Consumption, Globalization, and Development

Consumption, Globalization and Development Jeffrey James

Consumption, Globalization and Development

Also by Jeffrey James CONSUMER CHOICE IN THE THIRD WORLD CONSUMPTION AND DEVELOPMENT ECONOMICS OF NEW TECHNOLOGY IN DEVELOPING COUNTRIES (with F. Stewart) GLOBALIZATION, INFORMATION TECHNOLOGY AND DEVELOPMENT IMPROVING TRADITIONAL RURAL TECHNOLOGIES TECHNOLOGICAL SYSTEMS AND DEVELOPMENT (with H. Khan) TECHNOLOGY, INSTITUTIONS AND GOVERNMENT POLICIES (with S. Watanabe) THE IMPACT OF WORLD EMPLOYMENT PROGRAM RESEARCH ON TECHNOLOGY (with F. Stewart, A. Bhalla and others) THE STATE, TECHNOLOGY AND INDUSTRIALIZATION IN AFRICA THE TECHNOLOGICAL BEHAVIOUR OF PUBLIC ENTERPRISES IN DEVELOPING COUNTRIES THE TRANSITION TO EGALITARIAN DEVELOPMENT (with K. Griffin)

Consumption, Globalization and Development Jeffrey James Professor of Development Economics Tilburg University The Netherlands

First published in Great Britain 2000 by

MACMILLAN PRESS LTD Houndmills, Basingstoke, Hampshire RG21 6XS and London Companies and representatives throughout the world A catalogue record for this book is available from the British Library. ISBN 0–333–77266–0 First published in the United States of America 2000 by ST. MARTIN’S PRESS, INC., Scholarly and Reference Division, 175 Fifth Avenue, New York, N.Y. 10010 ISBN 0–312–23056–7 Library of Congress Cataloging-in-Publication Data James, Jeffrey. Consumption, globalization, and development / Jeffrey James. p. cm. Includes bibliographical references and index. ISBN 0–312–23056–7 (cloth) 1. Consumption (Economics)—Developing countries. 2. Technological innovations—Economic aspects—Developing countries. 3. Competition, International. 4. Markets. I. Title HC59.72.C6 J364 2000 339.4'7'091724—dc21 99–052077 © Jeffrey James 2000 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 0LP. Any person who does any unauthorised act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The author has asserted his right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988. This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. 10 09

9 08

8 07

7 06

6 05

5 04

4 03

3 02

2 01

Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, Wiltshire

1 00

Contents List of Tables

vi

List of Figures

vii

Preface

viii

Acknowledgements 1

Introduction

Part I 2 3 4 5

7 8

1

Consumption Theory and the Welfare Effects of Globalization

9

Globalization, Preference Change and Consumer Welfare in Developing Countries

11

Globalization, Conspicuous Consumption and the International Demonstration Effect Reconsidered

28

From Global Products to Individual Functionings: Medicinal Drugs in Developing Countries

61

Do Consumers in Developing Countries Gain or Lose from Globalization?

85

Part II 6

ix

Countervailing Influences over Patterns of Global Consumption

103

Can Appropriate Products Capture Mass Markets in a Globalizing World? a Case Study from India

105

Cultural Advantage Reversal: the Case of Telenovelas in Brazil

122

Globalization and the Potential for Social Marketing in Developing Countries

131

Index

141

v

List of Tables 1.1 2.1 3.1 4.1 5.1

5.2 6.1 7.1 7.2

An outline of Part I Growth in the number of television receivers Economic vs. sociological approaches to taste change Source of individual-specific variations in health functionings following drug usage The spread of advertising in the developing world – top 10 countries in advertising expenditure as a share of GDP, 1986 and 1996 Globalization, frustration and consumer disappointment The top 10 advertisers on television in India, 1984–5 A comparison of television ratings Selected technological achievements at TV Globo

vi

4 13 34 67

89 100 115 124 128

List of Figures 2.1 2.2 2.3 3.1 3.2 3.3 3.4 4.1 4.2 4.3 4.4 6.1 6.2 6.3

The quasi-cardinal approach 15 The welfare effect of a new product 17 The direction of product innovations in the developed countries 19 a) Nurkse’s view b) The Hirsch/Veblen view 31 Positionality and the direction of change in demand 47 The appearance of modernity 48 The squeeze on essential characteristics 51 Functionings in Sen’s approach 63 Drug efficacy, toxicity and health functionings over time 65 Drug usage, characteristics and functionings 75 Sequentiality in the determinants of health functionings 77 Two types of markets in developing countries 107 The laundry detergent market prior to introduction of Nirma 113 A comparison of Surf and Nirma in characteristics space 118

vii

Preface During the 1980s and early 1990s I wrote a number of essays expressing my concern about the ability of traditional consumption theory to come to grips with the series of changes in developing countries that are frequently referred to as the modernization process. Jointly expressed in a volume entitled Consumption and Development (Macmillan 1993), those concerns have only been intensified by the changes in the relationships between rich and poor countries wrought in the 1990s by information technologies of various kinds. For as this volume seeks to demonstrate, the various processes of globalization render even more implausible the assumptions of traditional consumption theory, and so too the welfare foundations of free market economics that rely so heavily on those assumptions. Jeffrey James Tilburg

viii

Acknowledgements Chapters 3 and 4 appeared originally in other sources. The former was published under the title ‘Positional Goods, Conspicuous Consumption and the International Demonstration Effect Reconsidered’, World Development, 15(4), 1987. It is reprinted here with permission from Elsevier Science. Chapter 4 appeared originally in J. James and H. Khan, Technological Systems and Development, Macmillan 1998. It is reprinted here with permission from the publisher. The author and publishers wish to thank the following who have kindly given permission for the use of copyright material: the table, ‘What the World is Watching’, © The Economist, London, 15 August 1998; Figure 1 from A. Amsden, ‘The Division of Labour is Limited by the Type of Market: The Case of the Taiwanese Machine Tool Industry’, World Development, 5(3), 1977, pp. 217–33, with permission from Elsevier Science; Table 6.1, showing the top 10 advertisers on television for 1984–5 in India, by arrangement with the Economic Times, which may be viewed at www.economictimes.com © Bennett, Coleman & Co. Ltd.; UNESCO for Table 6.5 from the UNESCO Statistical Yearbook, 1997; Table 3.4 from Human Development Report 1998 by the United Nations Development Programme, © 1998 the United Nations Development Programme, used by permission of Oxford University Press, Inc.; Figure 3.1 from F. Stewart, Technology and Underdevelopment, 1977, with permission from Macmillan; Figures 2.1 and 2.2 from J. Harsanyi, ‘The Welfare Economics of Variable Tastes’, The Review of Economic Studies, 21(3), 1954, with permission of the journal.

ix

This page intentionally left blank

1 Introduction

The conventional consumption theory that is found in most textbooks on microeconomics is a part of demand theory and as such belongs to positive economics, but it also forms part of welfare economics, which is concerned with prescriptive or normative issues. 1 Indeed, with regard to this latter function, it is possible to argue that, The entire edifice of normative or welfare economics rests on consumers’ sovereignty, i.e. the assumption that individuals know what is best for them; that they make unconstrained decisions in the market; and that this will lead to maximum welfare not just for the individuals but also for society. Maximising consumer welfare thus provides the ultimate basis of every normative prescription in economics, including those about optimum trade policies, cost/benefit analysis, price regulation and so on. This is the foundation stone for the justification of the market model, with consumption as the driving force of the economic system. (Stewart, 1997: 1) Traditional consumption theory has, of course, been subject to considerable criticism from a variety of disciplines, not only as a theory of demand but also in the role (described in the quotation) as a basis upon which welfare economics fundamentally rests.2 It is not our intention here to review this large body of literature. Rather, what we would like to suggest is that traditional theory is especially unsuitable as a framework for analysing a key aspect of the current debate about 1

2 Consumption, Globalization and Development

globalization, namely, the ways in which advances in information technology enable the developed countries to exert an ever increasing influence on consumption patterns in the developing countries. Expressed more specifically, the basic argument we shall advance in the first part of the book, is that whereas traditional consumption theory allows of ‘no surprises’ (so that the consumer invariably gets what he or she expects), the influence thus exerted by globalization on developing countries is prone to yield precisely the feelings – of frustration and disappointment – that this theory assumes away. Underlying a good deal of this argument, in turn, is the view that because of globalization of the mass-media and huge increases in advertising expenditures, products from developed countries are penetrating increasingly into parts of the Third World where consumption systems diverge sharply from the conditions prevailing in the former group of countries, for whose consumption systems the products are initially designed. (By a consumption system we mean the related package of characteristics, such as income, location, consumer skills, infrastructure and so on, in which the consumer finds himself. Just as a particular technological system exhibits certain patterns among its constituent parts, so too are there discernible differences between consumption systems, which may, for example, tend to exhibit traditional as opposed to modern features.)3 Problems often arise, however, when products designed for one particular type of system are introduced into an entirely different context. For then, the chances are that the consumer in the recipient system will tend to experience disappointment and frustration rather than, or additional to, the utility that was expected from the product in question. Satellite technology, for example, enables the mass-media and advertising to reach many of the remotest areas in developing countries, such as a village in China, which is then ‘as likely to be linked to Hollywood movies and advertising … as by road or railroad to a village 50 kilometres away’ (Human Development Report, 1998: 63). The problem, though, is that consumers in locations such as these may often tend to experience frustration and disappointment (as well as feelings of satisfaction) from globalization, because the consumption system to which they belong lacks key elements of the system from which the globalizing influences are initially propagated. If one is to understand the mechanisms that give rise to this outcome, rather than the result that is predicted by traditional

Introduction 3

theory, it is necessary first to specify the assumptions in question and then to provide an alternative and more realistic framework for assessing the welfare effects of globalization on poor countries.

Traditional consumption theory and getting what one expects As explained in Chapter 5, the most obvious way in which traditional consumption theory eliminates possible discrepancies between expected and realized utilities is through the assumption of perfect knowledge, which applies not only to the preference orderings of the consumer but also to the range of alternatives from which he or she may freely choose. In addition to being known, however, preferences also need to be constant, since once these change, circumstances may be such that the consumer feels worse off in the new situation than the old. As regards the nature of individual preferences, the absence of any discrepancy between what is expected and what is realized is brought about by the assumption that individual utility depends purely on one’s own consumption. This assumption excludes not only the various types of social comparisons that so pervade the real world in which consumption decisions are made, but also the whole range of phenomena that economists refer to as negative externalities of consumption. Yet another way in which expected and actual utilities are made to converge in traditional theory is that consumption is assumed to take place at the point where goods are purchased. What this implies is that utility accrues to the consumer from the act of purchase itself, rather than from what is actually done with the product in question. Realistically, however, the latter assumption is usually far more relevant and it makes the relationship between consumption and utility a highly variable one that may often be difficult to predict at the time of purchase.

Relaxing the assumptions of traditional theory The assumptions that have just been described are listed in the first column of Table 1.1. Chapters 2 to 5 are concerned, however, to relax each of these assumptions and to replace them with alternatives that

4 Consumption, Globalization and Development

Table 1.1

An outline of Part I

Traditional assumption

Alternative approach

Relevant chapter(s)

Perfect information about Globalization and preference orderings and information imperfections consumption possibilities

Chapters 5

Individual utility depends Externalities in global only on own consumption consumption

Chapters 3 and 5

Preferences constant

Globalization, the Chapters 2 and 5 international demonstration effect and preference change

Preferences constant and known to the consumer

Unknown preferences and changing consumption patterns

Chapter 5

Consumption takes place at the point of purchase, where utility is derived

From global products to actual human functionings

Chapters 4 and 5

better reflect the realities of consumption by poor countries in a globalizing world. Corresponding to each assumption in column one, therefore, is an alternative point of departure which is summarized by the entries in the second column of Table 1.1. The final column indicates the chapters in which each such topic is discussed. The conclusion that emerges from these chapters is that there are a variety of mechanisms through which globalization affects the well-being of consumers in developing countries, mechanisms that are not allowed for in traditional consumption theory. In some cases, the outcome is that consumers will tend to gain from these mechanisms while in other cases they will tend to suffer instead from frustration and disappointment. The latter feelings, we find, tend to bear with particular severity on low- rather than highincome groups in developing countries. And this finding, in turn, seems to us to be less the result of mere chance and more the reflection of the hypothesis that was advanced at the outset of this chapter, namely, that globalization is penetrating increasingly into consumption systems that diverge from and are hence often lacking in, at least some of the elements required for modern products to yield the utilities that are expected of them.4

Introduction 5

Countervailing influences over patterns of global consumption Many of the welfare losses to which we refer in Part I were based, explicitly or implicitly, on the assumption that globalization intensifies the desire for goods and cultural communications from the developed countries, a tendency which, in turn, would be likely to lessen the demand for locally produced products. Yet, in Chapters 6 and 7 respectively we present cases – of detergents in India and television programs in Brazil – where precisely the opposite tendency could be discerned, namely, for products made by locally owned firms to capture over time markets previously dominated by foreign goods. Given the vast size of the markets concerned,5 neither of these cases can be easily dismissed as merely marginal exceptions to the more general trend that is being wrought by the globalization of consumption in the Third World.6 Nor were these cases where a product designed initially for high-income consumers somehow trickled down to the medium and low-income groups. Rather, in both cases the mass markets were captured by a product designed specifically for these latter income groups. In short, what occurred in India and Brazil seems to be sufficiently exceptional to warrant close scrutiny in any investigation of the effects of globalization on consumption and welfare in developing countries. Apart from various economic factors, whose explanatory power, we suggested, was exerted on both the supply and demand-sides of the market, each of the two case-studies clearly also contains an important cultural dimension. In India, for example, one cannot ignore the historical influence exerted by the ‘swadeshi movement’ which ‘developed as a theory as early as the nineteenth century’ (Leadbeater, 1993: 16) (the word ‘swadeshi’ literally means from ‘one’s own country’ and thus ‘swadeshism stood for … the use of Indian manufactures in preference to imported goods’). ‘After Gandhi’s return from South Africa and particularly his ascension to the leadership of Congress in 1920 it became a vehicle for mobilising India’s vast rural populace’ (Leadbeater, 1993: 16). In combination with the subsequent Nehruvian brand of socialism after independence, when ‘there was no greater evil than conspicuous consumption’ (Datta-Ray, 1998: 11), the many decades of ‘Gandhian simplicity’, ‘have instilled’ according to some observers ‘an almost puritanical sense of “value” in the country’

6 Consumption, Globalization and Development

(The Economist, June 27, 1998: 80). It is this attitude, one presumes, that helps to explain why India is probably the only country in the world where Coca-Cola is less popular than a local soft drink and why a locally made detergent offering a high proportion of relatively ‘lowincome’ characteristics was able to capture a dominant share of the market in that particular type of product, from an established multinational corporation.7 In Brazil, as well, the ability of the telenovela to displace well-known foreign imports was predicated in part, as we shall see in Chapter 7, on the congruence of this product with local popular culture. In sum, then, after examining these two cases, one cannot but agree with Featherstone when he argues that globalization ‘is not just a question of the everyday practical culture of local inhabitants giving way to globally marketed products’ and that ‘the scenario of ‘cultural dumping’ of obsolete American television programs on a powerless nation-state on the periphery is only one possibility from a range of responses’ (Featherstone, 1995: 116). The final chapter in the book, Chapter 8, also represents a countervailing force in the sense described earlier, albeit a force that is exerted not principally by market actors, as in Chapters 6 and 7, but rather by actors such as governments, NGOS and aid donors. For these non-market agents could, in principle, use the same globalizing forces as the mass-media and advertising to counter, rather than reinforce, some of the welfare losses that were described in the first part of the book. More specifically, it is argued in Chapter 8, on the basis of a good deal of empirical evidence, that countervailing influence of this kind can best be exerted by using methods and techniques borrowed from commercial marketing (that is, by practising so-called social marketing).

Notes 1. See James (1993). 2. Some of these criticisms are contained in James (1993). See also Stewart (1997). 3. For a detailed discussion of the concept of a technological system in developing countries see James and Khan (1998). Much of this discussion in turn is based on earlier work by Stewart (1977). 4. If one assumes that the poorest groups in developing countries tend to belong to consumption systems that are most unlike those prevailing in

Introduction 7

developed countries, then it is worth noting that in India ‘The upsurge in purchases of consumer durables and products reached even the 90 million lowest-income households in India. Although two-thirds of them had incomes below the official poverty line, more than 50% owned wrist watches, 41% bicycles, 31% transistor radios and 13% fans’ (Human Development Report, 1998: 47). 5. The Indian detergent currently sells to 70 million households on a regular basis (see Chapter 6), while in Brazil one telenovela enjoyed an audience of 80 million viewers (Neto, 1997). 6. Though of course such cases do indeed exist, as when, for example, firms from developing countries find niche markets in certain developed countries. For further discussion see Reeves (1993). 7. See also a discussion of the difficulties that Nike and Reebok sports shoes encountered in competition with the local footwear maker, Bata India, which sells at much lower prices (Business Week, April 27, 1998). See also ‘The Day of the Desi Brand’, Business India, May 18–31, 1998.

References S. Datta-Ray (1998). ‘In India, Things Go Better With Self-Reliance’, International Herald Tribune, May 27. M. Featherstone (1995). Undoing Culture, Sage. J. James (1993). Consumption and Development, Macmillan. J. James and H. Khan (1998). Technological Systems and Development, Macmillan. S. Leadbeater (1993). The Politics of Textiles, Sage. R. Neto (1997). ‘Case Study: TV Globo – The Brazilian Media Giant’, The International Executive, 39 (2). G. Reeves (1993). Communications and the ‘Third World’, Routledge. F. Stewart (1977). Technology and Underdevelopment, Macmillan. F. Stewart (1997). ‘Consumption, Globalization and Theory: Why There is a Need for Radical Reform’, Background Paper for the Human Development Report, 1998.

This page intentionally left blank

Part I Consumption Theory and the Welfare Effects of Globalization

This page intentionally left blank

2 Globalization, Preference Change and Consumer Welfare in Developing Countries

Introduction Among numerous other things, globalization means a greater exposure of developing countries to products and product innovations occurring in the developed countries. And partly as a result of this increased exposure and partly for other reasons that are described below, globalization often also entails a change in preferences among consumers in developing countries in favour of such products. Traditional micro-economic theory based on the notion of ordinal utility and welfare economics, however, provides remarkably few tools for assessing the welfare effects of these important phenomena on consumers, partly because traditional theory finds it difficult to come to grips with product innovations 1 and partly because in welfare economics ‘there is no fixed standard for value judgment when tastes are variable’ (Dixit and Norman, 1978: 1). It is true that the normative aspects of preference change have received more attention from non-mainstream economists such as Galbraith (1973), but his approach requires one to assume that consumers are not the best judges of their own well-being.2 The purpose of this chapter is to provide an alternative framework for assessing the welfare effects on consumers of the type of preference change that is associated with globalization. Known as the ‘quasi cardinal’ approach it is not bound by the strictures imposed by ordinal utility,3 nor does it require one to assume that the consumer is incapable of judging what is in her own best interest. Furthermore, this approach focuses on what for the ‘globalized consumer’ in a 11

12 Consumption, Globalization and Development

developing country is likely to be the key welfare issue, namely, whether the particular circumstances at hand allow her to indulge her new preferences or whether they are such as to frustrate this achievement. Let us begin however by analysing the relationships between globalization, modernization and preference change in developing countries.

Globalization, modernization and preference change ‘Globalization researchers’ according to Sklair, have focused in particular on two new phenomena that have become significant in the last few decades: the spread of TNCs through processes such as ‘the globalization of production and capital’ and ‘the new international division of labour’ (NIDL); and transformations in the global scope of the mass media. (Sklair, 1991: 166) With regard to the first of these two phenomena, what is especially important for our purposes is the recognition that ‘developing countries have succeeded in attracting growing investment flows, reaching $84 billion in 1994 to account for 37 per cent of world FDI inflows. This is a continuation of a trend that began in 1990 and has propelled developing countries to become a major force in world FDI’ (UN, 1995: xxiii). As to the question of how far globalization of the mass-media has affected the developing countries, one can point first to the six-fold expansion in television receivers between 1980 and 1995, as shown in Table 2.1. Impressive as these growth rates in television ownership undoubtedly are,4 however, they do not convey the full extent to which the mass-media in developing countries have become globalized. For what these figures do not reveal is the emergence in the 1980s and the subsequent diffusion throughout the world of multi-channel television, a phenomenon that was precipitated by the rapid fall in the cost of launching communication satellites. (For a variety of reasons, moreover, these extra channels tended to be filled not with local programming but rather with material provided by American media groups such as films, cartoons and older vintages of soap operas.)5

Preference Change and Consumer Welfare 13

Table 2.1

Growth in the number of television receivers

Continents, major areas and groups of countries

World total Developing Countries Sub-Saharan Africa Arab States Latin America and the Caribbean Eastern Asia and Oceania Southern Asia Least developed countries Developed Countries

Total number of television receivers (millions) 1980

1985

1990

561

748

918

201 7.1 16 55 96 15 2.1

340 13 22 72 187 34 4.5

512 19 28 91 304 56 6.2

547

578

646

86 4.4 9.3 35 24 6.1 1.3 475

1995 1,157

Source: UNESCO, Statistical Yearbook, 1997

Though they may perhaps seem to be, the two dimensions of globalization that have just been described are not in fact independent of one another. On the contrary, they are closely related, among other ways, by the fact that policies of liberalization in many developing countries have given rise to the commercialization of the mass-media and in particular to an increase in television advertising by multinational corporations, which, as noted above, have been investing rapidly growing amounts in at least some parts of the Third World (Mattelart, 1991).

Globalization and modernity The empirical findings of sociologists such as Inkeles and Smith (1974) give one some reason to suppose that the globalizing tendencies described above have hastened the process by which individuals in developing countries acquire modern values, a process which usually includes the acquisition of a preference for modern consumer goods. For what these authors found was that variations in individual modernity (as defined by a range of attitudinal variables) among a large sample of individuals in developing countries could be best explained by differences in the degree of exposure to massmedia and work experience in large-scale (e.g. foreign) factories,

14 Consumption, Globalization and Development

factors that are both heavily influenced by the forces of globalization described above. And to the extent that individual preferences are being changed in this way, we need to assess the effects on consumer welfare to which it gives rise.

The welfare economics of variable preferences As already noted, though it forms the basis of traditional welfare economics (as this topic is described for example by Ian Little 6), ordinal utility theory is incapable of dealing with the impact of preference change. For there is then no way of judging whether to adopt the old or new preferences of the consumer as the relevant frame of reference. One way out of this seemingly intractable problem has been described by Harsanyi as the ‘quasi-cardinal’ approach 7 and for reasons given below, it applies particularly well, we feel, to the fundamental type of preference change with which this chapter is primarily concerned. As described by Harsanyi, this approach turns on the assumption that: [Consumers] are able to compare the utilities they would derive from different groups of commodities at a given moment (on the basis of given taste), but also that they are able to compare the utility (subjective satisfaction) they derived from one group of commodities at the time t1, when their taste was T1, with the utility (or subjective satisfaction) they now derive from another group of commodities at the time t2, when their taste is T2. (Harsanyi, 1954: 204) It is assumed, that is to say, that utility comparisons can be made by the consumer before and after she has undergone a change in her preferences. This assumption may not apply very readily to the type of preference change on which economists have tended to focus, namely ‘impulsive, uncomplicated, haphazard, publicity-induced and generally minor (apples vs. pears) changes in tastes’ (Hirschman, 1986: 145). As opposed to this ‘wanton’ type of preference change, which according to Hirschman is ‘not really a change in tastes at all’ (Hirschman, 1986: 145), are what he refers to as ‘non-wanton’ changes, which are really about changes in values. The

Preference Change and Consumer Welfare 15

distinction is that ‘A taste is almost defined as a preference about which you do not argue – de gustibus est non disputandum. A taste about which you argue, with others or yourself, ceases ipso facto being a taste – it turns into a value’ (Hirschman, 1986: 145, emphasis in original). And it is precisely in relation to the latter type of change, we would argue, that the quasi-cardinal approach best applies, for the consumer whose preferences undergo so fundamental a change is most likely to have a well-defined view of the desirability of her new, as opposed to her initial situation.8 Though it may seem to, the key assumption of this approach does not in fact require that utility can be cardinally measured. What is required instead is somewhat less stringent, namely that ‘if certain objective situations have the same value for the individual on the basis of his present taste T2 as certain other situations had on the basis of his previous taste T1, they must be assigned the same value of the utility function as the value ascribed to the latter situation’ (Harsanyi, 1954: 204). Exactly how this assumption is used in welfare comparisons with variables tastes (or, more correctly, with variable values) can most conveniently be shown in terms of Figure 2.1.

(a)

C

(b)

E

I

K

G

u=3

F A

u=4

u=2 u=3 L

u=2

u=1 D B

0 Figure 2.1

The quasi-cardinal approach

Source: Harsanyi (1954)

01

H

J

16 Consumption, Globalization and Development

Figures 2.1(a) and 2.1(b) represent the indifference maps corresponding to the consumer’s initial and new preferences, respectively. Since we can infer from ordinal utility analysis of this change only that indifference curves further from the origin of each map represent higher levels of satisfaction, there is no way of comparing the utility level associated with the curves belonging to the two maps. The other extreme, the assumption that utilities can be cardinally measured in the same way as height or weight, would enable these curves to be directly comparable since one can then ascribe absolute utility levels to the indifference curves associated with both maps. As its name suggests, the ‘quasi-cardinal’ method falls between these two forms of measurement in that any indifference curve in Figure 2.1(b) that has an equivalent in Figure 2.1(a) receives the same amount of utility as the latter. ‘This piece of information’ writes Harsanyi, ‘may be called the correspondence rule between the two indifference maps’ (Harsanyi, 1954: 204). As an example, the indifference curves in Figure 2.1(a) have arbitrarily been assigned the values u = 1, u = 2 and u = 3. Two of the curves in Figure 2.1(b), GH and IJ, correspond in the sense that has just been described to the curves CD and EF in Figure 2.1(a). The former two indifference curves, accordingly, are assigned the same values as the latter. On the other hand, indifference curves of the second graph [(b)] that have no equivalent on the first graph – in so far as they represent a utility lower, or higher, than any point of the first graph [(a)] – can be assigned an arbitrary value of the utility function, except that this value must be lower than the one corresponding to any preferred position and higher than the one corresponding to any less preferred position. Therefore KL has been given an index higher than the index of any indifference curve drawn on the left-hand graph. (Harsanyi, 1954: 205) Let us now examine how this approach can be applied to the particular case where changes in preferences are associated with the introduction into a developing country of a new product X. In general, X represents a product that is imported from the developed countries or produced locally by multinational corporations, but which, in either case is widely advertised on television and other mass-media as a

Preference Change and Consumer Welfare 17

(a)

(b)

V

K u=3

C

D J

B

u=2

A

u=1

E

F

W I

U=3 Z U=2

H

U Q S

U=1

G

P N

R 0 Figure 2.2

0′

L

M

The welfare effect of a new product

Source: Harsanyi (1954)

result of the processes of globalization that were described earlier. Figure 2.2 shows, as before, the indifference map of the consumer before and after her change in preferences. The difference now is that the change is associated with the introduction of the new product X. Money is measured in this case along the vertical axis and the horizontal axis measures the amount of the new product. In the initial situation, represented by Figure 2.2(a), since the consumer cannot by definition have any demand for the new product X, the indifference curves are all necessarily horizontal. We assume that the money income of the consumer is given by 0B, which is equal to 0'I. Whereas the consumer is initially in equilibrium at point B in Figure 2.2(a), in the new situation she moves to point U, where the budget line is tangential to the highest available indifference curve (KZ). Whether the new equilibrium denotes a position of higher or lower utility than the initial situation depends then ‘on the correspondence rule’ between the two sets of indifference curves and ‘in particular, on which indifference curve of the second map is equivalent to the line BE of the first map’ (Harsanyi, 1954: 208). Though he identifies three possible outcomes, Harsanyi argues that in practice the most common occurrence is the one in which the indifference curve equivalent to BE lies between IM and VW. In this case,

18 Consumption, Globalization and Development

U is a position preferred to B if it lies above the curve equivalent to BE, is a position less preferred to B if it lies below that curve, and is equally preferred to B if it lies exactly on that curve. Since the position of U depends on the slope of the price line IQ, this implies that U is a position preferred to B if the price of X remains below a certain limit and a position less preferred to B if the price of X is higher than that. In other words, … the consumer enjoys his new taste if the price of X permits him to gratify his new taste to a sufficient extent, but dislikes it in the opposite case. (Harsanyi, 1954: 204, emphasis added) Following precisely the same reasoning, one arrives at a similar conclusion in regard to income. That is that a ‘change in a consumer’s indifference map would … benefit him if his income is high, and would harm him if his income is low, and would express a shift in his taste away from simpler and less expensive commodities towards more refined and more expensive ones’ (Harsanyi, 1954: 213, emphasis added).

Application to consumption and welfare in developing countries In using the above framework to assess the welfare effects of globalization and preference change in developing countries, one needs to be guided, we feel, by two main recognitions. One is that the vast amount of product innovations take place in the developed rather than the developing countries, and the second is that – as a consequence thereof – those innovations tend to embody characteristics that match the incomes of rich rather than poor consumers. As shown in Figure 2.3, this means that ‘over time in advanced countries, as incomes rise, there is a tendency for the product characteristics to shift up and to the left, as shown by the arrows, as later products are designed for higher-income consumers’ (Stewart, 1977: 80). To this extent, therefore, among the growing numbers of consumers living in the Third World whose values, we believe, are changing in response to globalization, only the relatively affluent are generally likely to enjoy a welfare gain by virtue of being able to purchase the types of (‘high-income’) products that satisfy their new values. The vast majority of consumers, by contrast, will tend to

Preference Change and Consumer Welfare 19

High-income characteristics

Low-income characteristics Figure 2.3

The direction of product innovations in the developed countries

suffer the loss in welfare that results from an inability to purchase those same products. The problem, it seems, may be particularly prevalent in developing countries, where the modernizing effects of globalization have been relatively widely diffused among the population, but where income is heavily concentrated among a very small minority of the population (as in a number of Latin American countries for example). For then a large number of low-income consumers with the modern values and preferences described earlier will be confronted with products whose characteristics make them accessible to those with high rather than low incomes. The following summary findings of a study conducted in Belo Horizonte, Brazil, clearly indicate that this is not a purely theoretical conjecture. For what the researcher found in that location was that: The lower the family income, the more family members tend to watch television and listen to AM radio. These audiences have relatively little consumer power. It appears that television viewing is likely to stimulate the desire to consume, but consumption does not materialize if one does not have the means to purchase the advertised goods. When preference for manufactured items was

20 Consumption, Globalization and Development

compared to television viewing, the correlation was positive and reached statistical significance. One would expect, then, that poor viewers would buy the advertised products if they had the means to do so. When actual behavior was measured, the data did not show the expected relationship between exposure to television and consumption, but when preference was tested, the association was positive. These results indicate a discrepancy between behavior and attitude. Between the desire to own a product and actual consumption, there is the strong intervening variable of income. Therefore, in this case, attitude is translated into consumer behavior only among those who have higher socioeconomic status. Unfortunately heavy viewers in peripheral contexts are among those who have the highest desire to consume and the least possibility of experiencing their wishes. (Oliveira, 1991: 208, emphasis added) The last sentence of this quotation notwithstanding, it is important to recognize that there are in fact a number of ways in which the preferences of low-income consumers in developing countries can be indulged to varying degrees. And these offsetting mechanisms need thus to be included in an assessment of the welfare effects of globalization and preference change in the Third World.

Countervailing welfare mechanisms a) Expenditure substitution effects Especially among low-income consumers living in urban areas of developing countries, preferences for modern, Western, products are known to be indulged on occasion at the expense of other household budgetary expenditures that are more promotive of human development indicators such as health and nutrition. Recently, for example, Ger and Belk have described how the diet and health of such consumers is being endangered by cutting on food to afford foreign cigarettes, or a pair of much-desired jeans, or by replacing local foods, even milk, with tempting, attractively packaged, and well-advertised candy, and other less nutritious or carcinogenic new and modern world foods. (Ger and Belk, 1996: 282)9

Preference Change and Consumer Welfare 21

Substitution effects of this kind, which allow the consumer to indulge her tastes for relatively expensive products, register, by definition, as a welfare gain according to the ‘quasi-cardinal’ approach that was described earlier. On the other hand, there are two important reasons which, in combination, might lead one to question this result. The first is that the consumer may be unaware of the consequences of thus indulging her new preferences (as might be the case, for example with respect to the health and nutritional issues raised in the quotation from Ger and Belk). The second reason is that one could argue that what should count from a welfare perspective is the degree to which a person’s informed, as opposed to her uninformed preferences are satisfied.10 ‘Recent utilitarian philosophers’, for example, ‘have stipulated that to qualify as a good, satisfaction must be informed about the satisfying event or object and the consequences of attaining it’ (Lane, 1991: 460–1). The problem at this stage, however, is that with the partial exception of the infant formula case, there is very little systematic research on which one can draw in order to ascertain the extent to which preferences for modern goods are informed rather than uninformed among low-income consumers in developing countries. b) Low-income product innovations Essential to our previous analysis and the conclusion we drew therefrom, was the notion that product innovations occur overwhelmingly in the developed countries for whose relatively high-income consumers they tend accordingly to be designed. Against this, however, has to be set the fact that adaptations of these products in favor of low-income consumers are known to occur in at least some developing countries. On occasion, for example, adaptations of this kind take the form of attempts by multinational corporations to reduce the problem of indivisibility that is often associated with developed country products. Thus, ‘in LDCs such products as cigarettes, razor blades and chewing gun are sold in small or individual units so that consumers with limited discretionary incomes can afford to buy them. Warner Lambert, which sells Chiclets chewing gun in 12-piece packs in the United States, decided to sell 2-piece packs for the equivalent of a few cents each in Latin America’ (Hill and Still, 1984: 95). Similarly, ‘Less affluent consumers in Mexico cannot afford to buy detergents in large packages and manufacturers

22 Consumption, Globalization and Development

find cardboard too expensive for small detergent packages. Smaller units are therefore put into plastic bags’ (Hill and Still, 1984: 98). In India, on the other hand, low-income product adaptations often seem to emanate from local firms seeking to undercut the products made by multinational corporations that are targeted at high rather than low-income consumers.11 Probably the best known case in this regard is ‘Nirma’, a locally manufactured laundry detergent that (as shown in Chapter 6) captured a dominant share of the Indian market through a series of cost-cutting measures that included the elimination of a number of ‘high-income’ product characteristics such as optical whiteness, scent and expensive packaging and the use of highly labour-intensive methods of production. More specifically, at a price less than half that of ‘Surf’, the market leader, at the end of the 1980s, ‘Nirma’ has been able to capture a market amounting to some 70 million households and in the process has achieved what for most Indian firms is still only a distant goal, namely, ‘of hitting upon a consumer product affordable and attractive enough to penetrate the mass market of the rural hinterland’ (Far Eastern Economic Review, June 1, 1989, p. 60). As with the previous case, however, the actual extent of the countervailing welfare influence thus exerted on behalf of low-income consumers in developing countries remains an empirical question that has yet to be settled. c) Illegal copies of developed-country products In various forms, illegal copies of developed-country products enable poor consumers in developing (and indeed also developed) countries to gain access to at least some characteristics of these goods at what are often vastly reduced prices. Counterfeit copies of well-known brands of watches, fashion goods, sunglasses, and handbags, for example, are intended to look as much like the original as possible, not least by using the same brand name, although these products are usually of much lower quality than the authentic items (Grossman and Shapiro, 1988). What counterfeits of this kind confer on the user, therefore, is a certain status rather than ‘any utility deriving from their function’ (Grossman and Shapiro, 1988: 82) (to the extent, of course, that observers are mistakenly impressed by her conspicuous consumption behaviour). As such, the status characteristics commonly associated with foreign brands

Preference Change and Consumer Welfare 23

are made available to poor consumers who would otherwise be unable to purchase also the quality characteristics (that is, to purchase the original product as a whole). Low-income consumers also gain (and again at the expense of consumers who possess the authentic product) from the closely related phenomenon of making product ‘look alikes’, which involves (as before) the close copying of another product, without however, going so far as to actually use the same brand name (thus, for example, products labelled Chanel No. 5 or Chanel No. 6 are clearly intended to pass off as the original item).12 The best documented cases of brand piracy in the Third World, however, occur in the areas of computer software, music and films. For example: An explosion in the video compact disk industry in China in the last two years has created an enormous demand for cheap videos, and pirated versions have been openly advertised and sold in cities all over China. … ‘Titanic’ was sold on Shanghai street corners for $2 even before the film’s theatrical release in the United States. (International Herald Tribune, June 25, 1998, p. 17) Similarly, it appears that as a result of foreign programming piracy in the Eastern Caribbean region, ‘VCR and cassette viewing are coming within reach of even the poorest people in these nations’ (Cuthbert and Hoover, 1991: 274), while in addition, ‘Many people, of all classes … are regularly viewing pirated satellite programming’ (Cuthbert and Hoover, 1991: 276). d) Giving the appearance of modernity In some of the examples that were cited under the previous heading, we suggested that illegal copies of foreign brands might enable the poor consumer to acquire the status afforded by those goods, independently of the other (functional) characteristics that they embody. In order to ‘unbundle’ product characteristics in this way, however, low-income consumers do not necessarily have to rely on illegal copies of foreign brands. For, in some instances consumption behaviour that can best be described as ‘giving the appearance of modernity’ can serve much the same purpose.13

24 Consumption, Globalization and Development

This type of behavior usually involves using only part of a particular modern good to convey the impression to others that the consumer owns the entire good (as in the case of counterfeits, therefore, the efficacy of this behavior relies on the deception of other consumers who are duly but mistakenly impressed by what they perceive). The use of a discarded ballpoint pen top, for example, when inserted into the consumer’s pocket, gives the appearance that she owns the entire good. Other examples which have been observed in the African context, include ‘the broken wireless or gramophone, the sunglasses without lenses and the shirt with no back’ (Southall, 1961: 21). e) Second-hand foreign goods Still another means by which low-income consumers in developing countries may be able to indulge their preferences for foreign goods, is through the purchase of second-hand versions of these products. To the best of our knowledge, however, this practice has been systematically investigated in only one particular context, namely, the market for clothing in Pakistan during the early 1990s (Castelijn, 1994). In that context, it appeared that low-priced second-hand clothing from the West formed a higher proportion of total clothing purchases for the household groups defined as living in poverty than for groups with higher incomes. Such clothing, that is to say, exerted an egalitarian influence on the inter-household distribution of welfare in Pakistan.

Conclusions Either because it tends to assume that tastes are fixed, or because it has tended to focus on minor (‘apples vs. pears’) changes in tastes, as Hirschman has suggested, conventional economics has largely ignored the fundamental changes in values that are being wrought ever more rapidly in the Third World by the forces of globalization (which include the phenomenal growth of television receivers and channels, the growth in television advertising by multinationals and the increased exposure of developing countries to product innovations from the developed countries). Certain other disciplines (such as sociology), on the other hand, have recognized the importance of these changes but they tend to

Preference Change and Consumer Welfare 25

evaluate them normatively from a perspective other than that of the consumer herself. Using an approach that relies instead on the consumer’s own perception of her welfare (and in particular her perception of the situation before and after she has undergone a fundamental change in values), and by going beyond the confining effects of ordinal utility analysis that underlies conventional welfare economics, we have argued that the crucial issue turns on whether the consumer is able to actually indulge her new preferences. Those that can, we suggested, would tend to experience a gain in welfare, while those that cannot would tend to experience instead a sense of frustration and disappointment. Especially since the products that are favoured in this process of globalization originate in the developed countries, for whose high-incomes they are mostly designed, we concluded initially that only a small, relatively affluent minority in the Third World would be able to benefit, while the vast majority of consumers would suffer a welfare loss. The reach of the globalizing forces in the Third World, that is to say, extends not just to the affluent minority but also to income groups that are far from being able to realize their new consumption aspirations, as one country case study (of Brazil) clearly showed. Subsequently, however, we recognized that there are a number of potentially important mechanisms through which even those consumers with relatively low incomes might also be able to indulge their preferences for modern, Western goods, or, more correctly in some cases, for certain characteristics of those goods rather than the products in their entirety. What are needed in future research on this topic are estimates of the prevalence of these countervailing welfare mechanisms in particular developing country circumstances.

Notes 1. A discussion of this point can be found in James and Stewart (1981). 2. This is implied by his notion of ‘producer sovereignty’. 3. In terms of conventional welfare economics which is based on ordinal utility, a change in preferences would imply that the consumer has intersecting indifference curves (and hence inconsistent preferences). 4. For a disaggregation of the growth in advertising expenditures in Asia, see the International Journal of Advertising, 17 (2), 1998.

26 Consumption, Globalization and Development

5. For a full discussion of this point see The Economist, November 29, 1997, p. 91. 6. Little (1957) remains one of the classic works in this area. 7. Harsanyi (1954). The following summary of this approach draws heavily on James (1993). 8. The point is not only that major changes in tastes are usually easier to compare than minor changes, but also that the former are more likely than the latter to give rise to large differences in welfare which are easier to compare than small differences. 9. Ger and Belk (1996), p. 282. In Brazil this phenomenon had already been documented by Wells (1977). 10. This point of view is argued, for example, by Harsanyi (1992). 11. Rao (1989), contains a detailed discussion. 12. For a discussion of this distinction, see, for example, Business Week, December 16, 1985. 13. This term is used by James (1987) in a somewhat different context.

References I. Castelijn (1994). Second-Hand Clothing to Developing Countries, Master’s thesis, Tilburg University. M. Cuthbert and S. Hoover (1991). ‘Video Beachheads: Communication, Culture and Policy in the Eastern Caribbean’, in G. Sussman and J. Lent (eds), Transnational Communications, Sage. A. Dixit and V. Norman (1978). ‘Advertising and Welfare’, Bell Journal of Economics, 9, Spring. J. K. Galbraith (1973). The Affluent Society, Pelican. G. Ger and R. Belk (1996). ‘I’d Like to Buy the World a Coke: Consumptionscapes of the “Less Affluent World”’, Journal of Consumer Policy, 19. G. Grossman and C. Shapiro (1988), ‘Foreign Counterfeiting of Status Goods’, Quarterly Journal of Economics, Feb. J. Harsanyi (1994). ‘Welfare Economies of Variable Tastes’, Review of Economic Studies, 21 (3). J. Harsanyi (1992). ‘Utilities, Preferences and Substantive Goods’, UNU, WIDER, Working Paper No. 101. J. Hill and R. Still (1984). ‘Adapting Products to LDC Tastes’, Harvard Business Review, March–April. A. Hirschman (1986). Rival Views of Market Society and Other Recent Essays, Viking. A. Inkeles and D. Smith (1974). Becoming Modern, Harvard University Press. J. James (1993). Consumption and Development, Macmillan. J. James and F. Stewart (1981). ‘New Products: A Discussion of the Welfare Effects of the Introduction of New Products in Developing Countries’, Oxford Economic Papers, 33, March. R. Lane (1991). The Market Experience, Cambridge University Press.

Preference Change and Consumer Welfare 27

I. Little (1957). A Critique of Welfare Economics, Oxford University Press. A. Mattelart (1991). Advertising International, Routledge. O. Oliveira (1991). ‘Mass-media, Culture and Communication in Brazil: the Heritage of Dependency’, in G. Sussman and J. Lent (eds), Transnational Communications, Sage. S. Rao (1989). ‘Innovative Marketing Strategies: Small Enterprises Fight Large Established Companies’, Economic & Political Weekly, August 26. L. Sklair (1994). ‘Capitalism and Development in Global Perspective’, in L. Sklair (ed.), Capitalism and Development, Routledge. A. Southall (1961). ‘Introductory Summary’, in A. Southall (ed.), Social Change in Modern Africa, Oxford University Press. F. Stewart (1977). Technology and Underdevelopment, Macmillan. UN (1995). World Investment Report. J. Wells (1977). ‘The Diffusion of Durables in Brazil and its Implications for Recent Controversies, Concerning Brazilian Development’, Cambridge Journal of Economics, 1 (3).

3 Globalization, Conspicuous Consumption and the International Demonstration Effect Reconsidered It has long been recognized that consumption behaviour in poor countries is influenced by the consumption standards prevailing in the richer, more industrialised societies. Indeed, as far back as 1957, Nurkse referred to the notion of an international demonstration effect, by which he meant that ‘the attraction of advanced consumption standards exerts itself fairly widely, though of course unevenly, among the poorer two-thirds of mankind’ (Nurkse, 1957: 65). Especially in the past ten years or so, however, the international demonstration effect has been greatly intensified by advances in information technology, which, via the mass-media, international trade, foreign investment and other factors that collectively define globalization, expose advanced consumption standards ever more widely across the Third World. The growing importance of this phenomenon means that it needs to be conceptualized more clearly and its welfare effects better understood, than has hitherto been the case. In particular, what is seldom recognized is that Nurkse adopted a highly specific view of the mechanism that gives rise to this effect and as a result, he was predisposed to favour just one possible description of its consequences. The purpose of this chapter accordingly is to argue, first, that Nurkse’s view – especially his attempt to disassociate the demonstration effect from Veblen’s notion of conspicuous consumption – is not at all well founded, because it is based on an excessively narrow vision of the process of taste and value formation in developing countries. On this basis, and second, it is contended that a reappraisal of the relationship between these concepts needs to be 28

Conspicuous Consumption 29

undertaken within a considerably broader socio-economic framework than that which Nurkse employed. Finally, the chapter is concerned to stress that the need for such a reappraisal is based on far more than mere theoretical curiosity; it derives instead from demonstrations by Hirsch (1976), Scitovsky (1976) and Frank (1985a and b) that a Veblenian approach to consumer behaviour has welfare implications that are not only profound but also sharply different from those associated with the conventional theory of consumer demand. So far, none of these authors has considered the specifically developmental implications of their analyses – indeed all of them have confined their attention to developed market economies – but in so far as a Veblen-type approach increasingly has relevance also to consumption in the Third World, and to the demonstration effect in particular, equally if not more important policy implications may arise. The first part of the chapter describes Nurkse’s formulation and shows how the welfare implications of his view for individual consumers differ from those which would result from the alternative view of how individual preference functions are related. For this purpose, the product characteristics approach to demand theory is used. Following a discussion of the respects in which Nurkse’s formulation of the issue is thought to be deficient, an alternative framework is proposed. Within the context of this alternative approach, the competing interpretations are then described and assessed.

Nurkse’s formulation The point of departure of Nurkse’s formulation is that the preference functions of individual consumers in rich and poor countries are interdependent rather than independent. What is envisaged in particular is that ‘When people come into contact with superior goods or superior patterns of consumption, with new articles or new ways of meeting old wants, they are apt to feel after a while a certain restlessness and dissatisfaction. Their knowledge is extended, their imagination stimulated; new desires are aroused’ (1957: 59). Nurkse is at pains to stress that this particular form of interdependence in consumption functions allows no room for any concept of social emulation through conspicuous consumption. As he himself put it:

30 Consumption, Globalization and Development

We can leave out Veblen’s point that the propensity to spend is partly based on the desire for conspicuous consumption. I do not think that on the international plane the effect of unequal living standards depends on the idea of ‘keeping up with the Joneses.’ All it depends on is demonstration leading to imitation. Knowledge of or contact with new consumption patterns opens one’s eyes to previously unrecognized possibilities. It widens the horizon of imagination and desires. It is not just a matter of social snobbishness New products constantly emerge from the course of technical progress, which modify existing ways of life, and frequently become necessities. (Nurkse, 1957: 61–2, emphasis added) Before proceeding to describe the limitations of Nurkse’s formulation and to show why it is not thought to be an adequate basis on which to judge the issue that is raised in the previous paragraph, it is necessary to demonstrate why the issue itself is important.

Why the issue matters Use of the product-characteristics approach (which views products in terms of the characteristics that they embody) reveals that there are two distinct possible consequences for the welfare of the consumer in a poor country who is exposed to the demonstration effect à la Nurkse. These two possibilities – corresponding to what may be called a Galbraithian view and a Becker/Stigler household production function approach – are illustrated in Figure 3.1(a). The figure assumes a world of two (indivisible) goods, two types of characteristics (‘modern’ and ‘traditional’) and prior to contact with the advanced countries, the individual in the poor country is consuming good A at point P. According to the one (Galbraithian) version of how Nurkse’s effect operates, the knowledge that is transmitted (via advertising for example) alters the tastes of the consumer (in favour of modern characteristics), so that his entire indifference map shifts (from IC1 to IC2) and he moves to Q by replacing good A with good B. In this case, as is recognized in formal welfare economics, without a fixed frame of reference, it is impossible to assess the effect on the welfare of the individual consumer. (See, however the discussion in Chapter 2.)

Conspicuous Consumption 31

The alternative view of the individual welfare effect of contact with the developed countries originates in the household production function approach espoused by Stigler and Becker (1977). These authors argue that the effect of advertising (and presumably also other forms of the transmission of knowledge) is to alter the information available to the consumer, rather than his tastes. And since the welfare effect of the information that is conveyed depends only on its accuracy and in so far as the demonstration effect on balance improves rather than detracts from the knowledge available to the consumer, his choice of good B in Figure 3.1(a) will enable him to move on to the higher indifference curve IC11. If the product characteristics approach is useful in revealing these two distinct variants of Nurkse’s formulation, it is also helpful in elucidating the approach that he chose to eschew. In particular, it enables one to reconcile Veblen’s early views on consumer behaviour with the similar ideas that have more recently been advanced by Hirsch. The latter argued that in so far as consumption is based on positional competition – that is, ‘competition that is fundamentally for a higher place within some explicit or implicit hierarchy and that thereby yields gains for some only by dint of losses for others’ (1976: 52) it is collectively

(a)

(b)

Modern characteristics

Positional characteristics

B

P

Q IC 2

B l1

B1

P1

A P

IC 1

IC11

Traditional characteristics

Figure 3.1

A

N l2

0 L L1 M Non-positional characteristics

(a) Nurkse’s view; (b) the Hirsch/Veblen view

32 Consumption, Globalization and Development

self-defeating. The fundamental problem is that ‘The choice facing the individual in a market or market-type transaction in the positional sector, in a context of material growth, always appears more attractive than it turns out to be after others have exercised their choice’ (1976: 52, emphasis in original). Hirsch referred to goods whose consumption is based on positional competition as ‘positional goods’ or, more specifically, as ‘those things whose value depends relatively strongly on how they compare with things owned by others’ (Frank, 1985a: 101). Conversely, non-positional goods ‘depend relatively less strongly on such comparisons – the nonpositional category includes, but is not limited to, goods that are not readily observed by outsiders’ (Frank, 1985a: 101). Like Hirsch, Veblen believed that much of consumption behaviour is driven by ‘emulation – the stimulus of an invidious comparison which prompts us to outdo those with whom we are in the habit of classing ourselves’ (1899: 103). Veblen also contended – and here a clear analogy with positionality begins to emerge – that the outcome of this emulative behaviour would be ‘discrimination in favour of visible consumption’ (1899: 112), a concentration of consumption ‘upon the lines which are most patent to the observers whose good opinion is sought’ (1899: 112, emphasis added). And although he did not formulate the matter in the same rigorous zerosum terms as Hirsch, Veblen nevertheless seemed to perceive that this conspicuous consumption was inherently economically wasteful, as the following quotation suggests. ‘Relative or competitive advantage of one individual in comparison with another does not satisfy the economic conscience’ (1899: 98). Unlike Hirsch, however, Veblen maintained that it is the characteristics of goods and not the goods themselves that are required for ‘positional’ or ‘conspicuous’ purposes and he suggested, furthermore, that these are rarely the only reasons why particular products are in demand. In general, that is to say, the demand for goods has both a positional and a non-positional component. Thus: It is obviously not necessary that a given object of expenditure should be exclusively wasteful in order to come in under the category of conspicuous waste. An article may be useful and wasteful both, and its utility to the consumer may be made up of use and waste in the most varying proportions. Consumable

Conspicuous Consumption 33

goods, and even productive goods, generally show the two elements in combination, as constituents of their utility. (Veblen, 1899: 103) By combining the ideas of Hirsch and Veblen it is possible to represent the welfare implications of this synthesis – the type of approach that Nurkse chose to reject – in terms of Figure 3.1(b). Unlike Figure 3.1(a), the two goods A and B are now defined by their embodiment of positional and non-positional, rather than modern and traditional, characteristics. The individual initially chooses good A but given his new tastes, which, because of the demonstration effect, reflect the increased orientation to the positional aspect of demand, he changes to good B. Following Hirsch, however, the positional characteristics turn out (ex post) to be entirely illusory. That is, ‘after others have exercised their choice’, all the consumer ends up with is 0L of intrinsic or non-positional characteristics, which is less than was obtained (0M) with the original preference ordering. It is worth noting that the extent of the absolute decline in non-positional characteristics (and hence the degree of the consumer’s ex post loss in welfare) depends on the nature of good B (i.e. on the slope of the ray representing this good). If, instead of B, the consumer had chosen good B,1 for example, it is clear that the diminution of non-positional characteristics would have been much less severe. We shall explore the welfare implications of this point below, when a detailed interpretation of the demonstration effect along Hirsch/Veblen lines is attempted. For the present purpose, it is sufficient to note that if this effect is represented as causing an enhancement of positional tastes, the welfare impact is quite different from that which is associated with Nurkse’s view. 1

The limitations of Nurkse’s formulation If the issue raised in this chapter seems therefore to bring up a potentially important question of welfare, it is also fairly plain that Nurkse’s formulation – and in particular his dismissal of a Veblenian interpretation – is inadequate. For it entirely ignores the manner in which individuals become responsive to, and predisposed to search for, the knowledge about advanced country

34 Consumption, Globalization and Development

products that is transmitted through the demonstration effect. And it also fails to address the question of how, in the process by which tastes for products are altered, other proclivities of individuals may also be affected. Yet, these factors are likely to constitute essential determinants of the type of interdependence between individual utility functions that is created by the process that Nurkse describes (i.e. whether the process is better described by Figure 3.1(a) or (b)). A more adequate assessment needs to incorporate these omitted factors in a much wider analysis of how tastes and values are formed in developing countries. And this in turn means that one has to enlarge the scope of the enquiry beyond the confines of economic variables (with which Nurkse was mostly concerned) to include the findings of sociologists, and particularly those sociologists who have studied the modernization process.

Modernization theory As it is formulated by modernization theorists, the sociological approach differs from that taken by economists not merely with respect to the specification of the independent or explanatory variables but also (as Table 3.1 indicates) with regard to the range of variables that are to be explained. Whereas in the former approach changes in taste are isolated from any other accompanying (psychological and socio-cultural) changes, sociologists tend to a view in which all these various dimensions of change occur simultaneously, and often inextricably, as part of what Table 3.1

Economic vs. sociological approaches to taste change Economics

Sociology (modernization theory)

Independent variables

Income (absolute and relative), advertising

Urbanization, factory employment, education, mass-media exposure, etc.

Dependent variables

Taste change in isolation

Value change, openness to new ideas, ambition, secularism, participation, etc. (the ‘modernity syndrome’)

Conspicuous Consumption 35

is referred to as a ‘modernity syndrome’. In particular, together with the acquisition of a taste for modern goods, ‘modern man’: identifies with the newer, larger entities of region and state, takes an interest in public affairs, national and international as well as local, joins organizations, keeps himself informed about major events in the news, and votes or otherwise takes some part in the political process. The modern man’s sense of efficacy is reflected in his belief that, either alone or in concert with others, he may take actions which can affect the course of his life and that of his community; in his active efforts to improve his own condition and that of his family; and in his rejection of passivity, resignation, and fatalism toward the course of life’s events. His independence of traditional sources of authority is manifested in public issues by his following the advice of public officials or trade union leaders rather than priests and village elders, and in personal matters by his choosing the job and the bride he prefers even if his parents prefer some other position or some other person. The modern man’s openness to new experience is reflected in his interest in technical innovation, his support of scientific exploration of hitherto sacred or taboo subjects, his readiness to meet strangers, and his willingness to allow women to take advantage of opportunities outside the confines of the household. … In other words, psychological modernity emerges as a quite complex, multifaceted, and multi-dimensional syndrome. (Inkeles and Smith, 1974: 290–1) What is striking about the vector of factors that make up the syndrome described in the quotation is not only that it seems to appear consistently in studies that are independently conducted, but also in a variety of diverse populations. And because some sociologists have devoted a considerable amount of time to the empirical investigation of the factors that contribute to the ‘modernity syndrome’, much can be learnt about the causes of value change (which, in the transition from traditional to modern values, comprise an integral part of this syndrome) from an analysis of their results. This knowledge in turn, ought to throw a good deal of light on the assessment of the competing interpretations that were described above. Such an assessment, as noted earlier, derives particular importance from the

36 Consumption, Globalization and Development

growth in importance of the demonstration effect that derives from globalization.

The determinants of modernity Using a summary measure of modernity that incorporates a very wide range of attitudinal variables, Inkeles and his collaborators sought to identify the factors that were most important to explaining the variation in this measure among a large sample of individuals in six developing countries. What they found was that approximately 50 per cent of the variance in the data could be accounted for by a set of eight variables of which education, experience in a modern large-scale factory and exposure to massmedia were the most important. But modernity also appeared to be influenced to an important degree by the country in which these separate influences were exerted. In particular, the researchers: found no way to escape the fact that the men from some countries consistently scored higher in individual modernity than those from other countries, even when the individuals compared were apparently alike in certain characteristics which had previously been shown to be the most powerful determinants of such scores. … A man from Argentina who had completed more than eight years of schooling and a Chilean with about seven years of schooling did only as well as an Israeli who had only been to school for three years. (Inkeles, 1983: 178) The authors concede that the reasons for this ‘national character’ effect on modernity are not at all clear but suggest that the most likely explanation has to do with a view of modernization in which ‘individuals living in a more modern setting should become more modern merely by sharing a generally modern ambience. And one important factor in making a modern ambience may be the average level of modernity of the individuals who live in the environment’ (Inkeles, 1983: 182). Because this average is also highly correlated with more objective indices of development such as GDP per head, ‘the richer countries will enjoy a double advantage.’ For not only will they be better able to supply more of the factors (such as education,

Conspicuous Consumption 37

factory jobs, etc.) that contribute to modernity but their very affluence in itself contributes to the process.

The social learning view of taste change Earlier we noted that one reason for the inadequacy of Nurkse’s formulation is its failure to provide insight into the question of why individuals become favourably disposed towards or inclined to search for the knowledge that is conveyed about developed-country products via the demonstration effect. The evidence presented in the previous section indicates the variables that are important to this transformation, but provides no view of what sociological processes underlie the explanatory variables. Inkeles and his collaborators assert that the causal factors underlying their results (as well as those of other modernization theorists) mostly derive from the principle of ‘social learning’. That is, in their view individuals become modern by internalizing the values which are ‘embedded in the organizational practices of the institutions in which they live and work’ (Inkeles, 1983: 19). Consider, for example, the way in which education and work in large-scale factories are thought to promote modern attitudes according to this approach (which derives its rationale principally from the Marxist view that it is the relationship to the mode of production that is fundamental in shaping one’s consciousness). Thus: School and factory produce the same result because they both expose individuals to certain common principles of organization, procedures for assigning power and prestige, modes for allocating rewards and punishment, and approaches to the management of time. Individual modernity then becomes a quality learned by the incorporation into the self-system of certain qualities characteristic of particular institutional environments. (Inkeles, 1983: 20) What is stressed by this approach, therefore, is the general, rather than the specific nature of the impact that is imparted by the modernizing institutions. In the case of education, for example, the process by which the school inculcates modern values has mostly to do with ‘the distinctive nature of the social organization as such’

38 Consumption, Globalization and Development

and little to do with ‘the curriculum as such’ (Inkeles, 1983: 54). As an example of how the inherent characteristics of this institution work to produce this effect, consider its relation to the notion of ‘a sense of efficacy’, which, as noted earlier, is one component of the ‘modernity syndrome’. Thus: In the acquisition of this feeling generalization plays a substantial role. Generalization occurs when an individual enjoys so satisfying an experience in one specific relationship or performance that he is led to believe that he can attain comparable success in other contexts … The child who learns to read his school books later finds himself able to read directions and instructions and to follow events in the newspaper … By extension and diffusion, or what we have called generalization, a heightened general sense of personal efficacy results. (Inkeles, 1983: 55, emphasis in original) Or, to take another example, it is argued that ‘principles directly embedded in the daily routine of the school teach the virtue of planning ahead and the importance of maintaining a regular schedule’ (Inkeles, 1983: 56). It follows from this social learning view that there is an essential similarity in the way that individuals become modern, regardless of the context in which the process occurs. Indeed, even the vast diversity that is brought about by different economic systems (capitalist vs. socialist), cultural and political variations and so on, are thought not to have a ‘great impact on the process of individual modernization. The critical factor in that process will be the extent to which individuals are exposed to the “schools for modernity” such as formal education, factories, modern offices’ (Inkeles, 1983: 318). In brief, then, according to the social learning view the ‘emergence of modernity is primarily a consequence of internal dynamics of each society rather than of external forces. With law-like regularity, the same psychological traits arise in different and relatively independent national settings in response to the same structural pressures. Modernity is the natural response of personality and culture to incipient processes of structural transformation wherever they may occur’ (Portes, 1973: 270). In so far as it is valid, this view lends support to

Conspicuous Consumption 39

Nurkse’s formulation of the demonstration effect for it shows how individuals can become predisposed to favour advanced country products in a way that need have no Veblenian positional component at all.2 However, as Portes (1973) has pointed out, even on the basis of the same explanatory variables an alternative interpretation of the modernization process is possible. In particular: as an alternative to the dramatic portrait of a new personality system arising spontaneously and independently in different social contexts – the universality of modernity is an obvious consequence of the universality of diffusion of modern, i.e. Western, values and behaviour patterns. Many people in many countries have had the opportunity to be socialized in, essentially the same set of orientations … Massive aspirations, media-consumption and other such modern traits do not arise naturally from internal processes of structural change, but artificially from the impact of Western cultural diffusion centered in urban areas. Factors which have been repeatedly found to promote modernity – education, industrial occupation and urban residence – may function less as carriers of self-evident psychological implications discovered anew by each developing population than as vehicles for closer and more intensive exposure to Western values and lifestyles. (Portes, 1973: 270–1, emphasis added) It is on the basis of this view that one can formulate an alternative interpretation of the implications of the demonstration effect – an interpretation that will rely upon, rather than eschew, the ideas of Veblen and Hirsch3 and one that will emphasize that these factors need to be considered in an increasingly global context.

An alternative interpretation of the demonstration effect: the transfer of positional values Following Maslow’s (1954) concept of a hierarchy of needs, the striving for status (or the acquisition of high standing relative to others) becomes a predominant motivation only after other, more basic needs (for hunger, thirst, safety and belongingness) are satisfied. In this sense, Hirsch’s positionality or Veblen’s ‘conspicuousness’ may be described as a ‘high-income’ taste that is indulged principally by

40 Consumption, Globalization and Development

those with incomes sufficient to have satisfied their more fundamental needs. Yet, according to the alternative interpretation of the demonstration effect that this section will present (that corresponding to Figure 3.1(b)), it is precisely this high-income taste that (among other things) is transferred by the demonstration effect to the poor countries of the Third World. This alternative formulation differs from that described in the previous section not in its view of the set of variables that are thought to be important to the modernization process, but rather in the manner in which they are said to contribute to this process. It is to the elucidation of this crucial distinction that we now turn.

The instruments and mechanisms of the transfer of positional values in a globalizing world Education and work in large foreign factories In the alternative view the role of education and work in large foreign firms is to transfer developed-country values, including positional values, to developing countries. Wilson’s social anthropological study of the English-speaking societies of the Caribbean provides a fascinating example of this view. Wilson found that for these communities, ‘The reference group – for high social status and for respectability – is the urban middle-class Afro-American and white community of the United States or the Panama Canal Zone. This is the model where proper English is spoken, where manners come naturally, where a house is immaculate and furnished to the best taste to the utmost convenience, where sophistication has its roots and modernity its zenith’ (Wilson, 1972: 14). Wilson shows how this role orientation produces a pattern of consumption that can only be described as highly positional. For instance, ‘Economic differentiation becomes a public and social matter through the assumption of a life style, marked in part by the acquisition, use, and display of goods. … Economic differentiation is clearly a large and vital part of social differentiation – the reclassification of people in terms of their status’ (Wilson, 1972: 94). Or again, and more specifically, ‘Respectability at its material level, at its level of signification, is a show of luxury. It is a grand, well-furnished home, well equipped with modern appliances, fine furniture, china and linens, good stylish clothes, an expensive education, manners and

Conspicuous Consumption 41

deportment. These signifiers are also the foci of ambition for the population as a whole’ (Wilson, 1972: 226).4 What drives the patterns described in the previous paragraph according to Wilson is primarily the system of education. ‘EuroAmerican ideas and ideals enter the society most pervasively and influentially through the educational system. Here they are instilled, often obliquely, as the standards of right and wrong. It is here that youthful hopes and ambitions are engendered and directed – but are all too often left unfulfilled’ (Lieber, 1981: 231). A similar role appears to be played by the church and: in the professional standards and moral expectations that EuroAmerican institutions, firms, hotels, and employers impose on their employees, beginning with standards of dress and address and going deeper by nurturing ambitions and setting the terms by which those ambitions might be realized, including, a ‘respectable’ outlook on life and business. To the extent that companies and employers do this, they invade the value system of the Caribbean, and the more powerful they are, the more successful their invasion. (Lieber, 1981: 223) More generally, Sauvant has suggested that: the ability of foreign affiliates to influence and shape a host country’s business culture – and through it the entire cultural and social system – is based on two factors: first, on their direct capacity to add to and shape the production apparatus of the host country and to introduce, promote, and disseminate new modes of operation and behaviour; and, second, on the impetus they give to the creation of a supporting (foreign-controlled) business service structure geared to meet the needs of foreign affiliates. The crucial common element in these factors is that the capital investments underlying international business are accompanied by sociocultural investments, that is, values and behavioural patterns associated with the parent enterprise (or agency) and its home country. (Sauvant, 1976: 68–9) Among the many multinational corporations that have increased their presence in developing countries over the past decade, the last

42 Consumption, Globalization and Development

sentence of this quotation applies perhaps most compellingly to the McDonald’s Corporation.5 For, as Featherstone (1995) has rightly emphasized, The [McDonald’s] burger is not only consumed physically as material substance, but is consumed culturally as an image and an icon of a particular way of life. Even though McDonald’s do not go in for elaborate imagistic advertising, the burger is clearly American and it stands for the American way of life. … For those on the periphery it offers the possibility of the psychological benefits of identifying with the powerful. (Featherstone, 1995: 8) It can similarly be argued that, Coca Cola is the symbol of that glittering consumer paradise widely imagined to exist in the US. Accordingly, in many countries like post-Communist Romania, Coca Cola makes it a practice to advertise in English, despite the small number of English speakers in the country. For the point is not to convey information, but rather to sell Coke as a high status symbol of modernity and Western consumer culture. (Ger and Belk, 1996: 272) Advertising As already noted in previous chapters, there has been a phenomenal growth of advertising in the developing world over the past ten years. Indeed, from the standpoint of consumption and welfare, this aspect of globalization is surely one of the most important. In the present context, the growth of advertising derives its importance from its probable contribution to an enhanced preference for the positional component of demand. There are two reasons for this. First, as Hirsch points out, it tends to ‘encourage the strengthening of self-regarding individual objectives and makes socially oriented objectives more difficult to apply. The reason is that interests of self-concern and self-regard can be enlisted much more effectively in support of commercial sales efforts’ (Hirsch, 1976: 82). Second, the appeal to self-interest is often made with direct reference to positionality – that is, through linking consumption of the advertised product to ‘getting ahead’ of the rest of society. Together, these features of advertising

Conspicuous Consumption 43

foster not merely the existence of an individualistic ethic (which would remain true, incidentally, even if only information is conveyed), but also the notion that self-interest is best pursued with respect to advancing one’s rank in society. In developing countries, much of the advertising and sales promotion over the past ten years has been influenced by the concentration of the advertising industry into an ever smaller number of hands. Even by 1991, for example, it was apparent to some observers that there were only some 13 or 14 major agencies that were able to service the requirements of the major multinational buyers of advertising services. These more global agencies, according to Leslie (1995) have begun to ‘develop and articulate notions of global advertising for multiple reasons: to rationalize their growing size and transnational structures, to gather business, to cut costs, and to increase profits’ (Leslie, 1995: 423). The resulting use of global advertising campaigns is particularly apparent in industries such as soft drinks, jeans, pharmaceuticals and alcohol, and it implies that the nature of the promotional activities thus undertaken on behalf of multinational corporations remains essentially unaltered from its original, host-country form. And this, in turn, means that these techniques are likely to have the same effect on the positional component of values that was described in the previous paragraph. For example, in the context of the promotion of tonic foods in the Third World, Jelliffe has noted that ‘The content of advertising and the techniques used are those widely employed in the Western World, with emphasis on status and convenience. The tropical mother is in many cases even more vulnerable to these motivating forces. … She is as anxious to emulate the well-to-do elite, both local and foreign, as is any Western mother’ (Jelliffe, 1972: 201). A very similar process is described by Ledogar in the case of the marketing of soft drinks (Ledogar, 1975).6 The globalization of mass-media The growth in advertising expenditures noted above has been made possible in large part by the no less impressive speed with which the mass-media have spread among developing countries over the past ten years. (See Chapter 1.) Aside thus from its indirect impact on the positional component of demand, however,

44 Consumption, Globalization and Development

the mass-media also have a direct effect on this phenomenon which needs to be considered. What is perhaps most telling in this regard is the fact that ‘America’s role in global media … is unique. Other countries are big producers of entertainment. … But the giants of American media, such as Time Warner, Walt Disney and Viacom, dominate entertainment export markets and lead joint ventures which are creating new television businesses around the world’ (The Economist, Nov. 29, 1997, p. 91). This fact is important in the present context because the individualism portrayed so starkly in American films and soap operas is precisely what underlies the drive towards positionality that Hirsch described so clearly. 7 Furthermore, the status-intensive goods shown as being part of this process in these media exports also tend to become the instruments of positional competition thus induced in the developing countries. Direct historical contact: the impact of colonialism A study of occupational prestige and social status in (what was then) Northern Rhodesia by Mitchell and Epstein provides some interesting observations on the impact of colonialism on urban consumption patterns and in particular on those that might be described as positional. The authors point, for example, to the acceptance in urban communities of the prestige scale introduced by the Europeans. ‘In other words, the social grading of occupations reflects the more generalized prestige system which manifests itself as the emulation of the way of life of the socially dominant Europeans. … Success in achieving this “civilized” way of life is demonstrated conspicuously by the physical appurtenances of living. The most important of these is clothes, but personal jewellery (especially wrist-watches), furniture, and European-type foodstuffs are also important’ (Mitchell and Epstein, 1959: 32, emphasis added). It is important to note that the propensity for this kind of positional consumption in the urban (as opposed to rural) areas of developing countries may derive not merely from more numerous contacts with the developed world (the factor that in Nurkse’s formulation endows the city with its modernizing influence), but also from pressures that are inherent in the process of urbanization itself. In this regard Veblen himself believed that ‘the serviceability

Conspicuous Consumption 45

of consumption as a means of repute, as well as the insistence on it as an element of decency, is at its best in those portions of the community where the human contact of the individual is widest and the mobility of the population is greatest’ (Veblen, 1899: 87). 8 For this reason he asserted that ‘Conspicuous consumption claims a relatively larger portion of the income of the urban than of the rural population and the claim is also more imperative’ (Veblen, 1899: 87).9 It is also worth noting his insistence that ‘It is not that the city population is by nature much more eager for the peculiar complacency that comes of a conspicuous consumption, nor has the rural population less regard for pecuniary decency. But the provocation to this line of evidence, as well as its transient effectiveness, are more decided in the city. This method is therefore more readily resorted to, and in the struggle to outdo one another the city population push their normal standard of conspicuous consumption to a higher point’ (Veblen, 1899: 87–8, emphasis added). Whether or not globalization fosters urbanization in developing countries is not a question that can readily be answered (though one can imagine several possible mechanisms through which this might occur). What is clear, however, is that the urban population in many such countries is expected to grow very rapidly in the coming decades. According to one estimate, for example, the annual growth rate of the urban population in the least developed countries between 1995 and 2015 will be of the order of 5 per cent (Human Development Report, 1998).

The welfare consequences of positional taste transfer10 So far in this section fragments of evidence have been adduced in support of the interpretation that the demonstration effect occasions a change towards positional values and behaviour on the part of (some of) those who live in the Third World. But as Figure 3.1(b) was concerned to demonstrate, the welfare effects of this change depend on the type of good that is used to give effect to the new values. And in order to get some sort of analytical handle on this issue it is clear that one needs to convert the concepts of positional and non-positional characteristics (representing the two axes of Figure 3.1(b)) into more concrete categories and to investigate how these categories are combined into the different

46 Consumption, Globalization and Development

goods that are available for the satisfaction of enhanced positional values.

Mapping positional demand into its constituent categories of product characteristics We have already taken note that visibility is a part of the definition of positional consumption and it is also a characteristic that forms an essential part of the promotional activities (packaging, advertising and brand differentiation) of multinational firms in developing countries. But though it is necessary, visibility is not a sufficient characterization of positional demands. It requires, as an essential complement, the quality of ‘superfluousness’. In Veblen’s words, ‘Throughout the entire evolution of conspicuous expenditure … runs the obvious implication that in order to effectually mend the consumer’s good fame it must be an expenditure of superfluities. In order to be reputable it must be wasteful. No merit would accrue from the consumption of the bare necessities of life’ (Veblen, 1899: 96–7). Or again, ‘The consumption of expensive goods is meritorious, and the goods which contain an appreciable element of cost in excess of what goes to give them serviceability for their ostensible mechanical purpose are honorific’ (Veblen, 1899: 154–5). Taken together, these two components of positionality imply the direction of change in the demand for characteristics that is broadly indicated by the north-westerly movement from the second to the fourth quadrant of Figure 3.2. The question that now needs to be addressed in assessing the welfare implications of the transfer of positional values is the nature of the means that are available to consumers to obtain the bundle of characteristics represented in the fourth quadrant of the figure. More specifically, one is required to ask whether and to what extent this bundle can be obtained without a sacrifice in other, nonpositional characteristics (for it is this that determines whether the consumer reaches L or L1 in the example of Figure 3.1(b)). Two general considerations would appear to be of special significance in answering this question. The first has to do with the historically distinctive condition of contemporary underdevelopment and in particular the near total dependence of developing countries on goods developed in and for advanced countries. New

Conspicuous Consumption 47

‘Visible’ Positional demand

‘Superfluous’

‘Necessary’ Non-positional demand

‘Invisible’ Figure 3.2

Positionality and the direction of change in demand

goods in the latter, whose members on average enjoy a rising standard of living, come to acquire an increasingly high proportion of ‘high-income’ characteristics over time (including those such as sophisticated packaging, advertising and brand differentiation that may be described as highly visible) and a lower balance of ‘lowincome’ characteristics. In so far as these developments continuously cause existing products to become outmoded it follows that consumers in poor countries are able to buy the bundle of characteristics that we described as positional, at an increasingly high cost in terms of non-positional characteristics. The second general consideration has to do with the closeness of the link that often exists between the way in which positional values are created and the manner in which they need to be met. The global promotional activities of multinationals, for example, may simultaneously alter values in favour of positionality and require the new values to be met by the specific products that are the subject of promotion. To the seemingly growing extent that these products are transferred unaltered from the form in which they are

48 Consumption, Globalization and Development

sold in developed counties (e.g. Coca-Cola, infant formula), the result is again that in order to acquire the package of positional characteristics, a high price has to be paid in terms of foregone nonpositional characteristics. These two general arguments amount to positing a substantial discontinuity in the process by which the consumer is able to give effect to the change in demand depicted by (the direction of the arrows in) Figure 3.2. This discontinuity, and the welfare loss with which it is associated, bears contrast with the more continuous process that seemed to occur during the historical experience of the now developed countries. Towards the end of the nineteenth century in the United Kingdom, for example, when powerful pressures to emulation were given by increasing urbanization, the growth of advertising and mass-media and the awareness of social class, ‘There could have been no large shift towards the range of “new” commodities associated with the “high mass consumption” of the twentieth century’ (Supple, 1981: 137). Emulative desires appear instead to have been indulged at that time through a relatively continuous or smooth process of enhancement to traditional patterns of consumption: ‘new instead of

B

Positional characteristics

R I1

A

S

Non-positional characteristics Figure 3.3

The appearance of modernity

Conspicuous Consumption 49

second-hand clothing, leather footwear, gas instead of candles, coal instead of wood, commercial instead of home-made soap, improved types of linoleum floor-covering, arm chairs instead of kitchen chairs’ (Supple, 1981: 137).

Giving the appearance of modernity So far, our argument has been concerned with the type of embodiment of the characteristics combination that was earlier defined as positional. We have tried to show that the range of goods available is rooted in the historical conditions of ‘latecomer development’ and that the welfare implications of positional consumption in the Third World have to be understood in these terms. Implicit in our analysis, however, is the assumption that the positional package of characteristics can only be obtained by consumption of a more ‘positional-intensive’ good. To some (unknown) degree, this assumption will not hold because consumers are able in some circumstances to acquire the positional characteristics of goods without actually consuming the goods themselves. That is, they merely give the appearance of so doing. Consider, for example, the situation described by Figure 3.3. Goods A and B represent two alternative writing instruments: respectively a pencil and a ballpoint pen. If the consumer were to switch to the latter because of its higher balance of positional elements, the welfare consequences would be those described earlier in relation to Figure 3.1(b). Let us assume, instead, however, that the consumer is able to acquire a discarded ballpoint pen top, which, when inserted into his pocket, effectively gives the appearance of consumption of this good. If we assume, further, that this ploy enables him to obtain an amount equal to RS positional characteristics, he can reach the indifference curve I1 while continuing to consume good A. In this case, therefore, the change towards positionality leads to no loss in welfare. (That is, for the consumer himself. In so far as they purchase complete ballpoint pens for positional reasons, other consumers will tend to be adversely affected.) That something like the process described in the previous paragraph actually occurs in developing countries is very nicely illustrated by the following description of the variation in the choice of building materials in a region near Buenos Aires:

50 Consumption, Globalization and Development

The principal variation was constituted by the nature of building material, the poorest huts being made of mud mixed with straw and reeds, the ‘richest’ being bricks and mortar. In between these two extremes … there was a series of combinations in which efforts were made to give the impression of brick construction, either by white-washing the mud walls in a way that made them appear to be rendered, by laying the bricks in mud rather than mortar and subsequently white-washing them, or by constructing the front wall which faced the roadway of bricks, and making the others, which were out of public view of mud. (MacEwan, 1974: 212, emphasis added) Other examples from Africa, include ‘the broken wireless or gramophone, the sunglasses without lenses and the shirt with no back’ (Southall, 1961: 21). More recently, Ger and Belk (1996) report how status is acquired in the Congo by displaying imported cans on the dashboard of local cars.

Low-income positional consumption and the squeeze on essential characteristics In so far as the device of giving the appearance of modernity is limited (and consumers have therefore to purchase good B in Figure 3.3 in order to satisfy their positional demands) and to the degree that the lowest deciles of the population in developing countries indulge in positional consumption, the problem of a ‘squeeze’ on low-income or essential, characteristics is raised. In a different context, and in somewhat different form, this problem was of concern also to Veblen because of his belief that ‘No class of society, not even the most abjectly poor, foregoes all customary conspicuous consumption. The last items of this category of consumption are not given up except under stress of the direst necessity’ (Veblen, 1899: 85). As a result, and especially in certain lines of consumption, he perceived that the outcome would be pressure on the essential elements of consumer demand. In particular, ‘It is true of dress in even a higher degree than of other other items of consumption, that people will undergo a very considerable degree of privation in the comforts or the necessaries of life in order to afford what is considered a decent amount of wasteful consumption; so that it is by no means an uncommon occurrence, in an inclement climate, for

Conspicuous Consumption 51

B R

Positional characteristics (i.e. advertising, packaging and other visible high-income characteristics)

Q

A P

0 L1 L Non-positional characteristics (e.g. food nutrients) Figure 3.4

The squeeze on essential characteristics

people to go ill clad in order to appear well dressed’ (Veblen, 1899: 168, emphasis added). For the poor consumer in a contemporary developing country, the problem that gives rise to the squeeze on essential characteristics can be described as in Figure 3.4. Following the alteration in his tastes between positional and nonpositional characteristics, the low-income consumer faced only with modern good B imported from the advanced countries suffers a decline in essential characteristics equal to L1L as a result of the consumption of this good. Only with an income level that would enable him to reach point R along the ray representing the modern good, would he be able to maintain the original level of essential characteristics 0L. The extent to which the process described in Figure 3.4 actually occurs is impossible to assess on the basis of available data. The most that can be pointed to is fragmentary evidence that the process seems to occur primarily in urban areas of the Third World. In the case of infant formula, it seems to be due in part to promotional activities that extend even to the poorest groups and in part

52 Consumption, Globalization and Development

to emulation by these groups of the positional values and behaviour of the relatively affluent.11 In the case of food advertising, a process very similar to that described in Figure 3.4 has been observed by nutrition experts. The following quotation is perhaps the clearest statement of this view of the consequences of advertising of imported food products: even with their limited means, the poor buy highly preferred or ‘fashionable’ foods at a premium in price. The urban poor may buy them for their status value, although they may be uneconomical from the nutritional standpoint. Foods that have low prestige in rural areas, such as fruits and vegetables which are gathered and not cultivated, may not be eaten by newcomers to urban areas because they cost money and are not thought of as ‘valuable’. On the other hand, foods may be eaten for status reasons. An example of such behaviour is the purchase of expensive commercial formula products, and other ‘prestige’ foods such as bread, soft drinks, sugar, tea, infant formulas, and canned milk. The acquisition of these foods by the poor may have a nutritionally detrimental effect due to the economic drain they place on the meager food budgets. (Austin et al., 1976: 88)12

Conclusions That spending patterns in poor countries are susceptible to influence from the consumption behaviour of the richer societies has always seemed self-evident. But just how this influence is transmitted and with what effect on the welfare of the individual consumer in the Third World are questions that admit of much less certainty. In a well-known contribution in the late 1950s, Ragnar Nurkse (1957) suggested that the process works through an ‘international demonstration effect’. By this he meant essentially the exposure of the poor to modern products – ‘demonstration leading to imitation’ – and he explicitly eschewed the notion that international emulation of the rich had anything to do with status seeking, or, to use Veblen’s phrase, with conspicuous consumption. This chapter has argued, however, that there are a number of reasons why Nurkse’s contribution needs to be reassessed. In the

Conspicuous Consumption 53

first place, the issue itself has become infinitely more important as a result of advances in information technology, which, over the past ten years or so, have greatly enhanced the extent to which developed countries are able to influence consumption behaviour in the less developed areas. Expressed more succinctly, the essential point is that globalization has vested the notion of an international demonstration effect with more relevance than in any previous historical period. Second, in the light of compelling analyses of the advanced countries (by Hirsch, Frank and Scitovsky) showing the powerful welfare implications of consumption based on status seeking, the issue of whether and to what extent this form of consumption behaviour also applies internationally takes on a much greater significance than might earlier have seemed to be the case. Third, it is contended that Nurkse’s approach is inadequate to appraise this important question and that a much broader framework – encompassing sociological as well as economic variables – is required. On the basis of this expanded framework, it was shown that underlying the different views of the way in which preference functions are related at the international level is a set of quite distinct hypotheses concerning the influence of a range of socio-economic variables on the formation of values and preferences in developing countries. Although a fairly substantial body of sociological and other evidence was brought to bear on the assessment of these competing causal mechanisms, and though there seemed sufficient material to cast doubt on Nurkse’s eschewal of status-seeking consumption behaviour at the international level, a clear verdict was not able to be reached. In part, this lack of conclusiveness is due to the inherent difficulty of conducting (and for this reason also the lack of) research that is able to elicit the ‘true motivations’ of consumers. It is now known from social psychology, for example, that direct and indirect methods of uncovering human motivation sometimes produce different results.13 But the lack of conclusiveness is probably also a reflection of the fact that the competing views should be regarded less as universal truths than as descriptions of consumption behaviour that are applicable to different socio-economic circumstances. Much may depend, for example, on the extent to which group values are retained to exert a

54 Consumption, Globalization and Development

countervailing influence on positional consumption behaviour during the modernization process. The importance of the group in this respect has been emphasized by Douglas and Isherwood: A strong group has its own characteristic ways of controlling the envy that might spoil the relations of its members and so threaten its permanence. The group imposes group values and so prevents deviant individual spending, defines what counts as too much conspicuous individual consumption, and proposes punishments. (Douglas and Isherwood, 1979: 36–7).14 The Japanese experience represents probably the clearest example of the workings of a social (the ‘Japanese ethos’) rather than a positional ethic during the modernization process.15 As Sen describes it, ‘If the invisible hand does a great deal of visible good in Japan, the hand does not seem to work through the relentless pursuit of self-interest’ (Sen, 1984: 104). Specifically, what the Japanese model suggests is that selfinterest based objectives can be more effectively achieved by an alteration in group attitudes (to, for example, team work). Indeed: when one considers how production takes place in a modern industrial establishment, it is quite incredible to think that being actively self-interested can be such a virtue. Success in production depends greatly on team work, and while that interdependent picture provides incentive for a group, it is not an incentive that can be effectively translated into rewards and punishments related to individual work and performance. (Sen, 1984: 104)

Notes 1. A very interesting example of the process described by Hirsch is the decline in prestige that is said to have been experienced by wearers of a Rolex watch. Whereas in the 1980s this watch was a signal that one ‘had arrived’, subsequently ‘something has happened to take the shine off the Rolex: too many other people wear them.’ ‘Wrist watching’, The European, Elan, 10–12 May 1991, p. 22. See also the case of Nike, described in Chapter 5. 2. It is essential to recognize that Nurkse’s view does not necessarily mean that the individual consumer is bereft of any sort of social influence on his consumption behaviour. His relationship to others may instead be

Conspicuous Consumption 55

one in which he communicates his modernity to them by consuming goods that are imported from developed countries. This wider view of ‘consumption as communication’ is very much in line with what Douglas and Isherwood propose. For them, in fact, Veblen’s analysis of conspicuous consumption ‘has much to answer for when we consider how widely his analysis of the leisure class is received’ (Douglas and Isherwood, 1979: 4). They advocate an alternative approach to consumption in which ‘Goods assembled together in ownership make physical, visible statements about the hierarchy of values to which their chooser subscribes … goods in their assemblage present a set of meanings, more or less coherent, more or less intentional. They are read by those who know the code and scan them for information. The great novelists have never doubted just how far removed this function of creating meanings is from the uses of goods for welfare and display’ (Douglas and Isherwood, 1979: 5). Csikszentmihalyi and Rochberg-Halton (1981) make a similar point by distinguishing between two meanings that goods convey. On the one hand, goods serve as a means of individual differentiation: that is, a means of separating the owner from the social context by emphasizing his individuality. This role of goods, they contend, has been greatly overemphasized. ‘Thus although status symbols are an extremely important aspect of the whole person-object interaction process, it is, nevertheless unfortunate that this one dimension has so overshadowed the rest that it is almost impossible to think of people’s possessions except as symbols of their social standing’ (Csikszentmihalyi and Rochberg-Halton, 1981: 31). The second (and less emphasized)meaning conveyed by goods, in contrast, involves social integration: ‘They might represent dimensions of similarity between the owner and others: shared descent, religion, ethnic origin, or life-style’ (Csikszentmihalyi and Rochberg-Halton, 1981: 40–1). In this case, the good expresses the belonging of the owner to the social context rather than his differentiation from it. Other important references on the communicative aspects of consumption include Leach (1976), Appadurai (1986), Solomon (1983), Mukerji (1983), Sahlins (1976), Goffman (1956), Holbrook and Hirschman (1980), McCracken (1988), and from an economics perspective, Spence (1974). 3. Yet another possible view – associated with Berger, Berger and Kellner (1974) – attempts to combine the general and specific modes of preference formation in developing countries. These authors argue that there are certain ‘intrinsic’ relations between modern technological production and modern consciousness such as the sense of ‘componentiality’ that is imparted to the factory worker. ‘This apprehension of reality in terms of components is essential to the reproducibility of the production process as well as to the correlation of men and machines’ (Berger, Berger and Kellner, 1974: 32). In addition, Berger, Berger and Kellner identify a set of ‘extrinsic’ processes which are linked to technological production more loosely through a variety of historical processes and

56 Consumption, Globalization and Development

4.

5.

6.

7. 8.

which, accordingly, will vary from one case to another. (In capitalist, but not socialist, countries, for example, there are extrinsic elements of technological production which derive specifically from ‘the motives of individual production which derive specifically from ‘the motives of individual competition and profit-making geared to the accumulation of private property’.) In general, the authors conclude that what ‘manifest itself empirically as modern consciousness is a highly complex combination of these “necessary” and “accidental” elements’ (p. 90). Another interesting anthropological study of the Caribbean region, by Lieber (1981), suggests that positional behaviour (in urban Trinidad) is apparent with respect to other dimensions of status than affluence alone. Because it ‘eludes precise definition’, ethnicity, for example, ‘involves the active use of cultural signifiers, or clusters of symbols which, serve as vehicles for marking a person’s (or group’s) social style and position.’ These serve in fact as ‘resources that the ethnically ambiguous actor may attempt to manipulate to his own advantage’ (Lieber, 1981: 104). Similarly, ‘Since jobs are related to class status, job seekers attempt to secure those sorts of positions which on the surface look good to other people. Hence a job in a modern, air-conditioned office which allows an employee to wear a white shirt is often preferred to what may be a far better and more highly paid job which has a less prestigious ambience. The signs of a position tend to be very conspicuously displayed’ (Lieber, 1981: 98–9). According to the Human Development Report of 1998, ‘McDonald’s restaurants expanded worldwide sales by $19 billion in 1986–96, 64% outside the United States’ (p. 62). Mattelart (1983: 38) quotes a suggestion from Business International regarding the sales promotion of Western products in the Third World that contains clear positional implications. The suggestion is to ‘try and give your products a Westernized appearance to give them social standing in regions undergoing rapid development wherein ideas of modernization and Westernization are linked.’ For a discussion of the cultural features of American soap operas see Barker (1997). Veblen (1899: 87). These are not, however, the only reasons why consumption in urban areas may be more prone to positionality. One needs to consider in addition the signifance of the fact that in traditional (rural) societies rank may be almost entirely excluded, or it may be ‘ascribed’ (Smith, 1966). In the latter case, rankings are ascribed to persons at birth and cannot subsequently be altered. The best-known case in point is the Indian caste system under which, ‘virtually every aspect of the individual’s future life was determined by his birth into a particular caste’ (Smelser and Lipset, 1966: 8). In another traditional form of ranking system, rank does vary but according to certain exogenous events. For example, in one of the South Indian villages studied by Scarlett Epstein (1962), ‘A man’s economic status is determined by the size of the ancestral estate and the number of heirs that have to share it; personal initiative can hardly help

Conspicuous Consumption 57

9.

10.

11.

12.

to raise his economic status. Although the relative economic status of successive generations of households may be quite different, this has nothing to do with the personal ability and efforts of the particular householder, but rather is a result of the accidents of births and deaths’ (Epstein, 1962: 324). Other examples are provided in Smith (1966). It might be possible to test Veblen’s hypothesis by comparing the consumption behaviour of urban and rural residents at given income levels. Following Frank’s (1985b) suggestion one could more specifically, examine whether spending by the former tends to have a higher ‘observable’ and a lower ‘non-observable’ component (since ‘conspicuousness’ is a necessary requirement of positional consumption). What fragmentary pieces of evidence are available seem to be consistent with this expectation. Consider first the ‘unobservable’, two of the most important of which are savings and (within limits) nutritional intake. With respect to the former, Gillis et al. (1983) find that rural households generally tend to save more than urban households ‘at comparable levels of income’ (Gillis et al., 1983: 282, emphasis added). With regard to nutrition, ‘It seems that for a given income group, the urban population has a significantly lower nutritional status as measured by caloric intake’ (Austin et al., 1976: 340, emphasis added). Corresponding to, and indeed partly accounting for these findings, is the tendency for urban consumers to spend more than their rural counterparts with comparable incomes on relatively expensive and observable symbols of status. Such a tendency has been noted, for example, in the case of infant formula (Landgren-Gudina, 1984), crystal sugar (James, 1977), sophisticated soap (Mubin and Forsyth, 1984) and imported detergents (James, 1983). But because they generally control only for income, these studies still leave open the question of whether other variables (such as product access and information) may account at least partly for what appears to be more positional consumption behaviour in urban areas. We should emphasize that our discussion here follows the Veblen/Hirsch view of status-seeking behaviour; other models of this behaviour, as Congleton (1989) has demonstrated, have different (and sometimes positive) welfare effects. (Favourable externalities, for example, may accrue to those who are not actively involved in statusseeking games.) Moreover, even the Veblen/Hirsch model may be associated with some offsetting welfare effects (such as, for example, on the supply of labour) which we do not address. In urban Upper Volta, for example, it has been observed that ‘many women cultivators are impressed with the fact that high-status women, whose babies are considered by all the town’s women to be the best cared for, use the bottle. (A corollary of this … is that low-status women use preserved milk because by so doing they feel and become un peu civilisée.)’ See Skinner (1974: 189). In Brazil a decline over time in nutritional status among the urban poor has been associated with the substitution for food expenditure of durable goods consumption (Wells, 1977: 5).

58 Consumption, Globalization and Development

13. McClelland (1976). It is interesting in this regard to note that the predominant reason often given by individual urban women for weaning in formal interviews (viz. milk insufficiency) differs from the motivations that are elicited during more informal discussions (Austin et al., 1976). 14. For details see Douglas and Isherwood (1979). The cooperative societies described in Mead (1961) also provide concrete illustrations of the various modes by which the group environment tends to constrain positional competition (as reflected in a weak emphasis on rising in status in these societies). And even in highly individualistic advanced economies such as the United States, the group, in the form of the trade union, is said by Frank (1985) to implement collective agreements to protect ‘nonobservables’ (such as savings and insurance). Policy too could seek to increase group influence on decisions, by, for example, reducing private control over consumption and increasing public (as for instance by state-provided goods and services). In addition, one might, as Layard (1980) has suggested, attempt to discourage positional as opposed to other forms of gaining status, by, for example, ‘getting adults to spend less time comparing the performances of young children’ (p. 749) and by granting honours, titles and other kinds of acknowledgments of diverse forms of achievement. Still other policies could be directed towards the status symbols themselves: according to our analysis, for example, there is a rationale for tariff protection against the import of products which are intensive in positional characteristics and low in their embodiment of essential characteristics. In some extreme cases (such as that of infant formula) these inappropriate symbols of status might be banned altogether, thus removing them from the relevant areas of choice. 15. See, for example, Lockwood (1954) who notes that ‘By comparison with the value standards prevailing among many other peoples, perhaps most of them, social position and prestige in Japan are less dependent on display of material wealth’ (p. 284).

References A. Appadurai (1986). The Social Life of Things, Cambridge University Press. J. E. Austin et al. (1976). Urban Malnutrition, a report submitted to the World Bank by Harvard University. C. Barker (1997). Global Television, Blackwell. P. Berger, B. Berger and H. Kellner (1974). The Homeless Mind, Penguin. P. Bohannan and G. Dalton (eds) (1962). Markets in Africa, Northwestern University Press. R. Congleton (1989). ‘Efficient Status Seeking: Externalities and the Evolution of Status Games’, Journal of Economic Behavior and Organization, 11. M. Csikszentmihalyi and E. Rochberg-Halton (1981). The Meaning of Things, Cambridge University Press. M. Douglas and B. Isherwood (1979). The World of Goods, Allen Lane.

Conspicuous Consumption 59

T. S. Epstein (1962). Economic Development and Social Change in South India, Manchester University Press. M. Featherstone (1995). Undoing Culture. Sage. R. Frank (1985a). ‘The Demand for Unobservable and Other Nonpositional Goods’, American Economic Review, March. R. Frank (1985b). Choosing the Right Pond: Human Behavior and the Quest for Status, Oxford University Press. G. Ger and R. Belk (1996). ‘I’d Like to Buy the World a Coke: Consumptionscapes of the “Less Affluent World”’, Journal of Consumer Policy, 19. M. Gillis, D. Perkins, M. Roemer and D. Snodgrass. (1983). Economics of Development, Norton. E. Goffman (1956). The Presentation of Self in Everyday Life, University of Edinburgh. F. Hirsch (1976). The Social Limits to Growth, Harvard University Press. M. Holbrook, and E. Hirschman (1980). ‘Symbolic Consumer Behavior: An Introduction’, Proceedings of the Conference on Consumer Esthetics and Symbolic Consumption, New York University, 16 and 17 May. A. Inkeles (1983). Exploring Individual Modernity, Columbia University Press. A. Inkeles, and D. Smith (1974). Becoming Modern, Harvard University Press. J. James (1977). Technology, Products and Income Distribution: A Conceptualisation and Application to Sugar Processing in India, ILO, World Employment Programme Research, Working Paper No. 61. J. James (1983). Consumer Choice in the Third World, Macmillan. D. Jelliffe (1972). ‘Commerciogenic Malnutrition’, Nutrition Review, Sept. M. Landgren-Gudina (1984). ‘Weaning Food and Low-Income Consumers in Ethiopia’, in W. van Ginneken, and C. Baron, (eds), Appropriate Products, Employment and Technology: Case Studies on Consumer Choice and Basic Needs in Developing Countries, Macmillan. R. Layard (1980). ‘Human Satisfaction and Public Policy’, Economic Journal, 90, Dec. E. Leach (1976). Culture and Communication: the Logic By Which Symbols Are Connected, Cambridge University Press. K. Lederer, (ed.) (1980). Human Needs, Oelgeschlager, Gunn and Hain. R. Ledogar (1975). Hungry for Profits, IDOC. D. Leslie (1995). ‘Global Scan: The Globalization of Advertising Agencies, Concepts and Campaigns’, Economic Geography. M. Lieber (1981). Street-Life: Afro-American Culture in Urban Trinidad, Schenkman. W. W. Lockwood (1954). The Economic Development of Japan, Princeton University Press. D. McClelland (1976). ‘Review of A. Inkeles and D. Smith, Becoming Modern’, Economic Development and Cultural Change, Oct. G. McCracken (1988). Culture and Consumption, Indiana University Press. A. MacEwan (1974). ‘Differentiation Among the Urban Poor: An Argentine Study’, in E. de Kadt, and G. Williams (eds), Sociology and Development, Tavistock. A. Maslow (1954). Motivation and Personality, Harper and Row.

60 Consumption, Globalization and Development

A. Mattelart (1983). Transnationals and the Third World, Bergian and Garvey. M. Mead (1961). Cooperation and Competition Among Primitive Peoples, 2nd edn, Beacon Press. J. C. Mitchell and A. L. Epstein (1959). ‘Occupational Prestige and Social Status Among Urban Africans in N. Rhodesia’, Africa, 29. A. Mubin, and D. Forsyth (1984). ‘Technology, Employment and Income Distribution: the Soap Industry in Bangladesh’, in W. van Ginneken, and C. Baron (eds), Appropriate Products, Employment and Technology: Case Studies on Consumer Choice and Basic Needs, Macmillan. C. Mukerji (1983). From Graven Images: Patterns of Modern Materialism, Columbia University Press. R. Nurkse (1957). Problems of Capital Formation in Underdeveloped Countries, Blackwell. A. Portes (1973). ‘Modernity and Development: A Critique’, Studies in Comparative International Development, 8 (3), Fall. M. Sahlins (1976). Culture and Practical Reason, University of Chicago Press. K. Sauvant (1976). ‘The Potential of Multinational Enterprises as Vehicles for the Transmission of Business Culture’, in K. Sauvant and F. Lavipour (eds), Controlling Multinational Enterprises, Westview. T. Scitovsky (1976). The Joyless Economy, Oxford University Press. A. Sen (1984). Values, Resources and Development, Harvard University Press. E. P. Skinner (1974). African Urban Life: The Transformation of Ouagadougou, Princeton University Press. N. J. Smelser and S. M. Lipset (eds) (1966). Social Structure and Mobility in Development, Aldine. M. G. Smith (1966). ‘Pre-Industrial Stratification Systems’, in N. J. Smelser and S. M. Lipset, (eds), Social Structure and Mobility in Development, Aldine. M. Solomon (1983). ‘The Role of Products as Social Stimuli: A Symbolic Interactionism Perspective’, Journal of Consumer Research, 10, Dec. A. W. Southall (1961). ‘Introductory Summary’, in A.W. Southall (ed.), Social Change in Modern Africa, Oxford University Press. A. M. Spence (1974). Market Signalling: Informational Transfer in Hiring and Related Screening Processes, Harvard University Press. G. Stigler and G. Becker (1977). ‘De Gustibus Non Est Disputandum’, American Economic Review, March. B. Supple (1981). ‘Income and Demand 1860–1914’, in R. Floud and D. McCloskey (eds), The Economic History of Britain Since 1700, vol. 2, Cambridge University Press. T. Veblen (1899). The Theory of the Leisure Class, Macmillan. J. Wells (1977). ‘The Diffusion of Durables in Brazil and Its Implications for Recent Controversies Concerning Brazilian Development’, Cambridge Journal of Economics, 1 (3). P. Wilson (1972). Crab Antics – The Social Anthropology of English Speaking Negro Societies of the Caribbean, Yale University Press.

4 From Global Products to Individual Functionings: Medicinal Drugs in Developing Countries Previous chapters and indeed most of the existing literature on the welfare effects of new products in developing countries are based on a form of analysis (developed primarily by Lancaster) that focuses on the specific characteristics embodied in those products.1 Those combinations of characteristics are then evaluated from the point of view of their appropriateness to household groups differentiated, most frequently, by levels of income. Because they are mainly produced in developed countries, new products, so the argument usually runs, contain a high proportion of ‘highincome’ characteristics that make them unconducive to the satisfaction of basic needs in developing countries. Or, put another way, the conclusion is typically that new products tend to confer their benefits disproportionately on rich rather than poor consumers in developing countries. From the point of view of individual welfare, however, the problem with this product characteristics oriented approach is that, as Sen (1985) has pointed out, it constitutes only part of the link between consumption and individual well-being and thus between new products and well-being. What also matters, according to Sen, is ‘what the person succeeds in doing with the commodities and characteristics at his or her command. For example, we must take note that a disabled person may not be able to do many things an able-bodied individual can, with the same bundle of commodities’ (Sen, 1985: 10). This comparatively neglected aspect of the relationship between consumption and welfare is what Sen calls functionings of persons, and our purpose here is to show how his approach 61

62 Consumption, Globalization and Development

can be analyzed in relation to one particular type of new product, namely, medicinal drugs. There are three reasons why we have selected this particular product. One is that the link between the characteristics embodied in modern medical drugs and their ultimate impact on health functionings is to a large extent objectively measurable. Indeed, there is a large medical literature which is available for this purpose, but which, until now, has gone largely unused in economic development or welfare economics, on account of what Sen refers to as a lamentable reluctance to go into medical matters (Sen, 1985: 46). The second reason is that the functionings to which modern drugs give rise in developing countries exhibit more individual variation than most, if not all other new products and this variation needs to be understood if one is to lessen the gap between those whose functionings are enhanced and those whose functionings are left unchanged or affected adversely. The final reason is that the products we have chosen so clearly illustrate a major theme of this book, as described in the first chapter. Modern medicinal drugs, that is to say, are, as a result of globalization, increasingly penetrating into areas of the Third World where consumption systems (e.g. in rural areas) diverge so markedly from the conditions prevailing in the (developed) countries, for whose consumption systems they are originally designed. Following a brief description of how the concept of functionings fits into Sen’s overall framework for analyzing the welfare effects of consumption, we show in the section thereafter why there is so much scope for individual variation in the impact of modern drugs on health functionings. Next, we discuss what factors determine whether that impact is positive, negative or negligible in particular cases. The final section examines the relationships between the variables thus identified.

Functionings in Sen’s model As set out in his Commodities and Capabilities (1985), Sen’s analysis is concerned with the relationships between commodities, characteristics (of those commodities), functionings and capabilities. It is not our intention here to describe all these complicated relationships in detail. Rather, what we wish to convey in simple graphical terms is the meaning of Sen’s statement that,

From Global Products to Individual Functionings 63

A functioning is … different both from 1) having goods (and the corresponding characteristics), to which it is posterior, and 2) having utility (in the form of happiness, resulting from that functioning), to which it is, in an important way, prior. The ‘posterior’ relationship, involving the connection between goods and characteristics – is familiar from Lancaster’s framework and is shown in the top left-hand panel of Figure 4.1. Indeed, as noted above, it is mainly this type of diagram that has been used in one form or another to analyze the welfare effects of new products. As was also noted previously, however, in Sen’s approach a second element needs to be added to the analysis in characteristics space, given his view that the welfare effect of a given bundle of characteristics depends on what is actually done with them. This second basic element of Sen’s model is reflected in the top right-hand panel of Figure 4.1, which, for the sake of simplicity, deals with the relationship between individual functionings and just one of the two characteristics used in the first panel.

char. 1 P

X char. 1

Y R

O

char. 2

happiness Figure 4.1

Functionings in Sen’s approach

T

S

functionings Q

64 Consumption, Globalization and Development

Consider for example the two goods X and Y, which, we assume, entirely arbitrarily, contain the amounts OP and OR of characteristic 1 respectively (one, rather than more characteristics, is chosen purely for simplicity). The extent to which these two amounts actually contribute to individual functionings, however, depends on the slopes of the lines connecting P and R to the horizontal axis in the top right-hand panel. Sen refers to these types of functions as ‘personal utilization functions’ and they depend on a variety of both ‘choice’ and ‘non-choice’ factors. The impact of a given amount of nutritional characteristics on nutritional functionings, for example, depends on a person’s metabollic rate (a non-choice factor) and the precise way in which he or she prepares the food (a choice factor). 2 Thus, depending on the particular combination of choice and nonchoice factors in our example, the utilization functions associated with good X may be more or less favourable to that individual’s functionings (PQ for example as opposed to PT). In the latter case (and assuming that the utilization function associated with good Y happens to have the same slope as PQ), then this good will have a more favourable impact on the person’s functionings than good X, even though the latter embodies more of the characteristic in question. By means of a similar line of argument, one can readily show that varying utilization functions produce correspondingly different effects on the functionings of two different consumers of the same amount of product characteristics. The last element of Sen’s model that can be captured in our simple graphical treatment is shown in the bottom right-hand panel of Figure 4.1. It reflects the fact that functionings are not an end in themselves but should be regarded instead as ‘prior’ events in the achievement of happiness. Sen refers to the function connecting these two variables as the ‘happiness function’ of an individual and in the illustrative example of Figure 4.1, this function is assumed to be such that happiness stands in a simple linear relationship to the achieved degree of functionings. In the rest of this chapter, we shall primarily be concerned with the nature of the utilization functionings connecting drug products and health functionings (that is, with the upper-right panel in Figure 4.1). First, however, we need to examine why the scope for variation in individual functionings is so considerable in the case of these particular products.

From Global Products to Individual Functionings 65

Modern drugs and the scope for individual variations in functionings Prior to the introduction of synthetic chemicals in the 1920s, the formulation of drugs relied mainly on natural plants and though the latter are still important in certain fields, the former are currently the major source of new drugs. Among the changes wrought by this transition was not only a dramatic increase in the number of drugs but also a striking change in the characteristics that they embodied. Most important from our point of view is the change that occurred in the relationship between the therapeutic efficacy of drugs and their potential toxicity. ‘Unlike many of the older agents’ that is to say, ‘which were by and large relatively ineffective but also relatively safe, the new drugs were often far more powerful – and far more toxic. … Physicians were accordingly faced not only with an astounding array and complexity of drugs, but with drugs possessing an equally astounding potential for curing – or for killing’ (Silverman and Lee, 1974: 6). Efficacy B

1990

A 1900

G

H

health functionings (negative)

D 1900

0

health functionings (positive)

E 1990

F Toxicity Figure 4.2

C

Drug efficacy, toxicity and health functionings over time

66 Consumption, Globalization and Development

Expressed slightly differently, what occurred was a greatly enlarged potential for variation in the impact of drugs on individual health functionings. Consider, for example, Figure 4.2, which contains characteristics on the vertical axis and functionings on the horizontal axis and which compares the case of a drug produced in (say) 1900 with a drug designed to treat the same illness in (say) 1990. Because the latter possesses potentially more of both efficacy and toxicity than the former (OB and OF versus OA and OE respectively), it increases the range of health functionings from HD (in the case of the 1990 product) to GC. The remainder of the chapter is concerned with the factors that determine whether the consumer of a modern drug product in a developing country ends up at point G, point C, or some intermediate position. Our discussion is based on a distinction drawn by G. A. Cohen (1993) between what a good does for a person as opposed to what the person does with the good. This distinction is important for policy purposes because to a considerable extent it overlaps with the distinction between factors that are and those that are not amenable to policy interventions designed to alter the relationship between consumption and individual functionings.

A. ‘What drugs do for people’ One reason why the individual has little or no control over the impact of drug consumption on her functionings is that this impact is heavily conditioned by factors innate to the individual herself. Table 4.1 lists the most important of those factors and provides examples of the mechanisms through which they operate. Many of the mechanisms and examples given in Table 4.1 are, of course, common to individuals in developed as well as developing countries. Other cases, however, are more prevalent in, or entirely specific to, the latter group of countries. Some of the variations listed in the table can be predicted (and, where unfavourable, avoided), whereas other variations are more idiosyncratic and less easy to predict. Predictability, one should emphasize, also depends on the nature of the drug itself. In the case of Lomotil, for example (a drug that is sometimes given in the treatment of infant diarrhoea), ‘The narrow range between therapeutic and toxic doses, and also the possibility of a child being abnormally sensitive … may account for the severe toxicity sometimes seen with low dosage’

Table 4.1 Source of individual-specific variations in health functionings following drug usage Source

Mechanisms/examples

Age

‘The main reason that age affects drug action is that drug metabolism and renal function are less efficient in babies and old people, so that with some exceptions, drugs tend to produce greater and more prolonged effects at the extremes of life’ (Rang, Dale and Ritter, 1995: 785).

Genetic factors

‘Genetic variation is an important source of pharmacokinetic variability’ (i.e. variability due to altered handling of drugs by the body, which leads to differing concentrations of the drug at the site of action) (Rang, Dale and Ritter, 1995: 789). Genetic variation can also cause differing physiological responses to the same drug concentration and some such responses may often be harmful (e.g. allergic individual responses even at low dosages). Glucose-6-phosphate dehydrogenase (G-6-PD) is a genetic disorder that is widespread in Africa, the Mediterranean, the Middle East and Southeast Asia. Certain types of drugs can produce hemolysis (a disorder of the blood) in G-6-PD-deficient individuals (some anti-malarial drugs, for example, produce this condition among Africans, Chinese and Thais (Weatherall, 1990), though it is generally well-tolerated in the general population.

Disease

‘Disease can cause altered handling of drugs by the body (pharmacokinetic variation) and/or altered sensitivity to drugs’ (Rang, Dale, Ritter, 1995: 42). Vaccines given to those suffering from protein-energy malnutrition may be ineffective, or less effective than vaccines given to those with adequate amounts of these nutrients (Silverman, Lee and Mydecker, 1982).

Physiological individuality

Pregnancy, weight (the therapeutic function of a drug may often be inversely related to the weight of patients), sex (female patients tend to react more strongly to most medications and to suffer from a higher incidence of adverse effects) (Martin, 1979).

Psychological individuality

The placebo effect: ‘It has been found that a placebo can potentiate, attenuate, or negate the active ingredients in a drug’ (Wickramasekera, 1985: 256).

68 Consumption, Globalization and Development

(Snyder et al., 1973: 47). The same is true of drugs such as digoxin (which are used in the treatment of heart disease), in that ‘The therapeutic level is 60 per cent of the toxic level. The toxic level, moreover, may vary considerably between one patient and another’ (Silverman and Lee, 1974: 275). The main conclusion to be drawn from Table 4.1, however, is that even without any differences in the way they behave, individuals can derive dramatic differences in health functionings from consumption of exactly the same therapeutic characteristics. Some individuals, that is to say, may end up near the one extreme point on the horizontal axis of Figure 4.2, while other individuals, behaving in the very same way, end up closer to the other extreme.

Environmental determinants of inter-individual variations in health functionings The examples that have just been cited are not the only ones to be considered under the heading of ‘what drugs do for people’. For there are also environmental factors that, on the one hand, influence the relationship between drug consumption and health functionings and on the other hand are unamenable to individual influence. ‘The effects of medications may be markedly altered’, for example, ‘if the physiological equilibrium is unbalanced by altitude, cold, heat, humidity, sunlight, vibrations, and other stresses’ (Martin, 1978: 103). But what comes perhaps most readily to mind in the current developing country context are the large differences between and within countries in the extent to which there is now resistance among the population to anti-malarial and antibiotic medications.3 Since these medications are amongst the most commonly prescribed in the Third World, such environmental differences as these clearly form an important part of any study that is concerned with the question of why given drugs are apt to elicit such marked differences in health functionings among individual patients in the Third World. What is less clear, though, is whether the drug resistance problem forms part of the category ‘what drugs do for people’ or whether it should be placed under the heading ‘what people do with drugs’. On the one hand, there are a number of studies which suggest that resistance is closely related to the way in which the drugs themselves are

From Global Products to Individual Functionings 69

used (and to this extent the issue should be dealt with below when we consider how functionings vary with drug usage). ‘Drug pressure’, concludes one such study of malaria resistance in Africa ‘is the main determinant of the rate of spread and development of resistance’ (Schapira, 1990: 43). Similarly, there is evidence of a direct and close relation between antibiotic use and antibiotic resistance in certain other parts of the Third World.4 On the other hand, it also seems unlikely that drug usage patterns alone can account for the entire resistance problem. ‘To some extent’, for example, the global incidence of genes that are resistant to antibiotics ‘may relate to the initial presence of the resistance determinant in a particular area’ (The Lancet, July 10, 1982: 86). Recent malaria research, moreover, indicates that ‘natural geographic variation in parasite mating habits may also play a role’ (Schmidt, 1995: 1670). That variation, it seems, has to do with differential degrees of inbreeding among malaria parasites in different parts of the Third World, with more pronounced inbreeding patterns tending to produce more acute resistance. Those patterns, in turn, appear to be heavily influenced by differences in local rates of transmission of the disease. In Tanzania, for example, the rate of transmission from mosquitoes to humans is relatively high: on average Tanzanians receive between 300 and 3,000 infected bites per year. ‘A single human host often carries many different strains of malaria at once, and mosquitoes are likely to take up mixed strains, allowing frequent cross-mating’ (Schmidt, 1995: 1670). In Papua New Guinea, on the other hand, ‘the transmission rate of malaria is 10 times lower, which means more mating between parasites of the same strain’ (Schmidt, 1995: 1670).

B. What people do with drugs Under this heading one needs to consider whether: 1) a given medication is capable (in principle) of effectively treating the illness or disease from which the patient is suffering (viral infections for example do not respond to antibacterial drugs) and 2) whether the medication is used in a manner that makes the maximum therapeutic contribution to the treatment of that particular illness or disease. This second issue, of course, only becomes relevant if the first condition is satisfied, the likelihood of which itself depends on the accuracy of the diagnosis, as

70 Consumption, Globalization and Development

well as the suitability of the medication that is chosen to treat the particular problem (Here again the issues are sequentially related: a suitable drug can only be chosen once an accurate diagnosis has been made. Note too that the term ‘suitable’ is being used here in a clinical sense and not in the sense of whether the characteristics embodied in a drug product are appropriate to persons with different incomes). Given the choice of a product that is capable of treating a particular illness effectively, a range of factors bearing on the use of the product then determines whether this potential is actually realized, or whether other, possibly negative impacts on health functionings will occur instead. Of these types of variables, the following three, somewhat overlapping categories, are perhaps the most important. 1) Compliance: To a greater or lesser extent, the efficacy of all drugs is dependent upon patient compliance with factors such as dosage (how and how frequently the drug is to be taken, at what times of the day, for what length of time, etc.); storage (under certain conditions of temperature, light, etc. drugs tend to lose their efficacy, a problem that assumes particularly serious proportions in tropical conditions);5 and simultaneous use with other drugs (see below). For some drugs, however, compliance is more of a problem than others. It is an especially problematic issue in the treatment of tuberculosis, for example, because of the length of the treatment (6 to 12 months) and the need to take several drugs. 2) Interactions: ‘Drug products may be made dangerously toxic or may be partially or wholly prevented from eliciting the desired response in patients because of interactions with other substances’ (Martin, 1978: 10). Many drugs, for example, ‘interact adversely with certain other drugs, food constituents, and other chemicals’ (Martin, 1978: 10). Such interactions, one should emphasize, are not confined only to prescription drugs; they frequently occur with ‘over-the-counter’ medications as well.6 3) Complementary inputs: We are referring here not only to the physical inputs that are often required for the effective use of modern drugs (such as refrigeration, measuring instruments, etc.) but also to the human capital inputs that are required for this purpose e.g. literacy, consumption skills. Even in the developed countries, for whose inhabitants the vast majority of modern drugs are specifically designed, there is variation in what people do with drugs and so too therefore is there variation

From Global Products to Individual Functionings 71

in the impact of those drugs on individual health functionings7 (see below, for example, on differing rates of compliance in developed countries). In the Third World, even more marked variations may be expected in the relationship between drug consumption and individual health functionings. For while a relatively small group of consumers in those countries tends to derive something akin to the maximum therapeutic benefits from modern drug products, in the vast majority of cases there are a number of factors that make such an outcome extremely unlikely. We can best consider these factors in relation to the variables we have already described in this section, namely, diagnostic accuracy and drug suitability, patient compliance, drug interactions and complementarities in drug use.

Diagnostic accuracy and drug suitability To a far greater extent than in developed countries, modern drug use in the developing world is unaccompanied by medical supervision. In large part, this is because drugs which in developed countries are available only on prescription, can often be freely purchased without any such restriction. But it also reflects the fact that for many poor consumers in the Third World, low incomes give rise to a conflict between purchasing medications on the one hand and medical services on the other. As is well illustrated by a study of drug use among rich and poor households in Brazil, this conflict is usually resolved in favour of the medication rather than the medical services. In particular, ‘the number of medicines in use in the peripheral shanty towns and in the best suburbs … did not vary much’ (Muller, 1982: 110). However, ‘The most important difference between the rich and poor was that the majority of the poor (52 per cent) obtained their medicine without a prescription, which the rich usually had. Further analysis confirms that ‘the real difference between rich and poor in Brazil is in their access to doctors’ (Muller, 1982: 110). A similar picture emerges from a study of drug use patterns among low-income rural households in the Philippines, who, it seems, ‘rely more on western medicines than on western doctors’ (Hardon, 1987: 287) and who use such medicines ‘even in minor self-controlling ailments’ (Hardon, 1987: 287). (All told, in this sample, modern pharmaceuticals, prescribed and non-prescribed, were used in about 50 per

72 Consumption, Globalization and Development

cent of the cases where no professional was consulted.) The problem with this pattern of usage is that in many cases, the drugs thus selected for home use turn out to be a highly inappropriate form of treatment, as when, for example, anti-diarrhoeal combination products are used against infant diarrhoea, even though they are generally not recommended for this purpose, not only because they expose patients to unecessary risks but also because they contribute to the drug resistance problem noted earlier.8 This comparative lack of correspondence between drug use and medical supervision is not, however, the only factor that makes the prospect of effective treatment comparatively unlikely in developing, as opposed to developed countries. For even when medical supervision is available, ‘Much -and perhaps everything- depends on whether the product is described in the labeling and promotion so that the physician can order the right drug for the right patient, in the right amounts and at the right time, and so that he can give each patient the proper instructions for use’ (Silverman, Lee and Mydecker, 1982: 13). In these respects, too, unfortunately, there is good reason to suppose that patients in the Third World tend to be severely disadvantaged in comparison with their counterparts in the rich, industrialized countries. Much of this difference, it seems, has to do with what ‘drug companies tell physicians in industrialized nations’ (Silverman, Lee and Lydecker, 1982: 14) as opposed to ‘what they say in the developing countries of the Third World’ (Silverman, Lee and Lydecker, 1982: 14). Consider, in particular, what one very detailed study of multinationals operating in a wide range of developing countries, has to say on the topic. Thus, Many products ousted from the market in such industrialized nations as the United States and Great Britain, or never approved for marketing, are readily available and widely promoted in the Third World. … With many important products, the dangers of serious or lethal side effects are minimized, glossed over, or totally ignored. … With many of these drugs, claims of effectiveness are wildly exaggerated. (Silverman, Lee and Lydecker, 1982: 85, emphasis in original)9 Possessed thus of information that is systematically inferior to what is propagated by the same firms in the developed countries, physicians in the Third World are correspondingly less likely to provide their patients with appropriate diagnostic and other care.

From Global Products to Individual Functionings 73

Compliance and drug interactions Rates of compliance with recommended dosages appear to be extremely low in developing countries. One Indian study, for example, suggests that out of the total number of antibiotics purchased without a prescription, only 11 per cent were taken in the correct dosage for a long enough period of time (and as many as 42 per cent of the drugs were taken for less than a day). 10 And even when the antibiotics were purchased from a medical practitioner, they were adequately prescribed in only a quarter of the cases. A separate study of Indian patients found a similar pattern: ‘among the self medicated, 30 per cent purchased antibiotics for less than 1 day and only 18 per cent purchased them for 5 days … on the doctor’s prescription only 40 per cent … were for a full course (minimum five days)’ (Krishnaswamy et al., 1985: 367). To a degree, findings such as these reflect some of the same factors that cause differences in patient compliance in developed countries, such as ambiguities in instructions or motivational differences among patients which affect their willingness to comply. 11 After all, even in those countries where conditions are comparatively favourable, rates of compliance with usage requirements in the case of antibiotics, are said to be no more than 50 per cent.12 To a probably greater degree, however, the findings reported in the previous paragraph, reflect problems that are peculiar to, or more pronounced in, the developing countries themselves. In those countries, for example, the poorest groups of patients often purchase one or two loose tablets from an open bottle whose package insert has already been removed, physicians often fail to provide any information about dosage requirements and such information, even when it is available, is often in a form that is inaccessible to poorly educated consumers.13 Because of the lack of information with which they are associated, these same factors, moreover, tend to raise the probability of unfavourable drug reactions in developing, as compared with developed countries. Against this, one could argue that patients in developing countries use fewer drugs simultaneously, thus lessening the likelihood of unfavourable drug interactions. Yet, while this argument may hold in some cases, it is well to emphasize that ‘in many developing countries up to six or seven drugs may be prescribed for conditions in which one or two would do’ (Hogerzeil, 1986: 436).

74 Consumption, Globalization and Development

Complementarities in drug use If it is to be effective, modern drug use often depends on the availability of certain inputs, some of which (such as refrigerators14 and sterilized syringes) take the form of physical capital, while others are better described in terms of the human capital that is necessary to read and understand dosage requirements, contra-indications, etc. In developed countries and those parts of the Third World with relatively high levels of income and education, these requirements are widely met and the translation of drug consumption into health functionings is consequently (from this point of view) not impaired to any significant extent. Among the poorest households in developing countries, by contrast, the all too frequent absence of the inputs we have just described means that a given drug will tend to have a less beneficial and sometimes even a highly unfavourable impact on health functionings (in much the same way that an absence of refrigerators and clean water has so adversely influenced the effect of infant formula in developing countries). Many of the problems that have just been described in relation to the consumption of modern medicines by the poorest members of developing countries, currently manifest themselves in a particularly interesting way in large parts of urban India. What is intriguing there is how ‘neighbourhood doctors’ are used in the medical treatment of urban slum dwellers. Consider, for example, the case of New Dehli, where, in at least one urban slum area (and probably a large number of other as well), although a qualified medical practitioner sees patients twice a week, the inhabitants, too poor to give up even a morning’s wages, usually rely [instead] on ill-educated amateurs who are open for business till 10 every right. In dingy little shops that swarm with flies, weary people, illuminated by dangling bare light bulbs, unquestioningly accept all manner of pills and shots from young men who have never gone to college. … As there is currently no law that explicitly prohibits their work, they operate with impunity in more than a thousand slum areas, known as jhuggies, that are home to about a third of

From Global Products to Individual Functionings 75

this capital city’s 10 million residents. (International Herald Tribune, Sept. 29, 1998: 2) Lacking any formal medical education, such ‘practioners’ are usually ill-equipped to dispense the powerful modern drugs which they apparently acquire and learn about from travelling drug company salesmen. In some cases, for example, infant diarrhoea is treated with a medication that is known to cause dehydration in young children, while in other cases antibiotics are prescribed for only two days, instead of a full course of treatment. And ‘While individuals treated so briefly with antibiotics often get better, the bacteria they were infected with can often become drug-resistant for a whole community’ (International Herald Tribune, Sept. 29, 1998: 2).

What people do with drugs and health functionings We suggested earlier that the role of functionings in Sen’s approach could be neatly explained in simple graphical terms. We now show that this same framework can be used to illustrate and compare the mechanisms through which differences in drug usage behaviour influence individual health functionings. Therapeutic characteristics X Therapeutic characteristics

Q

Y S T

0

W

Z

V

Health functionings Figure 4.3

Drug usage, characteristics and functionings

U

76 Consumption, Globalization and Development

As before, the issue is approached in terms of how the relationships in characteristics space (the left-hand panel of Figure 4.3) are mapped into differences in functionings (the right-hand panel). Consider to begin with the ‘bench-mark case’ where the individual derives the maximum therapeutic benefits from the drug products that he or she consumes (this begins in Figure 4.3 at point Q, which is translated, we assume, into functionings by the line SU). Deviations from this outcome occur either because the individual derives less than OS of the therapeutic characteristics embodied in the product, or because that amount of characteristics translates itself into functionings via a less favourable function than SU (Of course, a combination of these categories is also possible). Into the first category would fall cases, where, for example, the individual fails to comply with drug usage requirements (by taking the drug at say the wrong time of day), or where negative interactions between two drugs occur. In Figure 4.3, we assume that events such as these cause the individual to derive only the amount OT of therapeutic characteristics and hence to derive proportionately less in terms of health functionings (assuming, as we are, that TV runs parallel to SU). The second category includes the case where the individual takes the drug entirely as recommended, but because it is clinically unsuited for the particular illness from which she is suffering, no positive impact on health functionings emerges (and the function connecting therapeutic characteristics and health functionings then coincides with the vertical axis). The problem of drug resistance can also be placed in this category to the extent that it is caused by factors other than inappropriate use patterns, for what then occurs is an inward movement over time of the line SU (to say SZ), with no reduction at all in the amount of available therapeutic characteristics. To some degree, though, resistance is itself partly the result of misuse and the deviation from the ‘ideal’ outcome should then be depicted as a combination of fewer characteristics and an inward shift of the line SU, resulting, ultimately, in a point such as W on the functionings axis.

An integrated policy framework At various points we have suggested that the impact of drug consumption on individual health functionings depends on variables

From Global Products to Individual Functionings 77

that are sequentially related. From a policy point of view, it is useful to present these types of relationships in the graphical form of a ‘tree’, each of whose forks represent the outcome of a particular stage in the process by which drug consumption is converted into health functionings. At each of the first three forks in Figure 4.4 (collectively representing the category ‘what people do with drugs’), there is an outcome which is necessary (but not sufficient) for drug consumption to engender an improvement in the health functioning of an individual and an outcome which prevents this from occurring (regardless of the stage when it occurs and regardless of whether the upper-fork has a very high probability of occurring in any subsequent stages). Framing the issue this way is of course something of an oversimplification since in reality there can be degrees of satisfaction with variables such as correctness of usage; the point of the ‘tree’ diagram however is less to capture the details of the process and more to emphasize its sequentiality. The last fork, which is labelled ‘efficacy’, corresponds to what we referred to earlier as ‘what drugs do to people’ and it shows that even when the first three outcomes are favourable, there may still be some probability that a drug will fail to have the desired effect on health functionings.

efficacy (p4) correct usage (p3) correct selection (p2) non-efficacy (1-p4) correct diagnosis (p1) incorrect usage (1-p3) incorrect selection (1-p2) incorrect diagnosis (1-p1)

diagnostic stage

Figure 4.4

selection stage

usage stage

‘what drugs do to people’

Sequentiality in the determinants of health functionings

78 Consumption, Globalization and Development

If we assume that all of the variables in the figure are independent of one another, then the likelihood that a given medication will elicit the desired response (what in relation to Figure 4.3 we referred to as the ‘bench-mark case’), becomes a multiplicative function of the four individual probabilities, p1, p2, p3 and p4. 15 We have argued that the first three probabilities tend to be low in developing rather than developed countries and among low-rather than highincome consumers in those societies. Not all medications, however, conform to this typical pattern and the exceptional cases can also be explained in terms of the probabilistic framework we have just described. Measles vaccines, for example, are said to have an efficacy of 85 per cent in that they reduce the incidence of measles by that percentage and since the remaining probabilities are equal to 100 per cent (by definition), the overall probability is also 85 per cent.16 ORT (oral rehydration therapy) also has a very high efficacy (as measured by the decrease in case fatality rates) and unlike the vaccine it is susceptible in principle to probabilities of less than 100 per cent in the other variables. Field data suggest, however, that diagnostic accuracy is also very high (approximately 95 per cent) and patient compliance is said to be in the range of 60 to 80 per cent. 17 Thus, with efficacy estimated to be 95 per cent, as already noted, the combined probabilities are in the region of 54 to 72 per cent. It is worth emphasizing that outcomes such as these are only possible when all upper branches of the tree exhibit relatively high probabilities (that is, when there are no significantly weak links in the chain of sequentiality). This point is worth emphasizing because of its implication that the major policy challenge for the vast majority of other drug products, is to simultaneously raise the existing low probabilities at all stages of the process by which drug consumption is converted into health functionings in developing countries.

Conclusions Most existing discussions of the welfare effects of new products in developing countries are based on the appropriateness to different household groups, of the characteristics that are embodied in those products. Basing our argument on Sen’s notion of functionings, we take the view that this approach constitutes only part of the true

From Global Products to Individual Functionings 79

relationship between consumption of new products and individual well-being.18 What also matters is what the individual is capable of doing with the characteristics at his or her disposal. This additional consideration, we believe, is especially relevant to the case of modern drug products that, on account of globalization, are increasingly being used in developing countries, since there is so considerable a degree of variation in the way that different individuals respond to consumption of the very same medication. Our main concern in this chapter has been to examine the origin of these pronounced variations in individual health functionings and thereby better to understand the actual welfare effects of drug products in developing countries. What we find is that some such variations are due to factors over which the individual consumer of a particular drug product has little or no control (and these factors were accordingly described as belonging to a category entitled ‘what drugs do to people’). To a probably greater extent, however, the observed differences in health functionings appear to derive from what individuals actually do with the drugs that they purchase and it is in this respect (rather than in the factors belonging to the first category) that systematic differences between groups of households are seen to emerge. In particular, though our evidence is based on case studies from different countries we found that modern drugs tend to be least effectively used by low-income and poorly educated consumers in developing countries, who, indeed, often seem to derive the toxic rather than (or, in some cases, additional to) the therapeutic characteristics of those products. Because the variables that give rise to this conclusion are sequentially related, policy interventions need to be directed to all stages in the process by which drug consumption is converted into health functionings among these particular groups of consumers. Because the subject of this chapter has been new medicinal drugs designed in the rich countries, it is worth relating our conclusions to the notion of inappropriateness, which, as noted above, is sometimes said to reside in the transfer of products from developed to developing countries. Our view of the relationship between these two issues is that misuse of medical drugs in developing countries can be conceived in part as an additional dimension of product inappropriateness; additional, that is, to the other and better-known dimensions of inappropriateness such as product characteristics that

80 Consumption, Globalization and Development

(because they are designed for average tastes and incomes in the developed countries) are unsuited to the much lower incomes of the vast majority of Third World consumers. For as we see it, medicinal and other new products are designed to fit in not only with average tastes and incomes in the developed countries, but also with the available consumption skills, infrastructure and climatic conditions in those countries. New products, like technological innovations in general, are designed, that is to say, to form part of particular technological systems and where elements of those systems (such as medical supervision) are missing, product misuse is the probable outcome (as suggested in Chapter 1). To this extent there is a need for medicinal products to be redesigned for specific developing country conditions rather than the consumption systems prevailing in the rich countries. Nylon bed-nets for example could be used to combat malaria. Drugs could be redesigned for example with shorter treatment regimes to reduce the need for scarce medical supervisory support (Cowan and Heap, 1993) or with a lesser demand for expensive complementary inputs such as refrigerators. Or again, there is a need for medicines that deteriorate less rapidly under tropical climatic conditions. Apart thus from the need for product adaptations, in many developing countries a high priority should be given to policy innovation in the area of medical standards. What appears to be true of India, for example, as noted above, is a glaring gap between the regular medical profession on the one hand and a totally unregulated informal sector on the other. The problem with this situation, as we see it, is that it confronts the majority of poor patients in that country with an inappropriate choice, namely, between the high and expensive standards of regular medicine on the one hand and a cheap, but hazardous type of medicine, which is subject to no standards, on the other hand. What is needed, rather, is an intermediate solution which imposes limited standards on the practice of medicine in the informal sector, along the lines of ‘barefoot doctors’ in China. What also needs to be emphasized, however, is that drug misuse in developing countries cannot entirely be ascribed to product inappropriateness or lack of suitable policy. The reason is that even in the developed countries these products are improperly used – albeit to a lesser extent than in the Third World – and it appears that there are reasons for misuse common to both types of countries (such as,

From Global Products to Individual Functionings 81

for example, the lack of sufficient patient motivation for full treatment compliance). Where such common factors prevail, it is these rather than issues of inappropriateness per se that need to be addressed by policy. Still another observation has to do with implications for further research in the area of drug consumption and individual health functionings. In particular there is a need for research to show how the various different patterns of drug use and misuse we have described translate ultimately into aggregate differences in infant mortality, health, longevity and so on. (To what extent, for example, does the greater misuse of antibiotic medications in developing countries influence patterns of health and longevity in those as opposed to developed countries?) At present there is no such specific evidence known to us. There are however a few village-level studies that illustrate the kind of evidence that is needed to make the quantitative connections between product use and aggregate human functionings. Results from a community-based trial in Bangladesh, for example, suggest that use of oral rehydration salts (ORS) can lead to a 79 per cent reduction in diarrhoea case fatality rates (Bennett et al., 1990). Other evidence comes from the use of infant formula in developing countries which poses issues and problems that are closely related to those associated with medicinal drugs (especially the lack of certain necessary elements of the consumption system as a whole). There are in particular several studies that seek to quantify the impact of formula use on rates of hospitalization due to severe diarrhoea and on infant mortality at the village level (Jelliffe and Jelliffe, 1978). In each case, formula use is shown to have had a strongly negative impact on these indicators in contrast to breast-fed infants in the same villages who did far better. Since the use of infant formula in developed countries generally has few such negative effects, the sharply higher rates of hospitalization and mortality that are reported for the developing countries, can to a seemingly large extent be ascribed to the different patterns of use which are known to exist among resource-poor mothers in those countries. Finally, there is a need to acknowledge the potential of social marketing for improving the way in which modern drugs are used in developing countries. As suggested in Chapter 8, the same globalizing factors that give rise to an increase in the consumption of

82 Consumption, Globalization and Development

modern products (such as the rapid expansion of the mass-media) can also be used to improve the conditions under which they are sold and used.

Notes 1. See for example James and Stewart (1981), Helleiner (1975), Jenkins (1988), and Stewart (1977). 2. This distinction is made in Sen (1995) 3. See for example Schapira (1990). 4. As reported in The Lancet, July 10, 1982, p. 86: ‘A direct relation was found between antibiotic use and the frequency and kinds of antibiotic faecal strains in human beings in Toluca, Mexico’. 5. One study undertaken in Sudan, for example, showed that a number of pharmaceutical preparations underwent serious deterioration and degradation under tropical conditions. See Tropical Doctor, July 1980. 6. For a number of examples see Silverman and Lee (1974). 7. This is also discussed in Silverman and Lee (1974). 8. ‘The result is treatment failure, high costs and deaths’ (Hardon, 1987: 287). 9. Another study, of multinational pharmaceutical firms operating in Nigeria, found that most such firms ‘especially through their affiliates and subsidiaries, overpromote their drugs for extra indications and with mention of fewer hazards’, in comparison with information provided in the United States. See Osifo (1983: 8). 10. These data are taken from Greenhalgh (1987). 11. For a detailed review of these and other factors that contribute to patient compliance in the developed countries, see Blum (1981). 12. According to a survey released in 1995 by the American Lung Association. See USA Today, 13 October, 1995. 13. For the case of Sri Lanka, see Wolffers (1987); the Philippines, Hardon (1987); India, Greenhalgh (1987) and Krishnaswamy et al. (1985); Ethiopia, Sekhar et al. (1981). 14. Some antibiotics, such as amoxicillin, for example, deteriorate without refrigeration. 15. This approach is used by Bennett et al. (1990). 16. These data are taken from Bennett et al. (1990). 17. As for previous note. 18. The Sen approach is applicable in principle not just to medicinal drugs but to all products, which generally contribute to a wide range of human functionings, social and psychological, as well as physical. Food, for example, as Sen himself points out, contributes not only to one’s nutritional functionings but also plays an important role in a person’s social relationships. But whereas the former type of functionings can be objectively measured, many of the latter types of functionings are usually difficult to measure and are thus less amenable to

From Global Products to Individual Functionings 83

analyzing the effect of consumption on welfare in the same empirical way as we have done. For many products, therefore, it seems to us that future research should focus on measurement issues related to the subjective connection between consumption and psychological and sociological functionings.

References K. Bennett, P. Tugwell, D. Sackett and B. Haynes (1990). ‘Relative Risks, Benefits, and Costs of Intervention’, in K. Warren and A. Mahmoud (eds), Tropical and Geographical Medicine, 2nd edn, McGraw-Hill. R. Blum (1981). ‘Factors Affecting Individual Use of Medicines’, in R. Blum et al. (eds), Pharmaceuticals and Health Policy, Holmes and Meier. G. A. Cohen (1993). ‘Equality of What? On Welfare, Goods and Capabilities’ in M. Nussbaum and A. Sen (eds), The Quality of Life, Clarendon Press. G. Cowan and B. Heap (1993). Clinical Tropical Medicine, Chapman and Hall. T. Greenhalgh (1987). ‘Drug Prescription and Self-Medication in India: An Exploratory Survey’, Social Science Medicine, 25 (3). A. Hardon (1987). ‘The Use of Modern Pharmaceuticals in a Filipino Village: Doctor’s Prescription and Self Medication’, Social Science Medicine, 25 (3). G. Helleiner (1975). ‘The Role of Multinational Corporations in the LDC’s Trade in Technology’, World Development, 3 (4). H. Hogerzeil (1986). ‘Use of Essential Drugs in Rural Ghana’, International Journal of Health Services, 16 (3). J. James and F. Stewart (1981). ‘New Products: A Discussion of the Welfare Effects of the Introduction of New Products in Developing Countries’, Oxford Economic Papers. D. Jelliffe and E. Jelliffe (1978). Human Milk in the Modern World, Oxford University Press. R. Jenkins (1988). ‘Transnational Corporations and Third World Consumption: Implications of Competitive Strategies’, World Development, 16 (11). K. Krishnaswamy, B. Kumar and G. Radhaiah (1985). ‘A Drug Survey-Precepts and Practices’, European Journal of Clinical Pharmacology, 29. E. Martin (1978). Hazards of Medication, 2nd edn, J. B. Lippincott Co. M. Muller (1982). The Health of Nations, Faber and Faber. N. Osifo (1983). ‘Overpromotion of Drugs in International Product Package Inserts’, Tropical Doctor, Jan. H. Rang, M. Dale and J. Ritter (1995). Pharmacology, Churchill, Livingstone. A. Schapira (1990). ‘The Resistance of Falciparum Malaria in Africa to 4-Aminoquinolines and Antifolates’, Scandinavian Journal of Infectious Diseases, Supplement 75. K. Schmidt (1995). ‘Inbred Parasites May Spur Resistance’, Science, 269, 22 Sept. C. Sekhar, R. Raina and G. Pillai (1981). ‘Some Aspects of Drug Use in Ethiopia’, Tropical Doctor, July.

84 Consumption, Globalization and Development

A. Sen (1985). Commodities and Capabilities, North-Holland. M. Silverman and P. Lee (1974). Pills, Profits and Politics, University of California Press. M. Silverman, P. Lee and M. Lydecker (1982). Prescriptions for Death: the Drugging of the Third World, University of California Press. R. Snyder, H. Mofenson and J. Greensher (1973). ‘Toxicity from Lomotil’, Clinical Pediatrics, January. F. Stewart (1977). Technology and Underdevelopment, Macmillan. D. Weatherall (1990). ‘Common Genetic Disorders in the Tropics’, in K. Warren and A. Mahmoud (eds), Tropical and Geographical Medicine, 2nd edn, McGraw-Hill. I. Wickramasekera (1985). ‘A Conditional Response Model of the Placebo Effect: Predictions from the Model’ in L. White, B. Tursky and G. Schwartz (eds), Placebo: Theory, Research and Mechanisms, Guilford Press. I. Wolffers (1987). ‘Drug Information and Sale Practices in Some Pharmacies of Colombo, Sri Lanka’, Social Science Medicine, 25 (3).

5 Do Consumers in Developing Countries Gain or Lose from Globalization?

Drawing partly on the three previous chapters our main argument in this chapter will be that traditional consumption theory cannot serve as a framework for answering the question posed in the title because the theory cannot deal with the features of globalization that bear most tellingly on consumers in poor countries. More specifically, what we will describe is globalization as a process whose welfare effects on consumers are especially prone to the forms of uncertainty and disappointment that are assumed away in traditional demand theory.1 If one is to describe the actual gains and losses to consumers in developing countries from globalization, therefore, it is necessary to replace the assumptions of this theory with alternatives that describe reality more accurately. Before setting about this task, however, let us begin by examining how the assumptions of traditional demand theory preclude the possibility that consumers may experience feelings of frustration and disappointment, during acts of consumption that are undertaken because they are expected instead to yield satisfaction or utility.

The exclusion of frustration and disappointment in traditional consumption theory The assumptions of traditional consumption theory are such that there can never be a discrepancy between what the consumer expects to achieve and what is actually realized. That is to say ‘There are no surprises with respect to the amount of utility achieved’ (Cyert and de Groot, 1975: 225). 85

86 Consumption, Globalization and Development

Most obvious among the assumptions that this conclusion requires is probably perfect knowledge: that is, ‘In accordance with this postulate, people are supposed to calibrate their purchases and time uses by matching their preferences, which are fully known to them, against the equally well-known world of available consumption experiences’ (Hirschman, 1982: 17). The immunity of the consumer to possible ‘surprises’ in traditional theory, however, depends not only on the requirement that preferences be known, but also that they be constant. For once preference change occurs, there is always the possibility that the consumer judges herself to be worse off in the new situation than the old. ‘Another assumption that has the effect of excluding disappointment from the purview of traditional theory is that individual satisfaction depends only on one’s own consumption’ (James, 1993: 148). This assumption excludes not only the variety of ways in which the market mechanism is thought to promote ‘social comparisons rather than self-reference’ (Lane, 1991: 220), but also and more generally the entire area that economists refer to as ‘negative consumption externalities’. Still another way in which expected and actual utilities are made to converge in traditional theory is that consumption takes place at the point where goods are purchased. This assumption not only excludes the range of factors that make the characteristics derived from the same commodities a highly variable magnitude, but it also has the effect of excluding the variability in the way that characteristics themselves promote human functionings. (‘For example, if a person has a parasitic disease that makes the absorption of nutrients difficult, then that person may suffer from undernourishment even though he may consume the same amount of food as another person for whom that food is more than adequate.’2) Given all the assumptions of traditional theory that have just been described, one may reasonably presume that consumers in developing countries will tend to gain from (or remain unaffected by) globalization, inasmuch as the increased trade and foreign investment that are associated with this phenomenon not only expand the choice of available goods, but also tend to lower the prices at which such goods can in general be purchased. By relaxing each of these assumptions, however, we shall now attempt to show that the actual welfare effects of globalization on consumers in poor

Gain or Lose? 87

countries are far more complex than traditional theory allows. With regard to some assumptions, this has already been demonstrated in the previous chapters. By including the other relevant assumptions, however, this chapter seeks to provide a discussion of the issue as a whole.

Globalization and information imperfections In traditional theory, as already noted, there can be no distinction between the consumer’s revealed and her informed preferences, since, by assumption, no such discrepancy can arise. Yet, there are numerous scholars who see not only the possibility of a large deviation between the two types of preferences, but also the need, in such cases, to reject one’s actual preferences in favour of one’s (hypothetical) informed preferences.3 Harsanyi, for example, has argued that ‘a person’s utility function should be defined in terms of his hypothetical informed preferences rather than in terms of his actual preferences because some of the latter may be badly mistaken. … This approach … permits us to bypass his possibility mistaken actual preferences which are contrary to his own, more fundamental, informed preferences, which can be considered as his actual preferences as freed from the distorting effects of his factual errors’ (Harsanyi, 1992: 6–7). Consider, for example, the case of someone ‘who has a bad case of pneumonia, for which A is the best medication. But he erroneously thinks that B is the best medication, even though the latter would be quite ineffective against the type of pneumonia he has’ (Harsanyi, 1992: 6). In this case, therefore, the medication that was actually chosen contributes little or nothing to the consumer’s true welfare, defined, as it should be, by the degree to which his informed preferences are satisfied. For our purposes, the relevant question is then whether there are any reasons to expect systematic deviations of this kind to occur in relation to the products associated with globalization in developing countries. And the answer to this question, as we see it, needs to be based on the recognition that ‘vendors have incentives to provide reasonably accurate information about those features of … commodities that might attract … customers to … buy goods in their establishments if and only if the … customers have sanctions against deceit. Where they do not, as in … some purchases such as medicines

88 Consumption, Globalization and Development

and some foods, informed satisfaction is not encouraged by market structure’ (Lane, 1991: 461, emphasis added). Let us now apply this recognition specifically to advertising in developing countries, partly because this is the main source of information about many new products (or products newly introduced into developing countries),4 and partly because advertising has been so heavily associated with the recent processes of globalization in various parts of the Third World. More specifically, over the past ten years advertising growth has been most rapid in Asia and Latin America and within those regions, moreover, certain countries stand out. for China [advertising growth has been] more than 1,000 per cent, for Indonesia 600 per cent, for Malaysia and Thailand more than 300 per cent and for India, the Republic of Korea and the Philippines more than 200 per cent. And compared with GDP levels, the advertising expenditures in developing countries are particularly high [as shown in Table 5.1].5 Apart from constituting a response to the rapid growth in certain developing country markets, the change in advertising expenditures shown in Table 5.1 also reflects several other factors as well. One of them is the combined result of a rapid recent spread of television receivers to all parts of the Third World (see Table 2.1 above) and technological changes, primarily in the fields of cables and satellites, that have led to a massive increase in available channels per receiver (China, for example, ‘had 11 million cable television subscribers in 1990–35 million in 1995. … India had 7 million in 1993, and 16 million in 1995’).6 The second main reason why advertising plays so central a role in the globalization process of the past decade or so, is that the 1980s and 1990s have witnessed the de-regulation of public television, with a concomitant expansion of commercial programming in a large number of developing countries (the conversion of the state-owned television service in India is perhaps the most pronounced example of this general tendency7). To what extent, then, have such intensive advertising efforts in globalizing economies of the Third World (undertaken mainly, but not exclusively by multinationals) been subject to sanctions on the provision of misinformation described earlier? While one cannot of

Table 5.1 The spread of advertising to the developing world – top 10 countries in advertising expenditure as a share of GDP, 1986 and 1996 1996

1986

USA Australia United Kingdom New Zealand Hong Kong, China Switzerland Colombia Spain Venezuela Finland

Advertising as a percentage of GDP

Total advertising expenditure (US$ billions)

1.6 1.4 1.4

94.6 4.3 13.0

1.1 1.1

0.7 1.1

1.0 1.0 1.0 1.0 0.9

2.7 0.5 4.2 0.6 1.0

Country Colombia United Kingdom New Zealand Hong Kong, China Korea, Rep. of Venezuela USA Taiwan, China Brazil Australia

Advertising as a percentage of GDP

Total advertising expenditure (US$ billions)

2.6 1.4

1.4 16.6

1.4 1.4

1.0 2.2

1.4 1.4 1.3 1.2 1.2 1.2

6.7 1.0 101.2 3.4 8.2 4.7

Source: UNDP, Human Development Report, 1998

89

90 Consumption, Globalization and Development

course provide a complete answer to this question, what can be said is that in product categories where information is particularly important for the consumer (such as food, medicinal drugs and pharmaceuticals), effective sanctions in most developing countries are virtually non-existent and reliable information is invariably more scarce than in developed countries (where penalties for supplying misleading information do exist, even if they are not always applied as fully as they could be). With regard to medicinal drugs, for example, it is still the case that the information provided by drug companies in the Third World is inferior over a wide range of products to what is supplied by the same firms in the industrialized countries.8 Similarly, ‘Tobacco consumption is a growing problem in Africa, Asia, Latin America and the Arab States’ where, typically ‘information campaigns lag far behind, while marketing and advertising campaigns have intensified’ (Human Development Report, 1998: 63). And as with medicinal drugs, much of the blame for this regrettable situation can be ascribed to the absence of countervailing government action, especially in the areas of advertising and health warnings.9 In these and other markets where there are no effective sanctions against the provision of misinformative advertising in developing countries, what needs to be emphasized is that the accompanying welfare losses tend to be borne most heavily by poor, uneducated consumers in rural areas. For whereas the upper-income groups in those countries can usually gain access to countervailing sources of product information, such as consumer reports and the Internet, this is not true of the vast majority of the population (World Bank, 1998). More generally, as the most recent World Development Report has rightly emphasized, ‘Information problems are often at the core of the difficulties that poor people in developing countries encounter in their daily struggle to survive and to improve their lives’ (World Bank, 1998: 3).

Externalities in global consumption ‘It seems extraordinary,’ write Douglas and Isherwood in their wellknown contribution to the social anthropology of consumption, ‘but it is an outcome of the way that traditional utility theory has been used, that the consuming unit acts as if its decisions to spend

Gain or Lose? 91

on this or that were made in isolation and independently from those of all other consumers’ (Douglas and Isherwood, 1979: 44). In reality, of course, there are any number of ways in which consumption by one individual confers negative effects on other consumers, the so-called negative consumption externalities. From among these numerous possibilities we shall confine our attention to just two areas that have an especially close affinity with the globalization of consumption in developing countries. What concerns us first is an asymmetrical aspect of the increased consumption levels in developing countries that have accompanied -and indeed partly defined-globalization over the past decade or so. While, that is to say, much of the gains from this increased consumption have accrued to relatively high-income households in developing countries, the ensuing consequences for the environment are often borne instead mainly by the lower-income groups in those countries. Toxic emissions from automobiles owned primarily by the higher-income groups, for example, are usually most seriously felt by ‘The children of the poorest urban dwellers … because they tend to live near busy roads’ (Human Development Report, 1998: 70). Similarly, with the growth in affluence and consumption that has accompanied globalization in numerous developing countries has come a change in the composition of waste products ‘from primarily biodegradable organic materials to plastic and other synthetic materials, which take much longer to decompose, if they do at all’ (Human Development Report, 1998: 71). And in this case, as well, it is the poor that tend to suffer most, inasmuch as it is they, rather than higher-income groups, that live near waste-disposal sites.10 Even when – in contrast to the examples that have just been cited – the low-income majority in developing countries do participate in the expansion of consumption to which globalization gives rise, however, there is no certainty that their welfare will increase in direct proportion to the amounts they consume. The problem is that at least some of this consumption (as noted in Chapter 3) is likely to form part of what Hirsch (1976) refers to as ‘positional competition’: that is, ‘competition that is fundamentally for a higher place within some explicit or implicit hierarchy and that thereby yields gains for some only by dint of losses for others’ (Hirsch, 1976: 52). In brief, therefore, such consumption behaviour is collectively self-defeating

92 Consumption, Globalization and Development

and the welfare loss arises because ‘The choice facing the individual in a market or market-type transaction in the positional sector, in a context of material growth, always appears more attractive than it turns out to be after others have exercised their choice’ (Hirsch, 1976: 52). Goods which are chosen on the basis of their ability to compete in this type of competition are referred to as ‘positional goods’, or, more precisely as ‘those things whose value depends relatively strongly on how they compare with things owned by others’ (Frank, 1985: 101). Sporting goods such as Nike shoes and Izod shirts, for example, appear to fall into this category and they also well illustrate the problem of positionality that has just been described. For whereas these goods were once scarce enough to serve the positional goals of their consumers, they are now far too ubiquitous for this purpose (Nike, according to a well-known business magazine ‘has become too common to be cool’11) and a welfare loss ensues among those for whom such ubiquity comes as a surprise. Globalization tends to intensify positional consumption in developing countries because it is not merely a process that entails the import by those countries from the industrialized world. Rather, globalization is a much broader phenomenon that also encompasses, as we have seen, the immensely rapid spread of the massmedia, imported television programming and advertising by multinationals in developing countries, factors, which, together, help to ensure that consumption is undertaken not simply as an end in itself, but also, or alternatively, as a means of promoting one’s position in society. On the one hand, that is to say, the transfer of Western values, including positional values, tends to be fostered by the growth of predominantly American programming. 12 This general process, in turn, makes consumers increasingly susceptible to the advertising by multinationals of specific products that are designed to satisfy the values thus transferred (that is, to use Western status-intensive goods as a means of ‘getting ahead’ of the rest of society). In this way, the positional consumption behaviour that Hirsch described only in relation to the developed countries, is being rapidly transferred to the developing countries as well. And while it may not be immediately apparent, there is also some reason to think that the effects of this phenomenon are felt as much, if not more, by the lower-income groups in those latter countries. As Frank has formulated it, the reason is that:

Gain or Lose? 93

First, just as the poorly fed person derives more satisfaction than others from additional food, so too will the low-ranked person derive greater satisfaction from an advance in the rankings. The second reason is that any given positional expenditure will generally produce a larger change in rank for low-income than for high-income persons. This happens because people are bunched together more closely nearer the bottom of the economic ladder. For a low-income person, there are many people with consumption levels only slightly higher than his own, so even a small increase in positional expenditure will appreciably increase his rank. But in the upper reaches of the income distribution, where incomes are more spread out, the same expenditure will not affect rankings by much. A hundred dollars more spent on clothing, a hundred dollars less saved – all such expenditure shifts will produce larger advances in the ranking for persons near the bottom of the hierarchy than those near the top. (Frank, 1985a: 155–6). It may be at least partly for these reasons, that observers of consumption behaviour among the poorest groups in developing countries are prone to comment on how members of such groups often seem prepared to reduce their spending on food and other essentials, in order to purchase modern, status-intensive products.13

Globalization, the international demonstration effect and preference change It has long been recognized that the preference orderings of consumers in developed and developing countries are interdependent rather than independent of one another. More than forty years ago, for example, Nurkse (1957) advanced the notion of an ‘international demonstration effect’ of consumption in the developed countries on consumption behaviour in the developing countries. What he meant by this, (as was also noted in Chapter 3) was that ‘When people come into contact with superior goods or superior patterns of consumption, with new articles or new ways of meeting old wants, they are apt to feel after a while a certain restlessness and dissatisfaction. Their knowledge is extended, their imagination stimulated; new desires are aroused’ (Nurkse, 1957: 112).

94 Consumption, Globalization and Development

Globalization intensifies this effect not only because the massmedia have spread so rapidly in developing countries over the past decade (see Table 2.1 above on the growth of television receivers), but also because of the dominance of those channels of communication with programming imported from abroad (this seems to be especially true of the growth of satellite and cable television channels, which are dominated by American imports).14 Recent changes in consumption behaviour within the developing countries themselves have also helped to intensity the effects of globalization on the international demonstration effect. For not only do imported products now enjoy the increased exposure that results from the aforementioned growth in the mass-media and advertising expenditures, but they also benefit from the shift in local purchasing behaviour away from corner stores and markets towards supermarkets and shopping malls which ‘are places of conspicuous consumption, and have become entertainment centres and meeting places’ (Human Development Report, 1998: 62). To the extent then that the international demonstration effect has been intensified by the factors that have just been described, one would expect a corresponding shift in the preferences of developing country consumers in favour of modern, imported products. But this, in turn, creates a major theoretical problem for traditional consumption theory and welfare economics, which assume that preferences remain constant over time. In particular, the problem is that preference change gives rise to intersecting indifference curves and there is consequently no way of judging whether to use the old or new preferences as the relevant frame of reference for welfare comparisons. By somewhat extending the ordinal utility measure on which consumption theory and welfare economics are based, however, it is possible (as shown in Chapter 2) to gauge the welfare effect of preference change by using a technique proposed by Harsanyi (1954). By using a ‘quasi-cardinal’, rather than the usual ordinal utility measure, he shows that ‘the consumer enjoys his new taste’ if the price of the new good (in our terms the good which is the subject of the international demonstration effect), ‘permits him to gratify his new taste to a sufficient extent, but, dislikes it in the opposite case’ (Harsanyi, 1954: 209). The same logic prevails in relation to income, namely, that a change in a consumer’s indifference

Gain or Lose? 95

map ‘would … benefit him if his income is high, and would harm him if his income is low, and would express a shift in his taste away from simpler and less expensive commodities towards more refined and more expensive ones’ (Harsanyi, 1954: 213). In applying this criterion to developing countries, it is necessary to recognize in the first place, that the products to which the consumer is increasingly exposed as a result of globalization, emanate overwhelmingly from the developed countries and tend to embody characteristics that are appropriate to the income levels (and other features) of those countries. As such, the products in question are likely to be accessible to high, rather than low-income consumers in the Third World, although these goods are known to and highly desired by many of the individuals that fall into the latter group and who, consequently, will tend to experience frustration and disappointment, rather than pleasure and satisfaction, from the globalization of consumption and the intensification of the international demonstration effect. This problem, we should emphasize, is likely to be most acute in countries which are experiencing a rapid spread of contacts with and information about imported goods, but where, on account of an unequal distribution of income, the income growth of the poorest deciles is lagging behind. In Brazil, for example, where income inequality is especially pronounced, it appears from survey data that: The lower the family income, the more family members tend to watch television and listen to AM radio. These audiences have relatively little consumer power. It appears that television viewing is likely to stimulate the desire to consume, but consumption does not materialize if one does not have the means to purchase the advertised goods. When preference for manufactured items was compared to television viewing, the correlation was positive and reached statistical significance. One would expect, then, that poor viewers would buy the advertised products if they had the means to do so. When actual behaviour was measured, the data did not show the expected relationship between exposure to television and consumption, but when preference was tested, the association was positive. These results indicate a discrepancy between behaviour and attitude. Between the desire to own a product and actual consumption, there is the

96 Consumption, Globalization and Development

strong intervening variable of income. Therefore, in this case, attitude is translated into consumer behaviour only among those who have higher socioeconomic status. Unfortunately, heavy viewers in peripheral contexts are among those who have the highest desire to consume and the least possibility of experiencing their wishes. (Oliveira, 1991: 208, emphasis added)

Unknown preferences and changing consumption patterns Utility maximization requires that people be accurately informed about their own preferences and although this assumption is part of traditional consumption theory, its validity is far from being universally accepted. Hirschman (1982), for example, envisages an alternative world ‘in which men think they want one thing and then upon getting it, find out to their dismay that they don’t want it nearly as much as they thought or don’t want it at all and that something else, of which they were hardly aware, is what they really want’ (Hirschman, 1982: 21, emphasis in original). Underlying the disappointment-ridden world that Hirschman describes is Scitovsky’s well-known distinction between ‘comfort’ (the state of optimal arousal that is produced by consumption) and ‘pleasure’ (the process by which that optimal level is reached). This distinction is important for Hirschman’s analysis, because goods can then be classified according to the extent that they combine these two characteristics. On the one hand, there are consumer durables such as refrigerators and air-conditioning units that are highly comfort-intensive since ‘pleasure’ is experienced only initially when the good is first put to use. Thereafter, the consumer only derives ‘comfort’ because the process of moving from discomfort to comfort no longer occurs. At the other extreme, however, are the ‘truly non-durable goods’ such as food and fuel whose ability to provide pleasure arises precisely from their disappearance in the act of consumption itself. Hirschman argues that as countries make the transition from the one product category to the other, consumers are very likely to be disappointed because this transition entails a ‘drastic’ and ‘unforeseen’ change in the ratio of comfort and pleasure for the consumer, who has long been accustomed to the pleasure-intensive mode of subsistence living.

Gain or Lose? 97

As Hirschman puts it: The modern, mass-produced durable consumer good subverts the way in which consumers extracted pleasure from their purchases in the predurable era. Along with their extraordinarily useful services, these goods are likely to generate a great deal of vaguely felt disappointment, particularly at the time of their first massive diffusion in society. (Hirschman, 1982: 38) This is a stage, we should emphasize, which closely resembles the recent situation in many developing countries (such as India) where the consumption of durable goods has grown particularly rapidly. On the other hand, highly advertised global products such as pharmaceuticals and medicinal drugs, which, from the previous standpoint of imperfect information were regarded as being especially susceptible to welfare losses, fall instead into Hirschman’s category of ‘truly non-durable goods’ and as such are least subject to disappointment from this particular source. Although it may seem to be, this analysis is not in fact closely related to the positional process described above, since as Hirschman himself points out, there is a rather fundamental distinction between them with regard to the manner in which disappointment is thought to arise. For, whereas the consumer of positional goods ‘fails to get what he thought would be within his reach’, disappointments in Hirschman’s scheme, ‘are those that come from experience with goods that are available and affordable, but simply do not yield the sort of satisfactions which were expected from them’ (1982: 61).

From global products to actual functionings ‘Market theories’ writes Lane ‘identify the point of purchase as the moment when preferences are “revealed” and when utilities are achieved, but it is not the purchase of the good that yields utilities so much as the more private act of consumption’ (Lane, 1991: 474). As shown in Chapter 4, Sen (1985) has made perhaps the most of this distinction by examining the relationship between goods and functionings. He describes a ‘functioning’ as ‘an achievement of a person: what he or she manages to do or be. It reflects, as it were, a

98 Consumption, Globalization and Development

part of the ‘state’ of that person. It has to be distinguished from the commodities which are used to achieve those functionings. For example, bicycling has to be distinguished from possessing a bike’ (Sen, 1985: 10). As Sen describes it, the relationship between products and functionings is an entirely general one that applies to all goods and all consumers (though of course, the same product may give rise to entirely different effects on the functionings of alternative consumers). Our more specific concern here however is with the global products that are increasingly and rapidly being introduced into developing countries and the extent to which such products contribute to individual functionings in those countries. The answer to this question, we suggest, often depends on the degree to which the situation in the importing country approximates the technological system of the country in which the product originates. The reason is that innovations in general and product innovations in particular, are designed to fit in with a well-defined set of incomes, skills, infrastructure, complementary goods, climatic conditions and so on (that is, all the elements that jointly define a technological system).15 There are broadly two possibilities. In the first case, where the technological system in a developing country approximates that of the developed countries (from which most product innovations originate), the ensuing relationships between commodities and consumer functionings may not differ too much between the two types of countries. On the other hand, however, when certain developed country products are introduced into entirely different (or, traditional as opposed to modern) technological systems, the effects on functionings may be very different and often more negative. This distinction is perhaps most apparent, for example, in the case of infant formula, which, among a narrow group of relatively affluent consumers in developing countries, has an impact on infant health and nutritional functionings that is very similar to the typical case in the developed countries. Among the majority of poor consumers in the former countries, however, the lack of complementary inputs (such as clean water and refrigerators) often tends to engender a negative rather than a positive impact of formula on the health status of the infant (as manifest, for example, in severe diarrhoea and subsequent bacterial infections).

Gain or Lose? 99

A similar story can be told for modern medicinal drugs in developing countries, which, as shown in Chapter 4, often yield dramatically different effects on individual health functionings, depending upon whether the consumer has access to qualified medical practitioners, whether she can afford an entire course of treatment and whether she is capable of reading and complying with the recommended dosage regimes. Generally speaking, it appears that modern medicinal drugs are used least efficiently by low-income and less educated consumers in developing countries, who, to this extent, derive less of a contribution to their health functionings from these goods than consumers who form part of modern technological systems.16

Conclusions As its main thesis this chapter embodies the view that traditional consumption theory is inadequate as a means of assessing the welfare effects of globalization on consumers in developing countries, mainly because such theory excludes, by assumption, the unexpected outcomes – including feelings of disappointment and frustration – which inhere in the processes of globalization as we have described them. Following a systematic critique of the assumptions of traditional theory, our conclusion is that there are a variety of mechanisms through which some consumers in developing countries will tend to gain from globalization, while others will suffer instead from disappointment and frustration. These various mechanisms and the way in which they relate to the assumptions of traditional consumption theory are summarized in Table 5.2. In many, though not all the mechanisms summarized in Table 5.2, it is mainly the lower- income groups in developing countries that appear most vulnerable to the disappointments and frustrations that the globalization of consumption entrains. In some cases, this is the result of consuming the goods in question, which do not yield the satisfaction that was expected of them, while in other cases it is due to an inability to consume such goods and in still other cases it is due to the effects on one’s own utility of consumption by others.

Globalization, frustration and consumer disappointment

Area of consumption

Traditional assumption(s) rejected

Type of product

Consumers most likely to experience frustration and disappointment

Globalization and information imperfections

Complete information regarding product characteristics embodied in imported, highly-advertised products

Pharmaceuticals, cigarettes, cosmetics, medicinal drugs

Those exposed to product misinformation and enjoying little or no access to countervailing sources of information

Externalities in global consumption

Individual utility depends only on own consumption

Pollution-intensive goods from developed countries; positional (or status-intensive) products from those countries

Low-income consumers exposed to pollution caused by others; low-income consumers with most to gain from positional consumption

Globalization, the international demonstration effect and preference change

Preferences constant

Products advertised on mass-media

Those unable to indulge changed preferences in favour of high-income imported goods

Unknown preferences and changing consumption patterns

Preferences constant and known to the consumer

Durable (comfort-intensive) goods

Those in transition from subsistence to durable type consumer goods

From global products to actual individual functionings

Consumption is revealed at the point of purchase

Goods subject to wide variation in the relationship between commodities and functionings (e.g. modern drugs)

Those who belong to tradional consumption systems, lacking the necessary resources for a favourable relationship between modern goods and individual functionings

100

Table 5.2

Gain or Lose? 101

Notes 1. Parts of this article draw heavily on James (1993), which contains a critique, from various perspectives, of the relevance of traditional consumption theory in modernizing societies. That work, however, makes no reference to globalization. 2. Sen (1985: 9). 3. See the discussion on this point in Lane (1991). 4. This point is made with empirical support in the Indian context by Sachdeva (1988). 5. Human Development Report, 1998, p. 63. 6. Ibid. 7. See Barker (1997). 8. This was thoroughly documented by Silverman, Lee and Lydecker (1982). For more recent evidence see Chetley (1990). 9. See Shepherd (1985). 10. Human Development Report, 1998. 11. Fortune, March 30, 1998, p. 12. 12. For empirical evidence see Varis (1984) and Barker (1997). 13. See for example Wells (1977), Ger and Belk (1996). 14. ‘American media groups’, according to The Economist, Nov. 29, 1997, p. 91, ‘were … used to producing for a competitive market; they had the world’s most successful film industry; and they had huge libraries of cartoons, elderly soap operas, classic films and other programmes ripe for reuse’. 15. For a recent discussion on the notion of a technological system see James and Khan (1998). 16. According to a recent article, many slum dwellers in New Delhi, for example, are too poor to consult qualified medical practitioners. Instead ‘they unquestioningly accept all manner of pills and shots from young men who have never gone to college’ (International Herald Tribune, Sept. 29, 1998, p. 2). See the discussion in Chapter 4.

References C. Barker (1997). Global Television, Blackwell. A. Chetley (1990). A Healthy Business: World Health and the Pharmaceutical Industry, Zed Books. R. Cyert and M. de Groot (1975). ‘Adaptive Utility’, in R. Day and T. Groves (eds), Adaptive Economic Models, Academic Press. M. Douglas and B. Isherwood (1979). The World of Goods, Allen Lane. R. Frank (1985). ‘The Demand for Unobservable and Other Nonpositional Goods’, American Economic Review, March. R. Frank (1985a). Choosing the Right Pond: Human Behavior and the Quest for Status, Oxford University Press.

102 Consumption, Globalization and Development

G. Ger and R. Belk (1996). ‘I’d like to Buy the World a Coke: Consumptionscapes of the “Less Affluent World”’, Journal of Consumer Policy, 19. J. Harsanyi (1954). ‘Welfare Economics of Variable Tastes’, Review of Economic Studies, 21 (3). J. Harsanyi (1992). ‘Utilities, Preferences and Substantive Goods’, WIDER, Working Paper No. 101. F. Hirsch (1976). The Social Limits to Growth,, Harvard University Press. A. Hirschman (1982). Shifting Involvements, Princeton University Press. J. James (1993). Consumption and Development, Macmillan. J. James and H. Khan (1998). Technological Systems and Development, Macmillan. R. Lane (1991). The Market Experience, Cambridge University Press. R. Nurkse (1957). Problems of Capital Formation in Underdeveloped Countries, Blackwell. O. Oliveira (1991). ‘Mass-media, Culture and Communication in Brazil: the Heritage of Dependency’, in G. Sussman and J. Lent (eds), Transnational Communications. Sage. S. Sachdeva (1988). ‘Advertising in India: Some Characteristics and Trends’, Working Paper, Indian Institute of Public Administration, New Delhi. A. Sen (1985). Commodities and Capabilities, Elsevier. P. Shepherd (1985). ‘Transnational Corporations and the International Cigarette Industry’, in R. Newfarmer (ed.), Profits, Progress and Poverty, University of Notre Dame Press. M. Silverman, P. Lee and M. Lydecker (1982). Prescriptions for Death: The Drugging of the Third World, University of California Press. T. Varis (1984). ‘The International Flow of Television Programs’, Journal of Communication, 34. J. Wells (1977). ‘The Diffusion of Durables in Brazil and its Implications for Recent Controversies Concerning Brazilian Development’, Cambridge Journal of Economics, 1 (3). World Bank (1998). World Development Report, Oxford University Press.

Part II Countervailing Influences over Patterns of Global Consumption

This page intentionally left blank

6 Can Appropriate Products Capture Mass Markets in a Globalizing World? a Case Study from India

Introduction Historically the appropriate technology movement has been predisposed to favour non-commercial methods of designing and disseminating products and technologies specifically for the benefit of the low-income majority in developing countries. This particular bias can be understood, according to Carr (1985: 370), with reference to the fact that ‘the AT movement grew out of a concern for people with little or no cash, who must be helped to gain access to products and services in ways other than through the market place.’ In general, however, this non-commercial approach has enjoyed only a limited degree of success over the past two decades or so. For while one can readily point to examples in various developing countries where appropriate products and processes have indeed been successfully diffused at the local or village level, these successes have usually not been replicated at the macro or national level.1 The purpose of this chapter, by contrast, is to view the issue from precisely the standpoint that the AT movement for the most part chose to eschew. In particular, we assume that widespread dissemination of appropriate products and processes is only possible (in the long run at least) on the basis of inter-firm competition at ruling market prices. Whether or not such products and processes are able to capture a mass market then depends on whether producers are willing and able to supply them at a price that the majority of lowincome households finds attractive; it depends, that is to say, on whether there is an intersection of the relevant market supply and 105

106 Consumption, Globalization and Development

demand curves. This, in its turn, is dependent on the parameters that underlie, and hence determine the position of, the two curves in relation to one another. In the following section we argue that the values of these parameters are often such as to make it difficult, if not entirely impossible, for appropriate products and processes to capture mass markets in many developing countries. In the section thereafter, however, we describe how one firm in India was able to overcome these difficulties and capture the dominant share (amounting at present to some 70 million households) of the national market for laundry detergents, on the basis of highly labour-intensive methods of production and a product designed specifically to meet the needs of the low-income majority in that country. The final section addresses itself to the question of whether this important case is the unique result of a fortuitous combination of factors, or whether it is a model that can be replicated elsewhere in India and other countries belonging to the Third World.

Two types of markets in developing countries Reflecting the marked degree of economic and social dualism between modern and traditional sectors that exists in most developing countries, one can usefully distinguish between two major types of product markets in those countries. 2 The first market consists of a relatively small number of high-income consumers, whereas the second comprises a much larger number of consumers with relatively low incomes.3 Assume, for the sake of argument, that in the former type of market there are 100 economic units, each of which has an income of $10,000. Assume, further, that in the latter type of market there are 1,000 economic units, each of which has an income of $1,000. Thus, although the size of the market is the same in both cases, the type of market, as represented by the position of the market demand curve, will be very different. This distinction can be seen in Figure 6.1, where DA represents the market demand curve derived from the small number of highincome units and DB represents the market demand curve derived from the larger number of low-income units. DA may thus be said to denote a ‘high-income’ market in contrast to DB which represents a ‘low-income’ market.

$ per unit

A Case Study from India 107

DA (1)

DA (100)

M LRAC

DB (1000)

DB (1)

N

LRAC*

J 0

Quantity X

K

Figure 6.1 Two types of markets in developing countries Notes: DA = Demand schedule of market A (high income market) DB = Demand schedule of market B (low income market) (Numbers indicate consuming units in thousands) LRAC = Long-run average cost curve when product X conforms strictly with international standards of quality (modern technology employed). LRAC* = Long-run average cost curve when the quality of product X is below international standards and more labour-intensive technology employed. Source: Based on Amsden (1977)

In most developing countries the ‘high-income’ market is served by imports or multinational corporations, with products similar, or in some cases identical to, those consumed in the developed countries. The cost curve associated with this type of product is represented (we assume) in Figure 6.1 by LRAC and equilibrium in this market is therefore achieved at price OM and quantity OJ. At this price, however, the high-income product is totally inaccessible to the lowincome consumers, whose combined demand curve, we recall, is given by DB. The inaccessibility shown here is due partly to the fact that the’high-income’ product contains a relatively high proportion of inessential or luxury characteristics and partly because this type of product is usually produced on the basis of inefficient (high cost) methods of production (methods, that is to say, that make too much use of capital and too little use of the relatively abundant supply of labour in most developing countries).

108 Consumption, Globalization and Development

Equilibrium in the low-income market could in principle be achieved by changes in both these determinants of the LRAC curve. In particular, a product could be designed specifically for low-income consumers and such a product could, in addition, be produced by relatively labour-intensive methods of production. In terms of Figure 6.1, the cost curve could be lowered by these means to say, LRAC*, where equilibrium is reached at price ON and quantity OK. In practice, however, this rarely occurs and the reason is that the parameters underlying the supply and demand curves are often such as to move the curves away from, rather than towards one another. Let us therefore now examine these underlying parameters in the low-income market, beginning with those on the demand side.

The underlying parameters a) Demand Given the supply curve LRAC *, the existence of an equilibrium in the low-income market depends on the position of the market demand curve, which, in its turn, is derived from the individual demand curves by adding the amount demanded by all consumers at each particular price. Of the numerous parameters underlying the individual demand curves, income and preferences are perhaps the most important and in many of the poorer developing countries these parameters serve to frustrate, rather than to promote, the achievement of equilibrium in the low-income market. Many such countries, for example, suffer not only from a low GDP but also a skewed distribution of the total available income. 4 And in combination, these factors often imply that the effective demand for many non-essential commodities on the part of the majority of the population is negligible. Consequently, equilibrium in the low income market becomes unattainable even when commodities designed specifically for that market are produced at the lowest possible cost (for there would then be no point of intersection between LRAC* and the market demand curve DB). Even at the higher incomes prevailing in other developing countries, moreover, an adequate demand for a relatively inexpensive product embodying a high proportion of essential characteristics cannot be taken for granted. For what then also come into play are

A Case Study from India 109

the factors that determine individual preferences and more specifically, the factors that often tend to shift the demand curve for such a product to the left. What is probably most crucial in this regard is the prior existence of similar types of products serving the high-income market. For, although these products may initially be targeted mainly at the affluent minority of households in a developing country, there are nonetheless a number of mechanisms through which the preferences of low-income consumers are also likely to be influenced in the same direction. One such mechanism is driven by the fact that a growing proportion of this latter segment of the population has access to radio and television, as a result, partly, of the telecommunications revolution and the globalization of the mass-media.5 To this extent, therefore, the low-income population is also becoming increasingly exposed to the intensive advertising and marketing campaigns conducted via these media, on behalf of the products made in developing countries by multinational corporations. And some authors argue, furthermore, that it is low, rather than high-income consumers who tend to be most susceptible to these promotional endeavours. One study of a large Brazilian city, for example, found that ‘The lower the family income, the more family members tend to watch television and listen to AM radio. These audiences have relatively little consumer power. It appears that television viewing is likely to stimulate the desire to consume, but consumption does not materialize if one does not have the means to purchase the advertised goods. When preference for manufactured items was compared to television viewing, the correlation was positive and reached statistical significance’ (Oliviera 1991: 208). Related to, but nevertheless somewhat distinct from, the effect of advertising on individual preferences for high-income goods, are the pressures exerted by demonstration effects of various kinds. Perhaps the best known of these is the use of certain types of Western goods as a means of enhancing one’s position or status in society. Hirsch (1967) refers to this process as positional competition and the goods involved therein as positional goods. Though one may perhaps most readily associate this type of competition based on Western status symbols with relatively high-income consumers in developing countries, it may in fact be even more intense among the poorer groups for the following two reasons.

110 Consumption, Globalization and Development

First, just as the poorly fed person derives more satisfaction than others from additional food, so too will the low-ranked person derive greater satisfaction from an advance in the rankings. The second reason is that any given positional expenditure will generally produce a larger change in rank for low-income than for high-income persons. This happens because people are bunched together more closely nearer the bottom of the economic ladder. For a low-income person, there are many people with consumption levels only slightly higher than his own, so even a small increase in positional expenditure will appreciably increase his rank. (Frank, 1985: 155) To this extent, therefore, positional competition tends to compound the influence of advertising in predisposing low-income consumers in developing countries to favor foreign (and particularly Western) over local brands. And the combination of these factors, in turn, makes it all the more difficult to design and disseminate appropriate products on a large scale among the low-income majority in the Third World. b) Supply In Figure 6.1 costs of production in the high-income market are lowered (to LRAC*) by designing a simpler product that is produced on the basis of a relatively labour-intensive technology. By these means the low-income market is brought into equilibrium at price ON and quantity OK. As on the demand side, however, in practice there are a number of factors that frustrate, rather than promote, the achievement of equilibrium in this market. One such factor is the cost of designing (and bringing to the market) a product to meet the needs of low-income consumers in developing countries. Though there may indeed be cases where the design of such a product can be inexpensively copied from, say, an older and less sophisticated version of a brand that was originally designed for developed country markets, in other cases prospective suppliers will have to incur innovation and development costs of one kind or another. And where these costs are high in relation to the prospective benefits, product adaptations designed specifically for the low-income market are unlikely to be forthcoming. In this event, moreover, matters are made still worse by the fact that it will

A Case Study from India 111

then be difficult to effect a cost reduction by a change in the method of production; that is, by the second of the two means of effecting cost reductions to achieve equilibrium in the low-income market. The reason is that in most cases the choice of technology is known to be highly dependent on the prior choice of product. ‘In many cases’ that is to say, ‘adaptation of the processes requires product adaptation, to make fine tolerances less critical and frequently to lower the general quality of the product’ (White, 1978: 51).6 Even without the difficulties described in the previous paragraph, however, one has still to confront the difficult problem of distributing a low-income product to the large number of consumers for whom it has specifically been designed. For whereas the product sold to the high-income market has usually to be distributed to only a relatively small number of consumers in or close to large cities, low-income products have to reach a much larger and more dispersed group of households, many of whom may live in rural areas with limited access to transport and communications facilities. The final difficulty on the supply side relates specifically to the use of labour-intensive methods of production as a means of shifting the long-run average cost curve (LRAC) in a downward direction towards equilibrium with the low-income market demand curve DB. In particular, if those methods of production are to exert so powerful a positive influence on costs of production, the price of labour needs to be low in relation to the price of capital (that is, the prices of the factors of production should reflect their comparative scarcities). In many developing countries, however, it is still the case that factor prices are distorted in the sense that the price of labour is too high and the price of capital too low in relation to their respective shadow (or scarcity) values. Consequently, the search for and the use of labourintensive technologies is less intensive than it could be.

An Indian case study The purpose of this section of the chapter is to show how one Indian firm managed to overcome the difficulties described above and in so doing was able to capture a vast market for laundry detergents, a market that for many years had remained entirely unexploited. First, however, it is necessary to set this particular

112 Consumption, Globalization and Development

experience in the historical context of the Indian detergent market as a whole. The market for laundry detergents in India By the early 1970s the market for laundry detergents in India consisted of just two types of products, which could hardly have been more different from one another. On the one hand, there were a limited number of high-income brands, the most important of which (‘Surf’), was manufactured in India by the Hindustan Lever company. It was in fact ‘thanks to Surf’ that ‘all the other detergent powders, were ‘sophisticated’ in the sense that they were in heavy cardboard packing, very well printed and with powders which had a variety of ingredients to allow for optical whitening, high detergency, kindness to hands, etc.’ (Rao, 1989: 127). All these highlypriced products, accordingly, were aimed solely at upper-income level consumers in India. At the other extreme, was a much larger number of small-scale producers selling highly inexpensive products solely to regional markets. These products, however, ‘were poorly presented, of low quality and generally unbranded’ (Rao, 1989: 128). In terms of Figure 6.2, what was to be found in India during the early 1970s, therefore, was groups of detergents in only two of the four quadrants: that is, in the upper right and lower left segments of the diagram. What was clearly needed was a product positioned in the upper-left quadrant, namely, a detergent that could be sold at a low cost not on account of its low cleansing quality, but rather because it contained only the essential (functional) characteristics required for the purpose at hand. Just such a product, with the brandname ‘Nirma’, was introduced in the mid 1970s on an initially minute scale by a single entrepreneur, who ‘seized on the potential for a cheaper detergent of good quality’ (Far Eastern Economic Review, Jan. 1993: 47). In the ensuing four years annual production rose to 20,000 tonnes and by the end of the 1980s ‘Nirma’ had captured no less than 60 per cent of the entire powder detergent market with annual production equal to some 750,000 tonnes, an amount ten times greater than that produced by its main competitor, Hindustan Lever. With regular sales to some 70 million Indian households, Nirma today has achieved what for most firms in that country is still only a distant goal, namely, ‘of hitting upon a consumer product affordable and attractive enough to

A Case Study from India 113

High cleansing ability ‘Surf’

Low-income characteristics (e.g. Inexpensive packaging)

High-income characteristics (e.g. expensive packaging)

Regional unbranded products Low cleansing ability Figure 6.2

The laundry detergent market prior to introduction of ‘Nirma’

penetrate the mass market of the rural hinterland’ (Far Eastern Economic Review, June 1, 1989: 60, emphasis added). The question that now needs to be addressed is how in the process, ‘Nirma’ was able to surmount the many formidable difficulties described in the previous section. Following the order of that discussion, we shall begin with the difficulties encountered on the demand side of the low-income market. The demand for ‘Nirma’ The first such problem that was described there emanated from the combination of a low GDP and a highly skewed distribution of income. This combination, we suggested, would imply that in some developing countries there is a highly limited demand for nonessential items such as powder laundry detergent (clothes, after all, can be and in some places still are, washed with bars of soap rather than synthetic detergent in powder form). To some extent, of course, as one of the poorer developing countries, India has been

114 Consumption, Globalization and Development

subject to this particular problem. Against this, however, one can attribute the rapid growth in demand for ‘Nirma’ over the period noted above, at least partly, to the fact that many low and middleincome households were included among the beneficiaries of the high per capita income growth that India then enjoyed. ‘Between 1977–78 and 1985–86,’ for example, ‘rural income poverty fell from 51 per cent to 39 per cent; by 1989–90 it had fallen to 34 per cent. Income poverty also fell in urban areas, from 40 per cent to 33 per cent between 1977–78 and 1989–90’ (Human Development Report 1997: 51). More generally, it is fair to say that India has a low overall degree of inequality as compared to most African and Latin American countries (Deininger and Squire, 1996). As was also noted in the previous section, the demand for what we have referred to as appropriate products, is often negatively influenced by the prior existence of heavily promoted substitute brands imported from developed countries or produced locally by multinational corporations from those countries. ‘Nirma’ dealt with this problem in a manner that has been entirely ignored by the AT movement, namely, by an intensive advertising campaign of its own. As we have already noted, this movement has tended to exhibit an ideological bias against the commercial aspects of the adoption and diffusion of appropriate products and processes. Yet, the case of ‘Nirma’ suggests that television advertising on a national scale may be essential if such products and processes are to capture mass, as opposed to purely local markets in developing countries. Consider in this regard the amounts shown in Table 6.1 which represent the top ten advertisers on Indian television for the years 1984–5. In fifth position, ‘Nirma’ is one of the very few locally owned firms in India with advertising expenditures equivalent to some of the largest multinational corporations in India. Advertising on this scale may have favourably influenced the demand for ‘Nirma’ through several distinct mechanisms. In the first place, consumer-survey evidence from India indicates that television advertisements play a significant – and indeed often dominant – role in introducing new products to consumers (Sachdeva, 1988). Thereafter, and secondly, the brand familiarity thus acquired by consumers, often continues to be reinforced by television advertising, thereby raising the probability of repeat purchasing behaviour (Sachdeva, 1988). The final mechanism through which the

A Case Study from India 115

Table 6.1

The top 10 advertisers on television in India, 1984–5

Rank

Company name

1 2 3 4 5 6 7 8 9 10

Hindustan Lever Colgate-Palmolive Hindustan Cocoa Hindustan Machine Tools Nirma Pure Drinks Johnson and Johnson S. Kumar Food Specialities Glaxo

Total

Advertising expenditure (Rs. lakhs) 133.3 120.7 63.3 54.8 53.7 51.2 47.1 42.9 42.7 42.1 6218

Source: Sachdeva (1988)

advertising of ‘Nirma’ may have been influential is particularly relevant in a developing country context, where, typically, there is an existing predisposition to favour foreign over local brands (a predisposition, we should note, which itself arises partly on account of advertising that promotes an affluent life style along Western lines and thereby creates a standard for everyone to emulate). In this type of situation, the sheer fact that the local product is intensively advertised will help to convey the impression that it is worth purchasing. (In the case of ‘Nirma’, this particular mechanism may also have been reinforced by the content of the advertisements ‘which created an up-market image without referring to Nirma’s low price’.) The supply of ‘Nirma’ In a variety of different ways, ‘Nirma’ was able to effect remarkable cost reductions, which, at the end of the 1980s, enabled the product to be sold at less than half the price of an equivalent amount of ‘Surf’, the then leading high-income laundry detergent. Indeed, at that same time ‘Nirma’ was ‘barely more expensive than crude cottage-made soap cakes or loose detergent ingredients (which are sold in neighbourhood bazaars to be hand-mixed by house-wives at home’) (Far Eastern Economic Review, June 1, 1989: 60). One means by which cost reductions of this magnitude were achieved lies in the realm of product design and in particular the

116 Consumption, Globalization and Development

design of a detergent specifically for low rather than high-income consumers. Thus, in contrast to ‘Surf’ and other brands designed to fit in with developed country conditions, ‘Nirma’ does not contain characteristics such as optical whiteners and scents, or ingredients designed to reduce the harshness of the product on hands and fabrics. Packaging, too, is an area in which there are important differences that bear on the costs of producing ‘Nirma’ as opposed to the competing high-income products. Whereas, for example, the latter brands are packaged in heavy cardboard, on which there is also sophisticated and expensive printing, ‘Nirma’ is sold instead in simple plastic bags on which the printing is relatively crude and hence much less costly. In short, therefore, ‘Nirma’ may be said to retain just those characteristics of modern laundry detergents that are necessary for it to function effectively and by eliminating other, less essential characteristics, makes itself accessible to a much wider range of consumers than the highincome products with which it competes so successfully in the low-income market. If major cost reductions were thus effected in the product design, the method chosen to produce ‘Nirma’ further lowered the price at which it could profitably be sold. In particular, this product is mass produced not by a single large factory but rather by means of a large chain or agglomeration of small workshops ‘in which people mixed the ingredients of washing powder by hand, without using even electric power’ (The Economist, Sept. 10, 1988: 84). The effect of so pronounced a degree of labour-intensity on costs of production depends, of course, on the relative prices of capital and labour. And we noted in the previous section that the prospective benefits of a low capital to labour ratio are often muted in developing countries where factor prices are distorted to a greater or lesser extent. In the case of ‘Nirma’, by contrast, wages were kept low by the employment of non-unionized labour and by paying them on a piece-rate basis. And again in contrast to many other developing countries, small-scale production in India tends, in many ways, to be encouraged rather than discriminated against by government policies. In particular, since ‘Nirma’ is produced by an agglomeration of smallscale units, as noted above, each such unit is entitled to a full range of state benefits, such as excise-tax breaks, subsidies, privileged access to water and electricity and so on.7

A Case Study from India 117

Following our discussion in the previous section, what still needs to be discussed under the heading of supply, is the problem of actually reaching a large number of dispersed consumers in conditions where the available infrastructure leaves much to be desired (this, we should emphasize, is yet another problem that has been largely ignored by the appropriate technology movement). For ‘Nirma’ success in this area appears to have been predicated partly upon the establishment of ‘a low-cost distribution system of several hundred agents with exclusive territorial franchises. In a small company like Nirma, the agents had personal access to Patel [the founder of the company] himself’ (Far Eastern Economic Review, Jan. 14, 1993: 47). Through this so-called ‘stockist’ system, ‘Nirma’ is able to reach the thousands of very small, independent dry goods stores throughout the country, on which its mass market crucially depends. The combined implication of the cost reductions that have just been described was that by purchasing ‘Nirma’, consumers could obtain the characteristic ‘cleansing ability’ at about half the price of an equivalent amount of ‘Surf’ detergent powder. Or, put differently, for a given expenditure on laundry detergents, consumers were able to derive twice the amount of ‘cleansing ability’ from the former than the latter, though they would then also receive fewer non-essential characteristics such as scent and sophisticated packaging. These relationships are shown in Figure 6.3, which also depicts the distributional gains associated with the introduction of ‘Nirma’; gains which significant though they may have been for vast numbers of low-income households, are generally not reflected in data sets that pertain to the welfare of this group, such as estimates of the size distribution of income, or the income share of the bottom 40 per cent of households in total income.8 In relation to Figure 6.3, assume that a given amount of expenditure enables the consumer to reach point P on the ray (in characteristics space) representing ‘Surf’, or point Q on the ray representing ‘Nirma’ (where the former brand, as already noted, embodies a relatively high proportion of non-essential characteristics in contrast to the latter which combines the two types of characteristics in just the opposite proportions). By choosing ‘Nirma’ the consumer will therefore receive an amount of ‘cleansing ability’ equal to 0Q and an amount of non-essential characteristics equal to 0R. The choice of ‘Surf’, by contrast, yields only half as much ‘cleansing ability’ (0A as

118 Consumption, Globalization and Development

Non-essential characteristics (e.g. sophisticated packaging)

‘Surf’ i2

I2

I1

i1

P

S

i2

I2

I1 0

‘Nirma’

Q

R

A

i

1

B

Cleansing ability Figure 6.3

A comparison of ‘Surf’ and ‘Nirma’ in characteristics space

opposed to 0B) but substantially more (0S) of the high-income characteristics. Low-income consumers in India appear (from our earlier discussion) to have indifference curves that are relatively steep (that is, the II indifference map) and they will accordingly be better off purchasing ‘Nirma’, while high-income consumers with the opposite (ii) type of indifference map, continue to choose ‘Surf’. It is worth emphasizing how much this outcome deviates from the usual distributional impact of new products on consumers in developing countries. For, typically R and D on new products is conducted in developed countries, for whose (high-income) inhabitants the innovations are designed. And when they are introduced into developing countries, these products almost invariably benefit those with high rather than low incomes (James and Stewart, 1981).

Conclusions For understandable reasons, the appropriate technology movement has tended to favour non-market modes of generating and diffusing

A Case Study from India 119

products and technologies designed specifically to benefit those with low incomes in developing countries. Generally speaking, however, this non-commercial approach to the problem has produced few successes at the macro, as opposed to the local or village level. Partly for this reason, we have raised the question here of whether and under what conditions, the private sector may function more effectively from this point of view; that is, in enabling appropriate products and technologies to capture mass markets in developing countries. We began this task by examining the supply and the demand-side conditions for equilibrium in the low-income or mass market and concluded that the parameters underlying the market supply and demand curves are often such as to frustrate rather than promote equilibrium at this level of analysis. These difficulties notwithstanding, however, one firm in India has been able to capture the dominant share of the laundry detergent market on the basis of highly labour-intensive methods of production and a product designed specifically to meet the needs of the lowincome majority in that country. What makes this case all the more striking, we believe, is that the firm in question had to compete initially and subsequently, with a multinational corporation (Hindustan Lever) whose presence in India was already well established. So striking a success story (from the point of view of appropriate products and technologies) inevitably raises the question of whether and to what extent it can be replicated elsewhere. Within India itself, one can in fact already point to several other industries (such as toothpaste and soft drinks) where small-scale enterprises have also been able to capture a large (if not as spectacular a) share of the low-income market and seemingly, 9 for some of the same reasons that were described above in relation to ‘Nirma’ and the laundry detergent market. In most of these other cases, for example, there was an explicit attempt to design a product whose embodied characteristics were in proportions that were appropriate to low, rather than high-income consumers (as manifest, for example, in the use of the ‘new, cheap, simple and attractively printed packing material now available’).10 Or, again in parallel with the case of ‘Nirma’, ‘there was little initial investment in plant, machinery and buildings. However, there are many times, high levels of marketing investments, usually on advertising and promotion, many times on distribution, and sometimes on sales force’ (Rao, 1989: 127).

120 Consumption, Globalization and Development

In at least one important respect, however, the ‘Nirma’ experience may be more difficult to replicate in developing countries other than India.11 I am referring here to the fact that India appears to be culturally less dependent on the developed market economies than most of those other countries. This distinction is manifest, for example, in the extremely low proportion of imported television programming in India as compared to other developing countries, where programs from the United States in particular, tend to dominate the mass-media (Varis, 1984). It is also significant in this regard to note that the success of some of the afore-mentioned products designed for low-income households in India was based on the precedence of local over foreign cultural traits. In particular, the success of certain locally made toothpaste products appears to have been due in large measure to the recognition that cloves are ‘the accepted remedy all over the sub-continent for problems to do with the gums, teeth and mouth odour’ (Rao, 1989: 128). Accordingly, a toothpaste was introduced ‘whose major claim was that it contained clove oil. … While others tried to copy Colgate or to introduce product concepts picked up from their home markets in the US or elsewhere’ the brand in question (‘Balsara’) ‘instead went “native”. This success has been sustained over the years’ (Rao, 1989: 128). As has already been suggested in Chapter 1, examples such as these need to be understood in relation to the historical influence exerted by the ‘swadeshi movement’ (associated with Gandhi’s appeal on behalf of local as opposed to imported goods) and the Nehruvian brand of socialism in the post-independence period.

Notes 1. This observation is made by Carr (1985) and Stewart (1987). 2. This is, of course, an over-simplification of a more complex reality, which incorporates a much larger variety of different types of markets. 3. This analysis follows Amsden (1977). 4. Many countries in Sub-Saharan Africa, for example, fall into this category. 5. In India, for example, the number of TV sets grew from 400 in 1971 to 6.8 million in 1985 (Pendakur, 1991). See also the World Bank, Global Economic Prospects, 1996. 6. For a full discussion of this relationship see Stewart (1977). 7. See The Economist, Multinational Business, 2, 1989. 8. See James and Stewart (1981) for a discussion of this point.

A Case Study from India 121

9. One of the locally made toothpaste brands, for example, has gained a market share of 20 per cent (see Far Eastern Economic Review, Jan. 14, 1993). 10. Rao (1989: 129). 11. There may of course be other problems as well. Nirma’s strategy of selling a large volume of output on the basis of a low profit margin per unit of output, for example, will be less feasible in a smaller market.

References A. Amsden (1977). ‘The Division of Labour is Limited by the Type of Market: the Case of the Taiwanese Machine Tool Industry’, World Development, 5 (3). M. Carr (1985). The AT Reader, IT Publications. K. Deininger and L. Squire (1996). ‘A New Data Set Measuring Income Inequality’, World Bank Economic Review, 10 (1). R. Frank (1985). Choosing the Right Pond, Oxford University Press. F. Hirsch (1967). Social Limits to Growth, Oxford University Press. J. James and F. Stewart (1981). ‘New Products: A Discussion of the Welfare Effects of the Introduction of New Products in Developing Countries’, Oxford Economic Papers, 33 (1). O. Oliveira (1991). ‘Mass Media, Culture, and Communication in Brazil: the Heritage of Dependency’, in G. Sussman and J. Lent (eds), Transnational Communications, Sage. M. Pendakur (1991). ‘A Political Economy of Television: State, Class, and Corporate Influence in India’, in G. Sussman and J. Lent (eds), Transnational Communications, Sage. S. Rao (1989). ‘Innovation Marketing Strategies’, Economic and Political Weekly, August 26. S. Sachdeva (1988). ‘Advertising in India: Some Characteristics and Trends’, Working Paper, Indian Institute of Public Administration, New Delhi. F. Stewart (1977). Technology and Underdevelopment, Macmillan. F. Stewart (1987). Macro-Economic Policies for Appropriate Technology, Westview. T. Varis (1984). ‘The International Flow of Television Programs’, Journal of Communication, Winter. L. White (1978). ‘The Evidence on Appropriate Factor Proportions for Manufacturing in Less Developed Countries: A Survey’, Economic Development and Cultural Change.

7 Cultural Advantage Reversal: the Case of Telenovelas in Brazil

At several points in the first part of the book, we emphasized how rapid has been the growth of the mass-media in developing countries over the past ten to fifteen years. Such rapidity, we further suggested, was due both to developments in information technology (such as cable and satellite television) and a shift from a public to a commercial mode of television broadcasting. And following the prevailing view in the literature, we concluded that the extra programming capacity thus created would tend to be filled by imported serials and other low-cost programming from the developed countries, especially the United States. Indeed, it has long been axiomatic that the international circulation of television programming occurs from north to south and from west to east: that is to say, US and, to a lesser degree, European program producers maintain their hegemony over the global television market by selling their programs at a low cost to foreign broadcasters, particularly to broadcasters in the ‘developing’ nations of Latin America, Asia, and Africa. They can do so because their production costs have been recovered in the far larger and richer domestic markets. Thus prices for foreign sales can be kept at a level low enough to discourage domestic drama production elsewhere but still high enough to be profitable to the producer. (Allen, 1995: 13) Yet, while this axiom has accurately reflected patters of globalization in some parts of the Third World, it has turned out to be a 122

Cultural Advantage Reversal 123

highly inaccurate depiction of what has occurred in other parts, especially Latin America. Indeed, in some countries of that region quite the opposite tendency has been observed, namely, for locally produced television serials (known as telenovelas) and other entertainment programs, to displace competing imports from the developed countries, which imports now represent only a minor share of the total local market for these products. As such, this experience is no less exceptional than the Indian case described in the previous chapter, nor is it any less deserving of our attention as a countervailing force against developed country influence over patterns of globalization in developing countries. Rather than seeking to cover the experience of all relevant countries in Latin America, however, we shall focus our attention on Brazil, whose telenovelas have received particular attention in the literature.1

Extreme degrees of dependence on Western television programming in the Third World To a probably greater degree than any other region, Sub-Saharan Africa exemplifies the dependence of developing countries on television programming from the developed countries (and as such this region represents the clearest confirmation of the ‘cultural dependency’ view noted above). According to one estimate, for example, the 47 television organisations in black Africa ‘broadcast a total of about 90,000 hours of programs per year, more than half of which consists of imported material. Imported programs, especially shows from Western Europe and North America make up between 60 per cent and 90 per cent of all programs broadcast by most African television stations … A study of the flow of television programs … indicates that about 65 per cent of imported shows are entertainment oriented’ (Boafo, 1991: 115). Nearer the opposite extreme, however, are countries such as Brazil and Mexico, which possess the technological capabilities needed to supply the domestic television audience and which rely to a correspondingly lesser extent on television (and other mass-media) programs from the developed countries. The contrast between these two extreme groups of countries – Africa on the one hand and Latin America on the other – can perhaps best be made by considering the material contained in Table 7.1.

124 Consumption, Globalization and Development

Table 7.1

A comparison of television ratings

Ranking

South Africa (July 6–26, 1998)

Brazil (June 29–July 5)

1 2 3 4 5 6 7 8 9 10

Generations (soap opera) The Bold and The Beautiful (soap opera) New York Undercover (Police series) News (multi-Language news) Vicious Circle (Drama series) Note for Note (Afrikaans musical quiz show) Team Knight Rider (Action series) Mad About You (Sitcom) Pensacola: Wings of Gold (Action series) Cosby Show (Sitcom)

Journal Nacional (news) Braxil 500 Anos (variety show) Torre de Babel (soap opera) Corpo Dourado (soap opera) Praca TV 2a Edicao (news) Era Uma Vez (soap opera) Fantastico (variety show) Globo Reporter (news) Casseta e Planeta (variety show) Sai de Baixo (drama series)

Source: The Economist, Aug. 15, 1998, p. 74

In particular, by comparing the television ratings in South Africa and Brazil over a comparable recent period, it is clear that whereas South Africans are most attracted to American soap operas, in Brazil it is locally made telenovelas that dominate the ten most viewed programs on television (in company with other non-imported news and variety shows). Clearly, therefore, ‘The widely held belief that American shows drive out all else is not borne out’ (The Economist, August 15, 1998, p. 74). Let us consider next, then, how long it took for Brazilian telenovelas to achieve the degree of popularity that they now enjoy 2 and how in the process, they were able to displace imported programs of the same genre.

The growth of Telenovelas in Brazil Telenovelas were not always more popular than their American counterparts in Brazil. On the contrary data for the early 1960s show that the audience hours index (the product of a program’s length multiplied by its audience) heavily favoured the imported programs. In particular, the index for telenovelas at that time was only 2 per cent as against 25 per cent for television series imported from the United States. ‘But the situation changed in 1977: as telenovelas leaped to 22 per cent, US series declined to 17 per cent … In 1984, telenovelas reached 24 per cent of audience hours, while US

Cultural Advantage Reversal 125

series and feature films shrunk to a mere 8 per cent’ (Oliveira, 1993: 117). Nor, one should emphasize, has the popularity of the telenovela been confined just to the local market (large though this now is). For even by 1975 the major producer of this product in Brazil (TV Globo) had begun exporting to Europe and within the subsequent decade ‘its annual profits on foreign telenovela sales to nearly 100 countries had risen to US $20 million. Export revenue increased more than fivefold between 1982 and 1987’ (Allen, 1995: 2). Still later, by the mid 1990s TV Globo had emerged as ‘one of the most successful exporters of non-English products to the European markets. TV Globo Telenovelas are exported to Russia, South Korea, Portugal, China, Argentina and many other countries’ (Neto, 1997: 268–9). As such, Brazil derived gains from the expansion of the mass-media in the 1980s and 1990s (and the related increase in demand for television serials) that are often thought to accrue overwhelmingly in favour of the United States. As in the previous chapter, there appear to be factors on both the supply and demand sides of the market that help to explain how the locally made product was able to capture a mass market, partly by displacement of an already entrenched foreign competitor. It is to a consideration of these factors that we now turn, beginning again as in the previous chapter, on the demand side of the issue.

Demand-side factors The emergence of the telenovela on a large-scale in Brazil, as well as the form that it took, owes much to the economic growth and expansion of television ownership which saw the domestic audience change from a small, elite group to one that increasingly incorporated the working and middle classes as well. 3 Indeed, by 1971, no less than 70 per cent of the audience in Rio de Janeiro was said to be drawn from the ‘lower socio-economic classes’ (Straubhaar, 1982: 142). TV Globo saw locally produced telenovelas as the best way of capturing the emergent mass market (with its concomitant potential for advertising revenue), partly, it appears, because ‘programming had to respond to the popular culture, the symbols and interests of the new audience classes. These tended to be folklore and entertainment from the interior of the country, not the high culture favored by the elites of Rio and São Paulo’ (Straubhaar, 1982: 142). In thus

126 Consumption, Globalization and Development

seeking to create a product specifically for the mass audience, the Brazilian TV program was, in effect, behaving in much the same way as the Indian producer of Nirma detergent in the previous chapter, who, as we saw, was also concerned with mass rather than elite preferences.4 That telenovelas, rather than foreign serial imports were able to capture the major share of the mass market that had emerged, also has an affinity with the Indian case inasmuch as the locally made product more closely accorded with the specific characteristics of that market. In particular, ‘television viewers discovered something in the [local] soap operas which North American serials, despite their visual attractiveness and narrative skill, did not offer: a complicity with certain markers of cultural identity’ 5 (Barbero, 1995: 282). Telenovelas, after all, are ‘truly Latin American. In the Brazilian case, they are locally made and are written by local authors about Brazilian themes’ (Vink, 1988: 11). And the quality of this writing, in turn, was obviously enhanced by the fact that highly esteemed theatrical actors, film directors and prestigious left-wing writers were incorporated into the production of soap operas. However, in other countries, television in general and the soap opera in particular were rejected by managers and workers in legitimate culture as the most dangerous of traps and the most degrading of professional fields. In all Latin American countries, however, the crisis in cinema and politics has driven many artists, writers, and actors into television and the soap opera. In spite of the commercial controls on the genre, these artists are introducing themes and styles into the soap opera which deal with dimensions of national life and culture. (Barbero, 1995: 282) What, finally, bears emphasizing in this regard is that, in contrast to serials from the United States, which are generally targeted at a daytime, female audience, telenovelas, are ‘prime-time entertainment’ (Lopez 1995: 258) designed for a general audience. To this extent, therefore, the latter enjoys wider appeal during those particular viewing hours. It is true that the substitution of Brazilian for imported television programming was a state priority in the early 1970s, when,

Cultural Advantage Reversal 127

TV Globo also came under strong government pressure to reduce imported programs … both to eliminate “alien” cultural values and to eliminate an unnecessary import and foreign exchange expense … Nevertheless, TV Globo could not have responded if advertisers had not been willing to pay more for the cost of producing telenovelas and other Brazilian programs. The reason advertisers were willing to do so was that these programs were more popular than imports and drew larger audiences. (Straubhaar, 1988: 63)

Supply-side factors With regard to the technological capabilities that are needed to produce local telenovelas of a high quality, Brazil was fortunate, as noted above, in being able to draw on a wide range of artistic talent; talent that in many other countries is unavailable for the production of television serials. At the same time, however, it needs to be recognized that many of the relevant capabilities were not acquired fortuitously, but rather as a result of the continuous, deliberate and often expensive technological efforts made by TV Globo. As such, the successful production of local cultural goods does not seem dissimilar from the complex process of acquiring technological capabilities, that has been described in relation to industrial firms in developing countries.6 Indeed, according to one observer, TV Globo’s emphasis on professionalism, indigenous production capability, development of local technology and mastery of related know-how and an acute sense of market needs and opportunities coupled with quality products, increased TV Globo’s audience dramatically in its first ten years. (Neto, 1997: 260) Some of the main technological achievements to which these efforts gave rise are summarized in Table 7.2. To what extent the technological learning that has obviously taken place at TV Globo was facilitated by the state at different times is not at all clear, though according to at least one student of the topic, ‘Heavy emphasis by Brazilian policy-makers on domestic production of goods and services also helped TV Globo engage in the development of indigenous technological developments in several areas of Globo’s production’ (Neto, 1997: 268). This aspect of

128 Consumption, Globalization and Development

Table 7.2

Selected technological achievements at TV Globo

1. ‘One of Globo’s main features has been the struggle to remain up-to-date with technological breakthrough in the telecommunications industry and the world’s leading TV networks. An example is the 1994 network shift from U-matic to Beta Cam; new multi media and computer graphic divisions became operational and digital soundtracks were introduced’ (Neto, 1997: 261). 2. ‘TV Globo has also been one of the world’s leaders in a number of product technologies. It has developed its own formula to introduce interactive television, taking the lead not only in the domestic market, but also at the international level’ (Neto, 1997: 261). 3. ‘TV Globo not only innovates but also creates the “green house environment” to foster indigenous technological developments. For instance, TV Globo built Latin America’s largest studio complex [which] ranks among the world’s most modern television production centers’ (Neto, 1997: 261).

the Brazilian experience with telenovelas, however, deserves far more research attention than it has received thus far.

Conclusions At a time when increased foreign investment is accompanied by an expansion of the mass-media and advertising in developing countries, it is usually assumed that multinational corporations tend to drive out local consumer goods. It is also usually taken for granted that in the cultural sphere, American products will dominate the mass-media in those countries, as broadcasting moves increasingly from the public to the private sphere. The two cases that have just been described, in India and Brazil, indicate, however, that these assumptions are not universally valid. For, in both cases it was the local rather than the foreign product that was able to capture a mass market. On the demand-side this was achieved, again in both cases, by paying very close attention to the preferences of low, rather than is usually the case with multinationals, high-income consumers. In the Indian case, however, this demand was met by production based on highly labour-intensive methods of production, as opposed to the sophisticated techniques that were increasingly used to produce telenovelas in Brazil.

Cultural Advantage Reversal 129

Notes 1. According to Barker (1997: 88) ‘Brazilian telenovelas have received particular attention from scholars for three reasons. First, Brazil is the main exporter of telenovelas in the international market. Second, the rise of the dominant television company, TV-Globo, is intimately connected with the development of telenovelas, and third, because the social economic and political situation in Brazil has been refracted through changes in the genre’s format’. 2. Currently, for example, TV Globo produces 6 telenovelas per annum, the equivalent of some 350 movies (Neto, 1997). 3. According to Reeves ‘The size of the Brazilian audience and advertising market, which are linked to the levels and scale of industrial and commercial development, have provided a revenue base which is generally absent in smaller countries. This, certainly in the case of TV Globo, provided a strong base for local production (Reeves, 1993: 64). 4. The appropriateness of telenovelas for the mass domestic market is less clear-cut however than in the Indian case. On the one hand, it is true that these programs are rooted in and are reflective of Brazilian culture. It is also true that this cultural influence is exerted by the majority rather than the small minority of elite viewers. On the other hand, according to some observers, telenovelas are simply a ‘Brazilianized’ version of American imports, promoting images of consumption and social mobility that only a small minority can actually realize (Oliveira, 1993). Moreover, criticism of the Brazilian telenovela has arisen with regard to the practice of ‘merchandising’ – that is, ‘subtle advertising beyond that advertising which is explicitly presented as such’ (Straubhaar, 1982: 146). 5. The cultural aspects of telenovelas are discussed in Barbero (1995), Lopez (1995) and Vink (2988). 6. See for example Katz (1987).

References R. Allen (1995). ‘Introduction’, in R. Allen (ed.), To Be Continued … Soap Operas Around the World, Routledge. J. M. Barbero (1995). ‘Memory and Form in the Latin American Soap Opera’, in R. Allen (ed.), To Be Continued … Soap Operas Around the World, Routledge. C. Barker (1997). Global Television, Blackwell. S. Boafo (1991). ‘Communication Technology and Dependent Development in Sub-Saharan Africa’, in G. Sussman and J. Lent (eds), Transnational Communications, Sage. J. Katz (ed.) (1987). Technology Generation in Latin American Manufacturing, Macmillan Press. A. Lopez (1995). ‘Our Welcomed Guests: Telenovelas in Latin America’, in R. Allen (ed.), To Be Continued … Soap Operas Around the World, Routledge.

130 Consumption, Globalization and Development

R. Neto (1997). ‘Case Study: TV Globo – The Brazilian Media Giant’, The International Executive, 39 (2), March/April. O. Oliveira (1993). ‘Brazilian Soaps outshine Hollywood: Is Cultural Imperialism Fading Out?’, in K. Nordenstreng and H. Schiller (eds), Beyond National Sovereignty: International Communication in the 1990s, Ablex Publishing Company. G. Reeves (1993). Communications and the Third World, Routledge. J. Straubhaar (1982). ‘The Development of the Telenovela as the Pre-Eminent Form of Popular Culture in Brazil’, Studies in Latin American Popular Culture, 1. J. Straubhaar (1988). ‘The Reflection of the Brazilian Political Opening in the Telenovela, 1974–1985’, Studies in Latin American Popular Culture, 7. N. Vink (1988). The Telenovela and Emancipation, Royal Tropical Institute, Amsterdam.

8 Globalization and the Potential for Social Marketing in Developing Countries

Previous chapters have argued that the rapid recent growth of the mass-media and advertising in developing countries is occurring in a context where the market mechanism provides few if any sanctions against misinformation about products supplied overwhelmingly by the developed countries. Under these circumstances we further suggested that globalization raises the possibility of substantial welfare losses among consumers in developing countries, especially those with little or no access to countervailing sources of information. As the Human Development Report of 1998 has put it, ‘The information revolution, the media revolution and the spectacular rise of advertising in developing countries have all brought great imbalances in information to consumers’ (p. 65, emphasis added). What also needs to be recognized, however, is that the growth of the same globalizing forces mentioned in the previous paragraph could in principle be used instead, by non-market agents (such as NGOs, aid donors and government agencies), not only to counter the existing imbalance in information about say, cigarettes, but also to promote the consumption of goods (such as oral rehydration salts and nutritious foods) that are generally thought to promote the health, nutrition and other functionings of the poorest groups in developing countries. The purpose of this chapter is to examine these possibilities in relation to one specific technique, known as social marketing, that is so named because it draws heavily on commercial marketing principles for social (i.e. non-market) interventions in consumption behaviour.1 131

132 Consumption, Globalization and Development

The concept of social marketing Conceived more than twenty years ago, social marketing consists in essence of the application of commercial marketing principles to the promotion of social causes (where these may have to do with products, ideas or forms of behaviour). 2 Thus defined, social marketing should be distinguished not only from the marketing of private goods and services (which, primarily in the form of heavy recent advertising by multinationals has created the information imbalances noted above), but also from the numerous ways in which countervailing information is conveyed to consumers without any effort to embrace the techniques used in commercial marketing (Health warnings on cigarette packages represent perhaps the clearest example of this type of information).3 Of the various commercial techniques wich are employed by social marketing, perhaps the most important is detailed research on the group whose behaviour is the object of the exercise in question. Indeed, according to Kotler and Roberto, Research is what differentiates the marketing approach to social change from earlier impressionistic efforts to influence changes in social ideas and practices. Only by researching and understanding the specific needs, desires, beliefs and attitudes of target adopters and the specific characteristics of the social products that are being marketed can social marketers move toward the successful implementation of social change campaigns. (Kotler and Roberto, 1989: 62) One of the first applications of social marketing in developing countries was to a nutrition education program and this example well illustrates the essentiality of adequate research to a successful project outcome. For, ‘Although mass-media and other marketing methods had been employed for some years to deliver messages about nutrition, few previous attempts had been made to do the kind of research and other efforts that are widely employed in commercial marketing’ (Berg, 1987: 30). Such subsequent efforts incorporated research that was intended to identify ‘the beliefs, feelings, needs, constraints, and practices of mothers, with particular attention given to identifying those practices susceptible to change’

Social Marketing 133

(Berg, 1987: 30). On the basis of these and other findings the project officers were able to identify what could be achieved, for instance, by the introduction of a home-made weaning food prepared with locally available ingredients. According to Berg (1987), this project resulted in improvements in nutritional status that could not be accounted for by other factors, such as education. Similarly, in the Philippines, efforts to introduce oral rehydration therapy (ORT) as a treatment for infant diarrhoea went well beyond the simple transmission of factual information to mothers, an approach that in the past had been used with very limited success. Rather, the social marketing campaign relied first on focus groups to obtain the necessary information about how the ORT was actually being used by mothers. ‘This consumer research revealed that some mothers dissolved only the sodium and kept the glucose as high-quality sugar for use in guests’ coffee. Other mothers found the mixed ORT too salty and added sugar to make it more acceptable to their children, thereby imbalancing the treatment. These findings led to further consumer research and adjustments in the product and promotional messages’ (Austin, 1990: 309). Several examples illustrate how the mass-media in developing countries have been employed by social marketing campaigns to promote behavioural change among the target groups concerned. In Thailand, for example, CSM [contraceptive social marketing] programmes have offered new energy and creativity to sometimes pedantic family planning education programmes, mobilizing many non-traditional allies. Borrowing from commercial marketing principles, they have brought contraceptives and birth control front and centre, especially through the use of mass-media. Around the world radios are beaming out family planning messages with catchy jingles, competing for the listener’s ear with Coca-cola, Pepsi and soap ads. (McKee, 1993: 55)4 What appears to have been a particularly successful case of massmedia based social marketing, took place in Egypt between 1983 and 1991 with regard to the promotion of ORS (oral rehydration solution). Funded primarily by foreign aid, 63 television spots were aired over this period in a bid to promote ORS as the most appropriate

134 Consumption, Globalization and Development

treatment for diarrhoea (Television, one should note, is especially well-suited to the promotion of social causes in Egypt, reaching, as it does, a very high proportion of the total population and the television industry itself has a long history of producing high quality films and serialized dramas for the Arab-speaking world as a whole. Indeed, according to some evidence, television is the most effective means of communicating information in that country).5 In 1983, the first year of the program, ‘baseline information was gathered and five related messages about dehydration and oral rehydration solution were determined. Local indigenous terms for diarrhoea were identified by social scientists doing informal and focus group interviews. A term that had formerly described the yearly drought Egypt experienced before the construction of the High Dam … was chosen to describe diarrhoea-caused dehydration’ (Lane, 1997: 170). Following subsequent pretesting of the media campaign in 1983 and early 1984, nationwide interviews with 1,100 sample households were conducted annually until 1987. These households, were asked about their knowledge, attitude and practice (KAP) regarding dehydration and ORS. To assess whether or not they could correctly mix ORS, 100 urban and 440 rural mothers were asked to mix the solution in front of investigators. Women who reported having gotten their knowledge from the televised spots were the more likely to correctly mix the solution than women who said that their knowledge came from other sources. NCDDP’s [the National Control of Diarrhoeal Disease Project’s] impact on childhood diarrheal deaths has been considerable. Several studies report that by 1988 childhood diarrheal disease deaths had decreased substantially during the intervention period, although it cannot be stated precisely how much of the decrease is due to the program. (Lane, 1997: 171)6 Although the example just cited involved a series of one-minute television commercials, social marketing has also been successfully woven into a number of local soap opera productions in developing countries, thus exploiting the rapidly growing popularity of this form of television entertainment in the Third World. 7 In India, for example, a very successful ‘pro-development’ soap opera entitled ‘Hum-Log’ (or ‘We People’) was broadcast in a series of 156, 22-

Social Marketing 135

minute episodes during the mid-1980s. ‘The stories depicted every day life in India rather than the usual fantasies of Indian films. The characters provided realistic role models and viewers were affected through a social-psychological approach of social learning’ (McKee, 1993: 69). ‘In another developing country, the Philippines, issues related specifically to the transmission of the AIDS virus were integrated into the script of a popular soap opera, doubling the number of people visiting sexually transmitted disease clinics in the weeks following the broadcast’ (McKee, 1993: 71).

The particular potential that exists in the least developed countries From some points of view, social marketing is likely to be more difficult in the least developed of the countries that belong to the Third World. Inadequate infrastructure and exceptionally low incomes, for example, mitigate against the spread of mass-media and hence the opportunity to reach even a moderately high proportion of the target population within a country. On the other hand, as Harvey (1997) has recently emphasized, the least developed countries also present particular advantages to social marketers, not the least of which are that ‘there is very little clutter from other advertisers … and consumers have not been jaded by decades of competing advertising claims.’ (Harvey, 1997: 162). Put another way the point is that ‘Advertising in newly emerging economies has the advantage of working from a clean slate’ (Harvey, 1997: 162).

Counter-marketing In the introduction to this chapter we distinguished between two ways in which the growth of the mass-media in developing countries could be exploited by non-market actors for purposes other than the use of the market mechanism for the purpose of commercial gain. So far we have described how social marketing could be used by these actors to introduce, via the mass-media, what can be described as socially desirable product innovations (such as ORS therapy), or new consumption practices primarily to members of traditional societies in developing countries. The second purpose of social marketing, as described in the introduction, however, is to counter the imbalance

136 Consumption, Globalization and Development

in information about products that is exacerbated by the recent growth of television advertising in developing counties, which has led, for example, to a pronounced increase in cigarette smoking in many such countries. The use of social marketing in this second context is often referred to as ‘counter-marketing’.8 One of the most important applications of counter-marketing has arisen in response to the much-publicized promotion of infant formula in developing countries. In particular, Borden, Nestlé, and other infant formula manufacturers, faced with declining birth rates in developed countries, began aggressive marketing to mothers in Africa, Latin America, and other less developed regions characterized by poverty, illiteracy, lack of pure water, lack of facilities for sterilizing bottles, and lack of refrigeration for storing mixed formula … Social marketing campaigns have been carried out to counter such powerful slogans as ‘Give your baby love and Lactogen’ and to inform mothers of the advantages of breastfeeding. (Fox and Kotler, 1980: 27) In Brazil, for example, the counter-marketing of breastfeeding in 1985 included a mass-media campaign involving 100 television and 600 radio stations and when it was combined with other measures such as the formation of mothers’ groups at the community level, the result was an ‘important increase in the duration of breastfeeding’ and corresponding declines in the incidence of infant mortality resulting from diarrhoea, respiratory and other infections (McKee, 1997: 139). Even in developed countries, one should emphasize, where the information available to the consumer tends to be far more balanced than is now the case in most developing countries, efforts have long been made to counter-market cigarettes, alcohol and nutritionally harmful foods. For, even in the former countries, what needs to be recognized, is that ‘consumer sovereignty’ is poorly served in circumstances where ‘The large promotional budgets behind these [harmful] products tend to crowd out opposing views, because those who disagree are often too fragmented, too few in number, or have inadequate resources to present the counter position’ (Fox and Kotler, 1980: 27). Indeed, in the absence of this important recognition and countervailing actions that are designed

Social Marketing 137

to address it, governments are effectively choosing a market system, whose functioning, in the present context of globalization in developing countries, is only very distantly related to the textbook model of consumption theory and consumer sovereignty. Unfortunately, however, while there is a long history of intervention on the production side of many developing economies, 9 with regard to consumption there tends to be a presumption in favour of the status quo system, however undesirable this may actually be.10

Conclusions A major theme of the 1998 Human Development Report is that the rapid recent growth of the mass-media and advertising in developing countries, has caused a marked imbalance in the information available to consumers, a phenomenon which offsets the welfare gains that may arise from an expansion in the range of consumer choice in a globalizing world. The Report recognizes that various policies are needed to address this problem, including the provision of ‘correct and clear information, especially on the basics of food, drink, medicines, health care, household appliances and transport safety’ (p. 87). What the Report lacks, however, is an adequate discussion of exactly how such information is to be provided to consumers.11 This issue is crucial because information provided in one form can be assimilated very differently from the same information provided in another form, with correspondingly disparate effects on the behaviour of the consumer. In this chapter we have argued that the same mechanisms through which information imbalances arise in the market system, could also be used by governments, aid donors and NGOS, not only to provide countervailing information to consumers but also to promote the use of products that enhance the individual functionings of the poor in developing countries. We cited numerous instances where methods and techniques borrowed from commercial marketing (collectively known as social marketing), appear to have contributed to these objectives, often in circumstances where other more traditional modes of providing product information have not proved successful. It is not our intention to suggest that social marketing is invariably successful in endeavours such as these, since even a cursory review of the literature would confirm that this

138 Consumption, Globalization and Development

is not the case.12 Our goal has rather been to emphasize the potential afforded by this policy tool in a context where typically only the negative influences exerted by the growth of the mass-media receive attention in the literature on globalization and development.

Notes 1. Parts of this chapter draw on James (1993). 2. For a detailed description of the methods used in social marketing see Kotler and Roberto (1989). 3. In that they contain only a broad factual statement about the health risk of smoking. 4. In relation specifically to the social marketing of contraceptives, Harvey (1997) has emphasized the role of product branding as opposed to a purely ‘generic campaign’. Among the advantages of branding, as he sees it, are the following: First, brand advertising conveys quality. Although many individuals know intellectually that generic brands of some products are likely to be just as good as advertised brands, they still tend to believe that an advertised brand is of higher quality … in the case of condoms particularly, a heavily advertised brand can become generic, especially in underdeveloped economies like Vietnam and Ethiopia, with the result that embarrassment at point-of-purchase can be lessened. … Brand promotions do not patronize. An important and often overlooked point about brand advertising is that it positions contraceptives in the marketplace as normal consumer products, rather than something the government (or someone else) is promoting as socially useful. (Harvey, 1997: 155) 5. According to Lane (1997: 167), ‘studies comparing television, radio, posters, pamphlets, and public meetings as vehicles for health communication found that television was the most effective medium for disseminating information’. 6. More detailed estimates provided by Fox (1988: 102) suggest that in the one-year period between the first and second mass advertising campaigns, 1. Mothers aware of the symptoms of diarrheal dehydration increased from 32 to 90 per cent. 2. Mothers who knew about ORS increased from 1.5 to 94 per cent. 3. The percentage of mothers who had ever used ORS increased from 1 to 50 per cent. 4. After the second campaign, 40 percent of mothers reported using ORS in the most recent episode of diarrhoea. 7. See Barker (1997) and the Latin American telenovelas described in the previous chapter.

Social Marketing 139

8. The term is taken from Fox and Kotler (1980). 9. With regard, for example, to infant industry arguments in favour of import substitution. 10. Much of this has to do, we feel, with the unquestioning acceptance of ‘consumers’ sovereignty’ and its implication of the desirability of laissez-faire on the consumption side of socio-economic policy making. 11. With its emphasis on the general lack of knowledge in developing countries, and the consequent need for countervailing public action, the World Development Report for 1998/99 is also relevant in this regard. Furthermore, the Report recognizes, as we have, that ‘The way governments convey information to citizens, especially the poor, is often critical’ (p. 153) and it emphasizes, again as we have done, that for this purpose there is a need to learn from the target group itself. Yet, unfortunately, no reference is made in this regard to the already substantial literature on social marketing in developing countries. 12. For a review of some of the unsuccessful cases see Lane (1997).

References J. Austin (1990). Managing in Developing Countries, The Free Press. A. Berg (1987). Malnutrition: What Can be Done, Johns Hopkins. C. Barker (1997). Global Television, Blackwell. K. Fox (1988). ‘Social Marketing of Oral Rehydration Therapy and Contraceptives in Egypt’, Studies in Family Planning, 19 (2). K. Fox and P. Kotler (1980). ‘The Marketing of Social Causes’, Journal of Marketing, 44, Fall. P. Harvey (1997). ‘Advertising Affordable Contraceptives: The Social Marketing Experience’, in M. Goldberg, M. Fishbein and S. Middlestadt (eds), Social Marketing, Lawrence Erlbaum. J. James (1993). Consumption and Development, Macmillan. P. Kotler and E. Roberto (1989). Social Marketing, The Free Press. S. Lane (1997). ‘Television Minidramas: Social Marketing and Evaluation in Egypt’, Medical Anthropology Quarterly, 11 (2). N. McKee (1993). Social Mobilization and Social Marketing in Developing Communities, Southbound.

This page intentionally left blank

Index Advertising, 2, 13, 16, 19, 20, 24, 30, 31, 42–4, 47, 48, 52, 88, 90, 92, 97, 105–20, 131, 135 Africa, 5, 24, 44, 50, 69, 90, 114, 122, 123, 124 Allen, R., 122 America, 6, 12, 40, 41, 42, 44, 72, 92, 94, 122, 124, 125, 126 appropriate products, 5, 6, 49–50, 61, 70, 80, 105–20 Austin, J., 52, 133 Barbero, J., 126 Becker, G., 30, 31 Belk, R., 20, 21, 42, 50 Bennett, K., 81 Berg, A., 132, 133 Boafo, S., 123 Brazil, 5, 6, 19, 25, 71, 95, 96, 109, 122–7, 136 Buenos Aires, 49, 50 Caribbean, 23, 40, 41 Carr, M., 105 Castelijn, I., 24 China, 2, 23, 80, 88, 125 Cohen, G., 66 constant preferences, 3, 4, 11, 14–18, 34, 86, 94 consumer sovereignty, 1, 136, 137 consumption systems, 2, 4, 98 Cowan, G., 80 Cuthbert, M., 23 Cyert, R., 85 Datta-Ray, S., 5 Deininger, K., 114 Dixit, A., 11 Douglas, M., 54, 90, 91

Epstein, A., 44 Featherstone, M., 6, 42 foreign investment, 12, 13, 16, 21, 22, 24, 28, 41, 43, 47, 92, 109, 119, 128 Fox, K., 136 Frank, R., 29, 32, 53, 92, 93, 110 Galbraith, J. K., 11, 30 Ger, G., 20, 21, 42, 50 Groot, M. de, 85 Grossman, G., 22 Hardon, A., 71 Harsanyi, J., 14–18, 87, 94, 95 Harvey, P., 135 Heap, B., 80 Hill, J., 21, 22 Hirsch, F., 29, 31, 32, 33, 39, 42, 44, 53, 91, 92, 109 Hirschman, A., 14, 15, 24, 86, 96, 97 Hogerzeil, H., 73 Hoover, S., 23 Human Development Report, 2, 45, 90, 91, 94, 114, 131, 137 India, 5, 6, 22, 73, 74–5, 80, 88, 105–20, 123, 126, 128, 134, 135 information technology, 2, 12, 28, 109, 122 Inkeles, A., 13, 35, 36, 37, 38 Isherwood, B., 54, 90, 91 James, J., 86, 118 Jelliffe, D., 43, 81 Jelliffe, E., 81 Kotler, P., 132, 136 Krishnaswamy, K., 73 141

142 Index

Lancaster, K., 61 Lane, R. 21, 86, 88, 97 Lane, S., 134 Latin America, 19, 21, 88, 90, 114, 122, 123, 126 Leadbeater, S., 5 Ledogar, R., 43 Lee, P., 65, 68, 72 Leslie, D., 43 Lieber, M., 41 Lopez, A., 126 Lydecker, M., 72 MacEwan, A., 50 Martin, E., 68, 70 Maslow, A., 39 Mattelart, A., 13 McDonald’s Corp., 42 McKee, N., 135, 136 Mexico, 21, 123 Mitchell, J., 44 Muller, M., 71 Neto, R., 125, 127 Norman, V., 11 Nurkse, R., 28–30, 33, 34, 37, 39, 44, 52, 53, 93 Oliviera, O., 20, 96, 109 Philippines, 71, 88, 133, 135 Portes, A., 38, 39 positional goods, 32, 33, 43, 44–52, 91, 92, 109

Rao, S., 112, 119, 120 Roberto, E., 132 Sachdeva, S., 114 Sauvant, K., 41 Schapira, A., 69 Schmidt, K., 69 Scitovsky, T., 29, 53, 96 Sen, A., 54, 61–82, 97, 98 Shapiro, C., 22 Silverman, M., 65, 68, 72 Sklair, L., 12 Smith, D., 13, 35 Snyder, R., 68 Southall, A., 24, 50 Squire, L., 114 Stewart, F., 1, 18, 118 Stigler, G., 30, 31 Still, R., 21, 22 Straubhaar, J., 125, 127 Supple, B., 48, 49 traditional consumption theory, 1–4, 11, 85–7, 99, 100 United Kingdom, 48, 72 Varis, T., 120 Veblen, T., 28–33, 39, 44–6, 50–2 Vink, N., 126 White, L., 111 Wilson, P., 40, 41 World Bank, 90