Wellbeing and Development in Peru: Local and Universal Views Confronted (Studies of the Americas)

STUDIES OF THE A MERICAS edited by

James Dunkerley Institute for the Study of the Americas University of London School of Advanced Study Titles in this series are multidisciplinary studies of aspects of the societies of the hemisphere, particularly in the areas of politics, economics, history, anthropology, sociology, and the environment. The series covers a comparative perspective across the Americas, including Canada and the Caribbean as well as the United States and Latin America. Titles in this series published by Palgrave Macmillan: Cuba’s Military 1990–2005: Revolutionary Soldiers during Counter-Revolutionary Times By Hal Klepak The Judicialization of Politics in Latin America Edited by Rachel Sieder, Line Schjolden, and Alan Angell Latin America: A New Interpretation By Laurence Whitehead Appropriation as Practice: Art and Identity in Argentina By Arnd Schneider America and Enlightenment Constitutionalism Edited by Gary L. McDowell and Johnathan O’Neill Vargas and Brazil: New Perspectives Edited by Jens R. Hentschke When Was Latin America Modern? Edited by Nicola Miller and Stephen Hart Debating Cuban Exceptionalism Edited by Laurence Whitehead and Bert Hoffman Caribbean Land and Development Revisited Edited by Jean Besson and Janet Momsen Cultures of the Lusophone Black Atlantic Edited by Nancy Naro, Roger Sansi-Roca and David Treece Democratization, Development, and Legality: Chile, 1831–1973 By Julio Faundez The Hispanic World and American Intellectual Life, 1820–1880 By Iván Jaksic´ The Role of Mexico’s Plural in Latin American Literary and Political Culture: From Tlatelolco to the “Philanthropic Ogre” By John King Faith and Impiety in Revolutionary Mexico Edited by Matthew Butler

Reinventing Modernity in Latin America: Intellectuals Imagine the Future, 1900–1930 By Nicola Miller The Republican Party and Immigration Politics: From Proposition 187 to George W. Bush By Andrew Wroe The Political Economy of Hemispheric Integration: Responding to Globalization in the Americas Edited by Kenneth C. Shadlen and Diego Sánchez-Ancochea Ronald Reagan and the 1980s Edited by Cheryl Hudson and Gareth Bryn Davies Wellbeing and Development in Peru: Local and Universal Views Confronted Edited by James Copestake The Federal Nation: Perspectives on American Federalism Edited by Iwan W. Morgan and Philip J. Davies Base Colonies in the Western Hemisphere, 1940–1967 By Steven High Beyond Neoliberalism in Latin America?: Societies and Politics at the Crossroads Edited by John Burdick, Philip Oxhorn, and Kenneth M. Roberts

Wellbeing and Development in Peru Local and Universal Views Confronted Edited by James Copestake

wELLBEING AND DEVELOPMENT IN PERU Copyright © James Copestake, 2008. All rights reserved. First published in 2008 by PALGRAVE MACMILLAN® in the United States—a division of St. Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Where this book is distributed in the UK, Europe and the rest of the world, this is by Palgrave Macmillan, a division of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. ISBN-13: 978–0–230–60869–6 ISBN-10: 0–230–60869–8 Library of Congress Cataloging-in-Publication Data Wellbeing and development in Peru : local and universal views confronted / edited by James Copestake. p. cm.—(Studies of the Americas) Includes bibliographical references and index. ISBN 0–230–60869–8 1. Well-being—Peru. 2. Quality of life—Peru. I. Copestake, James G., 1960– HN344.W45 2008 306.0985⬘09049—dc22

2008017288

A catalogue record of the book is available from the British Library. Design by Newgen Imaging Systems (P) Ltd., Chennai, India. First edition: December 2008 10 9 8 7 6 5 4 3 2 1 Printed in the United States of America.

Contents

List of Figures

vii

List of Tables

ix

Foreword and Acknowledgements

xi

Notes on Contributors Abbreviations 1. Introduction and Overview James Copestake 2. Resources, Conflict, and Social Identity in Context José Luis Álvarez, Maribel Arroyo, Lida Carhuallanqui, James Copestake, Martín Jaurapoma, Tom Lavers, Miguel Obispo, Edwin Paúcar, Percy Reina, and Jorge Yamamoto 3. Subjective Wellbeing: An Alternative Approach Jorge Yamamoto, Ana Rosa Feijoo, and Alejandro Lazarte 4. Economic Welfare, Poverty, and Subjective Wellbeing James Copestake, Monica Guillen-Royo, Wan-Jung Chou, Tim Hinks, and Jackeline Velazco

xiii xv 1

31

61

103

5. Wellbeing and Migration Rebecca Lockley with Teófilo Altamirano, and James Copestake

121

6. Wellbeing and Institutions José Luis Álvarez with James Copestake

153

vi

CONT ENT S

7. Reproducing Unequal Security: Peru as a Wellbeing Regime James Copestake and Geof Wood

185

8. Conclusions and Implications for Development Policy and Practice James Copestake

211

9. Implications for Wellbeing Research and Theory Jorge Yamamoto

231

References

243

Index

265

Figures

1.1 Diagram for thinking about personal wellbeing as process 1.2 A framework for thinking about development discourse 1.3 Map of the research sites 3.1 From goals to latent needs 3.2 “Place to live better” importance and satisfaction by period of residence 3.3 “Raise a family” importance and satisfaction by age 3.4 “Raise a family” importance and satisfaction by locality 3.5 “Improvement from a secure base” importance and satisfaction by locality 3.6 Factor structure of resources perception 3.7 Factor structure of values 3.8 Variation in values by locality, individualism and collectivism 3.9 Variation in values by formal education, individualism, and collectivism 3.10 Factor structure of personality 3.11 Personality by migration history 3.12 Path model for “place to live better” 3.13 Path model for “raising a family” 3.14 Path model for “improvement from a secure base” 3.15 Integrated model of subjective wellbeing 7.1 Model for wellbeing regimes 8.1 A reflexive framework for appraisal of development interventions 9.1 Implications for wellbeing research and theory: Chapter summary

4 5 18 68 72 73 74 75 77 79 81 83 85 87 89 91 96 98 187 228 232

This page intentionally left blank

Tables

1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 2.1 2.2 2.3 2.4 2.5 2.6 4.1 4.2 4.3

4.4 4.5 4.6

Three contemporary discourses of development National poverty rates for 2000 Peru and the Human Development Index Human Development Index by department, 2000 Summary findings of Peru’s first national participatory poverty assessment Some Peruvian attitudes in regional perspective A brief description of the research sites in Peru WeD data collection methods Education above primary level by research site: Logit estimates Global happiness by research site: Ordered Probit estimations Comparative statistics for Lima, Junin, and Huancavelica departments Demographic details of the WeDQoL sample Self-categorization by site Self and community categorization of social identity compared Use of the term blanco (white) by site Seasonality in the Mantaro Valley: Highlights Poverty head count based on official and subjective poverty lines Subjective Wellbeing and income comparisons for an urban sample in Peru Measured and perceived income mobility, 1991–2000 of a representative sample of 500 households in Peru Average household income and expenditure by research site Official poverty lines for 2005 (Soles per person per month) Household poverty estimates (mean over 10 months)

6 10 11 11 14 16 19 21 22 24 37 40 41 41 42 43 105 105

106 107 109 109

x

TA B L E S

4.7 Comparison of income and expenditure-based poverty classifications 4.8 Distribution of responses to global happiness question by round 4.9 Average global happiness by research site 4.10 Ordered Probit analysis of happiness determinants 4.11 One way ANOVA of subjective wellbeing indicators against global happiness scores 4.12 One way ANOVA of subjective wellbeing indicators against household income poverty category 5.1 Place of birth of head of household and spouse/partner 5.2 Households providing or making transfers from/to relatives in last year 5.3 Demographic characteristics of household members who were away at the time of interview 5.4 Visits outside the community (involving staying away more than one night) 5.5 Migration motives and outcomes cross-tabulated 6.1 Checklist of questions for semi-structured interviews 6.2 A summary of community institutions by purpose 6.3 Link from components of “place to live better” to institutions 6.4 Link from components of “raise a family” to institutions 6.5 Link from components of “improvement from a secure base” to institutions 6.6(a) Characteristics of selected faenas 6.6(b) Characteristics of selected faenas continued 6.6(c) Characteristics of selected faenas continued 8.1 Necessity and satisfaction with components of wellbeing

109 111 111 113 116 117 125 125

126 129 131 157 159 160 160 162 166 172 173 213

Foreword and Acknowledgements

This book is the outcome of more than five years of intensive multidisci-

plinary research on what it means for us to “be well” and how this relates to the idea of development. Human wellbeing is explored not only as an idea that inhabits debates over development, but as something personally and uniquely experienced by us all as individuals. Central Peru’s complex cultural heritage, and the often staggering diversity of its social and physical geography, provided a fertile context for such enquiry. This was further enriched by our interaction with fellow members of the “Wellbeing in Developing Countries” Research Group in Bangladesh, Ethiopia, Thailand, and the UK, as well as many others who shared with us their thoughts about what wellbeing means to them. Our methodological approach was as far as possible to elicit what people living in our seven selected research sites themselves understood by wellbeing, to generalize inductively from these “local” views, and use them to interrogate different understandings of wellbeing implicit in “universal” discourses of development. This does not signify a retreat into cultural relativism, a rejection of the possibilities of meaningful measurement of wellbeing, nor an outright rejection of a universal view of human development. Rather, the book seeks to relate local views to global development discourses and policy perspectives in a way that contributes to more constructive intercultural and policy-relevant exchange between them. By far the most important debt of gratitude shared by all the contributors to the book is to the inhabitants of the seven research sites in Central Peru, where the bulk of primary data collection took place. We hope that many enjoyed sharing conversations, experiences, and resources with us as much as we did. But we know that many also displayed a willingness to put up with repeated and lengthy interviews beyond any expectation of personal gain. We salute their goodwill, and gratefully dedicate this book to them. Within the research team itself, thanks are due above all to our six full-time field researchers, collectively referred to as the Pumas: Maribel Arroyo, Lida Carhuallanqui, Martín Jaurapoma, Miguel Obispo, Edwin Paúcar, and Percy Reina. They displayed integrity, endurance, patience, flexibility, and many other fine qualities in abundance. We are also very grateful to colleagues in their alma mater, the Universidad Nacional del Centro del Perú (UNCP) in Huancayo,

xii

FORE WORD AND ACK NOW LEDGEMENT S

and other scholars in the region who participated in research workshops before and after the data collection period. As editor, I would next like to thank staff employed by the Pontificia Universidad Católica del Perú (PUCP). This includes coresearchers and coauthors José Luis Álvarez, Teófilo Altamirano, Ana Rosa Feijoo, Alejandro Lazarte, Jorge Yamamoto. But it also includes many others, who offered ideas, advice, and practical help to ensure administration, data collection, and logistics proceeded as smoothly as possible. Particular thanks are due to Adolfo Figueroa who enthusiastically participated in the research in its initial phase. Among colleagues past and present here at Bath there are more coauthors to thank: Wan-Jung Chou, Tom Lavers, Rebecca Lockley, Tim Hinks, Geof Wood, Monica Guillen-Royo, and Jackie Velazco. Among the much larger number of people who contributed to WeD in important ways I would particularly like to thank Allister McGregor (Director), Katie Wright (whose work on wellbeing among Peruvians living in London and Madrid could easily have been included in this book), and Kate Burrell, who took over responsibility for subediting and arranging the final manuscript. Producing this book has been a complicated, lively, often messy and sometimes fraught experience. Whether overall I lived better or worse as a result is hard to say: but I certainly lived more. The support of the UK Economic and Social Research Council (ESRC) in the production of this book is gratefully acknowledged, it being an output of the ESRC Research Group on Wellbeing in Developing Countries (WeD). For more about WeD go to www.welldev.org.uk.

Notes on Contributors

Teófilo Altamirano is professor of anthropology and director of postgraduate diploma course International Migration, Globalization and Development, at PUCP in Lima. José Luis Álvarez is a lecturer in anthropology at UNCP in Huancayo, and was employed by WeD for three years as a research officer. Maribel Arroyo is a graduate in anthropology from UNCP and was employed by WeD as a field investigator. Lida Carhuallanqui is a graduate in anthropology from UNCP and was employed by WeD as a field investigator. Wan-Jung Chou is a postdoctoral research officer in the Department of Economics and International Development at the University of Bath. James Copestake is head of the Department of Economics and International Development at the University of Bath. Ana Rosa Feijoo is a graduate in psychology from PUCP and was employed by WeD as a research officer. Monica Guillen-Royo is a postdoctoral fellow in the Department of Social and Policy Sciences at the University of Bath. She completed a doctorate on consumption and wellbeing in Peru in 2007 as a member of the WeD Group. Tim Hinks is a development economist and lecturer in the Department of Economics and International Development at the University of Bath. Martín Jaurapoma is a graduate in anthropology from UNCP and was employed by WeD as a field investigator Tom Lavers works at the UN Research Institute for Social development in Geneva. After completing a Masters in International Development at the University of Bath he spent two years working for WeD, first in Peru and then in Ethiopia. Alejandro Lazarte is an assistant professor in the Psychology Department of Auburn University in Alabama.

xiv

NOTES ON CONTRIBUTORS

Rebecca Lockley is a doctoral student in the Department of Economics and International Development at the University of Bath. Miguel Obispo is a graduate in anthropology from UNCP and was employed by WeD as a field investigator. Edwin Paúcar is a graduate in anthropology from UNCP and was employed by WeD as a field investigator. Percy Reina is a graduate in anthropology from UNCP and was employed by WeD as a field investigator. Jackeline Velazco is an agricultural economist in the Food and Agriculture Organization in Rome. Until 2006 she worked as a postdoctoral researcher in the WeD group at Bath, and previously as a lecturer at the Pontificate Catholic University in Lima. Geof Wood is professor of development sociology and Dean of the Faculty of Humanities and Social Sciences at the University of Bath. Jorge Yamamoto is a social psychologist and associate professor at the Pontificate Catholic University in Lima.

Abbreviations

ANOVA APR A CCT CFA CVR ECB EEG EPHR EPL

ESRC FONCODES GABA HDI HPA I&E IRM ISB MIR MRTA PGI PLB

Analysis of Variance Alianza Popular Revolucionaria Americana (Popular American Revolutionary Alliance) Conditional Cash Transfer Confirmatory Factor Analysis Comision de la Verdad y Reconciliacion (Truth and Reconciliation Commission) Encuesta de Bienestar (Wellbeing interview protocol). See Chapter 3 Electro Encephalogram Emic and Post Hoc Research. See Chapter 9 Extreme Poverty Line. Peruvian government estimates of income needed in each region to purchase food for a month with a daily calorific value of 2,200 calories per person. See Chapter 4 UK Economic and Social Research Council Fondo de Compensacion para el Desarrolo Social (Peru’s fund for compensation and social development) Gamma-aminobutyric Acid Human Development Index Hypothalamic-pituitary-adrenal Income and Expenditure Institutional Responsibility Matrix. See Chapter 7 Improvement from a secure base, one of three latent needs identified using the Peru WeDQoL. See Chapter 3 Movimiento de Izquierda Revolucionario (Revolutionary Movement of the Left) Movimiento Revolucionario Tupac Amaru (Revolutionary Movement of Tupac Amaru) Patient Generated Index Place to live better, one of three latent needs identified using the Peru WeDQoL. See Chapter 3

xvi PRONAA R AF R ANQ S/. SEM SLS SPL SWB WeD WeDQoL WHOQoL PRONAA PUCP UNCP UCV

A B B R E V I AT I O N S

Programma Nacional de Asistencia Alimentaria (Peru’s National Food Assistance Program) Raise a family, one of three latent needs identified using the Peru WeDQoL. See Chapter 3 Resources and Needs Questionnaire. A research tool developed by the WeD Research Group Nuevos Soles (Peruvian currency) Structural Equation Modeling. See Chapter 3 Subjective Life Satisfaction index Subjective Poverty Line is the amount of income perceived to be the minimum necessary to live Subjective Wellbeing Wellbeing in Developing Countries Research Group Quality of life survey, a tool developed by WeD. See Chapter 3 The World Health Organization’s Quality of Life measurement instrument Programa Nacional de Alimentos (Peru’s national nutrition program). Pontificia Universidad Católica del Perú (Pontificate Catholic University of Peru) Universidad Nacional del Centro del Perú (National University of Central Peru) Unidad Comunal de Vivienda (Planned neighborhoods in Nuevo Lugar)

Chapter 1

Introduction and Overview James Copestake

1.1. Reconnecting Development and Wellbeing in Peru The concept of development—local, national, and international—remains a preoccupation of people and politicians across the world. For some it signifies a general belief in human progress, linked particularly to the spread of market liberalism and the struggle for more democratic government. For others, this vision is soured by daily experiences of poverty and conflict, coupled with a sense that global economics and politics are making their problems worse not better. Indeed, popular use of the word development often has an ambiguous if not outright satirical edge: implying change imposed by others (or through some unknown, possibly sinister process) that possibly does more harm than good.1 This book seeks understanding beyond the rhetoric of both development optimists and their would-be debunkers. It does so in two steps. First, we reconsider afresh what development is about by confronting it with another concept. Wellbeing is a more firmly person-centered idea, while at the same time open to multiple and indeed holistic interpretations. It opens up space for reflecting on the often unstated and restrictive philosophical assumptions underpinning much talk about development. Wellbeing discourse encompasses how people think and feel, as well as what they have and do; it acknowledges the differences as well as links between personal happiness and a sense of life fulfillment; it encompasses the effect of people’s relationships to ideas and to other people as well as to money and goods. It also counterbalances the necessary but narrow focus in development discourse on the negative (poverty, insecurity, exclusion, harm and so on).2 Second, rather than colonizing this conceptual space solely with abstractions of our own making, we seek an empirical approach: to clarify how different understanding of the dream of achieving wellbeing squares with the reality of development for specific people in specific times and places. In exploring how development can be reconnected to wellbeing we acknowledge that both public and private agencies often act in a way that fails to give sufficient weight to the wellbeing of all those affected by their actions. The definition and measurement

2

C O P E S TA K E

of both development and wellbeing is unavoidably political, and their meaning can easily be tarnished through misuse. But people continue to dream of a better future for themselves and for society even in the most difficult situations. In acknowledging difference and conflict over these terms, particularly the appropriation of development discourse by more powerful minorities, we do not abandon the possibility of mutual understanding, compromise, and consensus. Why Peru? The answer is partially arbitrary: what matters is to be specific, to ground argument in real lives, wherever they may be. But Peru is arguably a particularly good site for this enquiry because it starkly reflects a number of interesting paradoxes. It is a middle income country, blessed with rich and diverse natural resources. Its economy has performed reasonably over the last fifteen years, during which time it has also experienced two relatively peaceful and democratic changes in government. Yet Peru is also persistently one of the most unequal countries in the world, with a higher rate of poverty than its economic status suggests it should have. Opinion polls suggest that many Peruvians are also less happy with many aspects of their lives than people elsewhere in Latin America, evidence supported by high rates of migration abroad. In addition, Peru’s complex racial-ethnic mix contributes to culturally diverse visions of wellbeing that draw on strong indigenous as well as Western traditions. This book is one product of the “Wellbeing in Developing Countries” (WeD) Research Group at the University of Bath, formed in 2003 with a grant from the UK Economic and Social Research Council (ESRC). The group’s officially stated purpose was to develop a conceptual and methodological framework for understanding the social and cultural construction of wellbeing in developing countries. 3 In addition to the UK, the founding group included researchers from Bangladesh, Ethiopia, and Thailand as well as Peru. The group’s academic backgrounds included anthropology, development studies, economics, psychology, politics, sociology, and social policy. In its first year the WeD group concentrated on conceptual and methodological development, as reported in detail in Gough and McGregor (2007).4 The Peru team also published a literature review on poverty, inequality, and wellbeing in Peru (Altamirano et al., 2004). This period was followed by two years of parallel primary research in the four countries. The field work carried out in Peru provides the empirical base of this book. 5 In brief, the book sets out to do five things. First, it presents a multifaceted picture of visions and realities of poverty and wellbeing as experienced and felt by inhabitants of the seven research sites (see especially chapter 2). Second, it contributes to the literature on how to construct general indicators of wellbeing to guide development. More specifically, it presents an original approach to identifying and analyzing people’s own subjective wellbeing (chapter 3). For relatively poor people in Central Peru, this is found to revolve around three latent needs: to find a place to live better, to build a family, and to progress with security. We argue that the use of orthodox indicators of development (such as income poverty reduction) is an inadequate substitute for monitoring satisfaction with achievement of such needs (chapter 4). Third, we explore how a holistic vision

INTRODUC TION AND OVERVIE W

3

of wellbeing informed by a better understanding of people’s own views provides insight into the wellbeing trade-offs arising from migration (chapter 5) and institutional change, including evolution of Andean traditions of community self-help (chapter 6). Fourth, we use the case of Peru to develop a framework for country-level analysis based on the idea of path-dependent evolution of wellbeing regimes (chapter 7). Finally, we draw general conclusions about the usefulness of wellbeing as a concept to thinking about development policy and practice (chapter 8) and review potential for further scientific research into subjective wellbeing (chapter 9). The remainder of this chapter is organized as follows. Section 1.2 elaborates further on the three key concepts of development, wellbeing, and reconnection. A broad conceptual framework for thinking about wellbeing is then presented and used to explore “growth first,” “needs first,” and “rights first” strategies of development, as well as the challenge to each posed by antidevelopment critics. Section 1.3 relates these ideas to the Peru context. Section 1.4 sets out the scope and methodology of the empirical research conducted in Peru, and section 1.5 presents a small selection of indicators for the seven fieldwork sites. Section 1.6 provides a more detailed chapter-by-chapter overview of the whole book.

1.2. Wellbeing and Development Discourse 1.2.1. Dimensions of Wellbeing Human wellbeing can be viewed in many different ways, and by defining it too rigidly development policy makers and practitioners risk alienating other stakeholders, including those they are trying to assist. While it would likewise be selfdefeating to propose too narrow a working definition of wellbeing at this point in this book a broad definition is a useful starting point.6 To this end wellbeing is defined here as a state of being with others in society where (a) people’s basic needs are met, (b) where they can act effectively and meaningfully in pursuit of their goals, and (c) where they feel satisfied with their life.7 Each of the three dimensions requires further clarification. The first raises the issue of what constitutes a basic need or, putting it negatively, something whose absence, when viewed in isolation, invariably constitutes harm.8 While arriving at a definitive full and final list may be impossible, agreement on at least some components (hunger, drinking water) is reasonably straightforward.9 Turning to the second part, goals can be viewed as potentially achievable expressions of a person’s values, hence embracing the idea of being fulfilled and living a meaningful life. It allows for changes in individual aspirations over time, differences based on context, and the likelihood of political conflict over wellbeing. Third, satisfaction introduces both positive and negative subjective feelings. Hedonic psychology (e.g., Kahnemann et al., 1999) tells us that these are not opposites, and are also affected by aspirations and adaptive preferences. This threefold framework for thinking about wellbeing leaves scope for further elaboration of each component, and also for exploring trade-offs between them in line with personal tastes, personality, culture, and context. It can also

4

C O P E S TA K E

accommodate universal and local perspectives to evaluating wellbeing: functioning meaningfully and feeling well within a specific context, on the one hand; having resources, capabilities, and opportunities to achieve goals that go beyond those that present themselves in local contexts, on the other. The threefold framework can be further clarified by comparing it with the more widely used twofold distinction between subjective wellbeing (SWB)— how people think and feel, and objective wellbeing (OWB)—what they can be observed to have and do.10 OWB is particularly associated with indicators of access to observable resources that contribute to meeting needs and to avoiding harm. However, success in achieving goals can also be objectively measured, as indeed can outward signs of happiness like smiling a lot. SWB is particularly associated with people’s reported feelings, but this concept extends beyond positive and negative emotions to include cognitive assessment of goal achievement, as well as subjective perception of the adequacy of available resources. This illustrates the more general point that subjective and objective aspects of wellbeing are in practice often very hard to disentangle, particularly when it comes to interpersonal relationships. The postpositivist rise of constructivism in social as well as the natural sciences reflects growing understanding that all assessment of objective states (wellbeing included) is also ultimately socially and culturally embedded, or intersubjective (Pieterse, 2001:142). In breaking with the tradition of regarding wellbeing as either clearly objective or subjective we place considerable emphasis on the process of goal formation, whether this takes the form of individual preferences, locally accepted norms, or universal theories. The feelings and motives elicited by goals are determined in part by their relationship to actual or perceived availability of resources to achieve them in a particular context. They in turn trigger actions whose outcomes affect future goals and resource availability (see figure 1.1). At the personal level, the framework can be used to explore subjective wellbeing

Personal goals

Induced motives and feelings of wellbeing

Personal resources and capabilities

Actions and interactions leading to Outcomes (material, social, cultural)

Wider context Figure 1.1 Diagram for thinking about personal wellbeing as process

5

INTRODUC TION AND OVERVIE W

defined as long-term satisfaction with personal goal achievement. Individual goals reflect personality and self-perception, which in turn reflect personal relationships, values, social identity, and culture (Yamamoto, 2006a). We also hypothesize that life satisfaction is influenced by individual perception of the gap between personal goals and the resources needed to achieve them.11 The same framework can also be used to assess collective or interdependent processes of production and reproduction of wellbeing: this being one way of defining development. We are interested in (a) the political process by which communities formulate collective goals, (b) the institutions through which resources are mobilized and distributed, and (c) practical action.

1.2.2. Dominant Development Discourse This starting point for thinking about wellbeing can be extended to the policy level and to the analysis of development discourse. Such discourse can be defined as a language that seeks to establish consistent relationships between three components. First, there is a normative or ethical position, embodying a definition of wellbeing. Second, there is a historical component, representing a view of development as an actual historical process determining availability of resources, opportunities, and constraints in any period. Third, there is a practical component, concerned with development as action or intervention. The three are connected by language, which embodies cultural values and assumptions; and to the extent that logical consistency is achieved then the discourse offers a reliable framework for meaningful action.12 Figure 1.2 also indicates how different forms of discourse are more or less successful in achieving such a synthesis—erring on the side of pragmatism at the expense of normative clarity, for example.

Normative dimension

Impractical; deterministic

Unrealistic; posturing Development discourse

Amoral; opportunistic Historical dimension

Figure 1.2

Practical dimension

A framework for thinking about development discourse13

6

C O P E S TA K E

It is beyond the scope of this book to explore the evolution of development discourse over time, including how far distinct discourses have risen to the status of paradigmatic hegemony in different policy arenas.14 But it is important to expose the existence of sharply distinct assumptions about wellbeing implicit in contemporary policy positions. A useful starting point here is to explore different views about the nature and existence of a post-Washington consensus. As a first approximation, three positions can be distinguished in this debate as shown in table 1.1.15 Income first emphasizes the goal of raising average incomes per person and hence the importance of economic growth. Over the years it has been supported by classical, neoclassical, and some heterodox economists to the extent that they have all regarded economic growth as the key means not only to increasing average incomes, but also labor absorption, productivity growth, and falling rates of absolute poverty (Easterly, 2002). Recently its most influential version is also narrowly promarket or neoliberal, as characterized by John Williamson’s Washington Consensus (WC), though this continues to evolve.16 For example, there has been renewed interest in reducing inequality, particularly to the extent that this can be shown to be based on market failures and to restrict domestic demand in ways that adversely affect economic growth (World Bank, 2006). However, many advocates of this approach remain wary of public intervention aimed directly at reducing poverty on the grounds that these are prone to distort incentives away from innovation and growth, encouraging rent taking and seeking instead (Easterly, 2006). Needs first is based on a more multidimensional view of wellbeing and poverty. Its political home has been in the UN system, but it can be traced

Table 1.1 Three contemporary discourses of development Development discourse

Economic growth first

Wellbeing goals and values (normative component) Historical perspective (historical component)

Individual material prosperity, leisure, and choices.

Public policy and practice (practical component)

Basic needs first

Poverty reduction; satisfaction of multiple basic needs. Capitalism first: Managed capitalism: growth, jobs, and rational public rising incomes are response to delivered first and deprivations arising foremost by free from or ignored by enterprise. capitalism. Create better Build capacity to conditions for provide all with the pursuit of private means to meet a material self-interest basic set of human (market-led). needs (state-led).

Human rights first Social justice, equity of esteem, and opportunity. Beyond capitalism: social exclusion, oppression, and class struggle.

Fight for basic rights; increase demand for satisfaction of basic human needs (society-led).

INTRODUC TION AND OVERVIE W

7

back to the earlier literature on basic needs, and it encompasses Sen’s capability approach and the human development movement. It has historically been particularly concerned with the role of the state, including official international development agencies, and of rational planning to supplement the market in ensuring entitlement to human needs—particularly services with public good characteristics such as health, education, social protection, and food security. A needs-first perspective lies behind the Millennium Development Goals through a sharp increase in international aid flows coupled with fairer trade and sovereign debt reduction (Sachs, 2005). This initiative also indicates a willingness to work to an expertly informed universal specification of what constitutes basic needs for the achievement of wellbeing. Rights first emphasizes the relational (social, political, and cultural) dimensions of development, the struggle against injustice and the potential of human rights discourse to mobilize poor and excluded citizens through social movements to become more active agents of their own development. Rights discourse has become particularly influential within international NGOs, and the attempt to extend rights from the civil-political sphere to the socioeconomic. Hickey and Bracking (2005:862) describe this as a bid to secure basic needs of “distant strangers” not as alms but by rights through duties of action on major social institutions underpinned by a theory of transnational justice. However, in NGO hands and compared to needs first discourse the rights first approach is less rationalist, materialist, aid-oriented, top-down, and paternal and is more focused instead on justice, grassroots action, power, and citizenship education.

1.2.3. Anti- and Postdevelopment Perspectives The previous section highlighted how different ways of conceptualizing wellbeing are bound up with different interpretations of history and justifications for public action. Advocates of the three discourses we distinguished find themselves competing with each other not only in policy arenas but also to capture the popular imagination: competing discourses and public perceptions of development coexist in dynamic tension, mediated by bureaucracies, mass media, politics, and popular culture. The next section explores some of these tensions in the more specific context of Peru. Before doing so it is important to emphasize the extent to which these discourses have become global in reach, and in so doing have created powerful tensions with local perspectives. An indication of this is the growth of anti- and postdevelopment movements. Dissenters from universal development discourse regard it as a Faustian threat to individual autonomy and cultural diversity (Berman, 1997). Such resistance is most apparent in relation to the growth first discourse, particularly with its recent emphasis on market-led economic growth (e.g., Mehmet, 1995). But the two other discourses described above also attract criticism. For example, Illich (1992:88) suggested “basic needs may be the most insidious legacy left behind by development,” while Esteva and Prakash (1997:283) argued that “. . . any conception of universal rights—to education, for example—is controversial and a colonial tool

8

C O P E S TA K E

for domination.” Hickey and Bracking (2005:862) also note that the (Western) “ethnocentric character of this discourse remains a problem as does the issue of how the weakest members of society mobilize to actively claim their rights, suggesting a need to look beyond rights based approaches.” Some of this criticism can be viewed as an attempt to redress the balance in emphasis between the three components of wellbeing identified above. In particular, zealous advocacy of universal needs and rights discourse risks overwhelming and distorting the local, vernacular, and idiosyncratic narratives of individuals, communities, and grassroots movements that are essential for effective and meaningful action in pursuit of their own goals. However, critics emphasize a unity of development discourse that transcends such detail. Rahnema (1997:ix) observes how the very diversity of voices participating in policy-oriented development discourse is part of its attraction, so long as debate does not “question the ideology of development” and the assumption of “its relevance to people’s deeper aspirations.” The key point here is that development is more than just discourse, but part of a powerful and professionalized bureaucratic nexus with its own interests that imposes its views of the world not only through language but also though its far-reaching practice. It is acceptable to suggest the leopard change its spots, so long as it remains a leopard. Development discourse is dangerous because it is aligned with a global apparatus that justifies its existence by using language (of problems, poverty, need, rights, underdevelopment) that are to some degree self-perpetuating, particularly when the labels are internalized by the people thereby stigmatized or labeled as lacking in some way (Escobar, 1995). At this point the discourse of antidevelopment acquires a strongly deterministic streak in its skepticism of the power of reform of global institutions. Ironically, this echoes both Marxist criticism of bourgeois charity and neoliberal critique of the possibility of a benign state. An unstated but even more fundamental assumption of antidevelopment criticism is that having neutralized or removed the development industry an alternative (postdevelopment) dynamic of grassroots action will emerge that both delivers human wellbeing in greater measure and is expunged of the tendency to create precisely the kind of bureaucracy that was wished away in the first place. Pieterse (2001:111) welcomes the “shift toward cultural sensibilities” but fears the “ethno-chauvinism” and “reverse orientalism” that would result from reification of indigenous and local culture. Too pure and dogmatic a critique of development, he points out, risks replacing it with its shadow, and is an abdication of the messier and more complex task of political engagement with the details of development both as discourse and as practice. This discussion takes us back to the central argument concerning the relationship between development and wellbeing. The antidevelopment and postdevelopment literature is useful in deconstructing development in its Western ethnocentrism, tendency to centralization and bureaucratic hegemony, selfinterest in intervention, and more. But the philosophical foundation required for reconstructing development is broader. Renewed reflection on the nature of human wellbeing is part of this, as is greater attention to culture.17 In particular, it provides opportunities to reassert development as a holistic endeavor, thereby

INTRODUC TION AND OVERVIE W

9

challenging the way it tends to be carved up into specialist subfields and disciplines oriented each with a bias toward some particular aspect of wellbeing that may be important but cannot be viewed in isolation.18 To realize the potential of a more reflexive approach to development it must be systematically informed by voices and narratives with firsthand knowledge of poverty in ways that are not sanitized by the same old bureaucratic, professional, and disciplinary machinery. Pieterse (2001:163) observes that the tendency toward “authoritarian high modernism” stemming from the Western Enlightenment accounts for much of the past failure of development effort, but that hope resides in its capacity for transformation through “reactions to and negotiations of the crises of progress.”

1.3. Peru Context This section briefly relates these different development perspectives to contemporary policy debates in Peru and to different ways of depicting its current state of development. A more detailed exploration of Peruvian political economy is presented in chapter 7.19

1.3.1. Growth First Peruvian economists and policy makers have generally agreed that sustained economic growth is important, but differed over appropriate macroeconomic policy, trade stance, level of state intervention, and scope for redistribution to bring it about. Since Fujimori’s accession to the presidency in 1991 policies have generally conformed to neoliberal, Washington Consensus views. Indeed two Peruvian economists have had a particular influence over this perspective: Kuczynski has not only served as Peru’s Finance Minister, but also contributed actively to debate over the Washington Consensus (Kuczynski and Williamson, 2003); while De Soto has championed the cause of microenterprise, particularly through consolidation of property rights (De Soto, 2001). Partially as a consequence, successive governments have pursued a combination of relatively conservative fiscal and monetary policies, domestic market deregulation, public sector reform, and further external trade liberalization, most recently in pursuit of a controversial free trade agreement with the United States (Crabtree, 2006). These policies, in parallel with improvements in the global economic context resulted in real annual GDP growth averaging 2.9 percent per year between 1995 and 2005 (1.2 percent in per capita terms). 20 Exports of goods and services over the same period grew by 8.5 percent per year, and with imports rising by only 1.9 percent per year Peru’s creditworthiness greatly improved. Inflation fell to single figures in the mid-1990s and remained there. Final consumption expenditure of households and government over the same ten year period grew in real terms by 2.4 percent and 3.0 percent per year respectively. However, the economy barely surpassed its size in the 1970s, and continued reliance on export growth left it vulnerable to future trade shocks. Moreover, given the legacy of high income inequality and outward

10 Table 1.2

C O P E S TA K E

National poverty rates for 2000 Total population No. (’000)

National Lima Other coastal urban Coastal rural Highland urban Highland rural Jungle urban Jungle rural

Share (%)

Absolute poverty incidence No. (’000)

Share (%)

Rate (%)

25,625 7,400 4,552 1,326 3,235 5,742 1,548

100 29 18 5 13 22 6

13,863 3,345 2,417 854 1,433 3,761 797

100 24 17 6 10 27 6

54.1 45.2 53.1 64.4 44.3 65.5 51.5

1,822

7

1,261

9

69.2

Source: UNDP (2002).

orientation (with growth concentrated particularly in irrigated coastal areas, mining, retailing, and other urban services) the effect of this economic performance on poverty incidence was modest. For example, in 2004, 54.3 percent of people were found to be below the official poverty line and 36.6 percent below the extreme poverty line. This represented a slight fall from 2001, when the corresponding figures were 51.6 percent and 31.8 percent. 21 Table 1.2 illustrates the disparity in poverty incidence between different zones of the country: the highest incidence and largest share still being located in the rural highlands, notwithstanding the rising incidence of poverty in Lima and other urban areas as well.

1.3.2. Needs First The main prescription of the donor community for overcoming the failure of economic growth to have a greater effect on poverty has not been to reject neoliberal economic policy but to augment it with more active state-led social programs. Aid and debt cancellation have increasingly been linked to support for such social policies alongside compliance with macroeconomic policy and good governance standards. Peru has been less susceptible to these pressures than more indebted and aid dependent countries, but the discourse of basic needs and poverty reduction has still been influential. For example, it has informed criticism of the very low proportion of government spending allocated to poverty and child welfare (e.g., Parodi, 2000; Vasquez et al., 2002), and many voices have been raised in criticism of the inefficiency of the state social programs that do exist (e.g., Tanaka, 2001; Copestake, 2006). Table 1.3 illustrates changes in Peru’s performance as measured by the human development index (HDI).22 The overall HDI has also improved steadily since recovery from the economic crisis of the late 1980s, as has the difference between its GDP per capita and HDI rankings.

11

INTRODUC TION AND OVERVIE W

Table 1.3

Peru and the Human Development Index

Year

1987

1993

1998

2004

HDI score

0.753

0.694

0.737

0.767

HDI rank

74

91

80

82

Life expectancy (year)

63

66

69

70

Adult literacy (% of population)

85

88

89

88

GDP per person rank minus HDI rank

0

⫺3

7

12

Source: UNDP Human Development Reports for 1990, 1996, 2000, and 2006.

Table 1.4 Human Development Index by department, 2000 Mainly coast

Mainly highland

Mainly jungle

Tumbes Piura

0.620 0.561

Cajamarca Huanuco

0.495 0.494

Loreto Amazonas

0.621 0.515

Lambayeque

0.625

Pasco

0.575

San Martín

0.553

La Libertad

0.613

Junin

0.578

Ucayali

0.565

Ancash

0.577

Huancavelica

0.460

Madre De Diós

0.650

Lima/Callao

0.747

Ayacucho

0.488

Ica

0.667

Apurimac

0.457

Arequipa

0.635

Cusco

0.537

Moquegua

0.666

Puno

0.512

Tacna

0.681

Note: Departments in each column are listed from North to South. Data presented later in the book comes from the three departments shown in bold (see chapter 2). Source: UNDP (2002).

Peru’s first national “Human Development Report” (UNDP, 2002) provided estimates of the human development index for each department and province in the country. The latter are reproduced in table 1.4, and indicate that the HDI is generally highest on the coast and lowest in the highlands, with predominantly jungle areas occupying an intermediate position.

1.3.3. Rights First One criticism of the needs first perspective in Peru is that it encourages a technocratic view of poverty—as an absolute state of deprivation—in place of a more

12

C O P E S TA K E

political emphasis on relative poverty and processes of exclusion. A deeper political analysis concerns the incentives of different stakeholders to address poverty. Figueroa (2001a, 2001b, 2003) develops a formal political economy model that first explains the persistence of unequal access to employment, financial services, and social protection, and then examines why it is difficult to construct a governing coalition committed to addressing the needs of the less educated and culturally subordinated racial majority.23 Tanaka (2002) adds a geographical dimension by pointing out that political incentives to address extreme poverty in more remote areas are much weaker than those in favor of tackling less acute and mostly urban poverty, particularly in and around Lima. He attributes this in turn to the weakness of political parties and networks linking government and society. Although the Velasco reforms greatly weakened the rural oligarchy in much of rural Peru, subsequent events have served to perpetuate the political dominance of richer, more educated and generally whiter Peruvians, particularly in Lima. Prospects for more progressive state social policy are weak so long as political capital can most effectively be accumulated through economic liberalism coupled with populist social policy and mass media management. The above suggests that a combination of cultural, political, and institutional constraints limits the incentive of political leaders, hence government, to instigate a stronger propoor development strategy; furthermore, external donor pressure on them to do so is also relatively weak. If so, then it can be argued that the critical driver for a more egalitarian development strategy is pressure from grassroots social and cultural movements. Figueroa (2003) describes this as the need for a “refoundational shock” to correct the colonial shock that pushed the country into a path of unequal development in the first place. In Peru, universal human rights discourse has spread in part through the damage wrought by the Shining Path (Sendero Luminoso) insurgency and the Fujimori government’s subsequent abuse of civil and political rights (Tanaka, 2003). Popular resistance to Shining Path guerrilla insurgency in the form of village militia (rondas campensinos), resistance to Fujimori’s authoritarianism, the work of the Truth and Reconciliation Commission, decentralization efforts, and gradual institutionalization of popular consultation in local government have all contributed to this process. However, consolidation of political parties around policies, interests, and grassroots organization rather than personalities and populist platforms remains at best shallow (Orias, 2005). Chapter 7 explores these issues in more depth.

1.3.4. Other Perspectives Rights discourse inherently entails making claims to universality, and in Peru this exposes those who employ it to the charge of imposing a Western ethnocentric cultural perspective that undermines the culture and aspirations of people from other traditions. A recurrent point of reference in this regard for Peru is the debate over the nature of Andean culture (lo Andino). Scholars who emphasize the importance of a distinctive Andean perspective in their analysis of rural poverty include Orlove (1974), Isbell (1973), and Flores (1977), while Doughty

INTRODUC TION AND OVERVIE W

13

(1970), Lobo (1982), and Altamirano (1988) are among those who brought the same analysis to bear on urban poverty. According to these authors, lo Andino reflects a distinctive historical legacy, language, values, and traditions with rich variation from one region to another. This “cosmo-vision” also provides the basis for a distinctive view of development practice, as exemplified by the work of PRATEC, for example (Apfell-Marglin, 2003). Another example of this perspective is provided by Masías (2002) in his rejoinder to Hernando De Soto, who he criticizes (along with Sendero Luminoso leader Abimael Guzmán) for advocating a false path of development, deviating from an authentic Andean path based on traditions of community, reciprocity, and a holistic balancing of different aspects of life and wellbeing.24 Other scholars have been quick to accuse those who defer too strongly to local worldviews of being paternalistic. Starn (1991), for example, criticizes Isbell (1973) for “essentializing” peasant experiences, and overemphasizing the unchanging nature of Andean culture. The position adopted by Starn has in turn been criticized, amongst others by Poole and Renique (1992) who argue that ignoring lo Andino risks allowing Western cultural discourses in Peru to predominate, thus invalidating ways of life and means of production that remain distinctly Andean.25 One potential benefit of wellbeing discourse is to provide a conceptual framework that is broad enough to accommodate these differences. De Vries and Nuijten (2003) accept that Andean peasants often act in accordance with universal Western models: of rational pursuit of material self-interest, for example. But they also argue that their behavior is informed by more complex values, including reverence to a cultural otherness linked to the uniqueness of the Andean environment and history. The work of anthropologists on the subjective and internal meaning of life to people and communities has also contributed to more applied research on the disconnections (desencuentros) that can undermine the goals of development agencies working in the Andes (e.g., de Vries and Nuijten, 2003; Vincent, 2004; Bebbington et al., 2007). Anthropologists have also helped to explore the importance of semantic differences for development. For example, the words waqcha and apu are arguably the closest conceptual synonyms in Quechua for “poor” and “rich” respectively. However waqcha translated literally refers to an orphan who lives without parents, relatives, or social networks. This conception of poverty suggests that close relationships and social networks are considered to be an important asset in Andean societies, with both intrinsic and instrumental value (Altamarino, 1988a:27). It reminds us that a person’s wellbeing is often less influenced by government policy or livelihoods dynamics than by the joys and sorrows of ongoing relationships with family and neighbors.26 A practical response on behalf of many development agencies has been to seek better forms of consultation. A leading example here is the “voices of the poor” program of the World Bank.27 DFID and World Bank (2003) reports on the first participatory poverty assessment for Peru that set out to be national in scope. Primary data collection was carried out in nine communities (in Lima, Puno, Ayacucho, and Piura) and involved 730 participants. Findings are summarized in table 1.5.

Table 1.5 Summary findings of Peru’s first national participatory poverty assessment 1. Families confronting poverty 1.1 Women still bear the brunt of household reproduction. Migration of men for work and alcohol abuse make things worse, but male violence in the household is declining. 1.2 Access to health is a major problem, as is physical security, especially in urban areas (partly due to gangs, drugs, and youth unemployment). 1.3 In crisis (unemployment, illness, harvest failure) what matters most is the support of immediate and extended family. 2. Poverty and the world of work 2.1 Agriculture is stagnant, being affected by land shortage, uncertain weather, and market instability. Livestock rearing is generally more productive than growing crops. 2.2 Long hours, abuse from employers, and low and uncertain income are common experiences in urban areas. 2.3 Rural-urban links are crucial to consumption smoothing: rural areas as a source of food, urban areas as a source of income. 2.4 Land rights are a major issue in rural areas. Employment protection, education, and property rights are the main concerns in urban areas. Access to formal credit is an issue in both areas, despite the risks of indebtedness. 3. Poverty and institutions 3.1 3.2 3.3 3.4

3.5

3.6

3.7

Public and community institutions are uncoordinated. Discrimination and maltreatment is common. Public officials are inefficient and corrupt. FONCODES and PRONAA are the organizations that have the greatest impact and work most closely with community level organizations.a Water and electricity are valued highly, and even those on low incomes are willing to pay for these services so long as they get an efficient service. Education is the most important long-term solution to poverty, but the public system has many deficiencies. The high cost of uniforms, materials, and contributions for special events are a problem. Poor health is the main factor leading to increased poverty. Access to health services is impeded not only by high costs, but also by discrimination and verbal abuse from health workers. With respect to security, rural areas benefit from the presence of rondas campesinas and teniente gobernador (lieutenant governor). In contrast, the police are a threat, and the justice system is distant and corrupt. Municipal authorities do not respond to the concerns of the poor, whereas NGOs and church organizations are more reliable and effective. Communal organizations are important in rural areas, but less responsive to the interests of women and those in extreme poverty. Neighborhood organizations in urban areas are scarce and weak, especially once basic services have been secured. Their leaders are easily coopted by politicians. Women find mothers’ clubs and communal kitchens to be their main source of support in the struggle against poverty, although their involvement in food distribution brings unavoidable internal conflict. Continued

INTRODUC TION AND OVERVIE W

Table 1.5

15

Continued

4. Proposals of the poor 4.1 4.2

4.3

More government support is needed to reduce market uncertainty, and to improve access to credit and to technical assistance. There is scope for more participation in administration of schools and social programs, and for better coordination between government agencies, as well as more transparent legal processes for securing land titles. Women need more training opportunities on how to deal with male violence. Nutrition and food programs need to be more reliable. Health policies and programs need to work with traditional medicine and family networks, rather than ignore them.

Note: a FONCODES is the fund for compensation and social development (Fondo de Compensacion para el Desarrolo Social) and the main central government agency for organizing public works programs. PRONAA is the national nutrition program (Programa National de Alimentos) and supplies milk and other food supplements through comedores populares (community kitchens), nurseries, and mothers’ groups. Source: DFID and World Bank (2003): 222–27.

Participatory appraisal exercises of this kind that seek to identify wider public views on development provide a potentially important counterpoint to topdown policy perspectives. Apart from being focused on poorer people, they bear a close family resemblance to more mainstream opinion polls and focus groups. In Peru, as in other countries there is indeed a movement to incorporate broader questions about what people think and feel into national sample surveys, thereby generating a battery of so-called subjective indicators through which different aspects of development can be monitored (discussed further in chapter 4). A leading example is the annual survey of Latinobarómetro, from which the data reproduced in table 1.6 is taken. This indicates that Peruvians in general felt more negatively about their country than Latin Americans in general: indeed the first three rows indicate that they were more negative about their government, the state, and operation of the market economy than people in any other Latin American country. They were also distrustful of other people in general, though in this respect they are more typical of Latin America as a whole. They did not generally believe the country was progressing, that they lived better than their parents did, nor that the life of their children would be better than theirs. And they perceived themselves to be relatively poor. To sum up, this section has indicated that visions of development in Peru are complex, diverse, and contested. The three dominant global perspectives presented in the previous section are all influential, while the experience of Sendero Luminoso illustrates Peruvians’ openness to new and more radical imported visions, particularly among young people. Peru is also fertile ground for more home grown ideologies that are critical of any imported perspective. In seeking to understand how different views relate to each other we have also suggested

16 Table 1.6

C O P E S TA K E

Some Peruvian attitudes in regional perspective

Question (abbreviated form)

Do you approve of the current president? Yes Do you trust that taxes will be well spent by government? Yes In general would you say you are very/fairly satisfied with the way the market economy works in (country)? Would you say that you can trust most people? Yes Would you say that this country is progressing? Yes Do you believe your children will live better than how you live today? Would you say your parents lived better than how you live? Imagine a 10-step ladder, 1 is where the poorest people live and 10 the richest. Where would you stand?

Peru (%)

Latin America (%)

Peru’s rank (of 18)

16 10

49 21

18th 18th

12

27

18th

16

NA

8th

22

31

12th

49

54

13th

67

55

3rd

3.34

3.66

15th

Source: Corporacion Latinobarometro (2005).

that there is much to be learnt from systematic surveys of what Peruvians think and feel. It is appropriate therefore to turn now to a consideration of the empirical basis for this book.

1.4. Methodology 1.4.1. Overall Scope of Research This book reports on research conducted as one part of the wider WeD research project on wellbeing in developing countries described in section 1.1. A Latin American view, along with African and Asian views was regarded as essential to the research. Peru was selected within Latin America mainly on the strength of established research links between staff at Bath University, PUCP (Pontificia Universidad Católica del Perú), and UNCP (Universidad Nacional del Centro del Perú), through which a common interest in multidisciplinary research that locates poverty and inequality in wider political and cultural processes of social inclusion and exclusion had already been established (Figueroa et al., 2001; Altamirano et al., 2004). Guidelines for data collection under WeD were worked out within the research group over a period that started well before the onset of ESRC funding

INTRODUC TION AND OVERVIE W

17

in September 2003 and continued even beyond the planned date for starting fieldwork of May 2004. These entailed trying to reconcile different personal, disciplinary, and country-specific preferences with the use of comparable methods in each country. For example, the Peru team was keen to build on its earlier research on social exclusion (Altamirano, 1996) and it took time to explore how this related to the perspectives of UK participants. Different priorities also emerged over the cost, feasibility, scope, and importance of different methods, as well as the appropriate balance and sequencing of quantitative and qualitative data collection. These were recognized to be integral to the underlying research goal of developing new approaches to understanding wellbeing.28 The initial brief for data collection in Peru, as in the other three countries, was to obtain information on wellbeing over at least a year in at least two urban and four rural sites. The resulting database was expected to permit the following: (a) mapping of similarities, differences, and patterns in wellbeing during the year between different people and groups; (b) analysis of the relationship between different ways of defining and measuring wellbeing at individual, household, and group level; and (c) interpretation of the causal processes behind such variation, including the influence of government policies and other contextual factors. It was agreed that data collection should wherever possible be carried out through sustained and trusting relationships between respondents and field researchers living in the selected research sites. In the case of Peru, this entailed recruitment of a team of six graduate anthropologists from UNCP, each of whom took prime responsibility for data collection in one site. A seventh site was added later and shared between two of them. These researchers (four women and two men, including two native Quechua speakers) went through a rigorous selection and induction process. There was no turnover in this team during the research period, although budgetary constraints meant that the contracts of only four of them could be extended from July 2005 to February 2006.29 Section 1.4.3 reviews the different data collection instruments that were employed.

1.4.2. Selection of Research Sites Empirical research in Peru was not intended to enable general statements to be made about the country as a whole, but to reflect important dimensions of national diversity in order that the relevance of universal ideas about wellbeing could be tested against the views of people living in contrasting contexts. It was also accepted that site selection should build on prior geographical experience of the researchers and would need to take into account logistical constraints. Discussion of how to make the best selection led to the idea of adopting a corridor approach. This refers to the idea of identifying sites to reflect as far as possible the diversity of conditions along an interconnected East-West transect of Peru, linking coast, mountain, and jungle. By “diversity” we had in mind a broad and interconnected set of variables, including: (a) altitude, ecology, and

18

C O P E S TA K E

natural resources; (b) accessibility and integration with external markets; (c) degree of urbanization and quality of infrastructure; (d) proximity to centers of political power; and (e) the relative influence of Western and indigenous culture and values.30 The corridor selected for the research stretches from a large “shanty town” on the outskirts of Lima, through the Mantaro valley to small villages in the highlands of Huancavelica and the cloud forest on the Eastern slopes of the Andes. This corridor is polarized, since it links the richest part of Peru (Lima) with one of the poorest (Huancavelica) in a relatively short distance. Table 1.7 provides a very brief description of each site, which also reflects diversity in social organization and culture of participation.31 At the center of the research area is the Mantaro Valley and the city of Huancayo, capital of the Department of Junin. Progreso is a poor neighborhood located within the city itself. Many of its inhabitants are migrants who came from other parts of the highlands during the 1980s. Descanso is a not untypical Mantaro farming town: Spanish-speaking mestizo (mixed race), with irrigated land close to the river and pasture stretching high into the surrounding hills. The district town of Alegria (along with annexes Llajta Iskay and Llajta Jock) also lie within the economic hinterland of Huancayo city and close to the Mantaro River, but at a point where the river narrows into a gorge and enters the Department of Huancavelica. It is influenced by both the Spanish-speaking mestizo culture of Huancayo and the Quechua-speaking indigenous culture of Huancavelica. Selva Manta, in Junin Department, is inhabited mostly by people

Figure 1.3

Map of the research sites

Table 1.7

A brief description of the research sites in Peru

Name, altitude, and distance by road from Lima

Region, type, and population

Brief description

Llajta Iskay 3,400m 380km

Huancavelica (Rural– highlands) 365 Huancavelica (Rural– highlands) 212

Annex of Alegria with poor road access. Mostly Quechua speaking. High rate of migration to Huancayo, Lima, mines, and jungle: few immigrants. Annex of Alegria. A smaller and more closeknit community than Llajta Iskay. Mostly Quechua speaking. High rate of migration to Huancayo, Lima, mines, and jungle: few immigrants. Hamlet in a steep valley on the Eastern slopes of the Andes in the district of Monobamba. Spanish speaking. Comprises migrants from Huancavelica and other parts of Junin. Total evacuation during the violence, and since for education and business. Seasonal immigration for sugarcane and coffee harvesting. Farming town and district center in Tayacaja province with good road access to Huancayo city. Mostly bilingual. Some immigration from more villages as well as outmigration to Lima, Huancayo, the central jungle, and mines. Farming town and district center in the Mantaro Valley. Almost entirely Spanishspeaking, with easy access to Huancayo city. Some immigration, mostly for marriage. Migration out to Lima, central mines, and jungle, especially for education. Two neighborhoods on barren hillside overlooking the city of Huancayo. Bilingual. Residents mostly arrived in the 1980s from Huancavelica but also from Ayacucho and some highland villages of Junin. Large settlement (part of the district of Atí Vitarte) in hills to the east of Lima, founded in 1984. Mostly residents arrived in early 1990s from the Central Andes. Many are bilingual, but very few non-Spanish speaking.

Llajta Jock 3,300m 365km

Selva Manta 1,400–1,800m 290km

Jauja province of Junin (Rural–cloud forest) 560

Alegria 3,000–3,500m 355km

Huancavelica (Peri-urban– highlands) 5,440

Descanso 3,275m 290km

Junin (Peri-urban– highlands) 5,323

Progreso 3,275–3,325m 310km

Junin (Urban– highlands) 1,560

Nuevo Lugar 550–900m 35km

Lima (Urban– coast) 150,000

Source: Compiled from primary sources by the WeD peru research team.

20

C O P E S TA K E

born and raised in and around the Mantaro Valley. But its climate is more tropical, and access by vehicle is from the north through Tarma and La Merced rather than directly from the Mantaro Valley. Finally, Nuevo Lugar is part of the distant Lima-Callao metropolis some 300 kilometers from Huancayo. Being located close to the highway leading into the central highlands from which many of its inhabitants originated it is psychologically less distant from the other sites than pure geography might suggest. Chapter 3 provides a more detailed description of the seven sites.

1.4.3. Data Collection Methods Table 1.8 summarizes data collection activities in these sites between March 2004 and February 2006. These started with compilation of secondary data about each site, in the process of which field workers also introduced themselves to leaders and officials connected to each, taking great care to make it clear that they were not associated with any particular development agency or program, but that the fieldwork followed on from their anthropology studies at UNCP. The next phase of the work was to conduct open-ended interviews with a quota sample of men and women in each community designed to elicit broad perceptions of respondents about their quality of life. By the time this task was completed the researchers had become well-known in their fieldwork sites and this facilitated collection of more factual information, using what became known as the resources and needs questionnaire (R ANQ). This was a standard survey instrument, intended to permit comparisons of resources and needs not only across the seven Peru sites but also across the four WeD research countries.32 Where possible the researchers restricted further interviews to people covered by the R ANQ survey so as to permit subsequent crossanalysis. However, while this enabled them to build up personal relations with many respondents it also created a problem of respondent fatigue. This was addressed mainly by relying on smaller sample sizes in subsequent data collection exercises. The next major exercise was application of the Peru WeDQoL: a set of psychometric scales whose design was based closely on findings from the ECB, and which was designed to permit quantitative measurement and analysis of states of subjective wellbeing.33 This aspect of the work is described in more detail in chapter 3. The R ANQ and WeDQoL data together was intended to permit exploratory empirical analysis across the four countries of the association between measures of subjective and objective wellbeing. Subsequent fieldwork was designed to permit further analysis and interpretation of such relationships. Broadly, it focused on three mediating processes: migration, collective action, and the household economy. Accordingly, data collection during the latter part of the fieldwork focused on these three broad areas.

21

INTRODUC TION AND OVERVIE W

Table 1.8

WeD data collection methods

Activity

Dates

Community profiling

Mar– Compile secondary data about resources December and structures in each site. Conduct an 2004 inventory of all forms of social organization. Construct seasonal calendars (farming, festivals, and illness). Case studies of major conflicts. May–June Semistructured interviews with 419 2004 individuals across all seven sites to explore main values, goals, perceived resources, happiest and unhappiest experiences. July– Single visit questionnaire-based interview September of 1004 households across all seven sites 2004 (including all those covered under the ECB) to collect factual information on household resources, basic need satisfaction, and life satisfaction. Mar– May Two rounds of interviews with an initial 2005 sample of 550 individuals (including 302 from 208 households covered by the R ANQ) using three-point-scale closed questions on goals, values, adequacy of resources, personality, identity, and subjective wellbeing.b Mar–April Semistructured interviews with 71 key 2005 informants, including migrants, return migrants, and relatives of migrants (sample drawn from R ANQ sample). May–June Qualitative case studies of one faena 2005 (collective action initiative) in each site, and one “Glass of Milk” committee, using participant observation and key informant interviews. Jun 2005– Three round survey of 254 households February also covered by R ANQ. Supplementary 2006 sections on durable consumption goods (R1), migration (R2), and “Glass of Milk” program (R3).

Wellbeing study (ECB)a

Resources and needs questionnaire (R ANQ)

Quality of life survey (WeDQoL)

Migration study

Collective action case studies

Income and expenditure survey

Description

Use Ch. 2

Chs. 3,5

Chs. 1,2,6

Chs. 3,4

Ch.5

Ch.6c

Chs. 4,5

Notes: a ECB stands for encuesta de bienestar. b 176 of these respondents were also drawn from 134 households covered by the income and expenditure survey. c Copestake (2006) analyses and discusses the data collected on the Glass of Milk program. It is also discussed in chapter 7.

22

C O P E S TA K E

1.5. Some Illustrative Data from the Research Sites Whether data collected in the selected sites would reveal insights into the relationship between development and wellbeing depended in part on there being sufficient variation in wellbeing indicators between them. To test this we report below on data from the R ANQ about two variables: educational attainment as an indicator of satisfaction of human needs; and general happiness as an indicator of subjective wellbeing.34 The findings confirm significant differences between the sites. That respondents in the two urban sites were generally better educated but less happy than those elsewhere also encouraged further enquiry.

1.5.1. Education Figueroa (2003) argues that access to formal education in Peru is a critical factor in reproducing social exclusion and inequality more generally. R ANQ data permits an analysis of the extent to which the probability of advancing beyond primary school is associated with site location. 41.2 percent of all adults in the 1,004 households covered by R ANQ had not progressed beyond primary education, and this rate was much higher in rural areas. Table 1.9 shows results

Table 1.9

Education above primary level by research site: Logit estimatesa Site/variable 15–65 year Primary Coefficient olds (total education 2,282) or less (%)

Nuevo Lugar Progreso

Urban– coast Urban– highlands Descanso Peri-urban– highlands Alegria Peri-urban– highlands Llajta Rural– Iskay highlands Llajta Jock Rural– highlands Selva Rural– Manta forest Women (all sites) Constant

T-Test Marginal Effects

645

26.0

⫺0.630

⫺4.73* ⫺0.145

465

37.0

⫺0.033

⫺0.24

491

38.5

(Reference case)

387

52.2

0.581

4.12*

0.143

133

78.9

1.873

7.95*

0.421

62

74.2

1.628

5.26*

0.375

99

58.6

0.853

3.72*

0.210

0.766

8.31*

0.182

⫺0.895

⫺8.33*

(Included in figures above)

⫺0.008

Note: a 1 = primary or less; 0 = other levels of education. Asterisk indicates significance at 1 percent level. Reference case is men living in Descanso. Pseudo-R squared = 0.093. Results are for all 16 to 65 year olds not still in the education system.

INTRODUC TION AND OVERVIE W

23

of using a Logit model to estimate the likelihood of not completing primary school by research site. Differences from the benchmark case of working-age people in the Mantaro Valley district town of Descanso were significant in all cases except Progreso (the urban shanty town in Huancayo). Inhabitants of Nuevo Lugar in Lima were 14.5 percent less likely to exit schooling this early, whereas those in the rural sites were more likely to do so. In the extreme case, an inhabitant of Llajta Iskay was 42.1 percent more likely to exit formal education before entering secondary level. These estimates control for gender differences in the composition of each sample within each site, women being additionally 18.2 percent less likely to receive education beyond primary level.

1.5.2. Global Happiness As a rough and preliminary test of wellbeing within the sample, a global happiness question was addressed to heads of household during the R ANQ survey. The happiest heads of households were on average found in the Huancavelica district town of Alegria, where 25 percent reported being “very happy” and only 10 percent “not too happy.” At the opposite extreme, only 4 percent reported being “very happy” and 21 percent being “not too happy” in Progreso.35 In order to test for differences in responses between sites an ordered Probit model was estimated with global happiness being regressed onto a group of site dummy variables (see table 1.10). A dummy variable was added for womenheaded households in Model II to test whether women heads were significantly happier than male respondents. The coefficients can be interpreted by their sign and significance. In both models, household heads from Alegria report significantly more happiness than those in Descanso, whilst heads of households in Progreso were significantly less happy. Women heads of household were less happy than men (Model II).36 This confirmed that where people lived along the selected corridor did significantly affect how they felt about their wellbeing.

1.6. Outline of the Book Chapter 2 (Resources, Conflict, and Social Identity in Context) takes a historical and ethnographic approach to explaining how internal and external factors influence wellbeing in the seven chosen research sites. The first section briefly reviews the history of the region with particular emphasis on the racial/ethnic dimensions of political conflict, including the Spanish conquest, Independence, foreign investment, agrarian reform, and terrorism. It then reports on a small survey into how people in each research site perceived their own social identity. The main part of the chapter then takes the form of brief descriptions of the natural, material, human, social, and cultural resources in each site, interspersed with case studies of recent internal conflicts—between municipal officials and the leaders of communal farmers’ associations, for example.

24

C O P E S TA K E

Table 1.10 Global happiness by research site: Ordered Probit estimations a Variables

Model I Coefficient Standard error

Model II T-Test

Coefficient Standard error

T-Test

Nuevo Lugar

⫺0.180

0.116

⫺1.550

⫺0.065

0.127

⫺0.510

Progreso

⫺0.428* 0.458* 0.053 0.108

0.125

⫺3.420 3.680 0.280 0.540

⫺0.451* 0.453* 0.046 0.127

0.126

⫺3.580 3.640 0.240 0.630

⫺0.360

Alegria Llajta Iskay Llajta Jock Selva Manta

⫺0.079

0.124 0.192 0.202 0.218

Women heads Cut 1 Cut 2 Sample Pseudo-R Squared

⫺1.133 1.239 999 0.038

0.095 0.097

⫺0.110

0.125 0.192 0.202 0.218

⫺0.500

⫺0.231*

0.101

⫺2.290

⫺1.180 1.201 999 0.041

Note: a (0=Not too happy, 1=fairly happy, 2=very happy) Asterisk indicates significance at 1 percent level. Reference case in Model I is respondents living in Descanso, and in Model II is men living in Descanso. Additional models also found statistically significant links between (a) perceived adequacy of income and (b) self-perception of income relative to neighbors.

Chapter 3 (Subjective Wellbeing: An Alternative Approach) presents an original methodological approach to measuring wellbeing without reliance on externally dictated theories and indicators. The introduction explains the conceptual framework, the epistemological (emic and post-hoc) approach and methodology. After a brief discussion of initial qualitative findings it derives three broad categories of latent needs from the data: place to live better, build a family, and progress from a secure base. These three variables lay the foundation for linking personal development to wider processes of social and economic development taken up in the following three chapters. Variation in the importance respondents attached to each of them as well as in their satisfaction with achieving them is then analyzed statistically by research site and according to the sociodemographics of respondents. The chapter then uses these measures to develop an integrated and quantitative model of the quality of life, linking need perception with perception of available resources, prevailing values, and personality. This model provides an important antidote to reliance on predetermined assumptions about the psychological processes affecting wellbeing. Chapter 4 (Economic welfare, Poverty, and Subjective Wellbeing) compares a classification of households according to standard economic welfare measures, with how individuals within them thought and felt. The introduction includes

INTRODUC TION AND OVERVIE W

25

a brief review of secondary literature linking the two in Peru. Data is then presented for each site on income, expenditure, and poverty at the household level. These familiar welfare indicators are then compared statistically with (a) an index of global happiness and (b) the latent need importance and satisfaction indicators developed in the previous chapter. We find that material welfare and subjective life satisfaction indicators are not necessarily positively correlated with each other: indeed some are negatively correlated. Chapter 5 (Wellbeing and Migration) is the first of three chapters that moves from measuring wellbeing to exploring how it is generated and contested through specific social processes. Migration is an important determinant of wellbeing not only of those who move, but also those left behind. After setting out a conceptual framework for thinking about migration, the chapter provides a detailed profile of patterns of migration into and out of the selected research sites. In contrast to the usual emphasis on migration as a response to economic incentives it links migration to the three latent needs identified in chapter 3. It draws on qualitative data to explore in more detail the effect of migration on the relationship between parents and children. Chapter 6 (Wellbeing and Institutions) explores how a holistic understanding of wellbeing helps to explain the evolution of institutions in the selected research sites. In so doing it criticizes theories of institutions that emphasize their economic and political functions at the expense of their social, cultural, and emotional roles. The introduction defines what we mean by institutions (particularly forms of informal collective action) and briefly reviews the literature surrounding their origin and evolution both generally and in an Andean setting. This is followed by an analysis of variation in the mix of institutions by each research site, and discussion of how they support the different components of subjective wellbeing developed in chapter 3. Eight detailed case studies are then presented into the way faenas (community self-help projects) connect values, goals, resources, and peak happiness experiences. Chapter 7 (Reproducing Unequal Security: Peru as a Wellbeing Regime) returns to the wider policy context, locating the three different development discourses presented in the opening chapter within a general theory of national policy regimes. This starts by considering external “conditioning factors” affecting the research area including globalization, rising consumerism, and the evolution of the state. It then considers the changing “institutional responsibility matrix” including decentralization and second generation neoliberal reforms, as well as their interplay with rights-based, religious and paternalistic traditions of welfare provision. Third, it considers the relevance of an expanded inventory of “wellbeing outcomes” (material, social, and symbolic) to measuring and evaluating these changes. Finally, it explores regime “reproduction consequences,” particularly with respect to centralization of wealth and power, alienation, the political tolerance of inequality and the politics of social identity as well as of self-interest. The chapter argues that a wellbeing (rather than pure welfare) perspective on policy emphasizes both “freedom to” act and “freedom from” harm—increased personal security being critical to both.

26

C O P E S TA K E

Chapter 8 (Conclusions and Implications for Development Policy and Practice) opens with a summary of foregoing chapters and then highlights four ways in which the discourse of wellbeing they explore can be useful. First, it creates conceptual space within which conflicts and development disconnects can be traced back to fundamentally different views of the nature of human wellbeing. Second, it opens the way for more informed use of psychological techniques for assessing what matters most to people, how satisfied they are with their life and why. Third, it supports the case for a more subtle analysis of policy that acknowledges how rational self-interest is compromised by collective values, ideologies, and impulses. Fourth, faced with the complex outcomes of crosscultural action across diverse and contested wellbeing domains and discourses it supports the case for more reflexive, improvised, contextual, and devolved forms of development management. Chapter 9 (Implications for Wellbeing Research and Theory) suggests new lines for research into subjective wellbeing. First, it argues that there is scope for further use and extension of research methods and analytical approaches pioneered in this book. Second, it suggests lines for combining them with analysis of the history, archaeology, and evolutionary psychology of wellbeing. Third, it suggests connections with research in the fields of genetics, molecular biology, and neuroscience. Finally, it argues for using locally grounded subjective wellbeing models of the kind developed in chapter 3 to systematically plan, monitor, and evaluate experimental development interventions to suit particular communities and contexts. Notes 1. A UK example is the negative connotation attached to “property development.” This is also reflected in the rise of postdevelopment and antidevelopment discourse (see section 1). Even without referring to its pejorative usage Clark (2002:22) distinguishes more than thirty different ways in which the word is used in the social sciences. 2. Also people cannot be quite so quickly othered by being labeled (and often belittled) as the poor—as if they belonged to a different species. If poor people are mostly just like everyone else, only with less money, then it is understandable to treat them as people first and poor second. 3. The euphemistic term “developing countries” further illustrates the point about ambiguity of the term development. The more substantive point is that WeD funding was explicitly to conduct research in low and middle income countries, in contrast to the high income country context of the bulk of previous research explicitly into subjective wellbeing. 4. Dean (2003) provides an analysis of multiple understandings of wellbeing implicit in the different discourses that influenced initial planning and design of WeD research in the UK. 5. A cross-country synthesis book drawing on work in all four countries is also planned. Meanwhile working papers across the whole program are available at www.welldev.org. 6. For example, in reviewing secondary literature on wellbeing and development in Peru, Altamirano et al. (2004) found it useful to distinguish between studies that emphasized how wellbeing was molded by relationships to things (material), other people (social), and ideas (cultural).

INTRODUC TION AND OVERVIE W

27

7. Gough (personal communication) points out that these three dimensions can be identified in Aristotle’s Nicomachean Ethics. Aristotle identifies eudaimonia as the highest goal in life and this is normally translated into English as “happiness.” Happiness is such an end “because we always choose it for itself, and never for any other reason’ (Ethics I vii). However, recognizing that pleasures can be f leeting, he goes on to define the happy man as “one who is active in accordance with complete virtue, and who is adequately furnished with external goods, and that not for some specified period but throughout a complete life’ (I x). This introduces three further ideas. First, the idea of activity (energeia)—of exercising one’s powers and realizing one’s capabilities through time (though, given the accidents of fortune, this can take the form of enduring hardships). Second, it introduces the idea of virtue (arete) since “virtuous acts have the greatest permanence” (I x). Aristotle grounds virtue in good actions within the context that a man finds himself. Finally, he recognizes that happiness requires external goods, “for it is difficult if not impossible to do fine deeds without any resources” (I viii). 8. For example, building on the work of Doyal and Gough (1991) and of Ryan and Deci (2001b) we can identify needs for health, autonomy, competence, and relatedness. These in turn require inter alia a set of intermediate need satisfiers, such as food, childhood security, and so on, which have material foundations. 9. See Gough and McGregor (2007:11–16) for a review of different theories of basic needs. Reference to “in isolation” is important because it allows for situations where harm, even death, is voluntarily accepted in the name of some higher goal. 10. While OWB is by definition revealed through physical states and actions (including ownership of assets, allocation of time, the consumption of goods, and the use of services) that are in theory observable by others, its measurement in practice often relies on subjective statements of respondents (e.g., how much money and leisure they say they have). Hence it cannot be assumed that data on OWB is necessarily more reliable than that for SWB. Of course, both approaches to understanding wellbeing are real and important. 11. This contrasts with a rationalist analytical perspective, which emphasizes how decisions are made to maximize goal achievement subject to resource constraints. Recognition of the importance of individual and cultural constraints on such rationality is growing. For example, low wellbeing influences peoples’ actions and interactions through their influence on self-esteem, confidence, and the “capacity to aspire” (Appadurai, 2004). Similarly, preferences are bounded by cultural understandings, and “preference constraints” are themselves endogenous to development processes. 12. In a more comprehensive analysis of the different “foundations of knowledge” underlying academic disciplines Bevan (2007) identifies nine components: focus, values, ontology, epistemology, theorizing, research strategy, key conclusions, rhetoric, and praxis. 13. Lines show three sets of connections that render a discourse consistent and meaningful; arrows show three ways in which these can be weakened. 14. Useful examples of books that attempt this are by Hunt (1989) and Pieterse (2001). Copestake (2005) suggests that four dominant discourses can be distinguished in the past fifty years: comprehensive planning, basic needs, neoliberalism, and policy management. 15. In focusing on the main contemporary views I have not included a column for older, statist models of development—in contrast to Table 7.1 in Raczynski

28

C O P E S TA K E

16.

17.

18.

19. 20. 21. 22.

23.

24. 25.

26.

27. 28.

29.

(1998), for example. The last two columns echo the distinction made by Wood and Gough (2004:321) between the historical role of “far sighted elites” and popular social movements in building welfare regimes. See Williamson (2003) for ref lections on the original conceptualization of the Washington Consensus and Fine (2002) for a more critical view. Rodrik (2006) provides a useful update that, while continuing to emphasize the primacy of economic growth, suggests ten additional components of an “augmented” Washington consensus. Rao and Walton (2004) indicate the extent to which some people within the development industry are taking culture more seriously. Radcliffe and Laurie (2006) articulate some of the worries about the dangers of what they refer to as a “new paradigm of culture and development” that is neglectful of historical and geographical variation and contestation. This echoes arguments picked up by Pieterse (2001) in his chapter on “critical holism and the Tao of development.” It is surprising, however, how little this chapter refers explicitly to wellbeing: this being an indication of how recently wellbeing has entered into the lexicon of development theory. See also Altamirano et al. (2004) for a more extensive review of literature on poverty in Peru. Figures come from “Peru at a glance” on the World Bank web site. Figures were taken from the “interactive poverty map” on the National Institute of Statistics and Information (INIE) web site. This is a multidimensional indicator of human development that takes into account indicators of per capita income, health, and education. See any UNDP Human Development Report for a more detailed description. See also Copestake (2004, 2006) for a summary and critique of this model. This analysis has also been echoed by the World Bank, which regards factor market imperfections and the restricted size of the domestic market for goods and services arising from inequality as not only a source of ill-being but also a significant brake on economic growth (World Bank, 2005). In the terminology developed here this constitutes a bid to appropriate selected aspects of a rights first agenda into an income first agenda. For discussion of Andean reciprocity, kinship, and exchange see Mayer (2002), Mayer and De la Cadena (1989), and Golte (1980). Degregori (2000) provides a comprehensive review of the history of anthropology in Peru, including the political tensions between Peruvian and foreign anthropologists. For a Marxian critique of postmodern peasant studies see Brass (2002). Coxshall (2005) provides a telling example. Even the personal tragedies resulting from conflict between Shining Path and government can only be fully understood in the context of the way they affected (and were mediated by) kinship relations. These consultations were an input into the World Development Report 2000/01 on poverty. See http://www.worldbank.org/poverty/voices Although data collection started in March 2004, final details of data collection (including an extension to February 2006) were agreed only at a meeting of all WeD country teams in March 2005. The team (who are all coauthors of chapter 2) was supervised in the field up to June 2005 by an experienced anthropologist, who had previously worked as a lecturer in anthropology at UNCP and who was a graduate of PUCP. From July

INTRODUC TION AND OVERVIE W

30.

31.

32.

33.

34. 35. 36.

29

2005 to February 2006 they were supervised by a graduate sociologist from PUCP. The concept is also used in Peru, by USAID for example, to delineate and promote supply chains with growth potential stretching from the coast into the interior of the country. This is described by Abraham and Platteau (2004) as a spectrum from personalized relations between culturally differentiated members of large urban settlements (e.g., Nuevo Lugar) to culturally homogenous members of a small rural community (e.g., Llajta Iskay). Tanaka (2001) echoes this in his discussion of the relationship between community, clientelist, and broking forms of leadership and participation in sites of relatively high (Nuevo Lugar, Progreso), medium (e.g., Alegria, Descanso), and low (e.g., Llajta Iskay, Llajta Jock, Selva Manta) “complexity” respectively. The R ANQ was divided into six parts: household organization; global happiness; human resources; material resources; social resources; and cultural resources (language, social identification, and honorific titles). The global happiness part comprised one question: “taking all things together, how would you say things are these days; would you say you are very happy, fairly happy, not too happy?” Like other questions this was subject to translation and back-translation tests from English into Spanish and Quechua. The initial research proposal had been to adapt one of the quality of life instruments developed under the auspices of the World Health Organization. This was rejected in favor of an instrument design that relied more on semistructured interviews with potential respondents for scale construction, rather than compliance with the WHOQoL protocol for adapting its international instruments using focus group discussions with expert key informants. Statistical analysis of the education and happiness data was carried out by Tim Hinks. As is expected with this kind of question, the majority of respondents chose the middle category. Inclusion of this dummy variable affecting the coefficient on A in particular, where 182 household heads interviewed were women, compared with just 83 men.

This page intentionally left blank

Chapter 2

Resources, Conflict, and Social Identity in Context José Luis Álvarez, Maribel Arroyo, Lida Carhuallanqui, James Copestake, Martín Jaurapoma, Tom Lavers, Miguel Obispo, Edwin Paúcar, Percy Reina, and Jorge Yamamoto

2.1. Introduction Our relationship with physical things, other people, and ideas (hence material, social, and symbolic dimensions of our wellbeing) are profoundly affected by where we live as well as the “place” of different localities in history. Here we concur with Long and Roberts (1984:3) that local and regional analysis of development is an important complement to national and international. Central Peru has been profoundly affected, for example, by changing terms of engagement with the national and global economy: altering the link between livelihood and land, increasing labor mobility, and exposing people to other cultural perspectives through new forms of media. Moreover this engagement is quite different from that of other parts of the country. The main purpose of this chapter is to sketch the geographical setting and historical context of our research, allowing for deeper reflection into regional differences as well as similarities. This entails describing salient physical and social aspects of each research site as well as the wider region in which they are located. More ambitiously, the chapter also begins to explore how sites are perceived and connected in the minds of their inhabitants. In doing so, we draw also on the concept of “imagined communities” to consider the idea of “place” in a historical, social, cultural, and political as well as a geographical sense (Anderson, 2003).1 Section 2.2 provides a brief introduction to Huancayo and the Mantaro Valley, with some references also to Huancavelica. Section 2.3 draws on community profiles compiled by the field team to describe each of the seven research sites in turn. The chapter also draws on secondary literature and discussions by the authors with many local actors, including other social scientists working in the region.

32

Á LVA R E Z E T A L .

2.2. Regional Context 2.2.1. The Corridor Concept The idea of “economic corridors” has been promoted in Peru by government and aid agencies as part of a strategy for development of strategic supply chains linking coast, highland, and jungle.2 The concept was adopted by the WeD research team to facilitate selection of a set of research sites to reflect as much variation as possible in contextual factors within the region, including altitude and access to natural resources, population density and degree of urbanization, the relative importance of local, national, and global trade, proximity to centers of political power, and ethnicity and language. The concept is particularly applicable to the chosen research area, given the economic importance of the central highway between Lima and the interior of Junin and Huancavelica. However, flows of money, people, and goods can neither be neatly segmented into discrete channels, nor does the idea of a corridor exist within the popular imagination.3 Here we focus first on the Mantaro Valley, then briefly contrast this with a perspective from Huancavelica.

2.2.2. The Mantaro Valley in Historical Perspective Although not much is known about habitation of the Mantaro Valley in the preInca period, archaeological and linguistic evidence suggest it was the center of a unified Huanca kingdom immediately before the Inca conquest in 1460.4 More controversial is the argument, advanced by historian Waldemar Espinoza, that the Huanca were sufficiently united and strong to greet the Spanish in 1532 as liberators and potential allies. José Maria Arguedas built on this interpretation of the conquest by arguing that this alliance (a factor in persuading Pizarro to locate his first capital in the Mantaro Valley at Jauja, and then on the nearest part of the Pacific Coast at Lima) allowed the Huanca nobility and their subjects to escape the extremes of colonial servitude and exploitation experienced by other Andean groups. Others have argued that this simplified version of a distinctive Huanca history can also be seen as a more recent process of reinvention of tradition and identity (Álvarez, 2005; Romero, 2004). Studies of colonial history in the valley suggest that while the gradual establishment of a dominant mestizo (mixed race) landowning class was far from rapid or smooth, it advanced further and faster here than in the Southern highlands, where a stronger tradition of absolute rule by Spanish overlords (gamonales) emerged.5 Key to the process were matrimonial alliances across racial and ethnic barriers, the establishment of nucleated farming towns on each side of the river (such as Descanso), the role of fraternal alliances (cofradias) of landowners and peasants within each town, and the slow decline of a distinct indigenous landowning aristocracy of kurakas (Álvarez, 2005). The role of mestizo landowners in the nationalist struggle for independence from Spain during the 1820s remains a matter of debate, but the conflict is

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

33

likely to have created further opportunities for its consolidation as a class, as well as hastening the gradual disappearance from the valley of campesinos of pure indigenous ancestry. Consequently, the 1879–83 war and Chilean occupation did not pitch white landowning collaborators against Indian peasant guerrillas as starkly as it did further south. Rather, relatively successful resistance facilitated the emergence of a self-consciously patriotic mestizo social identity, in which hierarchy in control of land was conflated with communal and reciprocal traditions of work and festivity. In the 1920s, when anthropologists began to document the livelihoods and folklore of the valley, they could advance the idea of economically differentiated but culturally homogeneous communities: each with a distinctive identity linked to craft products, patron saints, festivals, weekly markets, and control of communal land. From there it was a small step for Arguedas and others to develop the idea of an even wider imagined community of mestizo peasants and landowners less deeply stratified by radicalized hierarchy than elsewhere in Peru (Álvarez, 2005). But if there ever was a golden age of harmonious communities in the Mantaro Valley then it didn’t last long. The penetration of capitalist relations in agriculture was greatly accelerated by the arrival of the Cerro de Pasco mining company and construction of the railway from Lima to Huancavelica at the beginning of the twentieth century. A substantial literature explores its impact on the regional economy, including the effect of wage income from miners on agrarian structure (Long and Roberts, 1978; Mallon, 1983). In rural areas, differential access to cash incomes stimulated the formation of a land market and the emergence of a class of commercial farmers selling food, wood, clothes, and shoes to meet growing demand from the mines and cities. These relied to varying degrees on reciprocal and communal forms of labor mobilization as much as on wage labor. In some villages, relatively equal land distribution combined with rising education to facilitate communal activities and infrastructural development: the hydroelectric project in Muquiyauyo (built in 1908) being a celebrated example (Adams, 1959). But more commonly smaller peasant farmers were unable to raise productivity to match rising off-farm employment opportunities. As a result their contribution to agricultural output became increasingly marginal relative to their dependence on seasonal migration. Population growth, farm differentiation, and informal small-business opportunities arising from increased demand for nonfood goods and services fuelled migration and urbanization. The city of Huancayo grew rapidly into the dominant regional trading, financial and supply center, linking mining and agricultural sectors, and controlled more by mercantilist than by landowning interests (Long and Roberts, 1978:70–87). Up to the 1960s it was still possible to view all of the above as part of a chaotic but progressive process of economic development, with the mining sector as the major engine of growth, pulling small-scale trading and more progressive farmers along behind it. However, by the 1970s direct and indirect employment creation as a result of mining and related activities was faltering. Major business interests in Huancayo began to switch their center of operation to Lima, weakening its small industrial base in favor of commerce and government

34

Á LVA R E Z E T A L .

services. The agrarian reform of 1969 did little to help small-scale farmers, being mostly restricted to upland pastoral zones and jungle estates (Long and Roberts, 1978:248–253). The agricultural sector throughout the country performed badly during the 1970s, with highland producers in particular finding it hard to compete with producers on the coast and abroad (Crabtree, 2006).6 Instead they became increasingly dependent on seasonal migration and remittances, including those sent back from rapidly growing towns in the jungle, such as Canchamayo and Satipo (CVR, 2003:136; INEI, 2005). Meanwhile in urban areas the struggle for secure employment accentuated the importance of ethnicity and education (Figueroa, 2003).

2.2.3. Violence and Poverty in the 1980s and 1990s The 1980s began with elections and heightened political expectations, but neither the government of Belaunde in the first half of the decade, nor that of Alan Garcia in the second, proved able to handle them. In the context of the wider “lost decade of growth for Latin America” the failure was in part economic: Peru’s GDP recorded an average annual fall of 1.2 percent through the 1980s, led by falling earnings from the mining sector (Sheahan, 1999:48). But even more traumatic was the growth of the Shining Path movement, fuelled in part by personal experience of poverty and exclusion from the prevailing political and economic system (Starn et al., 1995; Sheahan, 1999:33). The final report of the Truth and Reconciliation Commission (CVR, 2003) dedicates a chapter to the effects of the Maoist insurgency in the central highlands, which it defines as the Departments of Junin and Pasco, plus the Northernmost Provinces of Huancavelica. It starts by noting the region’s strategic importance (as the shortest transport link from Lima to both mountain and jungle, and its reputation as the most prosperous highland region. It also emphasizes its hybrid mestizo identity: a commercially oriented economy and popular culture that combined growing consumerism with many traditional Andean characteristics and had avoided the extremes of agrarian conflict experienced elsewhere in the country (CVR, 2003:137). These features, as well as the rapid defeat of the MIR (Movimiento de Izquierda Revolucionario) insurgency in 1965 led many people to assume it would be resistant to the revolutionary doctrines and tactics of both Shining Path (Partido Comunista del Peru—Sendero Luminoso) and MRTA (Movimiento Revolucionario Tupac Amaru). Events initially bore this out: during the early 1980s, acts of violence in Junin were less frequent than to the south. However, both terrorist groups greatly strengthened their presence in the region during this time: MRTA through the arrival of leaders from Cuzco in 1984, and Shining Path through the arrival of insurgents from Huancavelica, Ayacucho and Apurimac after the army entered these departments in 1982. Shining Path opened its first guerrilla zone in the Chaupihuaranga valley of Pasco, and built strong networks among students, particularly at the National University of Central Peru (Universidad Nacional del Centro del Peru), as well as in poorer settlements surrounding

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

35

Huancayo city.7 Although much smaller than Shining Path, MRTA infiltrated unions, student groups, and peasant associations. It established a strong presence in the jungle areas of Chanchamayo and in the cloud forest areas (including Selva Manta) stretching from there up to the Mantaro Valley. Destruction of infrastructure, hostage-taking, summary trials, and assassination of police, municipal, and communal leaders all increased rapidly, particularly after 1985. By 1987, Shining Path was present throughout the region; it controlled popular committees in many communities and its armed columns roamed freely. According to the Truth and Reconciliation Commission, the maximum number of recorded deaths and disappearances in the region in a single year was 785 in 1990, this being out of a total 3,618 recorded for the period from 1980 to 2000 (CVR, 2003:145). This includes 903 in the Mantaro Valley (including Huancayo city), 782 in the two Northern provinces of Huancavelica, and 1,556 in the central jungle areas. Local NGOs estimated that by 1990 approximately 15,000 displaced people were living in the Mantaro Valley (compared with more than 300,000 nationally). But it became increasingly hard to distinguish between internally displaced persons and other migrants, particularly as the “IDP” label became a means for securing support for relief and reconstruction (Stepputat and Sorensen, 2001:775). The government of Alan Garcia declared a state of emergency over the whole region in 1989. The presence of the army gradually spread out from the main barracks in Huancayo, Jauja, and La Merced into surrounding areas. Public confidence in the army was low, being fuelled by accounts of atrocities elsewhere, and this was initially reinforced by some indiscriminate punitive killings in communities in the Mantaro Valley (CVR, 2003:141). However, the Truth and Reconciliation Commission concluded that the tactics of the army in the region were mostly more restrained than further south, as well as more effectively directed toward a strategy of rebuilding confidence, accumulating intelligence, carrying out targeted actions, and supporting civilian self-defense committees through provision of training and firearms (CVR, 2003:143). This resulted initially in an increase in deaths and disappearances, but gradually the tide turned. MRTA’s strength waned first, after a series of setbacks in 1989, including the celebrated Battle of the Mills (Enfrentamiento de los Molinos) at which 67 MRTA militants were killed by the Peruvian army. By 1992, and with the capture of Abimael Guzmán, Shining Path also faced military defeat in the region. Violence nevertheless continued sporadically throughout the 1990s, with Shining Path retaining a minor presence in some jungle areas beyond 2000. Recovery from the national crisis at the end of the 1980s was gradual. The economic stabilization program introduced by Fujimori in August 1990 tackled the hyperinflation he had inherited, but resulted in two more years of economic stagnation and rising poverty (Diaz, Saavedra, and Torero, 2002). Although a long period of economic growth and gradually rising average incomes followed (interrupted briefly in the late 1990s) this was also associated with increasing inequality, mitigated only partially by the expansion of public investment and social protection programs.8 Agriculture in the highlands was also adversely affected by competition with the coast and from food imports, contributing to a

36

Á LVA R E Z E T A L .

sharp fall in real agricultural prices. For example, the five year average real farmgate prices for potatoes, wheat, and coffee for 1996–2000 was respectively 34 percent, 29 percent, and 36 percent of those a decade earlier, while the volume of food imports was 70 percent higher (Crabtree, 2006). Income was increasingly skewed in favor of farmers and traders with more land and better commercial connections, including Huancayo’s infamous “potato kings”—wholesalerscum-moneylenders with entrenched control over marketing of crops from particular localities. Meanwhile, poorer households sought to diversify their activities but were unable to stem the flow of out-migration (Escobal, 2001). In urban areas, the struggle to secure employment has spurred increased investment in private as well as public education, but often of poor quality. The availability of cheap imported goods has stimulated the retail sector, with fast-food chains and illegal imitation of branded consumer goods being other areas of growth. Some financial institutions (most notably the Caja Municipal of Huancayo) have flourished by attracting deposits built up from remittances and through investment in micro-businesses, mostly in the service sector. But the tendency for more successful business people and professionals to migrate to Lima and abroad means the city lacks a substantial resident business class. This social fluidity combined with unemployment and consumerism goes some way to explaining the growth of youth delinquency and gang culture. The presence of so many relatives outside the region, and the growing share of imported goods in the market, has also changed the regional identity: with Huanca nationalism persisting at least as much in the minds of the valley’s diaspora as those of its current inhabitants.9

2.2.4. A View from Huancavelica* The research sites in the district of Alegria are economically oriented toward Huancayo both through trade in goods and movement of people, but at the same time linked administratively and linguistically to Huancavelica. Hence some understanding of the latter is also necessary in setting the context for the research. Table 2.1 reveals some of the differences between the two departments and also Lima. The table shows that Huancavelica has a smaller and more rural population. Internal rural to urban migration is relatively low (Huancavelica city having a population of only 30,000) owing to the prevalence of migration out of the department altogether. GDP per person is also smaller and more reliant on agriculture than elsewhere. The Huancavelica regional economy contracted sharply in the second half of the 1990s, whereas it grew in Lima (along with much of the coast), and remained stagnant in Junin. Average formal education of those more than 25 years of age is less and illiteracy higher, particularly for women. The incidence of poverty and extreme poverty is also much higher. Cultural differences are reflected in the much higher proportion of children aged 5 to 14 who speak Quechua, as well as a slightly lower incidence of single mothers. The province of Tayacaja (where Alegria is located) occupies an intermediate position

37

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

Table 2.1

Comparative statistics for Lima, Junin, and Huancavelica departments

Department

Year

Peru

Lima

Junin

H’velica

Population (‘000)

2002

26,745

7,748

1,247

443

Population density (’000/sq. km) Population increase (% per year) Urban population (%)

2002

20

222

28

20

1993–2002

NA

2.0

1.5

1.1

1981

NA

1.0

60.0

25.0

Urban population (%)

2002

NA

1.0

67.0

29.0

Life expectancy (years)

2000

68.7

73.2

67.2

64.2

Adult illiteracy (%)

2000

10.7

3.9

11.9

27.5

Women’s illiteracy (%)

2000

16.0

6.2

17.6

38.3

Average adult education (years) Children speaking Quechua (%) Single mothers (%)

2000

8.1

10.0

7.6

4.6

GDP (million Soles 1994 prices) GDP (million Soles 1994 prices) GDP growth (%)

2000

NA

NA

8.0

59.8

2000

17.7

21.3

18.2

15.3

1995

107,039

50,155

4,420

940

2001

121,513

4,498

777

1995–01

57,462

13.5

14.6

1.8

⫺17.3

GDP per person

2001

13,815

6,473

3,545

2,122

Share of agriculture (%)

1995

13.3

1.9

13.9

24.8

Share of agriculture (%)

2001

NA

1.8

14.8

25.7

Poverty incidence (%)

2004

51.6

37.1

52.6

84.4

Extreme poverty (%)

2004

19.2

4.2

18.3

59.9

Source: UNDP (2002); PEISA (2003); INEI (2005).

between the departmental averages for Junin and Huancavelica: demographically close to the latter, but economically influenced by its commercial orientation toward the city of Huancayo, and in the more densely populated north of Huancavelica rather than the larger but more sparsely populated and pastoral south. Key informants emphasized a strong work ethic in Tayacaja, but also greater extremes of wealth, with many richer farmers being able to own their own transport and educate their children privately in the more Westernized schools in Huancayo city. The city of Huancavelica lies 150 kilometers south of Huancayo with direct road and rail links. It grew up originally to serve the nearby mine of Santa

38

Á LVA R E Z E T A L .

Barbara, which in the sixteenth century supplied the Spanish empire with much of the mercury it needed to extract and purify silver and gold. However, it is also the point of contact between the commercial economy and the more selfsufficient peasant farming communities to the east and south of the department. The city is also an entry point for growing numbers of government and nongovernment organizations, attracted there in part by Huancavelica’s dubious status as the poorest department in the country, with a Human Development Index of just 0.439.10 Part of Huancavelica’s low economic status can be attributed to the drain of educated people to Huancayo and Lima. In contrast, many of those who stay can recounted stories of being “badly treated” in these places and maintain their cultural orientation in the opposite direction. Cultural polarization is evident in the reluctance of people to use Quechua in the presence of strangers, and by separate queues for indigenous and white people outside banks and local stores. Among young and unemployed people there is a continued sense of oppression and cultural conflict between the communal peasant life and unfulfilled aspirations for better formal education and employment. Why is Huancavelica such a poor region? For many residents, the answer is both simple and profound: because the most important resources in the region (human, animal, agricultural, mineral, hydroelectric) are exploited to serve the interests of outsiders.11 The faded colonial style of the city itself testifies to this: aristocratic houses and lavish churches serving as reminders of the period of mining exploitation. During the colonial period, indigenous people were employed as servants or even as pongaje (effectively slaves). However, resources were directed toward foreign markets, with mine owners reinvesting very little in the region. Instead they encouraged centralism and the plunder of regional resources: the railway line to Huancayo and highways to the coast all built to reinforce these distribution channels. In the case of agriculture, merchants from Huancayo dominate the markets, extracting low prices from farmers (the word huanca is regionally used to mean “negotiator”). The trade in Alpaca and Vicuna wool is also controlled by outsiders. To conclude, although Alegria lies only on the edge of Huancavelica and is economically oriented toward Huancayo, it can be expected that many of its residents feel a stronger cultural and political ambivalence toward the modernization and globalization that the cities of Huancayo and Lima represent. This can be portrayed in stereotypes: of the more business-oriented and urban savvy Spanish-speaking mestizo in the Mantaro Valley, and of the Quechua-speaking Indian communero in Huancavelica. But it is more accurate to anticipate a many layered cultural landscape, providing people with opportunities to develop nuanced and indeed multiple identities, “cultural styles” and performances, deeply influenced but not wholly dictated by place of origin and race.12

2.3. Social Identities by Site By social identity we refer to how people are perceived or labeled by others. This is of course highly political. The same label can also carry multiple, ambiguous, and constantly changing meanings depending on who is using

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

39

it and in what context (Wright-Revolledo, 2007) and this presents a methodological dilemma. On the one hand, it is evident that social identities have an important bearing on wellbeing: both directly, through their inf luence on how people feel about themselves and are perceived by others; and indirectly, through their inf luence on status, power, and access to resources. On the other hand, use of such labels in field work is fraught with difficulties, both ethical and of interpretation. For example, a question in the R ANQ survey that asked people to describe themselves yielded data that was very difficult to interpret: most people opting simply to state where they came from in geographical terms. This section is based instead on closed questions included in the first round of the WeDQoL survey, comprising lists of social labels that the field team knew to be widely understood in the region. Although it is likely that some answers were influenced by the respondent’s perception of the social identity of the interviewer, most had been interviewed by the same person at least once before and were familiar with their presence in the locality. The data does also reveal significant differences between sites as well as insight into the underlying complexity of the issue. In our discussion of the Peruvian context, we have already alluded to the way the colonial settlement established a radicalized class hierarchy between blanco (white, of Spanish birth), criollo (white of Peruvian birth), mestizo, and indio (indigenous) categories (Manrique, 1999; Quijano, 2000). We have also described attempts to influence these labels by reviving serrano (highland) Andean cultural identity (lo andino). A further important evolution in social identity that is associated strongly with migration is emergence of the label cholo. Although its meaning remains fluid and much debated, a first approximation to a definition is of the peasant migrant in the city who gradually fuses indigenous and serrano identities with other influences (Manrique, 1999:6). The remaining labels used in the questionnaire are more explicitly racial in origin: negro (black), charapa (Amazonic Indian), and chino (East Asian, but not just Chinese as illustrated by reference to Fujimori as “El Chino” despite his Japanese ancestry). The sample comprised 550 individuals. Of these, 45 percent were men, 27 percent less than 25 years of age, 63 percent between 25 and 45 years of age, and 10 percent more than 45 years of age. Gender and age were not significantly different between sites. In contrast, there were significant differences between sites regarding residential status and religion (see table 2.2). In Llajta Iskay, Llajta Jock, Alegria, and Descanso, more than three-quarters of respondents had been resident there for more than 15 years, and a majority had been born there. In the two urban sites, in contrast, the majority were born in a “very different place” and had been resident for less than 15 years. Selva Manta represented an intermediate case; a higher proportion of respondents there were also Protestant. Turning to social identity, 329 of the 550 respondents were willing to describe themselves using 1 of 9 terms offered to them: mestizo, serrano, blanco, cholo, indio, negro, criollo, charapa, and chino. Of these respondents, 52 percent described themselves as mestizo, 25.2 percent as serrano, 11.2 percent as

40

Á LVA R E Z E T A L .

Table 2.2 Demographic details of the WeDQoL sample Llajta Llajta Selva Alegria Descanso Progreso Nuevo All sites Iskay Jock Manta Lugar Residence in site 1–5 years n

6–15 years

7

3

3

8

6

26

18

71

%

11.7

5.0

10.0

8.0

6.0

26.0

18.0

12.9

n

6

7

11

13

17

41

62

157

36.7

13.0

17.0

41.0

62.0

28.5

% Total

10.0 11.7

n

60

60

30

100

100

100

100

550

Where born? Here n

54

37

18

80

92

13

0

294

60.0

80.0

94.8

13.0

0.0

53.7

% In a very different place Total

90.0 61.7

n

1

3

1

10

0

71

65

151

%

1.7

5.0

3.3

10.0

0.0

71.0

65.0

27.6

n

60

60

30

100

97

100

100

547

49

8

90

91

79

83

449

28.6

95.7

95.8

80.6

84.7

84.4

28

94

95

98

98

532

Religion? (otherwise Protestant) Catholic

n %

Total

n

49

83.1 81.7 59

60

Source: WeD Peru: WeDQoL survey.

blanco, and 3.6 percent as cholo. (see table 2.3). Site differences were significant. The term serrano was used mostly by respondents in the urban and, to a lesser extent, the peri-urban sites. The term blanco was used more by people in rural sites, particularly those in Huancavelica. In 67.7 percent of cases, the respondent used the same category that they applied to themselves to describe their community as well. The main divergences were that 7.3 percent of respondents called themselves mestizo but their community serrano, and 8.5 percent described themselves but not their community as blanco (see table 2.4). The second of these divergences is explored further here, on the assumption that it might reveal more about how social identity differs between sites. Table 2.5 compares the use of the label blanco in response to six different questions, between-site differences in the frequency of responses being statistically significant in all cases. Responses to Question 1 reveal, quite contrary to what might be expected, that a higher proportion of respondents (one third) categorized their community as blanco in Llajta Iskay than in any other. More of them also categorized

Table 2.3

Self-categorization by site

Percent of site Llajta responses Iskay

Llajta Jock

Selva Alegria Descanso Progreso Manta

Nuevo All Lugar sites

Mestizo

53.3

70.0

90.0

53.3

56.7

41.7

30.5

52.6

Serrano

13.3

3.3

0.0

23.3

30.0

36.7

40.7

25.2

Blanco

26.7

20.0

10.0

15.0

6.7

5.0

6.8

11.2

Cholo

3.3

0.0

0.0

1.7

3.3

10.0

3.4

3.6

Criollo

0.0

0.0

0.0

0.0

0.0

1.7

15.3

3.0

Indio

3.3

0.0

0.0

3.3

1.7

1.7

0.0

1.5

Charapa

0.0

0.0

0.0

1.7

1.7

1.7

3.4

1.5

Negro

0.0

6.7

0.0

1.7

0.0

1.7

0.0

1.2

Total

30

30

30

60

60

60

59

329

Source: WeD Peru: WedQoL survey.

Table 2.4 Self and community categorization of social identity compared Community categorization

Self-categorization Blanco Mestizo Indio

Blanco

No

9

% Mestizo

No

Indio

Cholo

% Negro

Criollo

Serrano

Total

Cholo

All sites

Negro Criollo Serrano Charapa

6

0

0

0

0

2

0

17

2.7

1.8

0.0

0.0

0.0

0.0

0.6

0.0

5.1

17

137

1

1

2

2

10

2

172

%

5.2

41.6

0.3

0.3

0.6

0.6

3.0

0.6

52.2

No

1

1

2

0

1

0

2

0

7

%

0.3

0.3

0.6

0.0

0.3

0.0

0.6

0.0

2.1

No

3

3

0

7

0

2

2

0

17

0.9

0.9

0.0

2.1

0.0

0.6

0.6

0.0

5.1

No

0

2

0

0

0

0

0

0

2

%

0.0

0.6

0.0

0.0

0.0

0.0

0.0

0.0

0.6

No

1

0

0

0

0

1

0

0

2

%

0.3

0.0

0.0

0.0

0.0

0.3

0.0

0.0

0.6

No

6

24

2

4

1

5

67

3

112

%

1.8

7.3

0.6

1.2

0.3

1.5

20.4

0.9

34.0

No

37

173

5

12

4

10

83

5

329

%

11.2

52.6

1.5

3.6

1.2

3.0

25.2

1.5

100

Note: All percentages rounded to the nearest decimal point. Source: WeD Peru: WeDQoL survey.

42

Á LVA R E Z E T A L .

Table 2.5 Use of the term blanco (white) by site Percent of site responses 1. Community

Llajta Iskay

Llajta Jock

Selva Alegria Descanso Progreso Nuevo Manta Lugar

All sites

No

10

3

1

1

0

2

0

17

%

33.3

10.0

3.3

1.7

0.0

3.3

0.0

5.2

No

8

6

3

9

4

3

4

37

%

26.7

20.0

10.0

15.0

6.7

5.0

6.8

11.2

No

3

0

2

7

3

1

0

16

%

10.0

0.0

6.7

11.7

5.5

1.7

0.0

5.0

4. Good or very good?

No

28

19

24

52

47

34

49

253

%

93.3

63.3

80.0

86.7

78.3

56.7

81.7

76.6

5. Personal ideal?

No

18

18

13

26

9

12

8

104

%

60.0

60.0

43.3

43.3

15.0

20.0

13.3

31.5

6. Not content with category

No

5

2

2

9

4

2

2

26

%

16.7

6.7

6.6

15.0

6.7

3.3

3.3

7.9

7. Respected? (yes)

No

28

25

29

44

52

58

57

293

%

93.3

83.3

96.7

73.3

86.7

96.7

98.3

89.3

2. Self

3. Interviewer

Note: The actual questions were as follows. 1. In this community the people are [ . . . ]? 2. You are [ . . . ]? 3. Classification of research investigator [ . . . ]. 4. How do most people regard the following [ . . . ]? (very bad, bad, good, very good). 5. If you had the chance to be born again, what would you be like? 6. Are you content to be [response to question 2]? (very discontent, discontent, content, very content). 7. Are [ . . . ] respected in Peru? Source: WeD Peru: WeDQoL survey.

themselves in the same way.13 In both cases, the frequency of this response was next highest in the other rural Huancavelica site (Llajta Jock). This contrasts with the categorization of the interviewers who were half as likely to categorize respondents in any site as blanco.14 Respondents in Llajta Iskay were most likely to say they regarded the majority of blancos as “good” or “very good.” A higher proportion of respondents in Llajta Iskay and Llajta Jock also said that if it was possible to be born again, then they would wish to be blanco; whereas in all other sites, the majority said they would wish to be either serrano or mestizo. Their view of the level of respect accorded to blancos in Peru was not markedly higher, this being lowest in the two peri-urban sites. This data serves as a warning against a simplistic assumption that the closer people are to Lima the more they identify themselves with being blanco. A tentative interpretation is that in the rural sites of Huancavelica fewer people are content with the way they categorize themselves and are categorized by others.

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

43

2.4. Community Profiles This section provides a brief narrative overview of each research site (including physical features, demography, culture, history, livelihoods, social organization, and external services) on the basis of information collected between 2003 and 2005. Statistics, unless otherwise stated, are for 2002.15 We start with the three small hamlets, then describe the two peri-urban district centers, and finish with the two urban settlements.

2.4.1. Llajta Iskay Llajta Iskay is located at an altitude of 3,500 meters overlooking the gorge cut by the Mantaro River after it crosses the boundary from Junin to Huancavelica. It is an annex of Alegria district, whose headquarters is 25 kilometers away by unpaved road. Llajta Iskay has a population of 365, of whom 90 percent are Catholic and the remaining 10 percent evangelical. The latter are divided into two groups, who make regular use of adjacent chapels, while a priest only rarely visits the very dilapidated Catholic chapel. The district mayor offered to help renovate the church if all denominations agreed to share it, but this was not accepted. The most important festivals are the Festival of the Cross in May and Santiago in July, when many migrants return to the village from Lima and Huancayo (see table 2.6). Quechua is the first language in nearly all households, but three-quarters of the population are bilingual in both Quechua

Table 2.6

Seasonality in the Mantaro Valley: Highlights

Month

Seasons

Crop farming

Main festivals

January February March April

Wettest period. Limited work and money

Land cultivation and weeding

New Year Carnival

May

End of the rainy Main harvest period for rain season fed crops Dry and sunny, but cold at night

June July August September October Start of the rainy season November December

Easter Festival of the Cross Santiago

Land preparation and sowing

Source: Detailed seasonal profiles for each site compiled by the field team.

All Saints Christmas

44

Á LVA R E Z E T A L .

and Spanish: children and younger people put pressure on family members and teachers to speak Spanish so that they can learn it too. Most land is held by an association of communal farmers comprising 73 members, with land distributed among households and some plots retained for communal cultivation. The land controlled by the association has been reduced through conflict, including an unresolved dispute over grazing land with a neighboring hacienda. In the last 10 years, the community tried to reclaim control of some of this land through Courts of Justice in Tayacaja and Huancavelica, but the dispute remains unresolved. Membership of the communal association is effectively compulsory for everyone over the age of 18, except the very old and even they complain of being forced to do communal work after they have retired. A small amount of privately titled land also exists and can be purchased and sold, and there is a longstanding land conflict over some of this between two of the families. Agriculture is the main economic activity (followed by cattle rising) but none of the land is irrigated. The main agricultural products are potatoes, peas, barley, and beans. These are sold internally or carried by lorry to Huancayo or Lima. All households keep domestic animals such as cows, sheep, goats, donkeys, horses, rabbits, guinea pigs, pigs, chickens, dogs, and cats. Wild animals in the area include pumas and foxes. Communal working practices are very important to the farming system and include ayni (reciprocal help between two people), minka (e.g., everyone working together for one person at harvest time, during the sowing season or to build a house), and faena (communal work for the benefit of the community, including cleaning ditches and improving roads).16 Leadership of the community is divided between the teniente gobernador (representative of the government), the justice of peace, the municipal agent, and the president of the communal association.17 There is much comparing of current with past incumbents and competition between posts, fuelled by longstanding personal rivalries. For example, a faena was organized to build an adobe lockhouse in response to anger at young people returning to the village from outside and not being controlled by their families.18 But it remained unused for some time while argument raged about who should authorize its use. Meanwhile, another faena was organized to install new water pipes. These were provided (with much fanfare) by the municipality but their quality subsequently became a matter of fierce argument. A much bigger dream is to secure electricity, but most people are skeptical that the community will ever manage to organize itself effectively enough to do so. There are no formal links with outside political parties, indeed, most people perceive themselves to be unaffected by outside politics and view elections as a chore because they are forced to travel to Alegria to vote. Politicians only come to the community during municipal or presidential elections. The Government supports a preschool and primary school with 4 teachers and 113 pupils between them, but attendance is low. Many children do not complete primary school, leaving to work in agriculture instead. Many marry when as young as 14 years old, and almost all by the age of 20, by seeking permission within the community and without servinacuy (living together first). Although many dream of securing education for their children outside their

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

45

community they also complain about being unable to finance such an investment from farming income. Meanwhile, the nursery and school buildings have been eroded by rain and are at risk of collapse. People’s commitment to the school has also been undermined by disputes with one of the teachers, suspected of stealing school materials. Most houses have piped water, but there is neither electricity nor public sanitation. There is a health center in the community, but it is rarely open owing to the absence of staff and people tend to rely on traditional medicine. According to the visiting nurse, the most frequent health problems are acute malnutrition, pneumonia, bronchial infections, intestinal parasites, and skin diseases. Additional problems affecting women include vaginal infections, inflamed ovaries, and rheumatism. The death of a prominent member of the community (probably of AIDS) prompted widespread anger, fear, suspicion, and accusations of evildoing. The community also has an administrative building; there is one public telephone and four small stores that sell groceries, including liquor and coca. Government provides food assistance through free school meals and there is one “glass of milk” group.19 The government agency PRONAA also provides products such as tuna, rice, sugar, oil, yucca flour, corn, and milk for children each month through the school. Cáritas, a Catholic NGO, supports barley production and land conservation through terracing. The Ministry of Agriculture has promoted planting of eucalyptus and pine, and a faena was organized to establish the tree nursery, though not without some internal dispute, because the site selected had traditional Andean spiritual significance to the community.

2.4.2. Llajta Jock Llajta Jock is located 11 kilometers along the dirt road to Llajta Iskay from Alegria. It is also an annex of Alegria, with a similar mix of Catholic and Protestant, but the population is smaller (212). Most people speak Spanish and Quechua, while a few older people speak only Quechua and approximately 20 percent (all younger people) speak only Spanish. Important festivals include Santiago and Jalapato in July, for which many migrants return home.20 Other festivals have been abandoned as there are so few people living there. 44 houses are inhabited, all made of adobe with tiled roofing and mostly two storey. Llajta Jock is situated on land inherited by one of four children of the owner of a nearby hacienda. The settlement was formally recognized as a separate annex in 1976. As in Llajta Iskay, the main economic activities are crop and livestock farming, and there is also an association of communal farmers. More land is in private ownership and many people work as laborers for larger farmers, earning 8 to 10 Soles per day. Some lands are irrigated and so can generate two harvests each year. Previously the only crops were barley, wheat, beans, peas, and native potatoes. But in the last five years, richer farmers with links in Huancayo have bought land in the community and begun to grow a wider range of potatoes, also introducing new types of fertilizer, insecticide and fungicide. There is no access to bank credit for the population, but loans can be obtained in

46

Á LVA R E Z E T A L .

emergencies from the association of communal farmers or from neighboring families. The communal association possesses plantations of eucalyptus that are sold when required and also owns gypsum deposits. Cáritas technicians from Alegria offer some agricultural and nutritional advice. There are three small shops, but Alegria is the main market for selling produce. There is a preschool with 28 children and a primary education center with 53 students, but they have insufficient teaching resources such as text books and other materials. Adult illiteracy is 52 percent for women and 48 percent for men. As in Llajta Iskay, the community does not place great emphasis on education, preferring children to work on the farms and in the weekly market. But increasing numbers of children do continue their education by attending the secondary school in Alegria. This is linked to a higher incidence of migration, destinations including Huancayo, Lima, and the central jungle where people go to assist with coffee harvesting (especially between January and March). Many families receive groceries, clothes, and money from relatives who have moved to Lima. There is a health center in the community, staffed by two health technicians who work in coordination with the health center in Alegria. Seventy percent of the population use the health post, with the remainder relying on traditional medicine. Common ailments are similar to those in Llajta Iskay. The community has had an electricity supply since 2003, and a bus service runs to and from Alegria twice a week. To become a member of the community, one must be 18 years old and resident in the hamlet for at least 1 year. The community has an elected council, governor, and municipal agent, and an irrigation committee that manages water resources jointly with two neighboring communities. Fifty percent of women are married while forty-five percent live with a man outside of marriage. There is only one woman who has a position in the council and one working for Cáritas. As in Llajta Iskay, the community does not engage much in wider politics, the main influence of government being the school and clinic. Ingredients for school meals are provided by PRONAA and there is a “glass of milk” committee. Many traditional customs have declined, including respect for the authorities and participation in faenas. This is mostly attributed to out-migration and children’s exposure to inappropriate behavior through television. The community has had a long running legal dispute with a private organization over control of a lime quarry. Money has been raised to pay legal costs through faenas and by appealing to relatives in Lima. Residents dream that the quarry could provide them with better income than farming, and that this would encourage more people to live there. The community also fought and eventually won a 10 year legal battle with a large landowning family over disputed land. The same family provoked much anger when their grain mill caused a break in the village electricity supply. This was eventually resolved with mediation by the school teacher, and the mill no longer operates.

2.4.3. Selva Manta Selva Manta is located at an altitude of between 1,400 and 1,800 meters in cloud forest in the northeast of the department of Junin, and in the province of Jauja.

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

47

It is an annex, located 12 kilometers from the district center and more than 30 kilometers from the nearest asphalt road. The road to the hamlet was completed in 1970 at the initiative of a timber company from Lima, supported by the community and municipality. The population is 560, of whom 90 percent are evangelical Protestants, who have services 3 times a week in their own chapel. Most of the population originate from the Mantaro Valley and speak Spanish as their first language: only a few more recent migrants from Huancavelica and Huancayo know Quechua, but rarely use it in public. The village was first established by Franciscan missionaries, the site previously having fallen within the territory of nomadic indigenous communities. Italian immigrants arrived at the beginning of the twentieth century and colonized the zone, constructing six private haciendas and investing in sugarcane cultivation and livestock rearing. Their descendents have mostly moved to local towns or Lima, but continue to control much of the land. The most important festival for the community takes place in the district center in July, marking the anniversary of the district. It lasts three days, with jalapato and traditional dances such as carnivales de Jauja, and is attended by people from other annexes and many out-migrants returning from Lima and Huancayo. Typical food is pachamanca, with ingredients such as cassava, sweet potato, meat, and the drink warapo. Other festivals have stopped because of the influence of the evangelical majority, who on occasions have also interrupted the Catholic mass. Young people increasingly socialize separately, meeting people from neighboring annexes to drink chicha and other beverages. Most people (men, women, and children) work in the haciendas as agricultural laborers earning approximately 10 Soles a day, often paid in-kind. Sugar cane is the most important product, followed by coffee and aguardiente (sugar cane liquor). The climate is warmer than in Llajta Iskay and Llajta Jock, the alluvial soils are better and cultivation is more intensive. Selva Manta does not have a communal association of farmers and most people do not own land. Products are sold in San Ramon, Lima, and Huancayo. Forest land is important for food, building materials, tools, firewood, and medicine. Streams and rivers are abundant: water wheels power the sugar cane crushers, and fish are an additional source of protein. Selva Manta possesses a great biodiversity of flora and fauna, and its residents are quick to bemoan its unrealized ecotourist potential. Selva Manta has its own primary school, and in comparison with Llajta Iskay and Llajta Jock parents are much more supportive of their children’s education. But for secondary or higher education, children must travel outside the district. Most leave school early to work in agriculture and it is common for women to have their first children by the age of 16. Temporary migration is also common for trade and social purposes: indeed many families effectively operate with rural and urban household bases, the latter most commonly in San Ramon. Employment in production of sugar cane and aguardiente attracts seasonal migrants, mostly from elsewhere in Jauja. In contrast to the thriving primary school, the health post is not working and there is no access to electricity or piped water. PRONAA supports school meals and there is a “glass of milk” committee. But the most important organization

48

Á LVA R E Z E T A L .

in the settlement is the ronda campesinas (village militia) created at the behest of the Peruvian Army in 1990. During this period conflict between the army, MRTA, and Shining Path resulted in a curfew from 18.00 hours and forced women and children to hide at night in the fields, while many moved away to nearby towns and to Lima. Although most people have returned, this experience continues to cast a shadow of mistrust over the community, and men aged 17 to 40 years are still required to belong to the ronda. They are given weapons by the Peruvian army, who periodically visit for inspection and training.

2.4.4. Alegria The first references to Alegria are found during Inca times, as one of many resting places, approximately 10 leagues apart, along the main Inca highway between Cusco and Quito. Alegria has been a district center since 1912, and this is celebrated annually. The new municipal offices on the plaza are a striking homage to modernism, complete with tinted plate glass windows and rooftop satellite dish. The district consists of hills and valleys ranging between 2,500 and 3,600 meters above sea level. Its total population in 2002 was 5,440, nearly three-quarters of them dispersed among 16 rural annexes (including Llajta Iskay and Llajta Jock), with an overall density of just less than 35 per square kilometer. According to municipal figures, the total district population grew by the surprisingly high figure of 10 percent between 1999 and 2002.21 However, these figures also indicate that the proportion of men living in the rural annexes actually fell by 2 percent, whereas their female population grew by 13 percent. In contrast, the population of men and women living in the municipal area rose rapidly: by 26 percent and 19 percent respectively. The result was a fall in those living in annexes from 72 to 69 percent of the total population, and a rise in women’s share of the population (particularly in annexes) to nearly 54 percent. Having already described two of the district’s annexes in some detail, this section presents data on the population living in the municipality itself, this having been selected as a separate research site. It can in turn be divided into six quite widely dispersed neighborhoods (barrios) separated by fields and straddling the main road between Huancayo and Huancavelica. Although they share a single town cemetery, each barrio has its own dominant families, municipal representatives, community organizations (including “glass of milk” committees), and identity. Periodic conflicts rise up within and between them: one barrio would prefer to split into two; two others are in dispute over water rights from a common stream; and a third in conflict with a neighboring annex over water rights. With most households heavily dependent on agriculture it would be misleading to describe the municipality as urban—hence the term peri-urban. Most households do have access to electricity and piped water, but only one barrio has mains sewerage and more than 90 percent of houses are made of mud brick. The number of private houses (509) compared to the number of households (approximately 350) indicates much nonresident ownership of property by exresidents now living in Lima and Huancayo who also attract residents to work in

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

49

their urban businesses. But a growing number of older people (from outside and inside the district) have also chosen to settle in the town, renting a few rooms and often living off a small pension. At district level 4,000 hectares of land (of a total area of 17,000 ha) are cultivable, and 8 percent of this can be cropped more than once in the year through seasonal irrigation from streams. Within the 6 barrios of the municipality itself, 310 people belong to a single communal farmers association. Membership is open only for those more than 18 years of age who have a partner and have been resident (and active in meetings) for at least 2 years. Most households own less than three hectares of land and sharecropping is common: the tenant taking between half and three-quarters of the income, depending on the crops and on who pays for inputs. Cultivation mostly relies on oxen, and the main crops are barley, potatoes, wheat, peas, beans, and maize. Since 1998, a particular effort has been made by the municipality and Caritas to improve production and marketing of barley. An annual barley festival in the plaza attracts competitors from throughout the province, and a grain processing center was completed in 2004. However, farmers were wary of weakening often longstanding links with private traders, and the center procured less than 30 tonnes. The Caritas technicians have also set up revolving seed funds with farmers’ groups to encourage use of new seed varieties and techniques. But they admit progress had been slow, blaming this on both the farmers’ risk-aversion and (more vaguely) cultural misunderstandings. Farmers themselves complained particularly about the time burden of attending weekly faenas to help cultivate demonstration plots. Other livelihoods include farm laboring (earning 8 to 10 Soles per day), trading (mostly women, who can earn up to 20 Soles per day) and house-building, which can be much more lucrative. Although the town has neither a bank nor a post office (and only one very erratic public telephone) there is a major market each Friday (established in 1938) that attracts cattle and sheep traders from the surrounding area and from Huancayo city. Non-farm paid work is provided by the municipality, a sawmill, a metal workshop, a petrol station, a hotel with five rooms, and three small bakeries. Many people are unemployed or underemployed. Seasonal migrants, including many secondary school leavers go to Lima, Huancayo, and to the central jungle areas in January to find work harvesting coffee, returning to Alegria for festivals in July and August. Unpaved roads and tracks link Alegria with surrounding communities. The main road from Huancayo to Huancavelica was widened and upgraded in 2004. This was the source of various conflicts. Workers came mainly from outside, some bringing their wives, but others renting local lodging and causing problems with residents that resulted in two of the three disco venues in the town being closed down. Some local people were employed by a subcontractor (at a rate of 14 Soles per day) but on several occasions it lacked funds to pay them at the end of the month, causing them to go on strike. Meanwhile owners of small shops and restaurants, and those able to rent out rooms enjoyed a temporary increase in income, while at the same time fearing that completion of the new road would eventually result in fewer passers-by stopping. In contrast, the mayor has ambitions to develop the town as a recreational and ecotourism

50

Á LVA R E Z E T A L .

center. In addition to the plaza with its attractive view, there are fossil beds, the old Inca road, and limestone caves to visit. Those with houses adjacent to the road that had to be destroyed when it was widened received compensation, but conflict arose over the amount, particularly in the case of a local lawyer. Other residents argued that he obtained and extended his house deliberately to get more compensation. He then refused to vacate it until the Road Ministry agreed to give him more money, thereby delaying settlement and completion of the works for the whole community. The president of the communal association also claimed he had not paid the proper amount for the house. Others accused the lawyer and his wife of being greedy by putting their own interests ahead of those of the town. The lawyer in turn made accusations in the provincial court of aggravated robbery, usurpation of property, and physical aggression. Most of the adult population of Alegria town is bilingual in Spanish and Quechua, though a growing proportion of young people understand Quechua but cannot speak it. Approximately 80 percent of the population is Catholic, although only about a quarter regularly attend mass: most of the small chapels are generally in disrepair and used only for annual festivals, of which the most important are Santiago (July 25), the Virgin of Asuncion (August 15), All Saints (November 1), and Christmas (December 25). There are also three protestant churches with small congregations.22 Other community activities include football clubs and three community radio stations that broadcast from 06.00 a.m. to 09.00 p.m. each day. A small openly gay group of at least five men live in the town. Despite much disapproval one member of this group (a resident for only two years who runs a small restaurant) was appointed majordomo of the town’s anniversary celebration. There are two nursery schools, three primary schools, one secondary school, and an occupational education center. Secondary school children from the annexes live in the town during weekdays, and other children in the district go to Huancayo for their secondary education (total school enrolment in the district fell from 1,914 to 1,407 between 1993 and 2004). The road improvement makes it easier for teachers and health professionals to commute to work from Huancayo city (one way fare S./8), and also enables skilled workers to commute in the opposite direction. The health services are insufficient in terms of infrastructure, equipment, and staff. There is one health center with a doctor, two midwives, a nurse, and three technicians. The main problems it deals with are respiratory complaints, digestive problems, infectious diseases, genito-urinary infections, injuries, and nutritional deficiencies. According to the Education Ministry the malnutrition rate in children fell from 65 percent in 1999 to 55.8 percent in 2005, but these figures are much contested.

2.4.5. Descanso Descanso is located in the Mantaro Valley to the North of Huancayo, at an altitude of 3,275 meters. It is 17 kilometers from Huancayo city and a short

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

51

distance from a main road to Lima. Intact pre-Inca stone grain stores and other archaeological evidence indicate that there has been a settlement there for more than 500 years. But the town only acquired district status in 1952. It also lies only a few kilometers from the main town of a neighboring district and there is a strong rivalry between the two, sustained by unsettled land disputes, feuding between youth, and accusations of “cross-border” house burglaries. The 2002 census indicates the district had a population of 5,900 (distributed between four urban barrios and three rural annexes), with approximately 10 percent of households having a dwelling in both the town and upland areas. Ninety percent of the population is Catholic and ninety-eight percent speaks only Spanish. The majority of the 1,037 houses in the town are built from adobe, but many also use brick and cement. Ninety percent has access to electricity and drinking water, though lack of sewerage remains a widespread issue. Drinking water is provided to the town by a water users’ association with 762 members. It has been running for 30 years and is regarded as the most effective organization in the district, with water quality checked quarterly by the Ministry of Health. The 2005 census estimate of 4,114 implies a dramatic 30 percent fall in population over just 3 years, and is consistent with reports of a continued high rate of net migration away from the district. Agriculture is the main source of income for 80 percent of households, with another 10 percent relying mainly on livestock income. Casual work is provided by five small brickmaking businesses and commercial limestone quarrying. One of the annexes also has a small handicraft cooperative. Only 530 out of nearly 15,000 hectares of land in the district are cultivated, and 10 percent of this is irrigated by canal from the Mantaro River. Much of the irrigated land is owned by people who have migrated away and sharecropped. At the other extreme are undulating upland pastures stretching up into the Huaytapallana mountain range. The slopes are widely forested and support a wide variety of wildlife including medicinal plants. They are also used for grazing domestic animals and for rain-fed agriculture that causes localized erosion. The communal association was registered in 1938. It has 170 members and controls 9,000 hectares of upland. Small cultivable plots are distributed annually for cultivation by lot, and members also have grazing rights and share in revenue generated from forestry (mostly eucalyptus). The association has had a long-running dispute with the largest private landowning family in the town, which it accuses of acquiring land illegally at the time of the formation of the district. Several attempts to resolve the dispute legally (and as part of the agrarian reform) have failed, and it is said to have prompted accusations of terrorism during the 1980s. While facing criticism for not doing anything to resolve this issue, the president of the community association also played an active part in protests against environmental damage by the two companies extracting limestone in the district. For a long time the companies sought to “compensate” the community with gifts of cement and irrigation pipes, whereas the community demanded a comprehensive environmental assessment, followed by a legal settlement and payment of royalties. Meanwhile, production has been interrupted. In 2003, the president of the association

52

Á LVA R E Z E T A L .

also entered into an agreement with a Spanish NGO to construct a school for “Andean leadership.” But work was delayed by arguments among members over the terms on which the land was leased to the project. Rivals also accused the president of making unauthorized sales of timber, and though not proven this resulted in him being suspended a year ahead of his scheduled term. The town has a health center, staffed by eight professionals all of whom commute by bus from outside the district. Common problems include chronic bronchitis, diabetes, diarrhea, fevers, and general infections. Approximately half of households use the health center as their first point of consultation, while the rest rely on a mixture of herbal medicines or buy their own drugs commercially. There is also an independent midwife in the town. She was said to provide a warmer, home-based service, but she has attended few births since the clinic stopped charging for antenatal and obstetric services. Awareness and use of contraception is quite high, but adolescent pregnancy remains very common. Another foreign NGO has funded campaigns to uphold the rights of children and to keep the town clean. But its main function is widely regarded as providing a secure salary for its local staff. The district has 3 preschools, 4 primary schools, and 1 secondary school, with a total of 1,208 students and 10 teachers (although only 1 lives in the district). Approximately 70 young people also commute to neighboring towns and to Huancayo for further study. Adult illiteracy is much lower than in Alegria (9 percent among men and 16 percent among women according to the 1993 census), and there is an active parent-teacher association. In 2003, the longstanding secondary school head stood unsuccessfully for election as district mayor. His supporters criticized the successful candidate (also a teacher) for having lived away from the district for most of his working life, returning with a party political affiliation and money to spend on the campaign in pursuit of his own political career. Conflict flared up later when the municipality unveiled a plaque on the school wall to commemorate improved concrete-block fencing and drainage. Angry parents removed it because there was no reference to their own contributions to the work, and the municipality reacted by calling the police to investigate what they regarded as an act of vandalism. The most conspicuous acts of the municipality after taking office were to renovate its own offices and give the main square a makeover, which included removing most of the trees. This attracted further criticism and deepened a rift between those whose livelihoods were more rooted in the district (centered particularly on the communal association) and those with past or present working experience elsewhere. This was particularly evident at the main Santiago festival. While bigger than ever, this was dominated by visiting migrants, whereas local residents excluded themselves from the planning of it for fear of being asked to make financial contributions that they could ill-afford. In 2004, the municipality hired a facilitator to organize three meetings to promote wider participation in its own planning (in line with national guidelines for decentralization). Fifty attended the first, forty the second, and twenty the third. Many complained that the room was too cold and that decisions had been taken already. A year later the meetings were better advertised and explained.

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

53

A district advisory board was also established, comprising representatives of the community association, market committee, neighborhood groups, and NGOs. Although many people were still critical of the mayor, they were more reconciled to him completing his term of office. Nevertheless, older residents complain that the district is less united than it used to be, pointing to the conflicts between schools and the municipality, and within the communal association in particular.

2.4.6. Progreso The barrio of Progreso is in the southernmost of the three districts making up Huancayo city and has a total population of 3,540. The three poorest of its five sectors climb up barren hillside overlooking the city and it was these that were selected for primary research. Each hillside sector has its own elected management committee, involved (to varying degrees and not without internal conflicts) in improving water supply, sanitation, electrification, and land titling. Inhabitants originate from Huancavelica and Ayacucho as well as from Junin, the majority being bilingual in Quechua and Spanish. Approximately three-quarters are Catholic, but to attend mass they must walk to a neighboring barrio. Important religious events include the Festival of the Cross on May 24/25 and the Virgin Carmen Festival on the July 15/16. The remainder of the population belong to Pentecostal churches, of which there are four, one of which holds services in Quechua. Relations between Catholics and Protestants are bad. Religious buildings have been vandalized and house burglary (mostly of small animals and electrical appliances) is often attributed to youth from the other faith group. Tensions also exist between Huancas and Huancavelicans, and in one sector these were conflated with a longstanding feud between two dominant families.23 After 1943 the area was allocated to a communal (peasant) association, which still owns an office, some forested parts of the hillside and a cemetery. However, most of its property was appropriated by members in 1966, particularly by the family of the then secretary, who in turn illegally “sold” it to refugees from the violence of the 1980s and early 1990s. Inhabitants of two of the three hillside barrios have still not obtained formal land titles for their house plots. Doing so has been complicated by irregularities in the sale of the land in 1966 as well as subsequently.24 In one sector, further conflict arose when allegations of corruption were made against the president of the sector association. Although the majority of residents regarded him as capable and honest, another committee member (who may himself have had ambitions to become president) accused the president of using the money collected to pay for lobbying on the land issue for his own meals and transport. Associates of this rival broke into the president’s house in broad daylight, stole furniture, kitchen utensils, and electrical appliances. The president did not report the robbery to the police, but is said to know who the culprits are and to have promised vengeance. Cement roads link the lower neighborhoods of Progreso to the rest of Huancayo, with several minibus companies competing for the routes. In contrast, the higher sectors can only be accessed on foot, along steep unmade paths between the houses. Less than a quarter of buildings are made of brick and

54

Á LVA R E Z E T A L .

cement, the remainder being adobe, and only 68 percent of households have access to drinking water, mostly from wells dug 8 meters deep. The water is chlorinated and available without charge. In one sector, there is a water committee which, with the help of the FONCODES installed pipes to bring water from tanks higher up the hill. However, this water is rationed and only sufficient for 80 of the 150 households in the sector. Only 80 percent of houses have electricity, and those without have organized themselves into electrification committees to lobby for new connections. There is no farmland in Progreso itself, but some individuals cultivate vegetables within their house plots. Many also breed guinea pigs and chickens, and a few have sheep, pigs, and even cows. There are many informal livelihood activities in the neighborhood: firewood distribution, deshelling garlic, trading firewood, and sewing festival costumes, are examples. Most men work in and around the major wholesale market of the city, which is within walking distance, and also serves as a center for recruitment of casual labor. The main occupations are street vendors (40 percent), market vendors (26 percent), construction workers (12 percent), agricultural laborers (10 percent), and cobblers (9 percent). There is also an association of rickshaw/cart (vehiculos minores) owners, which strictly controls the number of operators. The barrio does not have its own school and only one sector has a nursery, with 3 staff and 80 children. Most children walk to schools in nearby areas, but there is a very high dropout rate, with children leaving to find casual work in the markets. Only 52 percent of all adults in the three upper barrios have completed primary school. Domestic violence and mistreatment of children are common, and the medical center estimates that 70 percent of children less than the age of 6 are malnourished. There are many unmarried couples living together, and many young single mothers. Most married couples are older, or belong to one of the evangelical churches. The most feared problems are crime, alcoholism, domestic violence, and drug-addiction. Progreso is recognized by external agencies as an area of extreme poverty. The government health center was relocated and expanded in 2005. It has a doctor, nurse, obstetrician, dentist, and social worker. Popular canteens and “glass of milk” committees operate in each sector where they are controlled by dominant families. There are several NGOs present, whose goals include protecting children’s rights, promoting village banking groups, and providing food assistance for malnourished children. There is very limited access to credit, especially in the two sectors where people have no formal land title. No national or regional political parties have branches in the neighborhood, and it is visited by politicians only during presidential, regional, or district elections. The neighborhood representative is elected for a period of two years and has organized the building of the medical post and faenas to clean the streets and drains.

2.4.7. Nuevo Lugar Nuevo Lugar is located approximately 20 kilometers east of the center of Lima, a few kilometers north of the main highway to Huancayo. The lower part of the

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

55

settlement, closest to the highway, is at an altitude of 550 meters, but houses stretch up the hillside to a height of more than 900 meters. The climate is warm, sunny, and extremely dry, rainfall is minimal and the soils are largely barren. The area has been occupied since pre-Inca times, some Inca ruins remain, and after the Spanish conquest it was part of a large colonial estate. Most of the current population have moved there only in the last 25 years: the recorded population for 2000 (100,025) was more than twice that recorded in 1993 (44,526). Plans for a large settlement took place in 1984 and were formally approved in July 1985. The project was led by the municipality of Metropolitan Lima (controlled at the time by the United Left Party) in close collaboration with a network of Lima-based associations of migrants already resident in other parts of the city. The settlement was divided into 23 areas (each known by a letter of the alphabet), which were further divided into 239 neighborhoods called UCV (unidad communal de vivienda). Each recognized housing plot is 90 meters square, with 60 lots per UCV. However, these guidelines have subsequently been undermined by continued unplanned arrivals of migrants both into and above existing neighborhoods. Zones are also informally classified into three by altitude: A to F being the lowest (and richest); G to I in the middle, and J to Z the highest (and poorest). In general more recent migrants are to be found in the higher zones, many of them arriving from the interior of the country (especially from Junín, Cerro de Pasco, Huancavelica, Apurimac, and Ayacucho) during the worst periods of conflict in these areas. A census in 1985 indicated that just more than half of household heads were born outside Lima and approximately 20 percent of the population was bilingual. Many inhabitants of the lower zones complain that the higher zones are chaotic havens of “gangs, thieves, drug addicts and prostitutes.” In contrast, inhabitants of the higher zones often refer to those living lower down as “selfish, evil and land dealers.” Older residents also complain that the settlement has become less well organized over time. Nuevo Lugar is legally a centro poblado (population center) of the district of Atí-Vitarte, but it also has a self-governing council elected by residents.25 This council has been campaigning for district status since 1987, but the municipality regards Nuevo Lugar as lacking a sufficiently diverse economy and skill base. When interviewed in 2005, the general secretary of the self-governing body fiercely denied this: “there is nothing you can’t get done here” he comments. He also criticized the municipality for prioritizing improvement of the main plaza, rather than allocating sums to improve hospital and other facilities for the growing proportion of residents who are elderly. Interviewed on the same day, the municipal agent described the settlement as immature and prone to assistentialismo (dependency culture). He dismissed the self-governing council as debilitated by political infighting. Behind these brief comments lies a complex history of conflict within the settlement, which is reviewed in some detail in an appendix of the final report of the Truth and Reconciliation Commission. Problems arose in part from the electoral success of APR A in 1985 soon after the formation of Nuevo Lugar. Although it temporarily gained control of the self-governing council, the

56

Á LVA R E Z E T A L .

municipality quickly and drastically reduced technical and financial support to the new settlement. When the United Left took back control of the selfgoverning council it was politically weakened and compromised by the need to negotiate deals with APR A to secure resources. At this time Shining Path cadres were also establishing their presence in the settlement. Many local people defied it (with strong support from within the Catholic Church), and criticized it for being more interested in its own power than in development of the settlement. Nevertheless, Shining Path built up a strong presence: infiltrating secondary schools; playing a strong role in popular protests for more resources in 1988; using Nuevo Lugar as a base for blocking the central highway in 1989; and stage-managing the seizure of the potato harvest of a local landlord, during which one of his employees was shot. Shining Path also benefited from and accentuated (through intimidation and murder) the organizational weakening and loss of legitimacy of both APR A and the United Left. On coming to power in 1991, the Fujimori government established a military base in Nuevo Lugar, giving soldiers a relatively free hand to search, intimidate, and arrest. It also quickly co-opted popular kitchens through use (and abuse) of its powers of patronage over food disbursement. As elsewhere, popular rejection of its message, and the resilience of local leaders and self-defense groups were instrumental in Shining Path’s loss of influence, though it retained a presence in Nuevo Lugar long after the capture of Abimael Guzman in 1992. A strong legacy also persisted of distrust, clientelism, and eroded local political autonomy. For example, rivals for control of the self-governing assembly resorted to violence and assassination attempts in 2003, and meaningful decentralization of municipal control appears to remain a distant goal. Economically, at least, the settlement has partially recovered. Private bus and taxi services operate in all zones, although quality of roads and frequency of service declines with altitude. The lower part of the settlement has electricity, water, drainage, telephone, and Internet connections. The higher parts are mostly electrified, and many now have at least some access to drinking water if not sanitation.26 Government also provides some night police and street cleaning. Private firms supply electricity, telephone, cable TV, education, health clinics, credit, security, and transport. All the main political parties have branches in the settlement, and many NGOs are also active. However, 70 percent of the working population leaves Nuevo Lugar daily or weekly for the center of Lima, to work as domestic servants, in factories or in retailing. Within the settlement, most employment is in retailing and services. Nearly a quarter of all households have a female head of household. There is virtually no commercial agriculture, and only a little formally regulated manufacturing (shoemaking, carpentry, and mechanical goods in an industrial park set up in 1986). Sand, clay, rocks and limestone are extracted on a small-scale for cement making and coastal defenses, but the work is dangerous and poorly paid. Education is available through 33 state schools and 34 private schools, with 90 percent enrolment of school-aged children, and 5 percent illiteracy among adults, only 60 percent of whom completed primary schooling. The state higher education institute offers three courses: car maintenance, computing,

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

57

and electrical trades. In 2005 the government also operated two public hospitals and six other health centers, and there are also more than 30 private health facilities and even more numerous pharmacies. The main health problems found in Nuevo Lugar are respiratory diseases, diarrhea, dehydration, nutritional deficiency, circulation problems, cancer, HIV/AIDS, and tuberculosis. As many as a quarter of households run short of food each month, depending for survival on communal kitchens and food distribution programs. There are 250 communal kitchens and 253 “glass of milk” committees provide food to 75,000 children. As in Progreso, many young couples live together without legally marrying. Domestic violence is a major problem, and there are a large number of young single mothers, particularly in the high zones.

2.5. Some Initial Conclusions The main purpose of this chapter has been to provide a “thick” description of the setting of the research on which this book is based, rather than to develop a specific argument. Inevitably this description will also reflect the partial (though far from uniform) perspectives of the field team and authors. However, some points can usefully be made at this stage about the tension between local variation and uniformity in experience of wellbeing within the research area. Not surprisingly, the community profiles highlight the diversity of the seven research sites and hence the difficulty of generalizing about them, let alone about the three departments in which they are located. For example, we have noted major locational differences in access to physical resources, prevailing livelihoods, urbanization, commercialization, population size, political designation, dominant cultural-linguistic orientation, and recent history. In an attempt to capture such diversity the original site selection was influenced by the “corridor” hypothesis: that these variables could to some extent be mapped onto a single underlying variable with a strong spatial component. The evidence presented here warns against this. Three examples illustrate why. First, despite being physically located within walking distance of the center of a major city most people in Progreso are politically more marginalized than the inhabitants of the two peri-urban district centers. Second, the inhabitants of Selva Manta are physically and politically more remote than those in Alegria district, but labor allocation and livelihoods are more strongly commercialized. Third, although physically close to each other and sharing a similar livelihood pattern, Llajta Jock and Llajta Iskay have many striking differences. Community cohesion appears to be stronger in Llajta Jock even though access from outside is easier and more of its inhabitants were born elsewhere. At the same time as confounding the idea of a simple linear corridor, the chapter has also revealed important commonalities between the sites. First, the struggle to build and sustain livelihoods combining self-employment with paid employment takes place within interconnected (if heavily segmented) labor and product markets. Temporary movement as well as longer-term migration are critical to these connections and vary according to geographical location. These patterns are explored further in chapter 5. Second, there are strong similarities

58

Á LVA R E Z E T A L .

as well as important variations in the informal institutions regulating social behavior (festivities, collective action, clientelism) as well as in the provision of services by government, NGOs, and community groups. These are explored further in chapter 6. Third, while a polar distinction between Western and Andean orientation is dangerously simplistic, there are marked differences in the balance of social identities among people living in each site. These three sets of variables (livelihood, institutional mix, and social identity) cannot be mapped onto each other in a simple linear way, and directly influence wellbeing in their own distinct ways: livelihood being more concerned with the material; institutional mix with the relational; and social identity with the symbolic. Nevertheless it is also worth exploring further the complex links that do exist between them. At the beginning of the research, we posited (in line with Figueroa’s formalization of social exclusion theory) a likely hierarchy: material entitlements dominating welfare, embedded within sociopolitical relations, and in turn protected by cultural/symbolic norms (Altamirano et al., 2004). Without rejecting this entirely, there is clearly scope for more complex theorization about the relationship between each of them and wellbeing. Notes * This section is largely based on field notes by Altimirano and Álvarez following a visit to Huancavelica in 2005, translated into English by Michelle McCrory. 1. Although primarily interested in nationalism, Anderson argues that “all communities larger than primordial villages (and perhaps even these) are imagined” (p6). Although he did not address explicitly the link between wellbeing and imagined political communities, the two are tacitly linked in the way he explains the rise of nationalism as a response to the cultural weakening of religion and dynastic hierarchies, and as a mechanism for enabling people to transcend their own temporal and spatial insignificance as mortal individuals. 2. The regional government of Junin (2003:72) uses the term to refer to the main road network linking Huancayo to Satipo via La Oroya and Tarma. USAID also adopted the term for a $35 million commitment for the period from 2002 to 2007 to a project entitled “increased economic opportunities for the poor in selected economic corridors of Peru.” Huancayo and Huancavelica were included as separate corridors and a number of value-chains with potential to support the livelihoods of poor people were identified within them, including barley production in and around Alegria. Rather ironically (given its capacity to depress agricultural prices) much of the funding committed was in the form of food aid. 3. An attempt by the national government of Alejandro Toledo to establish macro planning regions based on the idea of linking coastal, highland, and jungle regions was roundly defeated by referendum in October 2005. In Junin, 75 percent of voters said no to being linked with Ancash, Huanuco, Pasco, and the provinces of Lima. In 2005, 84 percent of voters said no to being linked with Ayacucho and Ica in Huncavelica, (Escuela Para el Desarrollo, 2007). 4. Previously it had been divided into three: Hatuna Xauxa, Lurin Huanca, and Hanan Huanca, whereas in cultural terms there had been two distinct areas (Xauxa and Huanca) since the twelfth century.

R E S O U R C E S , C O N F L I C T, I D E N T I T Y I N C O N T E X T

59

5. Such studies include Richard Adams in Muquiyauyo (1959), Gavin Smith in Huasicancha (1989), Gabriel Escobar in Sicaya (1973), and Florencia Mallon in Acolla (1983). 6. Sheahan (1999:48) reports an annual decline in value added in agriculture in the 1970s for Peru of 0.6 percent, compared to an average annual GDP growth rate of 3.7 percent. Its contribution to GDP fell from 16.6 percent to 10.7 percent and food output per person fell by more than 20 percent. 7. UNCP was the center of major infighting between MRTA, Shining Path, and paramilitary during 1989. The police and army entered the main campus in June 1990, detaining more than 100 students, and in 1991 the army took control of the whole university. The military presence on campus came to an end only in July 1998. 8. These include financial and technical support for community infrastructure projects through FONCODES, provision of food to popular kitchens and for children’s meals by PRONAA, and provision of milk powder and other food to “glass of milk” committees via municipalities. 9. Many emigrants retain their cultural affinity by returning to participate in fiestas. These originate in Kuruka assertion of power in colonial times, combined with Catholicism, the agricultural calendar and celebration of “the cycle of life.” 10. The manager of the President hotel, for example, remarked that “before our best clients were miners and business men, now its NGO workers and state civil servants.” 11. Antunez de Mayolo is one of the largest power plants in South America, located on the Mantaro River at Quichuas, Colcabamba. However, it does not generate any benefits for the region, apart from a disputed monthly levy of 8 million Soles for the province of Colcabamaba. Water to irrigate valleys along the Inca coast is also taken from the high lagoons of the department. 12. For a discussion of the idea of cultural styles see Ferguson, 1999. 13. The last column indicates that more of the respondents in Llajta Iskay were also discontent or very discontent with the social category they gave themselves. 14. Overall, the interviewers classified the respondents into three categories: 92.3 percent mestizo, 5 percent blanco, and 2.8 percent indio. In addition to raising questions about the social identity and categorization of the interviewers themselves it also serves as a reminder that respondents’ replies may also have been influenced by this. 15. For more detailed descriptions in Spanish see the WeD Web site http://www. welldev.org.uk/research/methods-toobox/cp-countries/cp-peru.htm 16. See chapter 6 and Mayer (2002) for a fuller description and analysis of these institutions. 17. The communal association also has an elected secretary, treasurer, vocales (to notify people of meetings), and a datarista (registrar of births etc.). Posts are generally elected by a show of hands and there is much rotation of individuals between them—anyone completing a turn at all receives a special certificate. 18. Cattle rustlers and bandits (abigeos) have also been a problem. Some villagers were arrested by the police (who came from Pampas) and imprisoned for three years after they took the law into their own hands by killing one such thief. Some people in the community had suggested they tell the judges that everybody killed him, but they were unable to maintain a consistent story.

60

Á LVA R E Z E T A L .

19. This is part of a national nutrition scheme that operated in all seven research sites; see Copestake (2006) for a detailed discussion of the program drawing upon data collected in each site. 20. Santiago celebrates St James the Apostle but incorporates many pre-Colombian elements, particularly associated with care of animals. Jalapata involves the ritual honoring and slaughter of a duck whose last will and testament then becomes a vehicle for commenting on people and events in the village. It is celebrated in many other villages as well as in Huancayo and Lima. 21. Figure from the National Statistical Office suggest it then declined again to 5,072 in 2005. 22. Two of these have been embroiled in disputes. In one case, the preacher was found to have two wives. In the other, the owner of the land on which the chapel was built died and his benefactor refused to accept their tenure. The leader of the congregation began to raise money to build a new one then ran away with it. 23. This flared up when the youngest daughter of a leading family became unexpectedly pregnant. Her father had been president of the sector association three times, and her mother was president of the sector “glass of milk” committee and of the children’s canteen. The head of the family of the suspected father had recently taken over as president, prompting an attempt to split the sector into two. Fortified by alcohol, members of his family visited the house of the first and became violent. The former president, his wife, and one son were hospitalized. They took legal action, which is yet to be resolved. 24. When interviewed he said “these people come from other places, they are foreign people wanting to invade my lands. I could not permit that and had to sell it, practically giving it away. But they are not grateful; they throw away their rubbish and spoil my crops, almost like they do it on purpose.” 25. The sub municipal office also covers a smaller neighboring settlement. However, inhabitants of this settlement violently contested the planned settlement of Nuevo Lugar in 1985, laying weak foundations for subsequent collaboration between them. 26. Water to half the population (all in the lower zones) is supplied through a treatment plant from the river Rímac, whereas higher areas rely on water delivered by truck by the municipality and private water sellers.

Chapter 3

Subjective Wellbeing: An Alternative Approach Jorge Yamamoto, Ana Rosa Feijoo, and Alejandro Lazarte1

This chapter presents an original Peruvian subjective wellbeing (SWB)

approach. The problem of imposition of theories, methods, and development practices from self-appointed “developed” countries to appointed “developing” countries is discussed and the alternative rationale and methodology for SWB investigation is presented and explained. Findings based on this approach are reported, providing empirical evidence for its validity and usefulness. These support an alternative theory of SWB that differs from the established Western theories and offers potential solutions to critical problems on development practice in developing countries.

3.1. An Emic and Post-hoc Methodology and Rationale It is well known that SWB is strongly influenced by culture, particularly by values. It is also widely recognized that culture and values differ around the world. However, SWB theories and international development practice do not start from the SWB conception of each culture; they are conducted with the assumption that Western SWB conception is best and should replace other conceptions in a process usually called “development.” This perspective ignores culturally distinct SWB conceptions—some with less interest in wealth for example; there are just underdeveloped and developed cultures. Ironically, in some studies “developed” countries report lower levels of happiness and “underdeveloped” countries appear to be happiest, particularly Latin America and the Caribbean (e.g. Marks et al., 2006). If this is correct two important consequences arise: to avoid unhappiness associated with developed country approaches to SWB, and to investigate happiness associated with alternative SWB models, particularly in developing countries.

62

YA M A M O T O E T A L .

In order to address these consequences, some degree of distance from established theories and methods is required, particularly a distance from the practice that starts with strong theoretical assumptions followed by simplistic self-confirmatory methodology, and an interpretation of the results based on the original theory: a quasi-circular rationale. Here we propose an emic and post-hoc approach, an inductive process that captures the SWB categories and contents of the population under research (emic), and then a theorization based on patterns identified through iterative exploratory to confirmatory analyses using methods that control for researcher bias. The first step in the process is ethnographic research. As described in chapter 1, a researcher migrated to each of the research sites to be resident there for more than a year. Participant observation and unstructured interviews were first used to establish basic understanding and to build the strong relationships with the community, a prerequisite to collection of reliable data that captures deep thinking and feelings (see chapter 2). In-depth, open-ended, semi-structured interviews then provided the emic material for identifying SWB components. This interview (the Entrevista de componentes de bienestar, ECB, or components of wellbeing interview) mimicked a casual conversation where people could express in their own terms the content of SWB. The results were processed through content analysis, reducing the casual conversation into categories that summarized the ideas of the people in their own context. At this stage, an emic list of categories for each wellbeing component was completed. A dichotomized database was created in order to start the post-hoc theorizing process with the aim to identify SWB patterns controlling the bias of the researchers through exploratory statistical analyses. Each category of each SWB component was a variable in the database: if a subject mentioned a category then a score of one was assigned in the database; if the subject did not mention that category then a zero was assigned. Through this process, a dichotomized database was obtained from the open-ended conversation, and exploratory statistical methods were conducted in order to identify SWB patterns in the qualitative data. These exploratory statistical results were also triangulated with the ethnographic data. In a second stage, the categories of each SWB component found in the previous research stage were converted into items for a closed question-based survey. The close-ended answers were scalars, so the respondents could express the degree of importance or agreement to each question. This survey was administered in the same sites, bringing quantitative data for confirmatory statistical analysis, adding to, and building on the qualitative data. Therefore, exploratory results provide hypotheses that were tested under confirmatory quantitative conditions. These results were contrasted again with qualitative data. The iterative process continued until a convergent and robust solution was achieved. All these processes were conducted at two levels. The first comprised analyses of each SWB component, and is reported in sections 3.4 to 3.8. The second took the form of integrative analysis of each SWB component within a coherent overall SWB model and is reported in section 3.9. Section 3.2 elaborates on the underlying conceptual framework, while section 3.3 provides further details of the methodology.

SUB JEC T IV E W ELLBEING

63

3.2. Conceptualizing Subjective Wellbeing Despite the post-hoc and emic approach, a starting definition of SWB is required. Different theories and particularly empirical approaches were revised, and led to the conclusion that there were at least five core components of SWB: needs, resources, need satisfaction, values, and personality. However, the detailed content of each of these components, and the relations between them were constructed through the emic and post-hoc approach. Needs are the universal conditions for human functioning and there is abundant evidence about their role in SWB (Brunstein, 1993; Brunstein et al., 1999; Sheldon, 2001; Sheldon and Elliot, 1999; Sheldon and Houser-Marko, 2001; Sheldon et al., 2002; Sheldon et al., 2004). Resources are means required to achieve those needs (neoclassical economics is virtually based in this assumption). The perception of need achievement or life satisfaction is an independent variable from needs and there is abundant evidence about its important role on SWB (e.g., Diener, 1984; Diener et al., 1999; Fujita and Diener, 2005). The role of values as a cultural dimension in SWB is widely stated as well (e.g., Diener et al., 2003; Oishi et al., 1999; Schimmack et al., 2002), as is the role of individual differences or personality (e.g., Diener et al., 2003; Schimmack et al., 2002; Weiss et al., 2002; Weiss et al., 2006). The current challenge is to integrate available evidence into a coherent model (Nesse, 2005). At the end of this chapter, we conclude that SWB can be defined as the process of satisfying universal needs considering personal, cultural, and contextual conditions. A goal is set in relation to individual (personality) and cultural (values) characteristics for need achievement, which is mediated by resources (material, subjective, social.). The achievement perception is also moderated by individual and cultural characteristics

3.3. Method 3.3.1. Participants Four hundred in-depth interviews were applied in the qualitative phase, 550 participants were selected for quantitative phase I and 330 for quantitative phase II. Phase I collected data on goals, resources, goal achievement, and values. Phase II collected data on personality and other related variables. Two phases were required in order to avoid respondent fatigue. Quantitative phase II had a smaller sample size due to budget limitations. The sampling was based on a corridor concept (see chapter 1). Two rural sites, two peri-urban, and two urban-marginal were selected. Five sites were located in the Mantaro Valley, starting at more isolated villages, ending in urban shantytowns. One urban-marginal site was selected in Lima. A quota sampling on those sites was applied. Neighborhood was the unit of sampling. Based on the size of each neighborhood, proportional size was calculated. Random selection from each neighborhood was conducted, however, due to seasonal variation in residence fully random assignment cannot be claimed. This procedure was

64

YA M A M O T O E T A L .

applied to both qualitative and quantitative phases. One additional rural site in the high rainforest was selected as strong migration to that place had been detected during the qualitative research phase.

3.3.2. Measures The components of wellbeing interview (ECB) is an in-depth, open-ended, structured interview protocol developed for this study in order to capture the emic categories of needs, resources, need satisfaction, and values (Yamamoto et al., 2004a–f). Application of the ECB required field researchers with specific characteristics. First, they should belong to the general cultural context so they could understand the symbols and the subtleness expressed through the interview and through observation. At the same time, they should not be from the same village or live close to it, as this would risk making it impossible for them to treat the research with sufficient analytical detachment. In addition, they should speak the native language of the site, Spanish and/ or Quechua. Each field researcher migrated to the site. The first month was dedicated to the ethnographic research and also served to contact people and build rapport. In-depth interviews were open-ended in order to capture the emic contents of each variable and the precise expression of them, which is important for subsequent emic psychometric scale development. The interviews were conducted in natural environments for the participants. One by one, participants were found and interviewed in their homes or in the fields while looking after their animals or during agricultural activities. In some cases, interviewers also participated in these activities. Field researchers also actively participated in the communal activities of the site, such as festivities, the school lessons, and faenas (see chapter 2). Each field researcher conducted the content analyses in order to reduce the open-ended answers into categories taking into account context, connotation, and meaning. Afterwards all the content analyses by site were merged into a single list, excluding site-specific categories. The latter were coded for further within-site analysis. A subjective Wellbeing Psychometric Battery (WQP: Yamamoto and Feijoo, 2005) was developed, integrating a goal scale, a resources scale, a goal achievement scale, and a values scale. In addition, an adaptation of the Goldberg personality scale to the Peruvian urban-marginal sample conducted by Calderón (2003) was used. This is a semantic differential scale widely used in evolutionary psychology research. In ideal terms, an emic native personality scale would have been developed, but budget constraints did not permit this. The WQP is an emic battery. The single merged list of categories of each SWB component obtained as the final result of the content analysis of the qualitative phase research was the list of indicators for measure construction. The questions related to each indicator were phrased using the terms collected in the interviews; however, in some cases, it was necessary to find new terms that could be understood transversally in all of the research sites.

SUB JEC T IV E W ELLBEING

65

3.3.3. Analytical Techniques Factor structure. Exploratory factor analysis was conducted in order to identify the underlying patterns behind the data. The list of items for each SWB component was reduced into the minimum number of factors that could explain coherently the variance in the sample. Each factor represents the natural combination of items in the sample. The output of this analysis comprised several alternatives from which to select a final preferred solution, with the final choice being based on triangulation with the qualitative data. In addition, the selected factor solution was tested with a more acid mathematical test: Confirmatory Factor Analysis (CFA). Using an identity matrix, fit indexes provide confirmatory evidence of the validity of the final solution. This procedure was followed for each SWB component. For the specialist reader, this chapter reproduces results of key statistical tests (shown in brackets and italics) but these can be ignored by others. It is important to comment that CFA procedure requires a theory as a starting point (Hair et al., 2004). However, as a core part of our epistemology, no theoretical starting point was used. In order to avoid this dilemma, CFA was supported not by a specific theory but by cross-reference to the qualitative emic evidence. In addition, we tested several models using different subsamples before arriving at the model presented here: a procedure that provides additional consistency to the CFA solution. After finishing each piece of analysis we looked again to available theories to corroborate findings: we do not deny the validity of previous theories, but sought not to impose them onto reality. Differences. MANOVA was used in order to identify differences in factor scores associated with differences in sociodemographic variables. For example, are there significant differences in the values of factors by site? In these analyses, factor scores were analyzed as multiple dependant variables, and sociodemographic cross-variables were used as fixed factors. Scalar cross-variables were analyzed as covariates. If differences were found, recoded scores were analyzed as fixed factors. If a variable showed significant differences, further analysis was conducted in order to identify in which of the specific factors the difference was located. In addition, post-hoc statistical analyses were conducted where pertinent, in order to identify homogeneous subsets underlying differences among fixed factor categories. Life Satisfaction Index. Subjective Life Satisfaction (SLS) index reports the contrast between need (expectancy) and need achievement (perception). Repeated MANOVA measures were used. Need and Need Achievement Perception were selected as time 1 and time 2 dependent variables. Where there was a significant difference, we interpreted it as a level of satisfaction or dissatisfaction, according to the direction of the difference. In order to analyze differences between SLS and demographic variables (education for example) the same procedure was used introducing fixed factors or covariates. The SWB definition stated in section 3.2 incorporated resources as mediators of life satisfaction. Resources are introduced only in the general SWB model due to the limitations of the MANOVA.

66

YA M A M O T O E T A L .

SWB Model. Structural equation modeling (SEM) was conducted in order to integrate coherently all the factors of all the SWB components. Like CFA, SEM requires a theoretical grounding, but we again rely instead on qualitative evidence. Amos 5.0 software was used for CFA and SEM; other statistical analyses were conducted using SPSS 14.0.

3.4. Goals and Needs Goals are the starting point for construction of SWB as defined in this chapter. People set goals that will ignite behavior. Goals are specific ways to satisfy needs, defining needs as the core requirements for complete human functioning and development. As satisfaction of needs is sensitive to contextual conditions, goals can be regarded as concrete, specific ways to achieve them, and the best way to measure them. From this perspective, latent goal structure could be interpreted as needs and can be identified through factor analysis as already described.

3.4.1. Goal and Need Latent Structure Goal structure is defined by three robust factors: Place to live better (PLB), Raise a family (R AF), and Improvement from a secure base (ISB). The first has three indicators: nice and clean neighborhood, quietness (without violence or delinquency), and salir adelante (to move ahead).2 The second has two indicators: partner/marriage and children. The third has five indicators: salaried job, household goods, children’s education, daily food and health, and superior education. This factorial structure has a good fit [chi2 (32, N = 500) = 40.765, p = .138, CFI = .990, RMSEA = .023] according to all the indices of fit used in the confirmatory factor analysis (see figure 3.1). All the factor loadings were significant. The Place to live better (PLB) factor or need indicates looking for a good place to live and is related to core issues in social sciences like migration and urbanism. Migration is a social pattern widespread in cultural and geographical terms and it is an important issue on the contemporary agenda (see chapter 5). Peru is not an exception: on the contrary, it is a country with strong internal and external migration patterns that elude simplistic explanations (Altamirano, 1985, 1988, 1992, 2000). Migration is not only a modern social issue, it is rooted in the ancient history of human kind; genetic anthropology suggests that modern humans began a massive migration in Africa 60,000 years ago that eventually populated the whole planet (Wells, 2003). In addition, PLB need is not only about finding a good place to live, but it also describes a continuous movement for improvement that is well documented by processes like the hedonic treadmill (Brickman and Campbell, 1971; Brickman et al., 1978). This first need seems to be related to Peruvian and ancestral history, bringing support for its validity. A clean and nice place to live is the first goal of this need (lambda = .79), identified by many with urban modernity (squares, paved roads, and streets) rather than the presence of unspoilt nature in line with Western bucolic or pastoral images. This is a possible explanation for a development practitioner’s

SUB JEC T IV E W ELLBEING

67

paradox. While the community lacks basic services, people prefer a new plaza or a paved road, even if traffic is low. For example, there are paved roads in some Amazonian villages where nobody even owns a car. International development practitioners might consider this goal absurd compared to health services, for example. However, health in a villagers’ mind can be a non-priority goal, managed adequately through traditional medicine and through cognitive mechanisms of coping with sickness and accidents. We are not stating who is right or who is wrong, rather emphasizing the ethical implications of imposing practitioner’s values and priorities on those of villagers without a deep understanding of their emic adaptive or nonadaptive function. Quietness, without delinquency and without violence is the second goal (lambda = .64). This resembles a minimum functioning condition for living, related to security. Move ahead (Salir adelante) goal (lambda = .48) introduces a forward-looking and development/ improvement element into this factor, whose precise content will vary from site to site. Further exploration of this variable and its variations is ongoing. Raise a family (R AF) refers to having a partner, within or outside marriage. It is not just about someone to live with, but also related to having children. In contrast to elsewhere, the number of partners choosing not to have children in this sample does not seem to be falling. Partner selection for successful reproduction is a core motivation of humankind (Buss, 2004; Shackelford and Buss, 1997c), as for other living species (Darwin, 2004). The fact that this goal is also rooted in deep and ancestral motivations could be viewed as cross-validation of our results.3 Andean culture includes social acceptance of living together before marriage (servinakuy). In syncretistic urban-Andean adaptations, an extension of this tradition helps to explain the high rate of people who live together, have children, and are not married. Hence in practical terms, the meaning of partnership and marriage are hard to disentangle in this sample. Having children is the second goal item loading onto this factor (lambda = .77). Partner attraction, selection, and retention, an underlying process behind this goal, is probably one of the most influential motivations in contemporary society. From different points of view, the importance of this goal is supported. ISB combines two independent motivational factors or states in psychology— security and improvement (for example, Alderfer, 1969; Maslow, 1943); also in economics see Schultz (1964). This analysis suggests that they are part of the same latent need. People’s motivation for improvement should not risk whatever base of secure functioning they already have. For others, their main motivation is to protect what they have and the improvement motivation remains latent. The specific content of what is improvement and what constitutes security is influenced by culture, or local understanding of the meaning of improvement. Thus an adequate car of 50 years ago is not so adequate anymore. Finishing secondary school in a village where the norm is to finish primary school could be a great improvement. For an individualistic society, taking advantage of opportunities for individual profit could be a means of improvement, while this would risk shame and social rejection in a more collectivistic Andean community. What remains the same is the dialectic tension and risk between the goals of security and improvement

.79

Nice and clean neighborhood

.64 Peacefulness

Place to live better

(No violence/ delinquence)

e1

Alfa .67

e2

.48

Move ahead

e3

Partner or marriage

e4

Children

e5

.32

.79

Raise a family

Alfa .75

.77

.78

.36

Salaried job

e6

House

e7

Improvement .50 from a secure base .50

Education for children

e8

.38

Food and health

e9

Get a degree

e10

Alfa .60

.55 .53

Figure 3.1 From goals to latent needs Note: Model fit: CMIN=40.765 DF=32; CMIN/DF=1.274; P=.138; CFI=.990; RMSEA=.023; RMR=.008; AGFI=.972; PGFI=.572; NFI=.956. Source: WeD Peru.

SUB JEC T IV E W ELLBEING

69

A salaried job goal (lambda = .55) is regarded as a source of security as well as a source of money. It is important to understand this point in relation to the fact that most respondents were living at the margin of the officially regulated economy. It means many do not have a secure source of income for food and shelter, but obtain these services directly through their own labor. However, such livelihoods are often very vulnerable to shocks: harvests depend on weather conditions, and it is not unusual for crops to be devastated by diseases or climatic conditions. Even when there is a good harvest, the price falls and income remains low. People need money for education, fuel, and many other things, but their capacity to earn it is very limited. In this context the salaried job goal can be better appreciated, even if the amount of the salary is very low. House (lambda = .53), and daily food and health (lambda = .50), also indicators of this factor are widely understood as basic necessities. Education of children (lambda = .50) and the quest for professional status (lambda = .38) are additional items, whose meaning is further clarified by the cross-analysis below. They also feature prominently in the ethnography on migration reported in chapter 5.

3.4.2. Cross-analysis This part reports on sociodemographic variables that show significant effect on need factors. Variables not shown here, including gender, have no significant differences. Age does produce significant differences on goals. The MANOVA result for comparing the three needs to differ significantly as a function of respondent’s age [Wilk’s Lambda = .861, F(15, 1572) = 5.447, p < .001]. The R AF Goal score is significantly different across age groups [F(5, 535) = 13.852, p < .001]. A Tukey HSD comparison shows that the younger the respondent the lower the score for R AF. The lowest average score corresponds to the youngest group (up to 35 years), while the highest score is obtained by the oldest group (more than 56 years) Sites also produce differences on all three need factors [Wilk´s Lambda F(18, 1531) = 28.845, p < .001. PLB: F(6, 543) = 50.868, p < .001. RAF: F(6, 543) = 23.388, p < .001. ISB: F(6, 543) = 37.403, p < .001]. However, the corridor concept is not supported. Using a Tukey HSD comparison, the PLB goal average score describes Llajta Iskay as significantly different from any other site. This pattern will be constant throughout the analysis. Llajta Iskay is a case of a rural community with strong goals of modernity. A second group of sites that do not differ in their PLB goal averages comprises Descanso, Alegria, and Selva Manta. These two peri-urban sites and the rural site closer to the peri-urban showed the second lowest score. Nuevo Lugar and Progreso, the two urbanmarginal sites, are the third group of sites with higher scores compared to the previous groups. Progreso and Llajta Jock are the sites with the highest score. These averages describe a typical cross-site analysis in this study: a mixture of urban, peri-urban, and urban-marginal that can be explained by the interaction of different influences not just by the economic geography of the corridor. Historical episodes that lead to specific adaptations may have a high impact.

70

YA M A M O T O E T A L .

Number of children also showed an effect on need scores [Wilk´s Lambda F(3, 525) = 20.763, p < .001], with R AF need [F(1, 527) = 57.440, p < .001] being higher for respondents with more children. Birth control programs assume that having fewer children is better for quality of life. Here is contrary evidence in terms of goal setting: those who have more children attach more importance to them. There could be different interpretations to this. First, in rural and periurban environments, food, shelter, and related needs are not a problem as so long as there is available land and houses can be built through community reciprocity (minka). Second, in these sites production is strongly related to family workforce, so more children are more of an advantage than a problem. Third, there is a higher risk of child mortality, so a higher number of children increases the probability of having family support in old age and also continuing the lineage, the latter being considered a fundamental motivation for living species by evolutionary theorists (Buss, 2004; Darwin, 2004). These three interpretations are speculative but subject to testing. Residence time also affects the need scores [Wilk´s Lambda F(9, 1324) = 5.192, p < .001]. The PLB need score tended to decrease with prolonged residence time [F(3, 546) = 8.462, p < .001], as did R AF need [F(3, 546) = 3.717, p < .011]. Religion also relates to significant differences in need scores [Wilk´s Lambda F(3, 528) = 3.704, p < .012]. Protestants show higher R AF need than Catholics [F(1, 530) = 9.700, df = 1, p < .002]. Educational level affects the R AF need score [Wilk´s Lambda F(24, 1559) = 3.924, p < .001]: the higher the educational attainment, the lower need [F(8, 539) = 8.727, p < .001]. Marital status also impacts the R AF need score [Wilk´s Lambda F(15, 1497) = 7.818, p < .001]: starting with the average score for singles it increases through living with a partner, being married, and being widowed. Divorced shows a score drop [F(5, 544) = 22.101, p < .001]. Finally, migration shows differences [F(9, 1317) = 10.226, p < .001] on PLB need [F(3, 543) = 28.637, p < .001] and on ISB need [F(3, 543) = 9.723, p < .001].

3.5. Need Satisfaction 3.5.1. General Satisfaction Need satisfaction is defined by the contrast between needs and their perceived achievement. Significant differences between these two will define satisfaction (perception of need achievement greater than need expectancy), dissatisfaction (need expectancy greater than need achievement perception), and equilibrium (no significant differences between need expectation and need achievement perception). Satisfaction is not presented here as an independent hypothetical construction whose structure was developed following the emic process: participants’ own categories of need satisfaction were not elicited separately in a qualitative study whose structure could be tested through exploratory SEM methods. Rather, the structure of need satisfaction is taken to be the same as the needs structure. Therefore “satisfaction” as used below can be defined more precisely as “need achievement perception,” and its structure mirrors the needs structure presented in the previous section.

SUB JEC T IV E W ELLBEING

71

The next step is how to contrast needs and satisfaction. This was conducted through correlated group t-tests comparing needs and satisfaction scores as the two dependent variables. Comparing all sites as a whole, this analysis showed a general dissatisfaction in the sample. PLB showed dissatisfaction [(M difference = –.367) F(1, 549) = 319.055, p < .001], RAF showed satisfaction [(M difference = .312), F(1, 549) = 107.196, p < .001], and ISB showed dissatisfaction [(M difference = -.696), F(1. 549) = 1688.70, p < .001]. However, specific cases of satisfaction are found, which are supported by the qualitative data, and they probably make the most interesting part of the need satisfaction analysis. In order to track these cases of satisfaction, the influence of different sociodemographic variables were analyzed through repeated measures ANOVA, introducing sociodemographic variables as fixed factors or covariates. Due to the density of the results not all the significant differences are discussed.

3.5.2. Place to Live Better (PLB) Satisfaction PLB by site reveals significant differences [F(6, 543) = 71.778, p < .001] around a general dissatisfaction. However, peri-urban sites do not show significant differences suggesting equilibrium between perception of need and of achievement. As the mean score is above the median in the scale, this equilibrium can be interpreted as a positive state. In contrast, urban sites reveal a nightmare level of dissatisfaction. Need is higher and satisfaction is lower, creating the biggest gap. Qualitative data provides support for the negative emotion in these sites toward the environment. Rural sites show a more complex pattern. Llajta Iskay was an outlier in needs, showing a lower level compared to other rural sites. Nevertheless, it appears as the least unsatisfied of them all (equilibrium at P.01) when needs are contrasted to satisfaction. Llajta Jock shows the highest level of achievement on PLB, consistent with qualitative analysis. However the expectations are higher, resulting in a dissatisfaction that was also supported by qualitative data. Only when goals and achievement were contrasted was qualitative data consistent with quantitative analysis. This also provides evidence of the importance of not analyzing needs without reference to goal achievement perception at the same time. PLB by Time of residence in the site has an effect on satisfaction [F(3, 546) = 27.313, p < .001]. Figure 3.2 shows this as an adaptive interaction: as time of residence increases, the goal expectation decreases but the achievement perception increases. Finally, Migration pattern also shows differences [F(3, 543) = 76.373, p < .001], whereas no differences emerged by age, sex, number of sons, religion, education, marital status, or role in local organizations.

3.5.3. Raise a Family (RAF) Satisfaction Age produced significant differences in satisfaction [F(5, 535) = 10.677, p < .001]. Figure 3.3 describes the average satisfaction along lifespan. At the younger years, R AF is not a problem: it is not a goal and it is not achieved. Later on,

72

YA M A M O T O E T A L .

Satisfaction

Need Perception of achievement

Estimated marginal means

2,6

2,4

2,2

2,0

1 to 5 years

6 to 15 years

16 to 30 years

more than 31 years

Period of residence

Figure 3.2 “Place to live better” importance and satisfaction by period of residence Source: WeD Peru.

the goal starts a stable climb. Achievement climbs abruptly, levels off slowly, and stabilizes in the 50s. This could be an ideal situation. Marital and family dissatisfaction is an issue in modern society (Calderón, 2003) due to the gap between modern society and ancestral evolution (Buss, 2000). The findings of the present study support this proposition because the less modernized respondents studied here show higher satisfaction on R AF—probably the most important need for wellbeing in this group. Cross-cultural studies with European or United States samples comparing this goal are required. Casual observation also suggests that PLB and ISB satisfaction would be higher in these countries compared to developing countries, but there is evidence of general dissatisfaction about raising a family there (e.g., Gottman and Levenson, 1992). This suggests a paradox about development: if R AF is the most influential need for happiness, this could explain why world happiness surveys (e.g., Marks et al., 2006) find that designated third world countries are happier than self-designated first world countries. Site also produces significant differences [F(6, 543) = 8.407, p < .001]. Figure 3.4 reveals a linear decrease from rural sites to urban-marginal. However, expectation shows a nonlinear function, with rural sites (excluding Llajta Iskay as an outlier) showing higher goals, that

73

SUB JEC T IV E W ELLBEING

2,8

Estimated marginal means

2,6

2,4

2,2

2,0 Satisfaction Need Perception of achievement

1,8

less than 25 years

26 to 35 years

36 to 45 years

46 to 55 years

56 to 65 years

more than 66 years

Age

Figure 3.3 “Raise a family” importance and satisfaction by age Source: WeD Peru.

in turn reduce the level of satisfaction. Llajta Jock is the only unsatisfied site with respect to this goal. By contrast, Nuevo Lugar, the more urban-marginal site, does not have a high conscious expectation about family; however the satisfaction is high. Probably, in poverty conditions raising a family could be contradictory in terms of the implicit costs and probabilities to afford it. Having children could be a non-desired issue. However, when it does occur, it seems to be a source of satisfaction. This could be related to the high rate of children and the early age of maternity and paternity. Some nonconscious processes can also affect these decision-making processes (Bernardi et al., 1992; Grant et al., 2002; Hobcraft and Kiernan, 2001; Kissman, 1998b). Further explorations and cross-validation in relation to the effect of the gap size between expectations and perceived achievement is required. Number of children also reveals significant differences [F(12, 516) = 16.297, p < .001] and partial contribution to the effect of satisfaction (part. eta2 = .51). As the number of children rises, so does satisfaction as well. This provides additional evidence relevant to the debate over family planning, discussed earlier in the chapter. Education is also significant [F(8, 539) = 2.583, p < .01]. The goal expectation decreases as education increases.

74

YA M A M O T O E T A L .

Satisfaction

2,8

Need Perception of achievement

Estimated marginal means

2,6

2,4

2,2

2,0

1,8

Llajta Iskay

Llajta Jock

Selva Manta

Alegria Descanso Progreso

Nuevo Lugar

Site

Figure 3.4 “Raise a family” importance and satisfaction by locality Source: WeD Peru.

Satisfaction is highest at lower educational levels and tends to decrease as education increases, stabilizing for groups with complete secondary education and higher. The group with incomplete higher education showed an important fall in achievement. Further exploration is required of this important and controversial topic: given the importance attached to education in development, what implications arise from a possible trade-off between ignorance and happiness? Marital status also shows differences [F(5, 544) = 72.708, p < .001] and a moderate effect on satisfaction (part. eta2 = .55). Dissatisfaction is found for respondents who are single and widowed. Equilibrium is found for those with an absent partner and divorced, and greatest satisfaction on living with a partner and being married. No differences by sex, time of residence, religion, or role in local organization were found.

3.5.4. ISB Satisfaction Site also produce differences in satisfaction [F(6, 543) = p < .001., (part. eta2 = .75)]. Figure 3.5 shows changes in need and need satisfaction across sites; however, the relative gap between need and need satisfaction remains unchanged. We can hypothesize that there is a positive change in objective indicators of ISB

75

SUB JEC T IV E W ELLBEING

3,0

Estimated marginal means

2,8

2,6

Satisfaction

2,4

Need Perception of achievement

2,2

2,0

1,8

Llajta Iskay

Llajta Jock

Selva Manta

Alegria Descanso Progreso

Nuevo Lugar

Site

Figure 3.5 “Improvement from a secure base” importance and satisfaction by locality Source: WeD Peru.

in more urban sites compared to more rural sites (salaried job, education, and health). This suggests an adaptive process where needs are kept low but not too low: enough to maintain some motivation but at the same time limit frustration. The satisfaction of partial needs is also of less consequence since it does not persist: given the huge costs of migration from rural to urban-marginal areas in hedonic terms it seems to be a questionable investment. This finding supports the hedonic treadmill hypothesis (Brickman and Campbell, 1971; Brickman et al., 1978) proposed in nonpoor urban sites. Further studies of the interaction between this process and learned helplessness, “change resistance” and relative privation could offer further insights. Education shows significant differences [F(8, 539) = 2.906, p < .004]. While the interaction between education and satisfaction is not significant in its effect on the dependant variable it does have a moderate influence on the pairwise effect (Part. eta2 = .75). This means the need shows a decrease as education rises along with achievement perception, with dissatisfaction remaining at all educational levels. Those who have completed higher education do show less dissatisfaction. Marital status shows differences [F(5, 544) = 3.114, p < .009] and the pairwise effect is considerable (Part. eta2 = .613): there is variation in achievement perception but dissatisfaction level tends

76

YA M A M O T O E T A L .

to remain stable. A similar pattern is found for Migration [F(3, 543) = 7.969, p < .001, Part. ␩2 = .638]. Lastly, no effects of age, sex, number of sons, time of residence, religion, and role in community organization were found.

3.6. Resources Perception Resources are defined as the perception of the availability of means required for goal achievement. In the preliminary qualitative phase already described, a comprehensive list of resources was obtained from respondents. To avoid duplication, the list was restricted to resources that were not also perceived as ends in themselves and therefore already included in the goal satisfaction scale. The quantitative phase used a resource scale based on this list.

3.6.1. Structure A unidimensional factor solution was supported by a confirmatory factor analysis [chi2(14, N=531) = 20.671, p = .110, CFI = .987, RMSEA = .030]. Seven indicators were included: to get loans, to rent/lease (land), saving, migration, inheritance, social contacts,4 and gestiones.5 Thus, an interrelated mixture of economic, social, familial, and migration indicators defines this unidimensional factor (see figure 3.6). For instance, migration assumes the existence of networks of family and friends. This has several implications. First, as was the case with the qualitative phase, the confirmatory factor analysis provides empirical evidence of the significance of nonmaterial resources. If this is correct, then any analysis that only includes material resources in a subjective wellbeing conceptualization is strongly limited. Objective wellbeing analysis must also consider the complexities and particularities of the subjective measurement of resources, since physical measurement of these resources is also generally based on the subjective perception of them. Second, this factor solution provides empirical support for refuting the assumption (underpinning programs based on a purely material conceptualization of development) that resources for development are fundamentally economic. While there is scope for more detailed analysis of perceived resources, the present findings should not be underemphasized: while money may be a relatively universal resource in some societies, the situation presented from these sites is very different, and it cannot be assumed that a more money-oriented societal model is associated with greater happiness.

3.6.2. Cross-analysis Sites significantly differ in their average resources [F(6, 543) = 48.640, p < .001]. In terms of gender [F(1, 545) = 8.698, p < .003], men reported higher resources than women. Individuals with more residence time in a community also showed higher resources [F(3, 87.622) = 4.271, p < .005]. Migration pattern produces differences [F(3, 543) = 14.036, p < .001]: individuals who migrated to different and very different places reported the lowest level of resources. Rural sites, excluding

77

SUB JEC T IV E W ELLBEING

.54

To get loans

e1

To rent/ lease (land)

e2

Good networks for seeking work

e4

Inheritance

e5

Contacts with organizations

e6

Saving

e7

Migration

e9

.57

.53

Resources

.53

.49

.56

.42

Figure 3.6 Factor structure of resources perception Note: Model fit: CMIN=20.671 DF=14; CMIN/DF=1.476; P=.110; CFI=.987; RMSEA=.030; RMR=.012; AGFI=.978; PGFI=.494; NFI=.961. Source: WeD Peru.

Llajta Iskay as an outlier, show higher levels of resources. This can be attributed to greater community organization, stronger familial networks, reciprocity, and mutual support. In addition, money could be scarce but as agricultural communities, lack of food is not an issue. This situation could be weakened in periurban sites and diluted in urban-marginal sites. No differences by age, number

78

YA M A M O T O E T A L .

of sons, religion, educational level, marital status, or role in local organization were found.

3.7. Values While there is no single definition of values, different authors agree that they include beliefs shared by a social group about the relative merit of different behavioral states or ends (Smith and Schwartz, 1996). Theoretical discussion of values is strongly related to morality and ethics, and this has important methodological implications. Personal morality and ethics conceptions are subjectively universal and generalized as to what is correct in any context: otherwise they would not fulfill their function as noncircumstantial behavioral guidelines. These personal conceptions can then filter through research design and interpretation, resulting in an implicit ethnocentric bias in theory and research: Said (1978) provides a classic discussion of this. Rigorous contemporary empirical research is grounded in testing alternative approaches, not rejecting some a priori. As social scientists rather than philosophers we are interested in the values behind social behavior more than what is behind politically correct discourse. And while this idea is not new, putting it into practice is difficult. The post-hoc and emic approach of this research provides a testable alternative. In relation to the social desirability problems, previous evidence shows that when we do not ask about a person’s own values but the values of people generally in the site, there is less bias to conform to socially desirable or politically correct responses. Following this methodology we obtained the following results.

3.7.1. Factor Structure Following an exploratory factor analysis that suggested two factors, a confirmatory factor analysis provided additional support for the bidimensional structure [chi2(4, N=522)=1.237, p = .872, CFI = 1.0, RMSEA