paradox of information’, Progress in Communication Sciences 5: 46–71. Luhmann, Niklas (1993) ‘Zeichen als Form’, in D. Baecker (ed.) Probleme der Form, Frankfurt: Suhrkamp: 45–69. MacCannell, D. and MacCannell, J. F. (1982) The Time of the Sign: A Semiotic Interpretation of Modern Culture, Bloomington: Indiana University Press. McCracken, Grant (1990) Culture and Consumption. New Approaches to the Symbolic Character of Consumer Goods and Activities, Bloomington: Indiana University Press. Miller, Michael B. (1981) The Bon Marché: Bourgeois Culture and the Department Store, 1869–1920, Princeton: Princeton University Press. North, Douglass C. (1981) Structure and Change in Economic History, New York: W.W. Norton. Oeler, Klaus (1981) ‘Idee und Grundriß der Peirceschen Semiotik’, in M. Krampen et al. (eds) Die Welt als Zeichen, Berlin: Severin und Siedler: 15–50. Prieto, Louis J. (1975) Pertinence et pratique, Paris: Les Éditions de Minuit. Riley, John G. (1987) ‘Signalling’, in John Eatwell, Murray Milgate and Peter Newman (eds) The New Palgrave: A Dictionary of Economics, London: Macmillan: 330–32. Rosen, Sherwin (1985) ‘The economics of superstars’, American Economic Review 71 (5): 845–58. Schücking, Levin (1962) Soziologie der literarischen Geschmacksbildung, Leipzig: Teubner. Sebeok, Thomas (1979) The Sign and its Master, Austin: University of Texas Press. Simmel, Georg (1957) ‘Fashion’, American Journal of Sociology 62: 541–548 (reprinted from International Quarterlyy 10, October 1904). Spence, A. M. (1973) Market Signalling: Information Transfer in Hiring and Related Processes, Cambridge: Harvard University Press. Spencer Brown, George (1969) The Laws of Form, London: Allen and Unwin. Strong, Roy (1986) Art and Power. Renaissance Festivals 1450–1650, Woodbridge: Boydell. Williams, Rosalind H. (1982) Dream Worlds: Mass Consumption in Late Nineteenth Century France, Berkeley: University of California Press. Williamson, Oliver E. (1985) The Economic Institutions of Capitalism, New York: Free Press.



Baudrillard, J. 181, 182, 184, 220, 227, 230 Bauman, Z. 231 Becker, G.S. 22, 48, 124, 162–3, 164, 168 Becker, H. 222 behaviour 14, 123, 179 Belk, R.W. et al. 193 Bell, Q. 83 Bellah, R. et al. 219 Benacerraf, P. and Putnam, H. 42 Bentham, J. 5 Berg, M. 83, 84; and Hudson, P. 166 Berlyne, D.E. 2, 14, 16, 73, 74; and Masden, K.B. 16 Besen, S. and Raskind, L.J. 253 Bettman, J.R. 179 Bianchi, M. 16, 40, 42, 43; and de Marchi, N. 254 Bienefeld, M.A. 153 Botham, F. and Hunt, E. 153 bounded novelty 74 bounded rationality 72, 75, 93 Bourbaki, N. 42 Bourdieu, P. 182, 183, 220, 223, 225, 229, 238 brand loyalty 131–4 Brewer, J. and Porter, R. 157 Broome, J. 31 Buridan’s ass 28–30, 37–40 Burnett, 144

adaptive models 10, 48–9, 61 addictions see habits Adler, M. 255 aesthetics 182–3, 226, 230, 231, 255 affective sociality 15, 220–2, 231, 232 agency 227–9 agents 9, 42, 70–1, 209 Aiginger, K. 93 Akerlof, G. and Dickens, W. 48 Allen, R. and Hicks, J. 30 Alsop, J. 181 Amsel, A. 16 anomalous consumption 179; antique/ discarded goods 178; growth in 180; handmade goods 177; Jurassic goods 176– 7; nostalgia goods 178; revival goods 177; see also consumption Archenholz, 159 Aristophanes 34–5 Aristotle 43 Arnould, E.J. and Price, L.L. 194 arousal levels 198, 201, 210 Arrow, K. 32–3, 40, 42, 91, 113 Ashton, T.S. 163 Austrian school 92, 97, 98 Baccini, A. 43 Bacharach, M.O. 48 Baer, E. 238 Barnard, C. 93 Barratt, A. 41 Bateson, G. 254

Calvino, I. 43 Campbell, C. 220, 225 capabilities 11, 16;


INDEX consumer revolution 157–62 consumer-collector 184–5; affect-driven ehaviour 185; explorative behaviour 186; and marginal inclination to consume 186– 7; seeking individualism 185–6; semiotics and aesthetics 185; and seriality of consumption 186 consumer-driven processes: collecting behaviour as mass phenomenon 190; and fragmentation of lifestyles 190–1; and globalization of markets 190–1; and institutionalization of collectionistic consumption 191–2; need for hyperdifferentiation 189–90; reduction of product life cycles 190–1 consumer/s: active/passive 53, 84, 209, 227; affect behaviour 179; as agent 70–1; assumptions and inferences 134–6; behaviour of 179; boundaries with producers 114–18; and brand loyalty 131– 4; decision procedures 10, 11–12; eterminative role of 68–9; goal-arrival problems 128–31; as ideal 229–30; as interdependent 228; in mediated reality 231–2; multiple strategies of 72–3; needs of 109–10; preferences 179; pressures on/errors by 12, 136–7, 138; and problem solving 10– 12, 109–10; as producers 7–8, 10–12; role of 175, 218; and routines 11; as selfconstructors 222–5, 227, 229, 230; setting/ attaining goals 124–8; social dimension of 8, 59–60, 82 consumption 46–7, 99; addictive 223, 228; anomalous 176–9; and capabilities 110–12; classification of studies 215–17; coordination and structure 112–14; defined 241; discovery in 3–5, 7; double role of 227, 227–9; imaginary 15, 225–6; and importance of time 209; markets 57; and needs 109–10; process 57; as production 107–9; rarefaction process 180, 189–92; role of novelty in 6–9; seriality of 186; shared 220–2; signs used for direct 243–4; skilled 74, 75, 84, 199; socio-semiotic meaning of 223–5; structuring capacity of 219–22;

boundaries between producers/consumers 114–18; co-ordination and structure 112–14; dynamic approach 108–9; and routines 110–12 Cassel, G. 6 Casson, M. 101–2 Castaldo, S. 194; and Mauri, C. 189 Chandler, A.D. Jr 117 change 53–4, 100 characteristics model 48, 50, 54, 56, 61, 70, 97, 103 Chaum, D. 256 Choi, Y.B. 91, 92, 94, 98, 103, 133 choice 8, 9, 11, 21–3, 52, 54, 56, 64–5, 72, 90, 202, 217, 228–9; as calculus of pleasure and pain 34–7, 42–3; divorce from utility 28–31; dynamic model 202–10; innovative 69–75; latent liaison with utility 31–4; logical 92; marriage with utility 23–8; and patterns of connection 99, 104–5; and preferences 37– 40; process of 32; random 133–4 Christopher, M. 179 Cipolla, C.M. 67 Clark, G. et al. 153 Clark, K.B. 113 Coase, R.H. 92 cognition 94–6, 179 collectable goods: aesthetics 182–3; history and time 184; reduced/lost-utilization values/ functions 182; semiotics 183–4; seriality 182; uniqueness and rarity/distinguishing features 181–2 collective goods, neglect of 217–19 Colonna, C.M. and Henry, L.H. 194 comfort 74 communication 14–16, 254–5 communitarians 219 competition 58–9 complexity 174, 199–200, 201 consultancy function 115, 118 consumer capital: and change 100; and creativity 99–100; and entrepreneurial activity 101–2; and environment 100–1; and preferences 98–9, 103; and prices 102–3


INDEX Etzioni, A. 130, 219

theoretical changes in 14–15; under postmodern conditions 230–1; see also anomalous consumption contract 25 Corner, D. 83 Cozzi, G. et al. 175, 188 Crafts, N.F.R. 143, 151, 158, 160, 163, 164, 169; and Harley, C. 151, 160, 167 Crouzet, F. 151 Crozier, W.R. and Greenhalgh, P. 210 Cyert, R.M. and March, J. 7, 123

Feinstein, C.H. 160, 169 Fielding, H. 159 Fine, B. and Leopold, E. 228 Fiocca, R. 194 firm–consumer relationship 180 firm-driven processes 188; communication strategies 189; positioning strategies 188; price/distribution strategies 189; product family strategies 188–9 Fisher, F.M. et al. 116 Fogel, R. 152 Foxall, G.F. 83 Freudenberger, H. 82; and Cummins, G. 148, 152, 167 Friedman, M. 130

D’Aveni, R. 174 David, P.A. and Bunn, J.A. 119 Davis, S. 175 Deane, P. and Cole, W. 152 Debreu, G. 31–3, 42, 91 decision process 52, 91, 122; costs of error 93– 4; and evolutionary cognition 94–6; and logic 92–3; neglect of 217–19; and pattern recognition 92, 102–3; pressures on 136–7; stimuli-response 91–2, 104 demand 97 Demaris, A. 147 demonstration effect 52 demonstrative publicity 232 detachment hypothesis 230 Douglas, M. and Isherwood, B. 183 Duesenberry, J.S. 52 Durkheim, E. 210 dynamic model 108–9, 202–6; discussion of 208–10; simulation of 206–7

Gay, A. 42 Gell, A. 162 Georgescu-Roegen, N. 5, 17, 28, 42, 48 Gershuny, J. 59–60 Giddens, A. 223, 228 Gilboy, E. 157 globalization 57, 58–9 goals 122–4; challenge of 137–8; and making sense of unfamiliar situations 131–4; and misplaced assumptions/inferences 134–6; pre-commitment to 127–8; pressure and error 136–7; setting/attainment of 124–7; universal 218; utopian 126 Gombrich, E.H. 210 Grandinetti, R. 194 Gronroos, C. 194 Gualerzi, D. 62 Guerzoni, G. 194 Gummesson, E. 194

Earle, P.E. 17, 49, 55, 72–3, 103, 119, 159 Eccles, Lord 181 Eco, U. 238 economics 25–7, 42, 46, 89–90, 215–16, 237–8 Edgeworth, F.Y. 23–7, 30, 33, 35–6, 40, 41–2 effectiveness 175–6 efficiency 7–9, 70–1, 75, 76–81, 100, 175 Egidi, M. 72; and Marris R. 72 Elsner, J. and Cardinal, R. 183 Elster, J. 127, 219; and Roemer, J.E. 43 entrepreneurs 90, 101–2, 115 ethics 25–6, 27

Habermas, J. 218, 219, 232 habits: adherence to 14, 200–1; defined 200–1; formation 14, 200 Hahn, F. 126 Hamel, G. and Prahalad, C.K. 174 Hardy, M. 147 Harley, C. 162, 163; and Crafts, N.F.R. 164, 167


INDEX 116–17; adjusting to 100; and chain of network possibilities 67–9, 71, 76; and consumer choice 69–71, 75–81; diffusion of 14; incremental 118; product 61; social 59– 60; trickle-down effect 117 institutional theory 7 Ironmonger, D.S. 48, 50–1, 62 Irwin, J. and Brett, K.B. 65–6, 82

Harris, M. 145 Haskell, F. and Penny, N. 194 Havlena, W.J., and Holak, S.L. 193; and Holbrook, M.B. 194 Hawtrey, 198–9 Hayek, F.A. 91, 93, 97, 113 Hecht, J. 159 hedonism 225–6 Heiner, R.A. 72, 119, 137 Henderson, R.M. and Clark, K.B. 117 Hesse, G. 243 Hicks, J. 30, 34 Hildenbrand, W. 41 Hirschman, A. 221 Hirschman, E.C. 186, 194; and Holbrook, E.C. 194 Hirsh, F. 224 Hofstadter, D. 254 Hogarth, R.M. and Makridakis, S. 138 Holbrook, M.B., et al. 194; and Grayson, M.W. 194; and Hirschman, E.C. 194; and Schindler, R.M. 193 hours of work: and backward-bending labour supply curves 13, 153–7; and capitalization hypothesis 150–1; and consumer revolution 157–62; and disappearance of Saint Monday 146–9; and food supply 152–3; increase in 143–4; and New Household Economics 162–6; and substitution effect 13, 162–4, 168; use of witnesses’ accounts 144–6, 167 household production 48, 55, 59, 98, 108 Houthakker, H.S. 61 Hunt, E.H. 153 Hutter, M. 246, 253, 254–5

Jones, E. 160 Juster, F.T. and Stafford, F. 162 Kahneman, D. and Tversky, A. 219 Karasek, R. 54 Kay, J.A. 133, 134, 136 Kelly, G.A. 72, 105, 131 King, P. 159–60 Kirzner, I.M. 3–4, 7, 76, 90, 97 knowing that/knowing how 11, 93 knowledge 96, 108; bucket theory 209; searchlight theory 209 Körner, S. 42 Krampen, M. 239 Krippendorf, K. 254 Kuhn, T. 217 Kusamitus, T. 161 labour: backward-bending supply curves 13, 153–7; changes in working hours 146–9, 167; and consumer revolution 157–62, 167– 8; decline in nutritional constraints 152–3; and New Household Economics 162–6; rise in capital–labour ratio 150–1, 167; supply 13, 143–4 Lacher, K.T. and Mizerski, R. 194 Lancaster, K.J. 7, 48, 50–1, 54, 56, 61, 69–71, 75, 103, 104, 107, 179 Langlois, R.N. 11, 115, 119; and Koppl, R.G. 107; and Robertson, P.L. 11, 108, 109, 113, 116, 118 Lash, S. 231 Latour, B. 232 Laudan, L. 217 Laurens, A.F. and Pomian, K. 184, 194 lean production 58

identification 244–5 income effect 13 Indian calicoes 65–7, 82–3 indifference curve 6, 9, 28–30, 37, 39, 78 individuality 54–6 industrial transformation 57 information 92, 202, 236 Ingrao, B. and Israel, G. 42 innovation 9–10, 46–7, 51, 54–5, 56, 104–5, 111,


INDEX Mokyr, J. 166 monetary theory 238 Morley, D. 227 motivation 14, 198; and adherence to habit 200–1, 208; and desire to imitate/conform 201, 208; and pursuit of novelty 199–200, 201, 208 Muensterberger, W. 182 Mukerji, C. 65–6; and Schudsor, M. 68 Mulgan, G. and Wilkinson, H. 168

learning 3, 16, 56, 62, 199, 201 Leibenstein, H. 117 Leijonhufvud, A. 102 Leiss, W. et al. 232 leisure revolution 166, 168 Lemire, B. 67, 68, 83, 160, 161 lifestyle 55, 57, 61 Lindert, P. and Williamson, J. 156, 169 Loasby, B.J. 4, 11, 111 Locke, J. 209 Loomes, G. and Sugden, R. 124 Luhmann, N. 239–40, 245 Lyotard, F. 218

Napoleoni, C. and Ranchetti, F. 42 needs 54, 56, 57, 62, 109–10 Nelson, R.R. and Winter, S.G. 108, 110 neoclassical theory 46, 47–8, 97, 98–9, 103, 107– 8, 119, 128, 216 new goods 50–1, 59–60, 83; adoption/diffusion of 12–14; selection/identification of 75–6, 78–9 New Household Economics 162–6 Newland, O. 129–30 Newman, P. 41 Noro, A. 223 North, D.C. 254 novelty 16, 64–5, 111, 116; choosing 75–6, 78–9; consumer 117; decline in 117; defined 1–2; double effect on pleasure 2–3, 76; dual role of 4–5; as focal point 74; inherent/ctualized 186; motivational importance of 3; producer 117–18; pursuit of 14, 73–5, 199– 200, 201; role of in consumption 6–9; taste for 65–7; time dimension 2; uncertainty/ unpredictability of 69–71

MacCannell, D. and MacCannell, J.F. 238, 239 McCracken, G. 255 MacIntyre, A. 219 MacKay, D. 113 McKendrick, N. 83, 159, 160–1, 166, 169; et al. 157–8, 159 Maffesoli, M. 221, 232 March, J.G. 7 marginalist theory 48, 122–3 Margolis, H. 219 market-driven processes 192–3 Marshall, A. 16, 24–5, 48, 91, 92, 94, 97, 100, 101, 107 Martindale, C. et al. 210 Marx, K. 143, 150 Mas-Colell, A. et al. 21 Maslow, A.H. 54, 57, 62, 123 mass customization 175, 179–80; anti-tech/ hyper-tech 13, 174, 180, 189 May, K. 42 meaning 15, 112–13 Menger, C. 99 Menger, K. 42 Middledon, E. 84 Miller, D. 217, 229 Miller, M.B. 255 Minsky, H.P. 129 Mintz, S.W. 68 Mirowski, P. 41, 43 Mitchell, W.C. 5 modernization 215

Oeler, K. 238 Ohmae, K. 58 Opie, R. et al. 190 optimization 90, 91, 93, 236 paradigm development 217 Pareto, V. 6, 28–30, 34, 39, 104, 107, 209 Parrinello, S. 49 Pasinetti, L.L. 62 payment 241; by immediate users 250–1; as form of sign use 249–50; through


INDEX Rheims, M. 181 Richardson, G.B. 11, 108 Riley, J.G. 238 Rizvi, S.A.T. 209 Robertson, P.L. 115 Rocha, J. 134 Roemer, J.E. 17 Rogers, E.M. and Shoemaker, F.F. 194 Rorty, R. 218 Rosen, S. 255 Rothstein, N. 83 routines 110–12 Rule, J. 150, 167; et al. 150, 151 Ryle, G. 93

advertising 251–2; through public purses 252–3, 256 Peirce, C.S. 238, 254 Penrose, E.T. 95, 123 personal construct theory 72 Phelps-Brown, E.H. and Hopkins, S.V. 155, 160 Pine, B.J. 175 play/s: emergence of new 246–7; medium of 245–6, 255 pleasure 13, 73–5, 76, 77–9, 82, 225 Plumb, J.H. 83 Polanyi, M. 108 Pollack, R.A. 49 Pomian, K. 181, 183 Popper, K. 209 positional goods 224 postmodernism 14–15, 215, 220, 227, 229–30; consumption under 230–1; doubts concerning 232–3 poststructuralism 230 pragmatism 218 preferences 30–2, 37–40, 64, 96–9, 103, 107, 128, 179; endogenous 49, 97; exogenous 52, 53, 62, 89 Prieto, L.J. 239 problems 3, 109–10 product differentiation 57, 70 production 107–9, 220; boundaries with consumers 114–18; defensive/creative goods 198–9; discovery in 3–5, 7; personalized goods 176; rarefaction process 180 productivity 98–9 psychology 216

Saarinen, A.B. 194 Samuelson, P. 30, 34 Sassen, S. 227 satisfactions 48, 50–1, 111, 117, 132, 163, 209 Saussure, F. de 238, 240 Schnapper, A. 181 Schor, J. 168 Schücking, L. 246 Schulze, G. 220 Schumpeter, J.A. 40, 42, 90, 95, 100, 103 Schwarz, L.D. 153–6 Scitovsky, T. 14, 73–5, 84, 137–8, 198–200, 201–2 Sebeok, T. 238 self-actualization 123 self-identity 15, 61 semiotics 238–40; of collectable goods 183–4, 185 Sen, A. 23, 39, 48, 104, 219; and Williams, B. 43 Sethi, R. and Franke, R. 210 Shammas, C. 152 Shapiro, N. 59 Shields, R. 220, 221 Shubik, M. 42 Sidgwick, H. 41 signs 14–15, 236–7, 253–4; in conventional economic approaches 237–8; defined 240–1; in the medium of play 245–7; and payment 249–53; recognition of 247–8; in semiotics and social systems theory 238–40; star

Raffaelli, T. 91 Ranchetti, F. 41, 42 rarefaction process 13, 180, 187, 193–4; consumer-driven 189–92; firm-driven 188–9; market-driven 192–3 rational choice theory 23, 32–40, 48, 49, 89–90, 95, 96, 99, 229 Rawls, J. 36 Regret Theory 124–5 Reid, D. 146 Rescher, N. 38, 43


INDEX pattern 248–9, 255–6; as used for direct consumption 243–4; as used to identify oneself 244–5; as used in transactions 242–3 Simmel, G. 255 Simon, H.A. 72, 92 Smith, A. 105, 156–7 social distinction theory 224–5, 230 social identity 54 social systems theory 238–40 society 219–20 sociology 216 Solomon, R.L. 201; and Corbit, J.D. 200–1 Spence, A.M. 237 Spencer Brown, G. 254 Stanback, T.M. Jr et al. 57 Stewart, S. 190 Stigler, G.J. 17, 42; and Becker, G. 22, 84, 98–9, 107–8 Stoller, M.A. 194 Stone, L. 166 Strauss, A. 222 Strong, R. 252 structuration theory 227–9 Styles, J. 161

total quality 58 transactions 108, 112–13, 114–15, 237, 241–3; control 242; negotiation 242; search 242 Tucker, R.S. 156, 160, 169 Tversky, A. and Kahneman, D. 219

taste 9–10, 12, 51–2, 62 Taylor, C. 219, 231 Taylor, F.H. 194 Taylor, L.D. 210 technical change 48, 50–2, 56, 57, 59–60 Teece, D.J. and Pisano, G. 108 Thaler, R. 128; and Shefrin, H.M. 127 Thayer, R.E. 16 Thirsk, J. 157 Thomas, G.Z. 66 Thompson, E.P. 146 Thompson, J. 193, 194, 232 Thornton, P. 83 time budgets 13, 143; beeper method 144; diaries 144; random hour recall 144; witnesses’ accounts 144–6, 167

Walzer, M. 219 Watzlawick, P. et al. 126, 131 Weatherill, L. 159 Whitehead, A.N. 93 Wicksteed, P.H. 42 Williamson, O.E. 7, 119, 254 Winston, G.C. 48 Witt, U. 16, 17 Wittgenstein, L. 36–7 women 224 Woo, H.K.H. 91, 95, 98, 100, 102 Wrigley, E.A. and Schofield, R. 152, 153 Wundt curve 73

Unger, L.S. and Kernan, J.B. 194 utility theory 5–6, 9, 57, 76, 78, 84, 107–8, 216, 229; abandonment of 28–31; ambiguous status of 31–4; goal 48; introduction of 23–8; ordinal/cardinal 30, 42, 47–8; pleasure and pain 34–7; process 48 Uusitalo, L. 192, 217, 219, 229; and Ahola, E.-K. 224; and Lindholm, M. 220, 229; and Uusitalo, J. 217 value 96–8, 102 Veblen, T. 169, 223 Vicari, S. 58, 174 Viner, J. 6 von Hippel, E. 114 Voth, H.-J. 152 Vries, J. de 69, 153, 158, 162, 167

Zajonc, R.B. 210 Zamagni, S. 42, 48, 49, 61, 62